SECTION 10 Environmental medicine, occupational me

10.1 Environmental medicine, occupational medicine

10.1 Environmental medicine, occupational medicine, and poisoning— Introduction 1637

The environment, both the occupational and the wider environ- ment, can affect health in many ways, both adversely and benefi- cially. While to date many of these effects have been well understood, most specifically for the work environment, the importance of the wider environment as a cause of ill health has been less well studied on the basis that there is not much one can do about general expos- ures, such as the weather, or specific events, such as volcanic erup- tions, apart from deal with the direct effects. However, increasing awareness of the importance of the environment on health, such as exposure to air pollution (both indoor and outdoor) and the short, medium, and long-​term effects of climate change, has resulted in better understanding of interventions that can improve health, or at least abrogate deleterious effects. Many environmental exposures are complex, often involving a range of potential agents. This has proved a real problem when trying to identify specific components that cause ill health, say in polluted air. Knowledge of these components would help in the design of interventions aimed at specific causal factors, ra- ther than going for a blanket reduction in air pollution as a whole. This optimal approach has been helped by an improving ability to measure population exposure to particular environmental agents, leading to the establishment of much more refined popu- lation or individual dose response relationships that are crucial in understanding the efficacy of any interventions. In the future, im- proved technology will allow personal—​rather than population—​ exposure measurement of a wide range of exposures, allowing population-​based studies to determine exactly how important in- dividual exposures are in conditions where multiple causal agents may be implicated. Climate change is another area where these sorts of problems occur, but is compounded by the fact that estimates of the effects of climate change on health are based on forward projections from known existing causal links between environmental influences and health. These projections largely assume that these associations are linear, hence they are open to some uncertainty when considering quantification of the health burden, but the likely impacts are at least broadly agreed, even if the quantification is uncertain. These are summarized in Box 10.1.1 and cover a wide area, ranging from the impacts of temperature change itself through to changes in vector-​ based disease geography. Management of many of these situations are covered elsewhere in this textbook, if not in this section itself. However, it needs to be understood that the physician has an ad- visory and exemplary role with respect to climate change and health by personal actions (e.g. walking or using a bike rather than a car where possible), and by putting pressure on governments to sign up to and implement recommendations of International Climate Change agreements. 10.1 Environmental medicine, occupational medicine, and poisoning—Introduction Jon G. Ayres Box 10.1.1  How this section is divided This section is divided into three broad areas: 1 Environment and health—​incorporating such major issues as air pollution and climate change. 2 Work and health—​both the well-​recognized effects of work on health (e.g. back pain, stress, and so on) and the less well-​recognized, but equally important effects of health on work. 3 Poisoning—​there are many types of poison, natural and man-​made, to which we can become exposed, either intentionally or accidentally. Included are the adverse effects of medications and treatments. The subject matter discussed in this section of the textbook often overlaps with that in other sections, hence several key areas are flagged here but also covered elsewhere (e.g. occupational lung disease).

10.2 Occupational health 1638

10.2 Occupational health 1638

10.2.1 Occupational and environmental health 1638

10.2.1 Occupational and environmental health 1638

10.2 Occupational health CONTENTS 10.2.1 Occupational and environmental health  1638 Raymond Agius and Debasish Sen 10.2.2 Occupational safety  1652 Lawrence Waterman 10.2.3 Aviation medicine  1656 Michael Bagshaw 10.2.4 Diving medicine  1664 David M. Denison and Mark A. Glover 10.2.5 Noise  1671 David Koh and Tar-​Ching Aw 10.2.6 Vibration  1673 Tar-​Ching Aw 10.2.1  Occupational and environmental health Raymond Agius and Debasish Sen ESSENTIALS Occupational diseases are those for which work or, specific- ally, exposures in the workplace are necessary causes. The most prevalent occupational diseases in developed countries today are
musculoskeletal and psychological disorders (usually stress-​ related conditions), but generally occupationally related malig- nancies (e.g. mesothelioma related to asbestos exposure) have the most serious outcomes. The proportion of all cancers attrib- utable to occupational exposures is about 4%, with occupation- ally related cancers almost exclusively concentrated in manual workers aged 20 or over in sectors such as mining, agriculture, and industry. When presented with a patient whose illness might possibly have been caused or aggravated by work or by other environmental fac- tors, the physician can usefully adopt an approach similar to that used for determining causation in epidemiological studies, with key issues being the temporality, reversibility, exposure-​response, strength of association, and specificity of the illness with exposure to the factor in question, also biological plausibility, consistency with other reports, and evidence of similar exposures producing similar illness. The prevention of occupational disease depends upon recogni- tion of the condition as occupational, assessment of ‘exposure’ and hence determination of risk, education of stakeholders, control of the problem at source, audit of the risk management procedures, and perhaps health surveillance of those exposed using suitable techniques for the early detection of disease and a check on the ef- fectiveness of the control measures. Extant health and safety legis- lation is driven mainly by risk assessment, and those who generate the risks are responsible for undertaking an assessment, the detail of which must be commensurate with the complexity of the situation and the ultimate risk. Worker compensation, or the financial recompense for harm done to an individual by work or workplace, can be for an injury or a disease and might be difficult to secure even in the 21st century. It is important that clinicians are aware of statutory compensation schemes, and patients should be advised to claim for compensation if their disease and work exposure seem related. If the disease and/​or work exposure are not scheduled, there may still be a case worthy of pursuit under common law. Around the world, most people in work do not have access to an occupational health service or an occupational physician, and this is despite the International Labour Organization’s recommenda- tions in 1985 for its members ‘to develop progressively occupa- tional health services for all workers’ and that ‘The provision made should be adequate and appropriate to the specific needs of the undertaking’. Introduction ‘Disease’ results from a combination of genetic, behavioural, and ‘environmental’ factors, generally cumulating with the passage of time. However, proportionately, genetic factors explain very little of the burden of ill health. Therefore, it behoves us as physicians to consider, as a first alternative or as a ‘default’, environmental fac- tors, ranging from the ambient air we breathe to our occupation, but including diet, micro-​organisms, and other exogenous factors as the prime determinants of disease.

10.2.1  Occupational and environmental health 1639 As physicians we should, first and foremost, aim to protect indi- viduals and society from such ill health through preventive meas- ures. We need to be aware of the sources and nature of the physical, ergonomic, chemical, microbial, and psychological hazards, how people are exposed to them and the risk or likelihood of this hap- pening, how they bring about adverse effects, and what structural, organizational, or behavioural interventions we should advocate to protect health and prevent ill health. When presented with a patient whose illness might possibly have been caused or aggravated by work or by other environ- mental factors, the physician might need to broaden their usual approach to history taking. Taking short cuts here can result in missing the diagnosis altogether and in losing a vital opportunity of making the patient better and of improving the fate of other coworkers. This additional information might be critical in at- tributing the illness to work. Bradford Hill postulated guidelines for determining causation in epidemiological studies, which lend themselves to be adapted for clinical purposes in an analogous manner (see Table 10.2.1.1). Moreover, when taking an occupa- tional history, it is also important to determine exposure, and a job title. For example, ‘engineer’, is not enough; a description of what the job entails and what agents or energies are involved are a bare minimum. The full job description might need to go back several years—​pleural mesothelioma, for example, can occur half a cen- tury after the first exposure to asbestos. The workplace is where many people spend a significant propor- tion of their lives. Work is important for self-​esteem and well-​being (physical and psychological) as well as for the economic well-​being of the individual and society. Work is good for us as long as it is good work and, therefore, does not impose unacceptable risks to our health. Thus, in dealing with the individual, physicians, whatever their specialty or interest, need to explore occupational and other envir- onmental causes of disease, to protect and better manage the pa- tient concerned, and indirectly protect others from the same fate. Furthermore, the physician has a responsibility as part of managing the patient’s health to do their best to help the patient achieve and maintain gainful and fulfilling employment, whether this in- volves rehabilitation back to work, resolving presenteeism, or even addressing health-​related issues standing in the way of achieving a job in the first place. Definitions and scope: Occupational disease What then is an occupational disease? Put simply, it is a disease for which work or, specifically, exposures in the workplace are neces- sary causes. The answer often has far-​reaching consequences, both in terms of suitability of employment, compensation costs, the in- clusion of controls against exposure, as well as legal or policy consid- erations. For silicosis (caused by airborne quartz), or mesothelioma (caused by asbestos), the disease is almost always caused by work and the key aetiological agent, silica, found in the workplace. But what about bronchial cancer in a joiner exposed to asbestos? Is it an occupational disease even though it is impossible to assess the precise contribution of occupational versus non​occupational factors (such as tobacco smoking) in any one individual? Epidemiologic research on the excess of illness that is attributable to work can be used to directly represent the impact of work on health at a popu- lation level, as well as occupational exacerbations of symptoms and disability in pre-​existing conditions, such as asthma brought on by irritants in the workplace. The most prevalent occupational diseases in developed coun- tries today are musculoskeletal and psychological disorders (usually stress-​related conditions) but generally occupationally related ma- lignancies have the most serious outcomes. However, when it comes to understanding the mechanism by which occupational diseases occur, two basic concepts immediately arise: the fact that in the case of two workers doing an identical job one will get the disease but the other may not; and the fact that with some workplace exposures, for example asbestos, the latency of the disease, in this case meso- thelioma, may mean many years before the disease appears. These factors can make the diagnosis of occupational diseases difficult. On the other hand, failures of safety aspects of work are usually very obvious, especially if they lead to a workplace accident resulting in an injury or even death. Fig. 10.2.1.1 guides the clinician to an outline of the hazards to health arising from the workplace or other environments. Table 10.2.1.1  Analogies between determining causality in an epidemiologic context (the Bradford Hill criteria) in population studies, and when dealing with an individual patient in a clinical context Epidemiologic criteria for causality Analogous clinical questions Temporality When in relation to exposure do/​did the symptoms start? Reversibility Do the symptoms improve when no longer exposed, e.g. after a weekend off, or by the end of a holiday? Exposure-​response Are the symptoms especially worse when undertaking tasks or in areas with high exposures? Strength of association Do other workers/​patients suffer from similar symptoms associated with the same exposures? Specificity What other exposures/​causal factors could be responsible for the same symptoms? (Smoking perhaps?) Other data, or information processing: Consistency Are there other reports of the same symptoms associated with or caused by the same exposure? Analogy Even if there is no evidence to hand of identical exposures or circumstances resulting in the same symptoms, have similar agents/​ chemicals of similar structure been implicated in the same symptoms of, for example ... dermatitis, ... or asthma. Biological plausibility Do the symptoms ‘add up’ in terms of what is known about the mechanisms of disease?

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1640 When faced with a case of ill health, to exclude work or the envir- onment, which might include the patient’s workplace, in the differen- tial diagnosis, it is important to determine, through the occupational history, the nature, duration, and intensity of any likely exposure, as illustrated in Fig. 10.2.1.1. The following are useful guides:

  1. How was the task done? For example, was the adhesive applied with a brush while the patient leant over it?
  2. Was the airborne concentration of a dust or fume such that it could be seen (although harmful particles are usually invisible to the naked eye), or the noise so loud that normal conversation was difficult?
  3. Is local exhaust ventilation/​extraction provided for the machine or area where work is taking place?
  4. Are the risks to health connected with work such that the em- ployer has (or should have) supplied workers with certain items of personal protective equipment such as gloves or a mask? Different categories of harmful agents require different and spe- cific forms of personal protection (e.g. rubber-​based gloves are not appropriate when handling solvents). The relationship between hazard, risk, and the eventual possible onset of disease is a very important one, since it goes beyond ‘clinching a diagnosis’ and extends to the crucial aspects of preven- tion of ill health, as illustrated in Fig. 10.2.1.2. The size of the problem The International Labour Organization in 2013 estimated 2.34 million deaths each year from work-​related accidents and diseases, the vast majority (2.02 million) dying from a range of work-​related diseases. This equates to 5500 deaths each and every day caused by various types of work-​related diseases. On top of this, the International Labour Organization (ILO) also estimates 160  million cases of non​fatal work-​related diseases occurring annually worldwide. However, these figures might well be serious underestimates. In the United Kingdom, estimates for 2016/2017 indicate around 13 000 deaths each year are thought to be linked to past exposures at work; primarily to chemicals or dust. (Asbestos-related deaths, often with historic exposure, make up the greatest number dying from work-related cancers). The types and trends of reported, non​fatal diseases worldwide vary widely although some common features do exist: occupational lung diseases from exposure to workplace dusts, gases, vapours, and fumes; musculoskeletal disorders particularly low back pain; psy- chological disorders. In 2016/2017 an estimated 1.3 million people in the United Kingdom were suffering from an illness (longstanding as well as new) they believed was caused or made worse by their current work; 516 000 were new cases among those working in the previous 12 months, accounting for 25.7 million working days lost due to work-related ill health in 2016/2017. Around 79% of new work-​related conditions were either musculoskeletal disorders or stress, depression, or anxiety. Fig. 10.2.1.3 shows how the incident cases distribute by major category from a general practice perspec- tive. Most of the incident cases are musculoskeletal, but mental ill health accounts for most of the sickness absence burden. The ILO estimates that, worldwide, work-​related accidents and diseases result in an annual 4% loss in gross domestic product, or about US$2.8 trillion in direct and indirect costs. The cost of work-​ related diseases in the EU has been estimated to be at least €145 bil- lion per year, excluding compensation costs (for asbestos-​related ERGONOMIC Heavy lifting, handling of loads Abnormal postures PSYCHOLOGICAL Work overload or underload High demand—poor control Other stressors BIOLOGICAL Work with infected patients or laboratory work: blood–borne viruses, tuberculosis, leptospirosis CHEMICAL Breathing in and/or handling chemicals or other hazardous substances; ingesting chemicals: mineral dusts, 'heavy metals', organic solvents, monomers, hardeners, etc. PHYSICAL High noise exposure, or vibration Hot work Radiation (ionizing and nonionizing) Fig. 10.2.1.1  Types of hazards, which may lead to occupational disease or other environment-​related ill health.

10.2.1  Occupational and environmental health 1641 diseases, for example, of around €1.5 billion per year) and per- sonal litigation/​compensation costs. In Great Britain, costs of new cases of workplace illness are estimated at £8.6 billion (2012/​13 data), shared between cost to the individual worker (pain, grief, and suffering), the employer (lost productivity), and government (healthcare). History of occupational disease Some industries, such as mining, have always been considered haz- ardous. The ancient Egyptians recognized this by restricting such work to slaves and criminals. Hippocrates emphasized the rela- tionship between environment (air and water) and health, but the HAZARDS Potential to cause harm May arise from many sources PRIMARY PREVENTION Reduction through intervention SECONDARY PREVENTION Reduction in risk to others by intervention TERTIARY PREVENTION Achieving improvement in capacity through occupational rehabilitation PRECLINICAL/BIOLOGICAL HEALTH EFFECTS CLINICAL ILLNESS Intensity of exposure leading to dose RISK or likelihood of effect/disease USUAL MEDICAL INTERVENTION PLUS REHABILITATION Fig. 10.2.1.2  From hazard, to risk, to disease, and the principles of prevention. Musculoskeletal Respiratory Hearing loss Mental ill-health Other Skin 1% 5% 4% 5% <1% 2% 3% <1% 2% 3% 55% 2% 9% 32% 51% % days sickness absence certified 48% 43% 35% % of cases reported % of cases sickness absence certified N = 6492 cases Fig. 10.2.1.3  Distribution of incident cases of work-​related ill health reported by general practitioners to the THOR-​GP surveillance scheme (University of Manchester) and associated certified sickness absence,
2006–​2015, by diagnostic criteria. The THOR-GP Surveillance Scheme, COEH, The University of Manchester, Manchester.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1642 earliest occupational physicians served military forces, Galen; for example, being a physician to Roman gladiators. By the Middle Ages, the plight of the free miner had been recognized by Georgius Agricola (1494–​1555) and Paracelsus (1493–​1541). Agricola not only described the ‘galloping consumption’ of Carpathian silver miners but also proposed ways of reducing the dust in mines by im- proved ventilation. The first authoritative treatise on occupational disease was written by Ramazzini (1633–​1764). His book De Morbis Artificium describes many occupational diseases ranging from mercurialism in mirror workers to repetitive strain injury in clerical workers. The Industrial Revolution in the United Kingdom brought occupational diseases to the attention of Parliament, largely through the work of phil- anthropists like Robert Owen, Robert Peel, and Lord Shaftesbury. Early legislation to control the worst vicissitudes of factory labour was emasculated by Parliament but the process had begun. The First Act of 1802 (which introduced the concept of limiting the hours of work and providing rest breaks in the working day) was followed by others leading to the 1833 Act which saw the start of His/​Her Majesty’s Factory Inspectorate—​the first enforcing, regulatory au- thority in this field anywhere in the world. By the early 20th century, the toxic effects of arsenic, mercury, phosphorus, and lead were common and understood in the West. Notification of these diseases became a requirement under health and safety law, and compensation for ill health was also granted. But the world of work has changed since then, particularly in the developed world. Here we now have a move away from manufacturing industries (heavy engineering, coal mining, and so on) to the provision of services, retail, and leisure. Consequently, the heavy, often dirty, industries such as mining and shipbuilding are, in some countries, like the United Kingdom, few and far between. With the growth in the service sector, whole new ways of working have also been developed (flexible working, short-​term and ‘zero hours’ contracts), and with this we now have a new set of occupational dis- eases (more biopsychosocial problems). While the working condi- tions in the developed world have improved steadily since the early 20th century, working conditions for many in developing countries remain hazardous, demonstrating an important tenet of occupa- tional health practice:  that is, while occupational disease can be preventable, the continued—​often necessary—​use of hazardous ma- terials and processes means that many such diseases are not elimin- ated, but need to be controlled. Prevention of occupational disease and other work-​related ill health The prevention of occupational disease depends upon recognition of the condition as occupational, assessment of ‘exposure’ and hence de- termination of risk, education of stakeholders, control of the problem at source, audit of the risk management procedures, and perhaps health surveillance of those exposed using suitable techniques for the early detection of disease and a check on the effectiveness of the con- trol measures. These procedures are outlined next. The legal duty on those who generate occupational and environmental health risks to manage them will vary from country to country. Occupational and environmental hazardous agents likely to harm human health might attack the body in various ways: through inhalation; ingestion; absorption through the skin; or a direct effect on organs of sense, for example, the eyes and ears. Occupational health risk management comprises recognition, evaluation, control, monitoring, and review. Fig. 10.2.1.4 illustrates the overall manage- ment of occupational health risks, and although a physician in a spe- cialty other than occupational medicine will contend directly with only a small part of this algorithm, awareness of it is important for the better management of the index patient and for the prevention of ill health in others. Extant health and safety legislation is driven mainly by risk as- sessment, and those who generate the risks are responsible for undertaking an assessment, the detail of which must be commen- surate with the complexity of the situation and the ultimate risk. Before embarking on any evaluation, especially where an occupa- tional cause is suspected, care must be exercised to determine what the known toxicological/​health effects are, previous evidence from similar circumstances and/​or environments, and any epidemio- logical evidence if available, as well as what the legislation requires. This is the ‘recognition’ aspect, and highlights the necessity not only to be well informed but also competent to interpret the available in- formation and act appropriately Table 10.2.1.2 shows how, after recognizing the possibility of a hazard being present in the workplace, evaluating the likelihood of harm arising, through the assessment of risk(s), is undertaken in a systematic manner. Similar approaches can be used in protecting the community at large such as in establishing air quality standards. Clinically, the issue of ‘what’ to assess might be self-​evident; for example, blood lead in a case of suspected lead poisoning. However, from the standpoint of preventive risk assessment, a systematic approach must be pursued to determine the specific purpose and value of the monitoring. For example, is biological monitoring or biological-​effect monitoring required? The former is the detection of a chemical or its metabolite in a biological sample as a measure of ex- posure (measurement of lead bound to red blood cells is a good ex- ample). The latter is measurement of a change in some biochemical or physiological variable to indicate the effect of the contaminant on the body (such as measurement of red cell or serum cholinesterase activity in relation to organophosphorus pesticides). However, even these measurements do not necessarily indicate harm but require an understanding of toxicology and epidemiology, as well as the pres- ence of relevant clinical findings. Often it is not possible to measure uptake or a biological effect short of ‘harm’, and one can only detect early manifestations of ill health, such as asthmatic symptoms from exposure to colophony fume in soldering. ‘Gold standard’ tests are rare in occupational health practice. Choice of an appropriate technique for making these measurements should be based, where possible, on indices of sensitivity and speci- ficity, as well as on the economics and acceptability of the technique. Remedial action: If the clinician suspects that work might be having an adverse impact or poses a significant risk to a patient’s health then they should advise the patient, and, with consent, the general practitioner and even the employer accordingly. If the clin- ician does not feel confident enough to do this, then advice can be sought from occupational physician colleagues in the relevant in- dustry, or in the health service, or from regulatory bodies, which also have a professional advisory role (such as the Health and Safety Executive in the United Kingdom). Advice given to the patient should include information on precautions that might be needed

10.2.1  Occupational and environmental health 1643 on returning to work, and whether any further investigations are required. Information on occupational rehabilitation and fitness to work is available from other specialists and agencies including voluntary/​charitable bodies (such as the Royal National Institute of Blind People—​RNIB, in the United Kingdom). European health and safety (H&S) legislation is clear: the em- ployer must adapt the workplace to be a safe environment for the employees. The employees should not have to adapt themselves to the stresses and strains of their work. Antidiscriminatory legis- lation, such as the Equality Act 2010 in the United Kingdom, now requires preplacement medical assessment of workers and not pre-​employment ‘medicals’, only after a job offer is made. Fitness to work within existing H&S legislation remains vitally important to fit the job to the individual. Occupational cancer Occupational cancer is the leading cause of work-​related deaths worldwide. Very often, it is difficult to establish the association be- tween occupation and cancer due to the long latency and multifac- torial causation of the disease. However, some of the associations Q1. WHAT HAZARDS DO WE HAVE? Q2. HOW DO WE ASSESS THE RISKS? Q3. HOW CAN WE MANAGE THE RISKS? Q4. HOW CAN WE CHECK TO SEE IF OUR INTERVENTIONS HAVE WORKED? Occupational health referrals Prior knowledge of Hazards Systematic surveys OCC hygiene measurements Employee/Union concerns Hierarchy of CONTROLS Risk-based HEALTH surveillance etc TRAINING & retraining on OH risks MONITOR all in Q1. (Active and Passive) AUDIT against standards & Occupational Health Plan Measurements/ Biomonitoring results Fig. 10.2.1.4  The overall strategy for managing Occupational Health Risks. Table 10.2.1.2  Specific considerations in assessing health risks Assessing the risks Where exposure can occur? The whole workplace? Where specific hazards exist? Who is likely to be affected? Those working full or longer shifts? Those working at increased metabolic/​breathing rate? Those undertaking unusual/​unscheduled tasks, e.g. maintenance work? Those already unwell? Are environmental measurements needed? . . . ‘background’ and/​or ‘personal’? How can the hazard enter the body? How likely is the exposure to cause a specific harmful
effect, and how severe? Route of entry and hence likely damage? Epidemiology and clinical toxicology, especially exposure-​response relationships When and what to measure/​ evaluate? Before, during, after exposure? Pre-​, postshift; end of working week? (Cumulative effect; different metabolic
half-​lives of hazards?)

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1644 are clear; for example, asbestos exposure and mesothelioma. Other associations are much more complex and less easily proven, such as the relationship between shift work and breast cancer. There is a considerable body of evidence on what causes oc- cupational cancer. The International Agency for Research on Cancer (IARC) based in Lyon, France, is part of the World Health Organization and has an established programme for the systematic evaluation of scientific evidence on the carcinogenicity of specific hazards and exposures. This it publishes in its extensive monograph series as an authoritative source of information on human carcino- gens. To date, IARC has classified nearly two hundred hazards as being established or probably carcinogenic to humans. Epidemiological evidence has been used to determine estimates of the proportion of all cancers attributable to occupational expos- ures. In round numbers the proportion is to the order of 4%, with a range of 2–​8% for a developed country like the United States or the United Kingdom. Occupationally related cancers are almost ex- clusively concentrated in 20% or so of the population, comprising manual workers aged 20 or over, in sectors such as mining, agricul- ture, and industry. In this group, perhaps as many as one case of lung or bladder cancer in every five might be attributable to workplace exposure. A wide range of industrial processes have been causally associ- ated with human cancer. Attribution of carcinogenesis by industrial process is often the first evidential step, and many legally recognized causes for compensation (‘prescribed’ industrial diseases) still rely on a definition by process. However, this approach has the short- coming of not necessarily being specific as to what chemical ex- posure needs to be controlled. Moreover, with the passage of time, once changes are implemented the risk associated with the original process could, and hopefully should, subside. The physician seeking to determine and thence prevent new circumstances with a risk of causing cancer must be aware that the causal agents, or analogues of them, might be responsible for causing cancer in other processes which are new and had not been recognized as risky. Therefore, in Table 10.2.1.3, the focus is on the chemical or physical hazard with processes listed by way of illustration while bearing in mind that these or similar hazardous agents might nowadays be manifest else- where. The agents named are not necessarily word for word as in the IARC documentation or as in legal schedules, partly for concise- ness and partly to consider more recent epidemiologic evidence. The IARC evaluation of carcinogenicity in humans sometimes applies to the group of chemicals as a whole, and not necessarily all individual chemicals within the group. Musculoskeletal disorders Statistical data from the European Union, including the United Kingdom, as well as the United States, indicate that musculoskeletal disorders have the highest reported incidence (Fig. 10.2.1.3) out of the major categories of occupational or work-​related ill health—​in the United Kingdom there were an estimated 507 000 workers suf- fering from work-related musculoskeletal disorders (new or long- standing) in 2016/2017. However, musculoskeletal symptoms are very common in the general population, a large proportion of which are in employment so making a work connection is often difficult to confirm. In some situations, a combination of occupational, psychological, personal, social, and home factors can be involved. This makes the diagnosis of the more common disorders associated Table 10.2.1.3  Some chemical and physical hazards causally associated with human cancer for which exposure has been mostly occupational. Examples of the index processes are in square parentheses while suspected target organs are in round parentheses Hazard Human target organ Arsenic and arsenic compounds Skin, lung (liver, haematopoietic system, gastrointestinal tract, kidney) Asbestos (most commonly in demolition nowadays) Lung, pleura, peritoneum, gastrointestinal tract, larynx Benzene (manufacture, petroleum refining) Haemopoietic (lymphatic) system Beryllium Lung Chromium VI (i.e. hexavalent compounds) Lung (gastrointestinal tract) Coal tars/​pitches (coke production and coal gasification) Skin, lung (bladder, gastrointestinal tract, haemopoietic/​lymphatic system) Diesel exhaust particulates or other particles containing polycyclic aromatic hydrocarbons (vehicle exhaust, some foundry processes) Lung, bladder Formaldehyde (disinfection, preservation, manufacture of resins and binders) Nasopharynx Inorganic acid mists (strong) containing sulphuric acid (isopropyl alcohol manufacture) Larynx Ionizing radiation (radiography, nuclear, and other industries) Haemopoietic/​lymphatic system, bone, skin, and other organs depending on exposure and type of radiation Mineral oils, untreated or mildly treated Skin (respiratory tract, bladder, gastrointestinal tract) 2-​Naphthylamine, and analogous aromatic amines (dye manufacture, rubber industry) Bladder (liver) Certain nickel compounds (nickel refining) Nasal sinuses, lung (larynx) Radon and its decay products (uranium mining, but also domestic/​residential) Lung Silica (crystalline, e.g. quartz) Lung Vinyl chloride Liver, lung, brain, lymphatic and haematopoietic systems (gastrointestinal tract) Wood dusts, especially hardwoods (furniture manufacturing) Nasal sinuses

10.2.1  Occupational and environmental health 1645 with work, such as low back pain, non​specific upper limb pain, and work-​related stress, particularly difficult because of the reliance on the subjective reporting of symptoms and the evidence of both cul- tural and psychosocial influences, many of which are unrelated to work, on such symptoms being reported. The epidemic of so-​called ‘repetitive strain injury’ in Australia during the 1980s highlighted the complex interaction of illness beliefs and behaviour, as well as employment in determining the workers’ symptomatic complaints. Musculoskeletal disorders can occur in all types of industry with high rates in post and courier services, agriculture, specialized con- struction work, and aspects of healthcare. The occupational risk factors implicated include:  awkward posture; manual handling of loads; keyboard and other repetitive work; and psychosocial factors. Musculoskeletal disorders cases are very variable in their clinical pro- gress, though generally more predictable in this regard than mental disorders. Thus, some affected workers might be unable to return to work for a time, primarily because of pain, and/​or need adjustments to their work. However, in general, cases are not very serious and treatment might be limited to pain management and adjustments in the workplace to reduce the risk of recurrence. People tend to recover and return to work, although conditions are often episodic. In the United Kingdom and elsewhere in the developed world, low back pain is possibly the most common musculoskeletal condi- tion experienced by people at work, making up 38% of musculoskel- etal disorders by affected area in the UK in 2016/2017. Poor lifting and manual-​handling techniques and sitting for prolonged periods in the course of work activities (e.g. professional drivers) are con- tributory factors. Nurses, porters, and bricklayers are groups with a high prevalence of low back pain. The total cost of sickness absence, early retirement, and treatment for low back pain in many coun- tries is considerable. Historically this was exacerbated by the belief by patients and their medical attendants that rest was needed for recovery. However, there is now very good evidence that, with the exception of serious spinal diseases and nerve root problems, special investigations are unnecessary, and rapid mobilization and return to work should be advocated. Moreover, the rising incidence of back pain reported to be work related, over a period of decades, cannot be adequately explained by physical conditions at work or by work practices. Cultural and psychological factors probably have to be in- voked satisfactorily to account for the high frequency of back pain. In the upper limb, other musculoskeletal disorders such as carpal tunnel syndrome and De Quervain’s tenosynovitis might be related to work, as well as to recreational activities, in tasks which involve strenuous gripping. Work can aggravate as well as cause musculoskeletal disorders and the physician has a responsibility to rehabilitate people with musculoskeletal disorders back to work. Fortunately, the advent of new ‘biologics’ has made rehabilitation easier for patients with arth- ritic disease, such as rheumatoid arthritis. They need to be assessed for the effect that the condition might have on the performance of their work duties. The fluctuating nature of most forms of chronic arthritis makes precise predictions difficult. Physical disability can improve despite persistence of the disease. This is due to the bene- ficial effects of treatment, to the patient’s adaptation to the conse- quences of the disease, and to successful rehabilitation at work. Mental or psychological disorders related to work The reported incidence of work-​related stress and mental disorders in the United Kingdom, the wider EU, and also the United States over recent years has increased dramatically (both in absolute rates as well as in comparison with other work-​related ill health), and in certain industry sectors such as health and social care, education, and finance, it is now the single biggest cause of sickness absence from work. Several factors in the working environment have been Sources of stress Intrinsic to job (DEMAND, CONTROL) Org structure/climate (CHANGE management) Role in org (CONTROL) Personal PERSONALITY PSYCHOLOGICAL HEALTH PSYCHOLOGICAL Physical & Mental ill health BEHAVIOURAL PHYSICAL HEALTH PHYSICAL Relationships at work (SUPPORT) Career development Individual characteristics Symptoms Disease Fig. 10.2.1.5  From stressors to disease. After Cartwright S and Cooper CL (1997). Managing Workplace Stress. Sage Publications Inc.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1646 identified as potential psychosocial hazards. Fig. 10.2.1.5 illustrates the broad categories of stressors and the key stages in the patient’s journey through symptoms of lack of well-​being to psychological or physical ‘disease’, such as depression or coronary ischaemia. There are various models to account for the ‘stress’ that people perceive when they are unable to cope. Karasek’s model is based on ‘demand’ versus ‘control’ wherein jobs with high demand and low control are the most stressful. In Siegrist’s model, the stress pro- voking imbalance adversely affecting employee well-​being is the one between ‘effort’ and ‘reward’. Prolonged exposure to one or more of the ‘stressors’ categorized in Table 10.2.1.4 can result in a range of symptoms of psychological dis- tress such as feelings of anxiety, irritability, or aggressive behaviour, lack of concentration, lack of confidence, and an inability to make decisions, sleep disturbance, and fatigue. There might also be asso- ciated physical symptoms, such as frequent headaches and nausea. Occupational stress is often identified as a result of the individual’s inappropriate (maladaptive) coping strategies, such as frequent short-​term absences, alcohol and other substance abuse, and poor time-​keeping, or by uncharacteristically poor work performance. Effective management of occupational stress usually requires an integrated approach which includes attention to both the workplace and the individual and should include intervention at the three fol- lowing levels: • Primary intervention focuses on the identification of particular sources of stress in the working environment and the institution of measures such as policies and procedures to eliminate or reduce these. These should not be viewed solely as ill health prevention, but as generally good management practices. • Secondary intervention focuses on improving the coping skills of employees, including managers, by the use of specific forms of stress management training (e.g. relaxation/mindfulness, con- flict management, assertiveness, time management) and health promotional activities. These are particularly appropriate where workplace stressors are intrinsic to the particular occupation, and therefore not removable (e.g. the potential for aggressive confron- tation with members of the public). • Tertiary intervention is concerned with rehabilitation of psycho- logically distressed individuals. When anxiety or depression is manifest, pharmacological intervention and behaviour therapy may be needed. Counselling might help, although its evidence base is weak. The source of stress can often be multifactorial, and not, therefore, solely work related, but it has an impact upon work performance and might be exacerbated by the demands of work. Infections Patients, notably those working in healthcare, are at an increased risk of occupational infections. Blood-​borne infections such as hepatitis B and HIV pose practical problems for the safety of staff during contact with infected patients, and the safety of patients during contact with infected staff, especially during so-​called ‘exposure-​prone procedures’. Numerous cases of HIV infection have occurred in healthcare workers following contact with in- fected blood or body fluids from patients. These have involved needlestick injuries, mainly from contaminated hollow-​bore nee- dles, or substantial blood contamination of damaged skin. Cases of HIV transmission from healthcare workers to patients are extremely rare, but hepatitis B is far more easily transmissible, explaining the higher incidence of outbreaks from doctor to patient and vice versa. Healthcare staff should be protected from hepatitis B by vaccination, but this does not cover other hepatitis viruses. Other occupational infections affecting staff in microbiological labora- tories and other healthcare workers include tuberculosis, salmon- ellosis, brucellosis, syphilis, and malaria. Working in tropical environments also exposes workers to the risk of tropical diseases. Occupational and environmental infections involve a range of organisms from viruses, rickettsiae, bacteria, and fungi to larger organisms such as parasites and insects. Occupations involved, together with the diseases involved, are forestry and gamekeeping (Lyme disease), sewage work (leptospirosis and some viruses), and farming (bovine tuberculosis, Q fever, and brucel- losis). Although rare in the United Kingdom, anthrax still occurs Table 10.2.1.4  Occupational factors that can act as stressors (psychosocial hazards) Work overload Quantitative: too much to do in the given time Qualitative: demands beyond the skills or organizational capacity of the worker Work underload Quantitative: not enough work to do Qualitative: monotonous, boring tasks, or below the skills of the worker Timing and control Shift work Limitations on organizing one’s own work Responsibility Role ambiguity, conflict Unclear responsibility and accountability Organizational culture, and relationships Lack of communication, participation Bullying and other harassment Financial and future prospects Inadequate reward or other recognition of ‘worth’ Job insecurity Poor training, personal development, or other prospects for advancement Hazards and comfort Physical, chemical, and biological hazards Other environmental discomfort or concern at work

10.2.1  Occupational and environmental health 1647 worldwide and can be responsible for fatalities typically following exposure to infected hides or bones, or indirectly during work such as construction on contaminated soil containing spores, where tan- ning or leather factories had once existed. Occupational dermatoses Since the skin offers such a large area to physical and chemical ex- posures in the workplace, it is not surprising that skin damage is a common occupational disease. An occupational dermatosis is a pathological condition of the skin for which occupational exposure can be shown to be a major contributory factor. The two most common skin conditions caused by/​made worse by work are irritant and allergic contact dermatitis (eczema), the ir- ritant form accounting for 95% of all dermatoses. In Great Britain it is estimated that there are tens of thousands of new cases of oc- cupational dermatitis seen by general practitioners (GP) each year, while some 2000 of the more serious cases are reported by derma- tologists. The diagnosis and management of skin conditions in gen- eral is covered elsewhere (Section 23) but some key facts are worth reiterating: •​ a very detailed chronological occupational and non​occupational exposure history is essential; •​ about one-​third of cases are thought to be allergic in origin; •​ in many cases, although it will be easy to decide whether the dermatitis (eczema) is allergic or irritant, this distinction between allergic and irritant can sometimes be difficult to make; •​ dermatitis that was allergic in origin may be aggravated and sus- tained by exposure to irritants at work and at home; •​ atopic subjects are at increased risk of irritant dermatitis. The best form of prevention of dermatoses is to eliminate the con- tact and/​or control exposure at source by applying the principles described earlier. In industrial situations, the hands and forearms are most at risk. The use of proper gloves (along with gauntlets or arm bands to prevent powders entering under the cuff), coupled with a high standard of hygiene, can minimize contact, and provide adequate protection. Where there is moving machinery, wearing gloves can pose a potential danger. Also when gloves are used, they might be taken off for tasks requiring manual dexterity, with the re- sult that contaminated hands are placed back inside the gloves. Many materials have been used to manufacture gloves, including cotton, leather, nylon, glass fibre, acrylonitrile, rubber, neoprene, butyl rubber, polyurethane, PVC, PVA, and tetrafluoroethylene. These confer specific protection for defined occupational exposures but the ingredients that go into their manufacture, for example ac- celerating agents such as thiurams, might also be a cause of derma- titis through skin contact. When cleaning skin, agents should be chosen that clean ad- equately in a short period of time without having too strong a degreasing effect. Detergents or solvents if used for cleaning will re- move fat from the skin, thus damaging its integrity and exposing it to further insults. Frequent handwashing, so-​called wet work, can have the same effect. No so-​called barrier cream actually provides a barrier to pene- tration of substances into the skin. In fact, in some situations they may actually enhance penetration and, occasionally, sensitization may occur to some of the constituents of the cream. There are many ‘after-​work’ creams, essentially moisturizers, which have the benefit of increasing the hydration of the skin following cleaning at the end of the day. They are of particular benefit in occupations where exces- sive drying of the skin may occur. Their use should be encouraged following ‘wet work’ and where hot air dryers are used, as these tend to dry the skin unduly (towels are preferred for hand drying). Besides occupational dermatitis, and skin cancers or other skin damage caused by actinic (UV) radiation, other dermatoses might arise from work. The more important are briefly listed in Table 10.2.1.5. Cardiovascular system Cardiovascular disease is a major cause of mortality and morbidity in industrialized countries (see Section 16). The association between personal risk factors and cardiovascular disease is well known, but now the evidence is also growing of occupational and environmental influences. There is good evidence from the classical studies of bus drivers and conductors that sedentary workers have a higher risk of is- chaemic heart disease than those who are more active. There is some evidence linking job stress and heart disease. The Whitehall II cohort studies suggest that (lack of) control over one’s job is an important factor in determining subsequent risk of myocardial Table 10.2.1.5  Benign occupational dermatoses, other than dermatitis Urticaria Type 1 hypersensitivity reaction and may be associated with rhinitis, asthma, and most importantly anaphylaxis. Itching, dermographism Glass and ceramic fibres, fibreglass, latex Psoriasis Identical to psoriasis from other causes. Can mimic dermatitis of the hands but without vesicles At sites of injury or friction in manual workers Infections Bacterial Viral Fungal Streptococcus to anthrax Poxvirus (Orf), human papilloma virus Tinea pedis, cattle ringworm, chronic candida (in wet work) Acne Typical acne features if slightly worse with multiple comedones, for example Oil, coal tar, chlorphenols, halogenated aromatics Vitiligo Areas of depigmentation often covering large areas of skin Para substituted phenols, and hydroquinones Scleroderma Skin changes may also include lesions Vinyl chloride, trichloroethene Pigmentation Altered pigmentation of skin Mercury and silver (argyria)

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1648 infarction. Overall, the risk is reported to be increasing by the ex- posure to various chemical substances, noise, stress, and working long hours and shift work. Exposure to chemicals in the workplace or general air pollution can contribute to cardiovascular ill health in several ways: provoking inflammation; damage to vascular endothelium; inducing dysrhyth- mias; renal effects and effects secondary to anoxia. Among physical agents, vibration is known to cause peripheral vascular disease and acute high exposure to noise is known to raise blood pressure. Workers on rotating shifts have an increased risk of ischaemic heart disease. Genitourinary system The kidneys play a crucial role in detoxification and excretion and, therefore, bear the brunt of many exposures to toxic chem- icals. Some toxic substances reach the kidney unchanged, but most are metabolized to some extent or other. Some, such as cadmium, become sequestered in the renal cortex while others, such as the carcinogenic aromatic amines or their metabolites, can be present in the bladder long enough and at a high enough concentration to induce malignant change in the transitional cell epithelium. Sudden, severe exposures to some chemicals can cause acute nephropathy. Such compounds may damage the kidney directly due to their intrinsic nephrotoxicity or may induce secondary damage due to prerenal effects, such as the haemolysis following arsine ex- posure, or oxalate crystals in the distal nephron after ethylene glycol poisoning. The metals most commonly implicated in renal disease are so-​ called heavy metals, such as mercury, cadmium, and lead (see Section 21). Mercury exposure resulting in acute tubular necrosis or the nephrotic syndrome is most unusual these days at least in the Western world. Similarly, modern industrial exposures to cad- mium rarely result in the proximal or distal tubular dysfunction or renal cortical damage that was more prevalent in the past. Lead nephropathy is also a rarity nowadays, but was not uncommon in the early part of this century. Lead is capable of causing damage to all parts of the nephron. Subtle tests of renal enzymes are needed to assess the effects of ‘heavy metal’ exposure on the kidney. Chlorinated aliphatic solvents such as carbon tetrachloride and chloroform can cause a hepato-​renal syndrome. The renal damage is largely an effect on the proximal tubular epithelium which can lead to tubular necrosis and acute oliguric renal failure. The weight of evidence from case–​control studies of workers exposed to solvents suggests an excess risk of chronic proliferative glomerulonephritis. The mechanism is unclear, but the demonstration of antiglomerular basement membrane anti- body suggests possible autoimmune damage to the glomerular basement membrane. Although the prostate possesses the curious ability to concen- trate (and excrete) heavy metals, inconsistent evidence exists of oc- cupationally related prostatic disease. Cancers of the urinary tract associated with occupational exposure to aromatic amines and polycyclic aromatic hydrocarbons were described earlier. Gastrointestinal tract Though ingestion is a means of entry for occupational and envir- onmental hazardous substances, and important when skin con- tamination is transferred to food or cigarettes, there are defence mechanisms that limit the damage to the gastrointestinal tract from such pollutants, and minimize their absorption. The mucous lining of the gut and diarrhoea and vomiting form part of these defence mechanisms. Acute gastroenteritis might follow the ingestion of chemicals such as soluble salts of heavy metals. The liver is frequently at risk from occupational and environ- mental exposures, as it is the target organ for detoxification and me- tabolism of absorbed compounds. A wide variety of infectious and chemical agents cause different types of hepatocellular injury, which can eventually lead to cirrhosis and liver failure, although there has been a decline in developed countries through improved working conditions, and a shrinking manufacturing industry. Haemopoietic system Lead poisoning is one of the oldest recognized occupational diseases and is still common, especially in the developing world. Exposures to inorganic lead are widespread, ranging from non​ferrous smelting to the burning off of old paint during refurbishment/​ renovation work in construction. In the domestic environment a particular hazard is the ingestion, usually by children, of lead com- pounds from old (now banned) paints. Lead causes anaemia mainly by inhibiting the enzymes involved in haem synthesis and also by haemolysis. The metal binds to erythrocytes and determination of blood lead levels is used in the monitoring of lead-​exposed workers. A diagnosis of lead poisoning is supported by symptoms of malaise, colic, and constipation, signs of anaemia, and peripheral motor neuropathy (rare, usually only in severe cases), and microscopic evidence in the erythrocytes of basophilic stippling (from abnormal haemoglobin), elevated blood lead, low haemoglobin, raised free erythrocyte protoporphyrin, and raised urinary δ-​aminolaevulinic acid. Indications of excessive lead absorption should lead to re- moval of the affected worker from further occupational exposure, with full investigation into, and control of, the circumstances of ex- posure to lead at work. Haemolysis and subsequent anaemia can follow occupational exposure to ionizing radiation in the industrial use of radioactive sources to test the integrity of welds, in the healthcare industry, and in nuclear power stations, as well as from chemical exposures (such as to arsine). Benzene is encountered in the petroleum industry, and is used as a starter chemical for the production of other aromatic organic compounds. Its serious effects on the haemopoietic system include aplastic anaemia, leukaemia, and probably multiple myeloma. Methaemoglobinaemia can result from exposure to occupational and environmental agents such as nitrates, and nitro and amino derivatives of aromatic compounds. Specific examples are aniline, aminobenzene, nitrobenzene, and nitrates in drinking water (from soil leachate). Babies are particularly susceptible to this. It is treated by the intravenous administration of methylene blue.

10.2.1  Occupational and environmental health 1649 Reproductive system Children can be affected by parental exposure to physical and chemical hazards. More recently, there has been much concern that various organic environmental contaminants ranging from pesticides to phthalates and other plasticizers, and compounds with oestrogenic properties might cause reproductive harm to the community. In men there are indications that exposure to these ‘endocrine disrupters’ has resulted in falling sperm counts, and an increased incidence of hypospadias and testicular cancer. Risk of adverse effects on human reproduction can include re- duced fertility, spontaneous abortion, low birth weight, and child ­development disorders. In the United Kingdom, for example, new and expectant parents make up a significant percentage of the work- force. It is estimated that there are some 350 000 pregnant, working women in any one year in the United Kingdom. Consistent evidence supports that work exposures to some chemical, biological, and physical hazards, and certain working conditions and some occu- pations have the potential to adversely affect reproductive health. Several potential risk factors have been identified:  exposure to chemicals (heavy metals, solvents, pesticides), radiation, local heat affecting sperm count and quality, having a heavy physical workload, and working irregular work schedules. Rubella can be a problem particularly for female healthcare workers and those working in microbiological laboratories, especially if adequate procedures are not in place for occupational health screening and immunization of this occupational group. Neurological disorders Damage to the central nervous system can be caused by a wide range of occupational or other environmental toxins, besides sub- stances of abuse such as ethanol and other solvents. There is very good evidence that exposure to lead (even at low level by occupa- tional standards), usually through drinking water but possibly also through inhalation, contributes to cognitive deficiency in children. Exposure to chemicals including lead, primarily encountered by workers in manufacturing, construction, and agricultural jobs, can cause transient and persistent effects on the central nervous system (Table 10.2.1.6). Transient central nervous system dysfunction is most commonly caused by exposure to volatile organic solvents, to organophosphate insecticides, or to carbon monoxide. Many low molecular weight fat-​soluble organic solvents, especially if chlor- inated, are chemically very similar to the halogen-​substituted an- aesthetic gases, and it is not surprising that they also have similar biological effects. In each instance, these substances, acting through different mechanisms, might cause central nervous system dys- function ranging from acute intoxication manifested by light-​ headedness and dizziness to loss of consciousness and even death. Persistent central nervous system sequelae can occur following one exposure episode if exposure levels are high and the duration of ex- posure is prolonged. Persistent central nervous system dysfunction, manifesting as neurobehavioural performance deficits, has been reported, par- ticularly in painters, following chronic exposure to moderate con- centrations of various agents encountered in the workplace, and occasionally in the environment. This syndrome, chronic toxic encephalopathy, consisting primarily of memory impairment, im- paired psychomotor function, and mood disorders, has been seen following chronic exposure to lead, styrene, carbon monoxide, and certain organic solvents. In more severe cases the deficits per- sist, but do not progress, following cessation of exposure. If be- havioural symptoms are present without evidence of abnormal neurobehavioural test performances (i.e. organic affective syn- drome) reversal of these manifestations usually occurs following cessation of exposure. Other rare central nervous system effects in- clude a Parkinsonian syndrome which might be a consequence of toxic exposures to manganese. Exposure to a range of toxic agents can cause damage to the per- ipheral nervous system causing either motor or sensorimotor poly- neuropathy (Table 10.2.1.7). Less commonly nowadays, exposure to lead at high levels for long periods can cause upper-​extremity motor neuropathy, consisting of wrist extension weakness or wrist drop. Other specific substances (e.g. acrylamide, n-​hexacarbon ali- phatic solvents, and certain organophosphorus compounds) can Table 10.2.1.6  Some central nervous system health effects caused by workplace or other environmental toxins Disorders Manifestations Causal agent Acute intoxication Light-​headedness Carbon monoxide Loss of consciousness Organic solvents Death (rare) Organophosphates Organic affective syndrome Fatigue Organic solvents Irritability Lead Depression Mercury Chronic toxic encephalopathy Impaired neurobehavioural function (symptoms as described) Organic solvents Lead (usually organometallic) Psychosis Marked emotional instability Carbon disulphide Manganese Mercury Toluene and related solvents Parkinsonian syndrome Tremor, rigidity, akinesia Manganese Carbon disulphide Carbon monoxide Visual disturbance Impaired acuity or peripheral field defect n-​Hexane Methanol Organic mercury Colour vision loss Organic solvents Cerebellar or other damage Ataxic gait Acrylamide Organochlorine insecticides Methyl mercury Seizures Lead (usually organic) Organic mercury Organochlorine insecticides Organotin compounds

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1650 act as axonal toxins causing a mixed sensorimotor polyneuropathy manifesting the asymmetrical, distal sensory loss. Occupational disease caused by vibration or noise Certain jobs involving the use of vibrating hand tools or pneumatic drills may be responsible for the occurrence of peripheral neur- opathy, and loss of motor function if progressive, as part of ‘hand–​ arm vibration syndrome’ (HAVS). These disorders arise from a combination of physical trauma to the nerve itself as well as damage to blood vessels. Some workers with HAVS suffer adverse effects on hand function and are no longer fit for work. Noise-​induced hearing loss is discussed in Chapter 10.3.4. Respiratory disorders The respiratory tract is, in many occupations, the most important route of exposure to hazardous substances. In many instances of oc- cupational diseases, it is the target organ. These conditions are dealt with in Section 18. Medically unexplained symptoms A growing number of occupational health complaints are charac- terized by the lack of a firm diagnosis or a clear occupational causal pathway. These include, in particular, complaints that consist of a range of non​specific symptoms, typically headache, fatigue, nausea, depressed mood, cognitive confusion, and sometimes eye and nasal irritations. They are reported in diverse situations such as in air-​ conditioned offices with poor control (accounting for many cases of ‘sick building syndrome’), proximity to low-​frequency electromagnetic fields, and perceived exposure to very low (often undetectable) levels of chemicals. Moreover, the prevalence of musculoskeletal complaints in many workplaces has been shown to be related to the presence of psychosocial hazards. Important elements are health beliefs and atti- tudes of the individual as well as the social and cultural environment. These influence their response to real or perceived exposure to haz- ards by determining their selection of which information to attend to and their subsequent interpretation of that information. Specific examples of syndromes with these features which occur in both an occupational and wider community setting may be ‘multiple chem- ical sensitivity’ and ‘chronic fatigue syndrome’. Current approaches to effective management of these and other similar conditions favour a ‘biopsychosocial’ approach, which rejects the artificial distinction be- tween a physically and a psychologically based complaint and treats both physical and psychological symptoms simultaneously. Compensation for occupational diseases Worker compensation, or the financial recompense for harm done to an individual by work or workplace, can be for an injury or a dis- ease and might be difficult to secure even in the 21st century. In the so-​called ‘first-​world’ nations (i.e. the industrialized society), a worker can receive compensation prescribed by the state, or through litigation in the civil courts. By the end of the 19th century, many nations in Europe and the United States had passed workman’s compensation laws of one sort or another. Such schemes were usually restricted to specified diseases or occupations. The United Kingdom, for example, has the Industrial Injuries Benefit Scheme by which, depending on the disease in ques- tion (from the list of Prescribed Disease) and the diagnosed degree of disability, compensation is paid by the state. There are over 70 Prescribed Diseases known to be a risk to health from certain jobs, and in 2011, figures from the Department of Work and pensions, United Kingdom, confirm 5920 cases of occupational diseases as being compensated, with pneumoconiosis, diffuse mesothelioma, and osteoarthritis as the three most common diseases. The principles underlying such statutory compensation schemes are that they should be ‘no fault’, that the disease should be, with reasonable certainty, caused by work, and that the benefit claimed should offset job loss, wage-​earning deficit, and disability, or pro- vide death benefit to the next of kin. Advice on proposed additions to the list of compensable diseases is made in the member states by government-​appointed advisory groups. In the United Kingdom, this group is the Industrial Injuries Advisory Council, which reports to the Secretary of State for the Department of Work and Pensions. It is important that clinicians are aware of such schemes. If the disease and work exposure seem related, and are scheduled, patients should be advised to claim for compensation. If the disease and/​or work exposure are not scheduled, there may still be a case worthy of pursuit under common law. Occupational health services The notion that employers should provide healthcare for workers is hardly new and the role of the physician in caring for, as well as advising on, diseases caused by work has been mentioned and goes back into the history of occupational medicine. Table 10.2.1.7  Peripheral nervous system syndromes caused by workplace or other environmental toxins Disorders Manifestation Causal agent Motor neuropathy Wrist weakness,
foot drop Lead Mixed sensorimotor neuropathy Symmetrical distal sensory loss, mild motor dysfunction Acrylamide Arsenic Carbon disulphide Carbon monoxide DDT n-​Hexane Methyl n-​butyl ketone Mercury Organophosphorus compounds (various agents including triorthocresyl phosphate) Thallium

10.2.1  Occupational and environmental health 1651 The first recognizable occupational health service in England began in the mid-​18th century when the London (Quaker) Lead Company recognized the adverse effect of lead mining on workers and provided health and welfare services in north-​west England. Since then, occupational health provision has expanded along dif- ferent lines in different countries. Around the world, most people in work do not have access to an occupational health service nor an occupational physician, and is this despite the International Labour Organization’s recommenda- tions, in 1985, under ILO Convention 161, for its members ‘to de- velop progressively occupational health services for all workers’ and that ‘The provision made should be adequate and appropriate to the specific needs of the undertaking’. In the United Kingdom, approximately one in 14 consultations with a GP by those of working age are for conditions due to work or that affects ability to work. GPs and secondary care specialists in hospitals where these patients are sometimes referred therefore need to think of work as a possible cause of ill health and ill health as a barrier to work. Both primary and secondary care physicians need to communicate freely with occupational physicians and occu- pational health services if these exist in their patients’ places of work in order to keep their patients happy, healthy and in work for as long as possible or to return them to work after an illness as quickly as the medical evidence justifies. In practice, delivery of good occupational and environmental health practice is a multidisciplinary approach. It might include physicians, nurses, or hygienists who monitor and control exposure to chemical, physical, and biological agents in the workplace; tox- icologists; ergonomists; and psychologists able to assess psycho- social aspects of work. These specialists will promote occupational health, but for long-​term success that is sustained, it is crucial that both managers and the workforce consider it an integral part of their working practices and philosophy. Unfortunately, occupational health services are not available universally and interpretation of the requirements varies greatly between countries and employment sectors. Initially, most services arose from a mixture of philanthropy and self-​interest; the theory being that the healthy worker was likely to be more productive, a concept that holds sway today. Present-​day services range from total healthcare including primary care and hospital medicine (as in some large multinationals operating in developing countries), to outsourced independent occupational health services. Services in the United States of America and much of Europe might in- clude general health promotion and education, but much inequity in healthcare exists even between enterprises within the same country. Recent increase in the provision of occupational health services has followed the enactment of effective health and safety legislation (e.g. the Health and Safety at Work etc. Act (1974) in the United Kingdom). Some countries, such as the Nordic countries, the Netherlands, and Australia, require the provision of occupational health services by law. Statutory provision of such services in the United Kingdom and the United States is limited to particular industrial sectors and specific occupational exposures, such as ionizing radiation, heavy metals, asbestos, and carcinogens. In general, primary legislation such as that produced in the United Kingdom then ‘enable’ a variety of government departments to create further secondary legislation in the form of ‘regulations’, requiring action from employers in par- ticular occupational and environmental circumstances. In the developing world occupational and environmental health often remains an almost unknown concept. National governments in the developed world have approached the question of preventing ill health at work in different ways, with the European Union’s dir- ectives increasingly dominating the scene in Europe. The EU dir- ectives increasingly drive the occupational health and safety agenda in member states, requiring them to adopt these principles and then modify or create national legislation in response. The EU phil- osophy has been to encourage the delivery of occupational health to all by ‘competent’ persons, although the mode of delivery, even the definition of competent, has been left to member states to inter- pret. Most important, the level of service provided should be based on a thorough risk assessment of the work processes in that organ- ization and a clear and logical procedure of risk management. To deliver even such a basic service requires multidisciplinary teams including trained physicians, hygienists, and nurses. Few com- panies have such services, and many are too small even to contem- plate such provision. Future development of occupational health services might de- pend primarily on the economic climate, but also on perceptions of what constitutes occupationally mediated disease, as well as political will. Perversely, an exponential rise in legal action, insurance costs, and compensation may also play a significant part in persuading employers, and even nations, that competent occupational health services are an absolute requirement of profitable organizations. If this is to encompass small and medium-​sized enterprises, provi- sion must come either from larger employers (such as the National Health Service) or from private providers. Conclusion Clinicians can make important contributions in two ways to ‘occu- pational health’ and hence to the health and economic well-​being of their patients and others. In the first instance they should have a high index of suspicion for occupational (and environmental) causes of disease in their patients. These can manifest to physicians in any specialty. Having reached a clinical diagnosis, especially in the case of work-​related ill health, they need to liaise with an occupa- tional physician or other occupational health professional, the rele- vant enforcing authority, or the employer directly so as to remove their patient from the harmful ‘exposure’ and to control exposure to prevent similar occurrence in the patient and in other workers. Secondly, whether or not the ill health is work related, the clinician has an important role in giving advice about return to work and achieving occupational rehabilitation. However, the delivery of ‘occupational health’ and of ‘environ- mental health’ is achieved through more than just a clinical ap- proach. Occupational health is concerned with managing the health of working people. While occupational physicians deal with the ef- fects of work on health, and the influence of health on work, other professional groups, including nurses, hygienists, toxicologists, psychologists, and safety engineers, also have important roles to play in keeping people healthy and at work. In common with other prob- lems in public health, the solutions mainly lie in a population based

10.2.2 Occupational safety 1652

10.2.2 Occupational safety 1652

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1652 and societal approach to achieve health promotion and protection, as well as ill health prevention. The strongest link to public health exist when work and the workplace (as well as the home, school, and other environments) are used as a setting to promote good environ- mental standards, personal health, and healthy behaviours so as to bring about the highest state of ‘well-​being’ for individuals, whether in the working population or not. FURTHER READING Adisesh A, et  al. (2013). Review:  UK standard of care for occupa- tional contact dermatitis and occupational contact urticaria. Br J Dermatol, 168, 1167–​75. Agius R, Seaton A (2006). Practical occupational medicine, 2nd edition. Hodder Arnold, London. Baroushi B (1993). Effects of the workplace on fertility and related reproductive outcomes. Environ Health Perspect, 101 (Suppl. 2), 81–​90. Baxter PJ, et al. (2010). Hunter’s diseases of occupations, 10th edition. CRC Press, Boca Raton, FL. Burton AK, Bastys S, Wright IA, Main OJ (2005). Obstacles to recovery from musculoskeletal disorders in industry, research report 323. Health and Safety Executive (UK), London. Cartwright S, Cooper CL (1997). Managing workplace stress. Sage Publishing, Thousand Oaks, CA. Dick FD (2006). Solvent neurotoxicity. Occup Environ Med, 63, 221–​6. Grandjean P, Landrigan PJ (2006). Review: developmental neurotox- icity of industrial chemicals. Lancet, 368, 2167–​78. Health and Safety Executive (UK) (2017/2018). Health and safety sta- tistics: annual report for Great Britain (2016/​17). http://​www.hse.gov. uk/​statistics International Labour Organization (ILO) (2013). The prevention of oc- cupational diseases. http://​www.ilo.org/​wcmsp5/​groups/​public/​-​-​-​ ed_​protect/​-​-​-​protrav/​-​-​-​safework/​documents/​publication/​wcms_​ 208226.pdf Langan-​Fox J, Cooper CL (eds) (2011). Handbook of stress in the occu- pations. Edward Elgar, Cheltenham. Palmer KT, Brown I, Hobson J (2013). Fitness for work: the medical aspects, 5th edition. Oxford University Press, Oxford. Rushton L, et al. (2012). Occupational cancer burden in Great Britain. Br J Cancer, 107, S3–​7. Sithamparanadarajah R (2008). Controlling skin exposure to chemi- cals and wet-​work: a practical book. RMS Publishing, Stourbridge. https://​www.personneltoday.com/​hr/​controlling-​skin-​exposure-​to-​ chemicals-​and-​wet-​work-​a-​practical-​book/​ Snashall D, Patel D (eds) (2013). ABC of occupational and envi- ronmental medicine (ABC series), 3rd edition. Wiley-​Blackwell, Chichester. Stellman JM (ed) (1998). Encyclopaedia of occupational health and safety (vol.2—​Hazards), 4th edition. International Labour Office, Geneva. Venables K (2013). Current topics in occupational epidemiology. Oxford University Press, Oxford. Waddell G, Burton AK (2000). Occupational health guidelines for the management of low back pain at work: evidence review. Occup Med (Lond), 51, 124–​35. Waddell G, Burton AK (2006). Is work good for your health and well-​ being? TSO, London. White RF, Proctor SP (1997). Review:  solvents and neurotoxicity. Lancet, 349, 1239–​43. 10.2.2  Occupational safety Lawrence Waterman ESSENTIALS Any approach to occupational health must acknowledge that acci- dents in the workplace result in many injuries. Construction, agricul- ture, and primary extraction provide the main worksites for fatalities and serious injuries, but many more minor injuries result from all types of work. Health and safety law has developed with an emphasis on accident prevention that is based on designing and managing the working environment by: (1)  defining appropriate processes and work practices that are safe; (2)  ensuring that the workplace and work equipment are suitable for the work to be undertaken; (3)  developing and maintaining a health and safety culture, including mechanisms to influence behaviour, so that everyone is focused on the best and safest way to do their work. Establishing this approach to safety management begins with an organization committing itself to a policy influenced by legal obli- gations and current good practice, such as the developing standards for corporate governance of risks and public reporting. Management systems based on the formula ‘Plan–​Do–​Check–​Act’ are central to accident prevention, with detailed decisions driven by risk assess- ments. Risk assessments are formed when the hazards associated with the work are identified; that is, the potential for harm, and the likelihood, severity, and impact of that harm is assessed. This forms the basis for defining precautions and mitigations to minimize likeli- hood and impact, in proportion to the degree of risk. A key ingredient to safety is genuine worker engagement, going beyond the legal obligations for consultation. Organizations can im- prove their safety culture when they recognize that this is the product of individual and group values, attitudes, competence, and patterns of behaviour that determine the commitment to, and the style and pro- ficiency of, their safety programmes. A positive culture requires appro- priate leadership, including genuine commitment of the most senior manager(s) in the organization, and an appropriate emphasis on com- petence, such that the right people, trained and skilled, are doing the right job in the right way, with their supervisors and managers having ready access to competent health and safety advice when required. These basic concepts hold true in whichever country you are working. While this chapter draws heavily on the UK situation, where there has been considerable experience and development of ap- proaches to health and safety, it is incumbent on any physician to consider the work environment and whether changes to the work- place might improve not only the lot of their patients but others po- tentially at risk in the workplace. Introduction Accidents, sometimes fatal, are an important cause of illness and harm to health at work. In 1995, the World Health Organization estimated that every year, worldwide, there were approximately 120  million

10.2.2  Occupational safety 1653 accidents leading to 20 000 fatalities. In 2014, the International Labour Organization estimated that of 2.33 millions deaths from occupa- tional accidents and work-related illness, 13.7% or 380 500 were due to accidents (Reference: Global Estimates of Occupational Accidents and Work-Related Illnesses 2017, Paivi Hamalainen, Jukka Takala, Tan Boon Kiat, WSH Institute Singapore for the ILO, 2017). China, whose workforce is expanding quickly through rapid industrializa- tion, reported a rise in fatal accidents from 73 500 in 1998 to 80 000 in 2010. According to official data the reported Chinese fatal accident rate was 21 times that for the United Kingdom. In industries with a poor safety record, such as mineral extraction and agriculture, the toll of accidental injuries is matched or exceeded by the numbers made ill by exposure to health risks. It is therefore appropriate to give ser- ious consideration to the risks of both accidents and illnesses at work. Accidents at work occur under varying circumstances. Globally, and in the United Kingdom, the largest group of accidents and fatalities occur in the construction industry. Worldwide, one in six fatal work- place accidents occurs on a construction site. However, in the United Kingdom in 2017/18, 24% of all such fatalities were associated with construction work with approximately 7% of the workforce, the higher proportion reflecting economic development away from traditionally hazardous industries such as heavy engineering and mining. But ser- ious accidents continue to occur: fires and explosions in chemical pro- cess plants, transportation disasters, and entanglement in machinery. Most accidents involve everyday events such as slips, trips, and hand-​ tool injuries that most commonly result in only minor injuries. Research since the 1950s has demonstrated that in any organiza- tion, many minor accidents occur for every serious one. This was developed into a strategy for preventing accidental losses, both fi- nancial and human, based on the concept of accident triangles illustrating the relationship between minor events and serious ac- cidents (Fig. 10.2.2.1). Despite the diversity of accidents and their outcomes, their causes, and the underlying principles of safety man- agement for their prevention are common to all accidents, and to the control of other health hazards at work. The three elements of safety management are designed to address: • working environment—​encompassing physical arrangements, equipment, materials, and the environment in which they are used • working processes and practices—​the way in which the work is expected to be carried out, often embodied in written standard operating procedures • culture and behaviour—​the human element, that takes the work- place and the procedures and brings them to life, summed up as ‘this is the way we do things around here’ In order to address these elements, employers have increasingly been encouraged to develop formal management systems. The key mechanism adopted for ‘encouragement’ has been the development of occupational health and safety law. British law from the mid-​19th century onwards seemed to reflect Parliament’s desire to take action only after a tragedy had happened. As industrial hazards were re- vealed by research or major accidents, regulators conducted a form of risk assessment and prescribed general control measures. As a re- sult, many different and highly specific sets of regulations ended up on the statute book, with overarching acts, such as the Factories Acts imposing general duties applicable only to defined workplaces, such as factories, workshops, offices, shops, and railway premises. The 1802 Health and Morals of Apprentices Act and subsequent legisla- tion were crucial in driving workplace controls. Many workers were excluded but it was estimated that the 1974 Health and Safety at Work Act brought up to 5 million workers within the ambit of the law. In the United Kingdom after the First World War, there were de- velopments outside legislation. The cooperative movement grew into a major developer of social commerce, and other organizations, such as the children’s organization Woodcraft Folk, were founded on ideas of social improvement. The public began to demand better management of industrial safety, both for moral and economic reasons. The British Industrial Safety First Movement, later the Royal Society for the Prevention of Accidents (RoSPA), promoted accident prevention through committees, workers’ participation, the employment of safety officers, joint accident investigations, and the promotion of positive attitudes. Enlightened industrial management that recognized the cost of accidental losses, legal obligations, and voluntary efforts based on worker engagement and morality was summarized in the 1972 Robens Report, which formed the basis for the Health and Safety at Work Act 1974. Risk assessment, initially implicit, was later made an explicit requirement. Worker engagement, through safety rep- resentatives and worker/​management committees (at least in the minority of British workplaces with recognized trades unions), was a new element. However, law always seemed to lag behind in- dustrial developments. The 1974 Act established the Health and Safety Commission and the Health and Safety Executive not only to be custodians, developers, and enforcers of the law, but also to be responsible for identifying and disseminating good practice. Although many people accept the need for both enforcement and encouragement in dealing with enlightened and obstinate employers, some call for more prosecutions of companies and individual dir- ectors. Employers are no longer expected to adhere slavishly to sets of statutory rules, but to develop their own safety organization and arrangements following good practice. It is now recognized that safety is a key challenge to be addressed by management. Other events have influenced organizations’ approach to risk. Financial scandals (Polly Peck, Mirror Group Newspapers) led to reports from committees chaired by Cadbury (on preventing malpractice), Greenbury (on the role of directors), and Hampel (outlining good governance) that laid the foundations for Turnbull and changes in the rules applying to publicly listed companies in the United Kingdom. Management of risk and safety are essential com- ponents of good corporate governance both in the public and vol- untary sectors. While these formal risk management requirements were being developed, corporate social responsibility was maturing through the formulation of the Global Reporting Initiative and local ventures such as corporate social responsibility indices, the launch of FTSE4Good, and ethical investment organizations. Their impact should not be underestimated. The history of companies such as Railtrack, where major accidents led to loss of control, should re- mind directors that they are expected to govern in order to protect the public, the organization, and its shareholders from such disasters. 1 20 30 600 Major Minor Property damage Near miss Fig. 10.2.2.1  Accident triangle developed by Frank Bird.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1654 The size of the problem During the year 2014/​15, some 142 workers, employees, and self-​ employed, were killed at work in the United Kingdom. This reflects both effective accident reduction and a dramatic change in economic activity; there are now many more people working in call centres than in mining. Numbers of non​fatal accidents are less certain, because of underreporting. In the period 2014/​15 61 100 workers reported in- juries through the self-​reporting schemes that the Health and Safety Executive uses to record injuries. According to figures recorded through Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 2013 (RIDDOR) 76 054 nonfatal injuries occurred in UK workplaces. Slipping and tripping are responsible for the largest group of major injuries, almost three times the numbers of injuries from moving machinery/​objects, from falls, or from direct handling. Statistics show that many employees suffer from mental ill health, followed by musculoskeletal disorders. However, an individual’s per- ception of what level of incapacity forces them to be absent from work is clearly influenced by their mental state. The nature of the injuries received and the process of rehabilitation and return to work are clearly important, but the safety practitioner is interested primarily in methods of prevention embodied in safety management systems and the associated work to develop and maintain the safety culture. Safety management Accident prevention requires a safe working environment and suit- able equipment combined with safe behaviour by the workforce, including the avoidance of error, by those doing hazardous work. Most practical accident prevention involves physical safeguards or safety rules designed to prevent recurrence of unsafe conditions or acts that have already led to accidents. Similarly, occupational hy- gienists anticipate, recognize, assess, and control hazards to health at work. Accidents usually result from a chain of connected events. Effective safety management addresses the physical conditions, the work processes, and the human behaviour that have combined to cause harm and conversely could prevent harm. Modern safety management is based on risk assessment and the definition and implementation of methods for preventing harm. Policy—​organizational commitment Effective health and safety policies set out a clear direction for the organization, indicating what is expected of managers and workers. This may represent no more than a commitment to formal, min- imum legal compliance but most bodies express themselves more in the manner of a mission statement. For example, Marks & Spencer is ‘committed to ensure the health, safety, and well-​being of all its employees, customers and others who visit or work on our premises’. The policy is an opportunity clearly to state the responsi- bilities and arrangements for delivering what is promised. Planning The foundation of safety management is planning:  to create the right physical conditions and shape the activities undertaken. This includes identifying accident risks inherent in the workplace and the work, assessing their significance and determining suitable pre- cautions. There is a hierarchy of such controls, promulgated across the European Union in the Framework Directive, which targets protection of the whole group of exposed people rather than per- sonal protection: • Elimination—​selection and design of facilities, equipment, and processes to prevent exposure to an identified risk. For example, fitting a long-​life plastic sign may avoid the need to repaint a high board. • Minimization—​reducing risk. For example, selecting a less-​ hazardous material, redesigning a tool, or replacing baskets and belts on a conveyor with softer, more resilient materials to reduce noise. • Personal protection—​as a last resort, selecting methods of work and personal protective equipment so that the exposed workers are protected to some extent. Risk assessment is a critical activity. It requires a degree of know- ledge and understanding of the work itself and of the principles of safety management. This means that in practice it is often best conducted by a team made up of workers, managers, and safety practitioners. Monitoring performance Residual risks that cannot be wholly eliminated require moni- toring. Key performance indicators may be used to explore training courses, supervisory inspections, and accident frequency rates. Accident frequency rates are typically related to workers’ hours (i.e. accidents per 100 000 h worked), allowing compari- sons between different-​sized workplaces and workforces. Active monitoring of key performance indicators is supplemented with reactive monitoring (accident investigations) to determine imme- diate and underlying causes of failure. Audit and review Good organizations learn from their experiences, and apply the lessons. Regular audits check that safety policies are being imple- mented, that people are discharging their responsibilities, and pro- cedures for safe working are documented and followed. Reviews are more searching and make comparisons with other organizations. They ask the challenging question, ‘Are we doing the right or best things?’ These processes contribute to an organization’s ability to communicate to its stakeholders inside and outside the organiza- tion, for example in the annual report. There are three types of cor- porate report on health and safety: • minimal reports on injuries and ill health, days lost, comparison with national or sector targets, and information on events such as awards and/​or convictions • comprehensive reports with statistics, trend analysis against per- formance indicators, director workplace visits, health and safety training days, inspections and audits, emergency drills, and so on • verified reports (i.e. comprehensive reports that have been re- viewed externally)

10.2.2  Occupational safety 1655 Safety management and safety culture The effort to create and maintain a physically suitable workplace and equipment, and implement documented safe working prac- tices, will be effective only if it is matched by the engagement of the whole workforce from director to shop floor. A healthy and safe culture is one in which the members of the organization: • understand and respect the hazards of their operations • are alert to the many ways in which safe working systems may be breached or bypassed • are committed to maintaining safe working • honestly report problems and opportunities for improvement A key ingredient is real worker engagement, going beyond the legal obligations to consult. Organizations can progress when they recognize that their safety culture is the product of indi- vidual and group values, attitudes, competencies, and patterns of behaviour that determine the commitment to, and the style and proficiency of their safety programmes. Those with a positive culture communicate on a basis of mutual trust, share percep- tions of the importance of safety, and are confident in the efficacy of preventive measures. This becomes possible only when there is appropriate leadership, including the commitment of the chief executive. As an example, the strategy adopted by the Olympic Delivery Authority for the planning and construction of venues, test events, and infrastructure for the London 2012 Games, the 2–​4–​1 approach, has allowed time to establish the right culture. The staff were com- mitted to weaving safe design and construction into all components of London 2012 and its legacy for east London and United Kingdom as a whole. This was made clear to all the contractors and other sup- pliers involved in this project. Competence Training is an essential part of any company’s safety arrangements. How well people have been trained to do what is required, from working safely to reporting problems, will influence the risks in the working environment. Competence is not purely defined by training and qualifications, but by the additional skills, knowledge, and experience of individ- uals in the workplace. The definition of competence has been up- dated in the Construction (Design and Management) Regulations 2015 reflecting this approach. Current issues Leadership There has been an increasing drive to promote strong leadership within organizations, to continually improve safety and health risk management performance. This is reflected in various activities. The Institute of Directors published guidance on safety leadership aimed at those holding senior corporate positions. Furthermore, there have been legal developments such as the Construction (Design and Management) Regulations 2015 that have put a greater emphasis on the duties held by a client and new role of Principal Designer to ensure safety management is a fundamental approach to construction projects. The Health and Safety Executive has re- vised its guidance on safety management by updating the HSG 65 Successful Health and Safety Management model. This revised model has a modernized approach that aligns more closely with the principles of Plan–​Do–​Check–​Act which was laid out in the management standard OHSAS 18001, now superseded by ISO 45001 management standard, published in 2018 and now used to evaluate, certify, and accredit management systems worldwide. The new standard is much more explicit and proscriptive about the leadership that organization are required to demonstrate and the evidence auditors will seek to be satisfied. The standard is written to align with recently updated management standards ISO 9001 Quality Management Systems and ISO 14001 Environmental Management. All of these new standards have stricter require- ments for organizations to demonstrate leadership to bring about changes and improvements in performance. Additionally, in the United Kingdom the Health and Safety Executive has published guidance aimed at small to medium-​sized enterprises, and cre- ated a self-​assessment tool to establish what leadership activity is in practice while identifying opportunities to improve leadership performance. Cost recovery and sentencing guidelines In light of government reviews to reduce the administrative burden of health and safety on employers, many pieces of UK legislation have been repealed, amended, and updated to ease bureaucracy, re- move unnecessary red-​tape, and reduce costs. Changes to accident reporting regulations (RIDDOR) brought in a change in threshold for when an employer must report injuries resulting in lost time from over 3 days to over 7 days. In addition, it was concluded that the cost burden for investigating and enforcing material breaches of health and safety legislation should be borne by the employer at fault and not the taxpayer. This has led to the Health and Safety (Fees) Regulations 2012 and the introduction of Fees for Intervention. Under these arrangements, HSE can charge employers, based on an hourly rate of around £127 to recover costs associated with inves- tigation, research, consultation with specialist advisors, and subse- quent enforcement action where there has been a material breach of legislation. Another major change to the enforcement process has been the publication in 2016 of revised sentencing guidelines in for health and safety offences, which has led to a significant increase in the financial penalties levied on larger organisations—for example in 2017 Laing O’Rourke was fined £3.8m following a fatal accident, and Warburtons the bakers £1.9m following an entrapment accident that resulted in severe friction burns requiring skin grafts for the worker involved. Prior to these changes, fines and custodial sentences were limited by fairly low thresholds. Furthermore, sentences were more likely to be based on the outcome of a breach in legislation. However, the Health and Safety at Work Act 1974 is based on risk of exposure and so the new sentencing guidelines reflect this, as the sentence issued should be based on the ‘harm risked’. The inflationary factors that can lead to significantly increased fines and custodial sentences include culp- ability for an organization or individuals, harm risked, number of people exposed, and the financial turnover of the organization. This can lead to fines in the millions of pounds for large organizations and a shift from a typical six-​month custodial sentence to a possible

10.2.3 Aviation medicine 1656

10.2.3 Aviation medicine 1656

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1656 18 months imprisonment. Although the Health and Safety at Work Act 1974 has remained unchanged in this time, the new guidelines represent the most significant legal impact in the last 40 years. The changing world of work An ageing working population, equality and diversity, and flex- ible working (sometimes referred to as the “gig economy” in which work is undertaken without a formal contract of employment guar- anteeing a “normal” working week—often resulting in employers treating the relevant workers as if they were self-employed and re- sponsible for making all the choices about the work that they do) are some features of the rapidly changing world of work. Over half of the British workforce is now employed in smaller organizations, and over 90% of businesses employ fewer than 10 people. Part-​time working has grown, and women now constitute half the workforce. Globalization has intensified competitive pressures, particularly on manufacturers. Public tolerance of accidents is very low and so there is pressure to make further improvements in occupational safety. In the United Kingdom, there will soon be more people over 65 than under 18 and many older people will continue to work past what was regarded as normal retirement age. Above all, we may be entering a new period of work shaped by Artifical Intelligence, robotics and other technological aspects. A number of studies have recently been undertaken exploring the impact of these issues on health and safety at work. Advice and assistance Those likely to be affected deserve to be properly protected against risks to their health and safety at work. Employers have a duty to protect their employees by sensible risk management. Workers also have a duty to protect their own and others’ health and safety. Sensible risk management requires effective systems to control those risks that arise frequently and have serious consequences. Balancing benefits and risks often requires expert help. The Management of Health and Safety at Work Regulations re- quire employers to appoint ‘ . . . one or more competent persons . . . ’ to assist them in meeting their duty of controlling risks. Employers and managers are in the best position to understand the health and safety issues in their business. Coupled with the knowledge of employees, this is often enough to ensure that risks are properly controlled, especially where the hazards are those commonly en- countered at work and methods for their control are already estab- lished practice. However, if the risks are complex or large numbers of employees are involved, expert help may be needed. Employers can rely on one or more of their employees to give them competent help, provided the employees have been given enough time, training, and access to information. The employer could: • train or develop the necessary skills in an existing employee • recruit someone with the necessary skills • make use of consultancy support staff Formally qualified health and safety practitioners can work with a team of risk managers, including occupational health advisers. In the United Kingdom, the Institution of Occupational Safety and Health sets standards and awards qualifications. Preventing accidents at work makes an important contribution to the health and well-​being of all who may be affected by an en- terprise, but achieving this aim requires a systematic approach and leadership. FURTHER READING Bird FE, Germain GL (1966). Damage control (a new horizon in ac- cident prevention and cost improvement). American Management Associations, New York, NY. British Safety Council and Robertson Cooper (2018). Future risk: The impact of work on health, safety and well-being. London. Eves D, Gummer J (2005). Questioning performance: the director’s essential guide to health, safety and environment. UK Institution of Occupational Safety and Health (IOSH), Wigston, Leicestershire. Frick K, et  al. (eds) (2000). Systematic occupational health and safety management—​perspectives on an international development. Pergamon, Oxford. Health and Safety Executive (2013). Managing for health and safety, 3rd edition. HSG65. http://​www.hse.gov.uk/​pubns/​books/​hsg65.htm Health and Safety Executive and Institute of Directors (2013). Leading health and safety at work, INDG417(rev1) 06/​13. http://​www.hse. gov.uk/​pubns/​indg417.pdf Reason J (1997). Managing the risk of organizational accidents. Ashgate, Aldershot. Sentencing Council (2015). Health and safety offences, corporate man- slaughter and food safety and hygiene offences: definitive guideline. https://​www.sentencingcouncil.org.uk/​wp-​content/​uploads/​HS-​ offences-​definitive-​guideline-​FINAL-​web.pdf Woolf AD (1973). Robens Report—​the wrong approach? Industrial Law J, 2(1), 88. 10.2.3  Aviation medicine Michael Bagshaw ESSENTIALS Travel by air is a safe means of transport, but puts people at various physiological risks and is a potential means of spreading infectious disease. Physiological risks associated with flying include hypoxia—​ atmospheric pressure falls with altitude. The minimum cabin pres- sure in commercial passenger aircraft (565 mm Hg, 75.1 kPa) brings a healthy individual’s arterial P along the plateau of the oxyhaemo- globin dissociation curve until just at the top of the steep part, but does not cause desaturation. By contrast, people with respiratory dis- ease and a low arterial oxygen pressure may desaturate, which can be overcome by administering 30% oxygen, this being equivalent to breathing air at ground level. Guidance for assessing a passenger’s fitness to fly is provided by the websites of the Aerospace Medical Association and the British Thoracic Society. A second physiological risk is increased exposure to cosmic radiation, although there is no evidence that this leads to abnormality or disease. Other medical problems associated with flying include (1) venous thromboembolism—​the relative risk is significant, but the absolute risk is very low. Medical practitioners need to be circumspect in advising preventative measures, taking account of the efficacy and risk profile of any intervention, but compression stockings and/​or a single prophylactic dose of low molecular weight heparin may be re- commended in high-​risk cases. (2) Jet lag—​there is no simple solution

10.2.3  Aviation medicine 1657 for combating the effects of jet lag: the individual must evolve strat- egies to suit their particular needs. There is no evidence of a causative association between the use of engine bleed air for pressurization and ill health of aircraft occupants. Transmission of disease—​there is no evidence that the pressurized aircraft cabin itself encourages transmission of disease, and recircu- lation of cabin air is not a risk factor for contracting symptoms of upper respiratory tract infection. It is important that individuals with a febrile illness should not travel on commercial aircraft. Restricting air travel will not prevent global spread of pandemic influenza, but might delay the spread sufficiently to allow countries time to prepare. Introduction To answer practical questions about the effects of flight on the body, it is necessary to understand the physics and physiology of flight, the discipline of aviation medicine. Aerospace medicine is very much a specialized discipline, with a history traced back to the descriptions of altered physiology during balloon ascent by Glaisher and Coxwell in 1862. Aviation medicine concerns the well-​being of humans in flight within the Earth’s atmosphere, whereas space medicine concerns the welfare of humans flying beyond the atmosphere and the Earth’s gravitational pull. Space medicine addresses the problems asso- ciated with very prolonged flight times and life support within a self-​contained environment, as well as weightlessness, exposure to high doses of cosmic radiation, and the psychological aspects of pro- longed spaceflight. Those seeking information on the specific effects of space flight are referred to the specialized texts in the ‘Further reading’ section at the end of this chapter. Physics of the flight environment The Earth’s atmosphere is an oxygen-​rich gas shielding the ground below from solar radiation above. Subjected to gravity, compressed under its own weight, the atmosphere is denser close to the ground than further away. Long waves of infrared light penetrate it easily but heat the ground below. Heated ground reradiates some of this heat at shorter wavelengths which are absorbed by carbon dioxide and water vapour, making the air close to the surface much warmer than that higher up. Short waves of ultraviolet sunlight, absorbed by oxygen molecules early in their journey, create a belt of ozone at high altitudes. Some rays intercepted in the same region generate secondary rays that extend lower down. Very few reach the ground. At sea level, the atmosphere exerts a pressure of about 760 mm Hg (101 kPa); it is variably moist, has a temperature that ranges from –​ 60°C to +60 °C, and moves at wind speeds from 0 to 160 km/​h. With increasing altitude, the temperature, pressure, and water content of the atmosphere fall and wind speeds increase (Fig. 10.2.3.1). Atmospheric pressure Total gas pressure falls with altitude in a regular manner, halving every 18 000 ft (5500 m) (Fig. 10.2.3.2). The oxygen content of the atmosphere (20.93%) is constant to very high altitudes, so the same curve can be used to obtain the ambient oxygen pressure by rescaling the ordinate (Fig. 10.2.3.2). The oxygen pressure of physiological importance is that which exists in ambient air when it is warmed and wetted on entering the bronchial tree. This raises water vapour pres- sure to about 47 mm Hg (6.3 kPa) regardless of the total gas pressure outside. The oxygen pressure in moist inspired gas (PiO2) fully satur- ated with water vapour at 37 °C is given by the relationship: PiO = FiO Patm 47 2 2 − ( ) O3 concentration (p.p.m.) 0 10 Temperature (°C) −60 0 40 Pressure (atm) 0 1 150 100 50 0 Heat lets molecules escape Some rays are absorbed Infrared UV Cosmic Gravity pulls atmosphere down Cosmic Altitude (ft × 103) Fig. 10.2.3.1  Some physical features of the Earth’s atmosphere, showing the variations in barometric pressure, air temperature, and ozone concentration with altitude. (NB: There is an international aviation safety convention that all altitudes are given in feet.) The shaded diagram on the left illustrates how the Earth’s atmosphere is compressed under its own weight. The atmosphere absorbs much solar radiation.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1658 where FiO2, the fractional concentration of oxygen in the inspirate, is 0.2093 and Patm = atmospheric pressure. Atmospheric temperature The atmospheric temperature reduces at 1.98 °C/​1000 ft (300 m) from the standard sea level temperature of 15°C, to the tropo- pause (40 000 ft (12 200 m)). It remains stable at –​56 °C up to about 80 000 ft (24 400 m) and then rises to almost body temperature at about 150 000 ft (46 000 m), but by then air density is so low that its temperature is unimportant. Atmospheric ozone Atmospheric ozone is formed by ultraviolet irradiation of diatomic oxygen molecules which dissociate into atoms. At very high alti- tudes, all oxygen exists in the monatomic form. Lower down, some of this monatomic oxygen combines with oxygen molecules to form the triatomic gas ozone, with concentrations up to 10 parts per mil- lion (ppm). The ozonosphere normally exists between 40 000 and 140 000 ft (12 200 and 42 700 m). Below 40 000 ft (12 200 m), the irradiation is normally too weak for significant amounts of ozone to form. Concentrations of 1 ppm at sea level can cause lung irritation. Modern passenger jet aircraft are fitted with catalytic converters in the environmental control system, which break down the ozone before it enters the pressurized cabin. Cosmic radiation Aircraft occupants are exposed to elevated levels of cosmic radiation of galactic and solar origin. The sun has a varying magnetic field, which reverses direction approximately every 11  years. Near the reversal, at ‘solar min- imum’, there are few sunspots and the sun’s magnetic field extending throughout the solar system is relatively weak. At solar maximum, there are many sunspots and other manifestations of magnetic turbulence. The Earth’s magnetic field has a larger effect than the sun’s mag- netic field on cosmic radiation approaching the atmosphere. The protective effect is greatest at the equator and least at the magnetic poles. At jet aircraft operating altitudes, galactic cosmic radiation is 2.5–​5 times more intense in polar regions than near the equator. The Earth’s surface is shielded from cosmic radiation by the at- mosphere, the ambient radiation decreasing with altitude by ap- proximately 15% for each increase of around 2000 ft (dependent on latitude). Protection against effects of cosmic radiation The International Commission on Radiological Protection (ICRP) recommended in 1991 that exposure of flight crew members to cosmic radiation in jet aircraft should be considered part of occupa- tional exposure to ionizing radiation. The ICRP limits for occupational exposure are a 5-​year average effective dose of 20 millisieverts (mSv) per year, with no more than 50 mSv in a single year. The annual limit for the general public is 1 mSv. Cosmic radiation doses The effect of ionizing radiation depends not only on the dose ab- sorbed, but also on the type and energy of the radiation and the tis- sues involved. These factors are taken into account in deriving the dose equivalent measured in Sieverts (Sv). However, doses of cosmic radiation are so low that figures are usually quoted in microsieverts (µSv) or millisieverts (mSv). Calculated and measured doses are well within the ICRP recommended limits. Health risks of cosmic radiation While it is known that there is no level of ionizing radiation ex- posure below which effects do not occur, current epidemiological evidence indicates that the probability of airline crew members or passengers suffering any abnormality or disease as a result of ex- posure to cosmic radiation is very low. Physiology of flight The physiological effects of flight are distinguished from those of terrestrial high altitude because exposures are relatively rapid, brief, and not cumulative. Flyers do not adapt to the hypoxic environment, unlike inhabitants of terrestrial high altitudes. However, the aircraft can be a means of transporting an individual to a high-​altitude destination. Hypoxia Oxygen has a dual role in most animal cells, being simultaneously life-​giving and extremely poisonous. In air, or dissolved in simple solution, it is benign and only ionized with difficulty. However, once an electron is successfully attached to an oxygen molecule it becomes a highly corrosive superoxide ion, forming a cascade of other very destructive oxygen radicals. This is an essential feature of oxygen toxicity, which is discussed in Chapter 10.2.4. Superoxide dismutase and various peroxidases have evolved to protect most cells from the effects of spontaneous formation of oxygen radicals by quenching the ions as rapidly as they appear. Other enzymes have evolved which harness this property in a controlled way. There are three types:  oxidases, oxygenases, and hydroxylases. Quantitatively, cytochrome a3 oxidase is the most important because, using oxygen as the ultimate electron sink, it allows many metabolic processes to proceed at the same time un- locking and trapping most of the energy the body needs (oxidative phosphorylation). Oxygenases introduce an oxygen molecule into organic molecules creating new compounds. Although these enzymes consume only a small fraction of the body’s total oxygen requirement, they are par- ticularly important for production and dismemberment of many critical compounds such as the amine transmitters of the brain. PB Pressure (atm) 1.0 0.8 0.6 0.4 0.2 0.00 20 40 60 80 0 20 40 60 80 PO2 0.20 0.15 0.10 0.05 0.0 Altitude (ft × 103) Fig. 10.2.3.2  The variations of barometric pressure (PB) and ambient oxygen pressure (PO2) with altitude.

10.2.3  Aviation medicine 1659 Hydroxylases insert one atom of oxygen and another of hydrogen into organic molecules. They too are responsible for many critical metabolic processes and for the denaturation of many drugs in the liver, kidney, and elsewhere. These enzymes differ in their affinity for oxygen, described by the Michaelis constant (for oxygen). This constant (KmO2) is that partial pressure of oxygen which, when all other factors are equal, allows an oxygen-​consuming reaction to proceed at half its maximum vel- ocity. The major oxidase (cytochrome a3), which is the cocatalyst of oxidative phosphorylation, has a very high oxygen affinity and thus a very low KmO2, of 1 mm Hg or less. Thus, this particular type of oxygen consumption, representing 80–​90% of the whole, can proceed at high rate down to very low levels of oxygen supply. By contrast (Fig. 10.2.3.3), the other enzymes, which are quantitatively less important but qualitatively critical, have Michaelis constants for oxygen that vary from 5 to 250 mm Hg (0.7–​33.3 kPa). A fall in oxygen supply will influence these processes long before oxidative phosphorylation is affected and at times when overall oxygen con- sumption is diminished little if at all. When humans are exposed to hypoxia, systemic and intracel- lular changes operate together to minimize hypoxic injury and re- store adequate oxygenation. Emerging evidence indicates that the hypoxia-​inducible factor (HIF) family of transcription factors plays a central regulatory role in these homeostatic changes at both the systemic and cellular levels. HIF was discovered through its action as the transcriptional activator of erythropoietin, and has sub- sequently been found to control intracellular hypoxic responses throughout the body. HIF is primarily regulated by specific prolyl hydroxylase-​domain enzymes (PHDs) that initiate its degradation via the von Hippel-​Lindau tumour suppressor protein (VHL). The oxygen and iron dependency of PHD activity accounts for regula- tion of the pathway by both cellular oxygen and iron status. Recent studies conducted in patients with rare genetic diseases have begun to uncover the wider importance of the PHD-​VHL-​HIF axis in systems-​level human biology. These studies indicate that, in addition to regulating erythropoiesis, the system plays an important role in cardiopulmonary regulation. Although Fig. 10.2.3.2 describes how ambient oxygen pressure is related to altitude, it does not convey the pressure of oxygen to be found in the lungs. That pressure is determined by two equations (Fig. 10.2.3.4). The alveolar ventilation equation states that alveolar CO2 pressure (PaCO2) depends only on CO2 excretion (CO2) and alveolar ventilation (Va), so: PaCO = k CO /Va . 2 2 ( ) The alveolar air equation states that since at any one time there is a fixed trading ratio between oxygen uptake and CO2 excre- tion (R = CO2/​O2), alveolar oxygen pressure (PaO2) can be calcu- lated from the moist inspired oxygen pressure (PiO2*) and alveolar PCO2, so: PaO = PiO * PaCO /R 2 2 2 −( ) Progressive hypoxia leads to a mild hyperventilation (i.e. a rise in Va and fall in PaCO2). Thus, it is possible to plot alveolar oxygen pressure against altitude (Fig. 10.2.3.5a). When arterialized blood leaves a healthy lung the oxygen pres- sure is some 10 mm Hg less than that in the alveoli, due to uneven The Michaelis–Menten equation when the substrate is oxygen: MO2/MO2 max = PO2/(PO2 + KmO2) Cytochrome a3 oxidase Other oxidases and oxygenases 1 5 25 100 250 KmO2 PO2 (mmHg) 1.0 0.8 0.6 0.4 0.2 0.0 1.2 MO2/MO2 max 0 40 80 120 160 Fig. 10.2.3.3  Curves of oxygen uptake (O2) as a fraction of the theoretical maximum (O2max) against the partial pressure of oxygen (PO2) for a family of oxygen-​handling enzymes with Michaelis constants for oxygen (KmO2) from 1 to 250 mm Hg. PACO2 ∝ MCO2/VA Time PAO2=PIO2 – PaCO2/R Partial pressure The Alveolar air equation pictures VA trapped in a bag, and notes there must be a link between the rise in PCO2 and the fall in PO2, so that, for most practical purposes: The Alveolar ventilation equation ignores dead-space, and supposes there is a stream of oxygen-rich CO2 free gas, VA and says, for practical purposes: VA Plus the CO2 output, MCO2 Minus the almost equal O2 uptake, MO2 VA Fall ∝ MO2 Rise ∝ MCO2 MO2 MCO2 Fig. 10.2.3.4  Graphical representations of the alveolar ventilation and alveolar air equations. PO2 (mmHg) (a) Inspired air Alveolar gas Arterial blood (b) Blood PO2 (mmHg) 0 0 1 0 Oxygen saturation (%) Whole-blood O2-Hb dissociation a. v 0 150 100 50 0 150 100 50 0 Altitude (ft × 103) a-v∆ MO2/Q 30 20 10 Fig. 10.2.3.5  (a) Variations in moist inspired, alveolar, and arterial oxygen pressure (PO2) with altitude in normal men. (b) The conventional oxygen–​haemoglobin dissociation curve of whole blood plotted to the same pressure scale as the left-​hand graph, so that arterial O2 content can be read directly (at the same horizontal level as the PO2 curve). It also emphasizes that the arteriovenous oxygen content difference (a–​vΔ) is proportional to the ratio of oxygen uptake (MO2) to local blood flow (Q).

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1660 matching of ventilation to perfusion, some anatomical shunting, and an almost nominal obstacle to diffusion. In resting people, the alveolar–​arterial oxygen gradient does not change much with alti- tude, although the relative importance of the factors contributing to it alter considerably; so subtracting a further 10–​15 mm Hg de- scribes the relation between arterial oxygen pressure and altitude (Fig. 10.2.3.5). The most important change is the loss of pressure driving oxygen from the alveoli to blood, as the fall in alveolar PO2 is much greater than that in mixed venous PO2 (because of the shape of the oxygen dissociation curve). As a result, the alveolar–​venous gradient for oxygen diffusion is smaller and equilibration slower than at ground level. People ascend to altitude in a matter of minutes, rather than over several days, and adapt to hypoxia by an increase in blood flow and a modest hyperventilation, limiting the effects of hypoxia. The effects are shown in Fig. 10.2.3.6. Individuals abruptly exposed to altitudes of 10 000 ft (3000 m) and above suffer mental and physical effects, and is the ceiling above which aviators are provided with oxygen. To allow a margin of safety, the maximum certified cabin altitude in civilian pas- senger aircraft is 8000 ft (2440 m), at which barometric pressure is 565 mm Hg and arterial oxygen pressure is around 55 mm Hg (see Fig 10.2.3.5b, the oxyhaemoglobin dissociation curve), and venous oxygen pressures have only fallen by 1–​2 mm Hg. Even at this altitude, there is a decrease in performance. The latest gener- ation of passenger aircraft are manufactured from newer materials which provide greater strength from a given mass, thus allowing a higher differential cabin pressure with a lower cabin altitude. Two physiological features of altitude hypoxia are important in aviation. The first is the total lack of awareness of cerebral impair- ment. The second is the time of useful consciousness, describing how rapidly consciousness is lost thus dictating how quickly the condition must be recognized and corrective action taken. The time of useful consciousness is the interval after the onset of hypoxia during which an individual can carry out some purposeful activity. The general relation between this time interval and the alti- tude of sudden exposure is shown in Fig. 10.2.3.7a. It diminishes from about 4 min at 25 000 ft (7620 m) to a minimum of roughly 15 s, which is reached at 35 000 (10 700 m) to 40 000 ft (12 200 m). This asymptote represents the sum of the 7 s or so required for blood to travel from the lungs to the brain and the time needed for the brain to utilize the oxygen already dissolved in its substance. In trained and healthy individuals breathing normally (i.e. with an alveolar PCO2 of 35–​40  mm  Hg (4.7–​5.3  kPa)), the dose of hypoxia acceptable before loss of useful consciousness is equivalent on a curve of alveolar PO2 against time, to an area of 150 mm Hg/s, where PO2 is less than 38 mm Hg (5.1 kPa) (Fig. 10.2.3.7b). However, this is sensitive to many other factors, such as the degree of hyperventilation and the acceleration to which the individual is exposed at the time. Hyperventilation causes cerebral vasoconstriction, and positive headwards accel- eration (+Gz) opposes the upward flow of blood to the brain. Exertion quickens loss of consciousness, because blood transits quickly through the lungs leaving insufficient time for oxygen equilibration. The minimum cabin pressure of 565 mm Hg (75.1 kPA) (8000 ft (2440 m)) in commercial passenger aircraft, will bring a healthy individual’s arterial PO2 along the plateau of the oxyhaemoglobin dissociation curve until just at the top of the steep part (Fig. 10.2.3.5), still saturated. At ground level, people with respiratory disease may have arterial oxygen pressures as low as 55–​60 mm Hg (7.3–​8 kPa). As they ascend to 8000 ft (2440 m) their arterial PO2 will fall further. If their hypoxaemia at ground level is due to a mismatch of ventilation to perfusion, as is usually the case, the drop in arterial PO2 will not be as extensive as in healthy people (about 40 mm Hg or 5.3 kPa), but if it is due to diffusion defect associated with desaturation on exertion, as in some fibrotic con- ditions, it may be greater. However, in either event, it can be re- versed completely by the administration of oxygen, 30% oxygen at 8000 ft (2440 m) being equivalent to breathing air at ground level. Given prior notice, most airlines can provide a personal oxygen supply for any passenger, although there may be a charge. (The altitudes of the patient’s destination and transit points en route should also be considered.) Oxygen equipment and pressure cabins Aircraft operating below 10 000 ft (3000 m) do not require oxygen equipment. Many sophisticated light aircraft which can cruise above 10 000 ft do not have pressurized cabins, so oxygen equipment must be provided. Other aircraft that fly higher usually have reinforced cabins capable of holding a high-​differential pressure between inside and Detectable losses in learning and night vision 0.20 0.15 0.10 0.05 0.00 Altitude (ft × 103) Must breathe pure oxygen at positive pressure to survive beyond here PO2 (atm) Degeneration in already-learnt tasks Physical weakness Coma on exertion Coma within minutes at rest Death 0 80 60 40 20 Fig. 10.2.3.6  A summary of the functional consequences of altitude hypoxia. Time (min) (a) Average Normal range Time of useful consciousness (b) Alveolar PO2 (mmHg) Time (s) Consciousness is lost once this area exceeds about 150 mm Hg/s 24 8 6 4 2 0 0 120 80 60 40 20 0 Altitude (ft × 103) 30 28 26 Fig. 10.2.3.7  (a) Variations in the time of useful consciousness with altitude. (b) One way of expressing the dose of hypoxia needed to bring about loss of consciousness.

10.2.3  Aviation medicine 1661 out. These are the high-​differential type, seen in passenger and transport aircraft generally, and the low-​differential variety found in military high-​performance aircraft. The former, holding a high transmural pressure, maintain cabin pressure above 565 mm Hg (8000 ft (2440 m)). They provide an environment in which the occupants breathe cabin air. However, it is possible that the pres- surization system can fail, allowing the cabin pressure to fall to the external ambient value. This can be limited by descent below 10 000 ft (3000 m), subject to air traffic control and terrain con- straints. An emergency oxygen supply is available for passengers and crew. The aircraft’s environmental control system automatically man- ages the internal cabin environment, providing healthy and com- fortable surroundings for all occupants. There are regulatory requirements for minimum cabin air pressure, maximum levels of carbon monoxide, carbon dioxide and ozone, and minimum ven- tilation flow rates. The cabin air must also be free from harmful or hazardous concentrations of gases or vapours. The cabin air supply is bled from the outside air entering the air- craft engine, or may be supplied from the outside air via electrically driven compressors. It is then passed through the air-​conditioning packs and mixed with filtered recirculated air before distribution to the cabin. The system provides approximately 20 cubic feet (566 litres) of air per minute per passenger, of which about 50% is re- circulated air (compared with up to 80% recirculated in buildings and other forms of public transport), giving a complete cabin air exchange every 2–​3 minutes. These high ventilatory flow rates maintain normal pressurization, as well as temperature control and the removal of odours and carbon dioxide. The high flow rates also ensure that the volume of oxygen far exceeds the requirements of the aircraft occupants (0.34 litre/​min at rest and 0.85 litre/​min when walking). The air is distributed to the cabin via overhead ducts and grills running the length of the cabin. The airflow circulates around the cabin rather than along the cabin and is continuously extracted through vents at floor level as shown in Fig. 10.2.3.8. The recirculated air is passed through high efficiency particu- late air filters of the same specification used in hospital operating theatres, giving 99.99% efficiency in the removal of physical contam- inants such as microbial particles. Aircraft cabin air has been dem- onstrated to be bacteriologically cleaner than the air in buildings, trains, or buses. Although clean, the aircraft cabin air remains dry. During the flight, moisture is derived from the metabolism and activities of the cabin occupants as well as from the galleys and washrooms, giving a maximum relative humidity in the order of 10–​20%. These levels are associated with surface drying of skin, mucous membranes, and cornea which may cause discomfort. Normal homeostatic mechanisms prevent dehydration and no harm to health has been demonstrated. A high-​differential cabin limits the aircraft’s range and manoeuv- rability and increases the risk of catastrophic damage if the fuselage is punctured. So, military high-​performance aircraft are fitted with low-​differential cabins, which prevent cabin pressure falling below 280 mm Hg (37.2 kPA) (equivalent to a pressure altitude of 25 000 ft (7620  m)). At this level decompression illness becomes a poten- tial hazard (see next). In such aircraft, oxygen equipment is used routinely. Mechanical effects of pressure change In civilian passenger and transport aircraft the climb to cruise alti- tude takes about 30 min and involves a maximum fall of about 200  mm  Hg (26.6  kPA) in cabin pressure (to the equivalent of 8000 ft (2440 m)). Descent to land takes much the same time. Body fluids and tissues generally are virtually incompressible and do not alter shape to any important extent when such pressures changes are applied. The same is true of cavities such as the lungs, gut, middle ear, and facial sinuses that contain air, provided that they can vent easily. Gas-​containing spaces that cannot vent easily be- have differently. The thoraco-​abdominal wall can develop transmural pressures of +100 mm Hg or so briefly, but is normally flaccid and has a trans- mural pressure of a few millimetres of mercury. Gas within will usually be at a pressure very close to that outside, and must follow Boyle’s law. Ascent from ground level (760 mm Hg) to 8000 ft (2440 m) (565 mm Hg or 75.3 kPa) will expand a given volume of trapped gas in a completely pliable container by about 35%. This may cause slightly uncomfortable gut distension in healthy people, but it is not an important problem. Even very diseased lungs can vent themselves over a minute or so. In consequence, the risk of lung rupture in normal flight is extremely rare (Fig. 10.2.3.9). Cabin air flow Fig. 10.2.3.8  Cabin air circulation and distribution. Each tube has an expiratory resistance, R Each balloon has a capacity, C The chest wall is very floppy The time-constant of emptying of any balloon is proportional to the product, RC. Maximum balloon volume Zero Transmural pressure (mm Hg) 0 0 1 0 Normal breathing Approximate range of bursting pressures C Fig. 10.2.3.9  A graphical summary of the factors determining lung rupture.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1662 The cavity of the middle ear vents easily, but sometimes fails to fill because the lower part of the Eustachian tube behaves as a non-​ return valve, especially when it is inflamed. As a result, the cavity equilibrates quite easily on ascent but does not refill on descent, and the eardrum bows inwards, causing pain that can be severe (otic barotrauma). Altitude-​induced decompression illness If ambient pressure falls quickly to less than half its original value, the gas dissolved in blood and tissue fluids may come out of solu- tion precipitously, forming bubbles and obstructing flow in small blood vessels. The time symptoms take to develop varies widely be- tween individuals and shortens markedly as the altitude of exposure rises. A guide to these times and variability is given in Fig. 10.2.3.10. Symptoms usually resolve quickly after a descent of a few thou- sand feet and rarely persist after descent to ground level, breathing oxygen. Should they persist, treatment should be along the lines de- tailed in Chapter 10.2.4. Atmospheric pressure halves at 18 000  ft and decompression illness occurs rarely, if at all, below this altitude. It is very rare below 25 000 ft (7600 m) and therefore is normally of no concern at normal passenger aircraft cabin altitudes, although the risk continues to be significant in some military flights. However, it does occasionally occur in those passengers who have been exposed to a hyperbaric environment prior to flight, such as divers and tunnel workers. Sub-​ aqua divers (q.v.) are advised to allow a minimum of 12 hours to elapse between diving and flight, or 24 hours if the dive required decompression stops. Clinical aspects of aviation medicine Travel by air is a safe means of transport. However, from the physio- logical point of view, flying is a means of putting people at risk as well as being a potential means of spreading infectious disease. Modern technology, coupled with stringent training requirements for flight crew, minimizes these risks but clinicians need to be aware of the ap- plications of physics and physiology to the flight environment. It can be difficult to apply epidemiological principles when con- sidering incidence and outcomes of medical conditions acquired during flight or the spread of infectious disease, because the passen- gers disperse after the flight before clinical symptoms or signs have become manifest. However, organizations such as the Aerospace Medical Association, the European Civil Aviation Conference and the World Health Organization have supported or undertaken epidemiological studies to establish the prevalence of conditions such as flight-​related deep venous thrombosis (DVT) and venous thromboembolism (VTE), spread of tuberculosis (TB), and spread of newly emerging infectious diseases such as severe acute respira- tory syndrome (SARS) and avian flu. Jet lag Besides sleep, the major influence on waking performance and alert- ness is the internal circadian clock. Circadian rhythms fluctuate on a regular cycle, which lasts something over 24 hours. The circadian rhythms are controlled by the suprachiasmatic nucleus of the hypo- thalamus. Many body functions have their own circadian rhythm and they are synchronized to a 24-​hour pattern by ‘zeitgebers’ (time givers), light being among the most powerful. Moving to a new light/​dark schedule (as in time zone changes) leads to a discrepancy between internal suprachiasmatic nucleus timing and external environmental cues. The internal clock can take days or weeks to readjust, depending on the number of time zones crossed (desynchronosis). Fatigue is defined as the likelihood of falling asleep. Therefore, in practical terms, there is little difference between chronic fatigue and acute tiredness. Fatigue can be caused by sleep loss and circadian desynchronosis, but it can also result from low motivation and low levels of external stimulation. Preventative measures Sleep scheduling: • At home the best possible sleep should be obtained before a trip; • On a trip, as much sleep per 24 hours should be obtained as would be at home; • Feelings should be trusted—​if the individual feels sleepy and cir- cumstances permit, then they should sleep. Good sleep habits: • A regular presleep routine should be developed; • Sleep time should not be reduced; • The individual should avoid going to bed hungry, but should not eat or drink heavily before going to bed; • Alcohol or caffeine should be avoided before bedtime. Caffeine consumption may be used to increase alertness. A cup of coffee usually takes between about 15 and 30 minutes to become ef- fective, and the effect lasts for between 3 and 4 hours. However this is less effective for individuals who regularly drink large amounts of caffeine-​containing beverages. Bright light (more than 2500 lux), used at the appropriate time in the circadian cycle, can help to reset the circadian clock. After flying east, the traveller should be exposed to evening light, but morning light avoided. Conversely, when travelling west, morning light should be sought, and evening light avoided. This makes the best use of the natural zeitgebers in resetting the body clock. Incidence (%) Frequency of decompression sickness at the end of 2 h exposure Rest Exertion Probable threshold 20 100 50 0 Altitude (ft × 103) 40 30 Fig. 10.2.3.10  The incidence of decompression sickness (percentage) at the end of 2 hours of exposure to various altitudes in men at rest, or exerting themselves.

10.2.3  Aviation medicine 1663 Temazepam is a short-​acting benzodiazepine with a short half-​ life. Many people find this drug helpful in promoting sleep and if used for two or three days after travel, can assist in resetting the sleep cycle. Melatonin is secreted by the pineal gland with a rhythm linked to the light/​dark cycle through the suprachiasmatic nucleus. It is ef- fective in inducing sleep when taken at the appropriate stage in the circadian cycle. However, if taken at the wrong stage, it can disrupt the sleep/​wake cycle and destabilize sleep patterns. This limits its usefulness in treating jet lag. There is no simple or single solution for combating the effects of jet lag. The individual has to evolve the strategies to suit his or her particular needs. Traveller’s thrombosis (DVT/​VTE) Long haul travel is associated with prolonged periods of immobility, a recognized risk factor for DVT first described by Virchow in 1856. However, there have been concerns as to whether there are other factors specific to air travel which further increase the risk. In the general population DVT occurs in 1–​3 per 1000 people per year, of which 20% give rise to pulmonary embolism. Increasing age is known to be a strong risk factor, possibly due to decreased mo- bility and reduced muscular tone. The pathogenesis of thrombosis still relies on the basic premise of Virchow who identified circulatory stasis, hypocoagulability, and endothelial injury as the risk factors. Several clinical studies have shown an association between air travel and the risk of DVT, with the risk of VTE in travellers increasing with the distance travelled. A recent case–​control study showed that all modes of travel increased the risk of venous throm- bosis about twofold, with an absolute risk of one thrombosis per 6000 journeys. It has been found that combinations of risk factors synergistically increase the risk of thrombosis. In people with factor V Leiden, the risk of thrombosis after flying was about 14 times increased and in women using oral contraceptives, it was around 20-​fold increased. It has also been shown that the risk rises with the number of flights taken in a short time-​frame, as well as with the duration of the flight. Most of these clots are asymptomatic and disperse naturally. Thus, even though the overall risk of venous thrombosis after air travel is only moderately increased, clear subgroups can be identi- fied in whom the risk is higher. The low humidity of the aircraft cabin does not in itself lead to de- hydration. Excessive alcohol consumption may cause dehydration, but there is no evidence that this is a significant risk factor leading to DVT. Two studies of reduced oxygen partial pressure with non​hypoxic control groups found no evidence of coagulation. There is no evi- dence that hypoxia or the hypobaric environment of an aircraft cabin is a significant risk factor for the development of DVT. Although there is good evidence for the value of aspirin in preventing arterial thromboembolic disease, its role in the preven- tion of venous thromboembolic disease is much less clear. The side effect profile is significant. There is no evidence to support the use of aspirin in preventing the development of DVT during flight. For those travellers at medium to high risk of DVT, there is evi- dence that the use of compression stockings appears to substantially lower the risk of asymptomatic DVT, but it remains unclear as to whether this reduction is clinically significant. One study has shown that for 20–​40% of travellers, the com- mercially available stockings do not fit adequately. It is essential for stockings to be correctly fitted so as to provide adequate compres- sion to stimulate venous return. Although the use of low molecular weight heparin for the preven- tion of DVT in the aviation setting is not supported by direct evi- dence, in a high-​risk traveller consideration may be given to a single prophylactic dose prior to flying. While the relative risk of developing venous thrombosis when flying is significant, the absolute risk of developing symptomatic DVT is very low. The absolute risk of developing a pulmonary em- bolus during or after a flight between the United Kingdom and the east coast of the United States has been calculated as less than one in a million. Medical practitioners need to be circumspect in advising any pre- ventative measures, taking careful account of efficacy and risk pro- file of the preventative method. Passenger fitness to fly Medical clearance is required when: • fitness to travel is in doubt as a result of recent illness, hospitaliza- tion, injury, surgery or instability of an acute or chronic medical condition; • special services are required (e.g. oxygen, stretcher, or authority to carry or use accompanying medical equipment, such as a venti- lator or a nebulizer). Medical clearance is not required for carriage of an invalid passenger outside these categories, although special needs (such as a wheel- chair) must be reported to the airline at the time of booking. It is vital that passengers remember to carry with them any essen- tial medication, and not pack it in their checked baggage. Deterioration on holiday or on a business trip of a previously stable condition, or an accident, can often give rise to the need for medical clearance for the return journey. A stretcher may be re- quired, together with medical support, and this can incur consid- erable cost. It is important for all travellers to have adequate travel insurance. Assessment criteria The passenger’s exercise tolerance can provide a useful guide on fitness to fly; if unable to walk a distance greater than about 50 m without developing dyspnoea, there is a risk that the passenger will be unable to tolerate the relative hypoxia of the pressurized cabin. A good source of guidance is provided by the web sites of the Aerospace Medical Association and the British Thoracic Society. Spread of infectious disease There is no evidence that the pressurized cabin itself makes trans- mission of disease any more likely, and it has been shown that re- circulation of cabin air is not a risk factor for contracting symptoms of upper respiratory tract infection. Data suggest that risk of dis- ease transmission to susceptible passengers, by person-​to-​person droplet spread within the aircraft cabin, is associated with sitting within two rows of a contagious passenger for a flight time of more than 8 hours.

10.2.4 Diving medicine 1664

10.2.4 Diving medicine 1664

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1664 Newly emerging infectious disease SARS is an atypical pneumonia caused by a novel coronavirus which first appeared in the Far East in 2003. Thousands of flights took place to and from what the World Health Organization (WHO) defined as ‘affected areas’ during the outbreak, but transmission occurred only on five flights involving 29 secondary cases (24 cases on one flight). In addition, a further 40 flights were identified, on which one or more probable cases (i.e. symptomatic at the time of travel) travelled, but where no secondary cases developed. Thus, the risk of transmission on board an aircraft is thought to be low. Avian influenza (‘bird flu’) is a highly pathogenic strain A/​H5N1 causing an epidemic among birds in Asia, Europe, and Africa. Human infection is very rare, but serious when it occurs. During 2006, WHO reported a total of 109 cases, of which 79 died. None of the reported cases occurred within Europe, and air travel is not thought to be a risk factor. On the other hand, pandemic influenza causes major morbidity and mortality in humans, with serious economic and social conse- quences. It usually affects a large proportion of the global population due to the absence of immunity, and spreads very rapidly throughout the world. Influenza pandemics occurred in 1918 (‘Spanish flu’), 1957 (‘Asian flu’), and in 1968 (‘Hong Kong flu’), all with high mortality. The WHO strategy for rapid containment of an emerging influ- enza pandemic aims to interrupt disease transmission by isolating and treating infectious individuals, treating and quarantining ex- posed people, and minimizing the exposure of uninfected persons. Modelling suggests that restricting air travel will not prevent the global spread of pandemic influenza, but might delay the spread suf- ficiently to allow countries time to prepare. Guidelines can be ac- cessed from http://​www.who.org or http://​www.cdc.gov. It is important that individuals should not travel on commercial aircraft with a febrile illness. Future issues Aerospace medicine is a subject that is largely understood. There is con- cern among some flight crew about health effects due to oil pyrolysis products in the cabin air. Evidence is conflicting and research is ongoing. The major peer-​reviewed journal in the field is Aerospace Medicine and Human Performance (formerly Aviation, Space and Environmental Medicine), published by the Aerospace Medical Association, which is worth reading to keep up to date with aviation medicine’s evolution and innovation. FURTHER READING Bagshaw M (2014). Air contamination health effects. Aerospace Medical Association, Alexandria VA, USA. http://www.asma.org/ asma/media/asma/Travel-Publications/Air-contamination-health- effects-report-v2-7-Apr2014.pdf Campbell RD, Bagshaw M (2002). Human performance and limita- tions in aviation, 3rd edition. Blackwell Science, Oxford. Coker RK (ed) (2004). Managing passengers with respiratory disease planning air travel: British Thoracic Society recommendations. British Thoracic Society Standards of Care Committee, London. DeHart RL, Millet KC, Murphy J (eds) (1996). Fundamentals of aero- space medicine. Williams and Wilkins, Philadelphia, PA. House of Lords Inquiry (2000). Air travel and health. Her Majesty’s Stationery Office, London. House of Lords Inquiry (2007). Air travel and health: an update. The Stationery Office Ltd, London. Kuipers S, et al. (2007). The absolute risk of venous thrombosis after air travel: a cohort study. PLoS Med, 4, 1508–​14. Gradwell DP, Rainford DJ (eds) (2016). Ernsting’s aviation medicine, 5th edition. CRC Press, London. Rosenberg CA, Pak F (1997). Emergencies in the air: problems, man- agement and prevention. J Emerg Med, 15, 159–​64. Thibeault C (1997). Special committee report:  cabin air quality. Aviation, Space and Environmental Medicine, 68, 80–​2. Zuckerman JN (ed) (2013). Principles and practice of travel medicine, 2nd edition. Wiley-Blackwell, Oxford. 10.2.4  Diving medicine David M. Denison and Mark A. Glover ESSENTIALS Diving remains the principal means of exploring and exploiting shal- lower underwater zones. Immersion and rapid change in pressure with depth cause most problems unique to diving. Effects of pressure on gases and ventilation Gas density, partial pressures, and solubility vary proportionately with ambient pressure. At elevated partial pressure, nitrogen be- comes narcotic, as can other inert gases, and contaminants barely detectable at the surface can become toxic. Hyperoxia irritates the lungs and the central nervous system, sometimes causing gener- alized seizures. A safe gas mixture at depth can become hypoxic as the partial pressure of oxygen decreases during the return to surface. Ventilation is compromised at depth and failure of CO2 elimin- ation increasingly limits activity. Some divers are not distressed by elevated CO2, but this does not protect them from its toxic effects. Clinical problems associated with diving and fitness to dive Immersion hazards include aquatic flora and fauna, water move- ment, impaired visibility and thermal control, and enhanced sound and blast propagation. Immersion predisposes susceptible individ- uals to pulmonary oedema. Aspiration of seawater can cause pul- monary inflammation and systemic manifestations. Water entering the external auditory meati can induce disabling caloric vertigo. The final common pathway in many diving-​related fatalities is drowning. Decompression illness—is caused by ascent from a dive which re- duces ambient pressure. This releases excess dissolved inert gas from tissues, often in the form of bubbles. Alternatively, gas in the airways can expand, rupture the lung, and force its way into the systemic cir- culation via the pulmonary veins. Typical symptoms caused by these bubbles include rash, limb pain and neurological deficit (often motor weakness, numbness and paraesthesiae, also disturbance of higher cerebral function which can impair the diver’s insight). Symptoms de- velop within a few minutes to 24 h of surfacing in most cases and can manifest before reaching the surface in arterial gas embolism arising

10.2.4  Diving medicine 1665 from pulmonary rupture and in decompression from deep, very long duration dives. Management requires exclusion of other diagnoses without delaying first aid treatment of decompression illness with oxygen (as close to 100% as possible) and rehydration, followed by definitive recompression. Intracardiac right–​left shunts, such as pa- tent foramen ovale, predispose to the condition. Extracardiac (pul- monary) shunts can also permit a similar paradoxical embolization of bubbles. Barotrauma—​gas-​filled spaces within, or surrounding, the body will be damaged unless they are flexible enough to accommodate pressure-​mediated changes in volume, or they are ventilated to pre- vent distortion. Divers’ ears, sinuses, lungs, carious teeth, or their masks and suits are vulnerable. Long-​term consequences of diving—​these include aseptic bone in- farcts, impaired higher cerebral function, and hearing loss. Fitness to dive—​unrestricted diving demands a high level of phys- ical and medical fitness. Potential disqualifying factors include con- ditions that might incapacitate, impair, injure, or distract a diver; predispose to decompression illness or barotraumas; or mimic de- compression illness. Introduction Divers are exposed to many hazards while remote from medical care. As a result, diving medicine is largely concerned with preven- tion. It requires a thorough understanding of the diver’s environ- ment and work. Some dives are conducted in dry pressurized chambers, but most involve immersion in fluids such as seawater. Immersion and rapid change in pressure with depth are responsible for most diving prob- lems. Ambient pressure in seawater rises by approximately 100 kPa for every 10-​m descent. Gas densities and partial pressures are pro- portional to ambient pressure. The amount of a chemically inert gas, such as nitrogen or helium that can dissolve in a diver’s body is pro- portional to its partial pressure. A typical shore (Fig. 10.2.4.1) slopes down to between 200 and 300 m at a gradient of about 1 in 50. Diving is largely confined to this continental shelf. Thereafter, the continental slope descends to be- tween 3 and 6 km at a gradient of roughly 1 in 15 to vast flat expanses of soft mud, the abyssal plains, interrupted by occasional mountains and chasms. The deepest point is just over 11 km below the surface. Currents, arising from differences in water temperature and sal- inity, course across the abyssal plains and well up the continental slopes as mineral-​rich streams supplying plant life in sunlit upper zones. Animals concentrate here to feed on these plants or on each other. Eighty per cent (80%) of the ocean biomass lies in the top 200 m, mainly close to the shore. Together, these sites form an area equal to Africa, infinitely more fertile and, as yet, virtually unfarmed. Limitations to diving Currents often exceed swimming speed (Fig. 10.2.4.2a) and may re- strict diving to an hour or two each day during slack water. High waves frequently prevent divers from being launched or recovered safely (Fig.  10.2.4.2b). Tidal currents tunnelled along marine Limit for breathing oxygen Salvage from HMS Edinburgh 1981 Probable limit to practical diving at ambient pressure Depth (km) (a) 0 12 100 Percentage of ocean area Continental shelves Continental slopes Abyssal plains Chasms (b) 0 0 Depth (m) 600 0 Percentage of ocean area Limit for air 10 Fig. 10.2.4.1  (a) A cumulative depth versus area plot of the oceans. (b) A similar plot of the top 600 m, including the continental shelves. Tidal current (knots) Usual Some Maximal swim Comfortable swim Tidal currents Borneo Cook Inlet Texas Louisiana North Sea Persian Gulf California Arabian Sea Incidence (%) 50 Texas Arabian Sea North Sea Month of the year Wave heights above 2 m J F M A M J J A D N O S 5 0 )b ( )a( Fig. 10.2.4.2  (a) A plot of the usual and the not uncommonly seen tidal currents in eight diving sites around the world. (b) A plot of the percentage incidence of waves exceeding a height of 2 m at different times of the year in three of the diving sites.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1666 canyons, and springs of fresh water or falls of cold ocean water, can carry divers in unexpected directions without them being aware. Dawn arrives late and dusk comes early to the sea. Light halves in intensity with every 1 or 2 m of descent, and it is effectively ‘night’ below 80 m. Most recreational diving takes place in clear, shallow, and placid waters. Professional diving occurs throughout the year, alongside or beneath large obstructions, and in turbid waters where finding the task, let alone completing it, can be very demanding. Backscattering often makes artificial illumination ineffective. Underwater, binaural localization of sound is poor. Pressure waves travel almost five times as fast and many times more efficiently through water than air. This increases susceptibility to blast injury. Loss of air conduction raises auditory thresholds by 30–​60 dB. Neoprene foam hoods raise thresholds by a further 30 dB or so. Only the surface waters of tropical seas are warm enough for indi- viduals to remain effective without insulation for any length of time (Fig. 10.2.4.3). Body temperature can be maintained at 37°C with minimal effort in air at 18–​24°C, the zone of thermal neutrality. In water, this zone is high and narrow (35–​35.5°C). Loss of tactile dis- crimination and manual dexterity are major problems when working in cold water. Exercise or excessive insulation in warm water rapidly leads to hyperthermia. Effects of simple immersion Water resists movement, making most tasks more tiring and less efficient than on land (Fig. 10.2.4.4). A  swimmer can sustain about 5 kgf thrust (c. 50 N), enough for propulsion at 1 to 2 knots (1.85–​3.7 km/​h). Full inspiration makes an adult swimmer about 2.5 kgf (c.25 N) positively buoyant, requiring half of maximum thrust to descend. Breathing out to residual volume results in about 2.5 kgf (25 N) negative buoyancy, requiring half of max- imum thrust to ascend. The neutrally buoyant diver can be poised at will but body weight can no longer be used to apply leverage or torque, or to stay in place against a current. Immersion opposes the effect of gravity and displaces blood from distensible vessels in dependent limbs. About 500 ml enters the chest, distending large veins, and right atrium. A larger volume of blood is displaced if immersion is in cold water, due to the per- ipheral vasoconstriction that results. Local stretch receptors in- terpret this central fluid shift as excess circulating volume and promote diuresis, resulting in hypovolaemia on emersion. Since gas is usually delivered to the mouth at ambient pressure, the pres- sure gradient across mouth, thorax, and upper abdomen can in- crease inspiratory or expiratory effort, depending on the diver’s attitude in the water. The displaced blood increases cardiac preload. This predisposes susceptible individuals to ‘immersion pulmonary oedema’ which can occur despite normal cardiac function, and usually after a dive in cold water or involving strenuous exercise. Prevalence is esti- mated at approximately 1% of the recreational diving population and it can recur. In one study, a history of pulmonary oedema after diving in cold water was associated with elevated peripheral vas- cular resistance, especially after a cold challenge, and an increased risk of developing hypertension. An acute increase in preload and afterload is presumed to cause the oedema. Treatment of persistent mild symptoms is with supplemental inspired oxygen. Diuretics and vasodilators have been used to treat more severe cases. It typically resolves within hours although some fatalities have been reported. Physiological studies and case reports suggest that sildenafil has po- tential for prevention of immersion pulmonary oedema. Divers and swimmers with cardiac compromise are also at increased risk of pulmonary oedema. Aspiration of small amounts of seawater can cause ‘saltwater as- piration syndrome’, characterized by productive cough, retrosternal discomfort, and haemoptysis during, or within 2 h of, a dive. Fever, aches, malaise, and even impaired consciousness can develop. The casualty is usually normocapnic, often hypoxic, and sometimes has a leucocytosis. Treatment is rest and supplemental inspired oxygen. Warming often helps extrapulmonary symptoms. Most cases resolve spontaneously within 6 to 24 h. Ergometer load (kpm/min) Water Air Constant load 3 2 1 0 Pedal rate (r.p.m.) (b) Water Air Constant speed 3 2 1 0 (a) 0 40 80 120 0 300 600 M O2(l/min) M O2(l/min) Fig. 10.2.4.4  A comparison of oxygen consumption (M˙ O2) when pedalling a cycle ergometer in air and under water, (a) at a constant speed (60 rev/​min) and (b) at a constant light load. Note the high cost of moving the limbs through water. Most people’s aerobic capacity is about 2.5 litres O2/​min. Temperature (°C) Depth (m) 100 200 Polar summer Temperate Tropical Polar winter 0 300 −5 5 10 15 20 25 30 0 Fig. 10.2.4.3  Variations in sea temperature with site and depth. Note that water temperatures of less than 20°C are too cold for unclothed individuals to stay in for very long.

10.2.4  Diving medicine 1667 If cool water enters one ear canal before the other, then a transient ‘caloric vertigo’ can result. Recurrent immersion can cause problems such as otitis externa. Problems of descent The chest wall can maintain a pressure difference equivalent to 1 or 2 m of water, so gas within the body is virtually at the same pressure as the surrounding sea. The lung of a breath-​hold diver is compressed from total lung capacity to residual volume at 30 m (400 kPa), so they will need half of their aerobic capacity to ascend. Variation in gas volume in clothing and equipment further complicates buoyancy control. Barotrauma of descent Barotrauma is the term used to describe mechanical damage caused by changes in gas volume as pressure varies. Compression will force a diver’s face into an unvented mask. The resulting facial oedema and subconjunctival haemorrhages usually resolve spontaneously. If gas is not added to a dry suit on descent, particularly if it is poorly tailored, it can pinch the skin resulting in linear wheals, commonly distributed around the neck, axillae, and groins. These require no active intervention but should not be con- fused with cutaneous signs of decompression illness (DCI). Severe suit squeeze can limit a diver’s movements. Blood is drawn into the chest vessels to compensate for reduced lung volume, so lung injury occurs only at very great depths in breath-​hold dives. When gas in obstructed sinuses is compressed, sinus walls become oedematous and may bleed. Epistaxis often oc- curs on ascent, as blood or clot is expelled by re-​expanding gas. Middle-​ear barotrauma is the most common problem in diving. Eustachian tube dysfunction or poor ‘ear-​clearing’ technique pre- vents ventilation of the middle ear. Compression of the trapped gas draws the round and oval windows of the inner ear and the eardrum towards the middle ear space. Eardrum perforations can occur. They normally heal spontaneously, but persistent ruptures require sur- gery. Diving should be avoided until the drums have healed. Strenuous Valsalva-​like efforts to ventilate the middle ear raise thoracic pressure. This transmits to the perilymph and can be suf- ficient to rupture the oval or, more typically, the round window. This is known as inner-​ear barotrauma. Immediate or delayed ver- tigo, tinnitus, and hearing loss (usually at high frequencies) ensue. Management is bed rest with the head elevated, avoidance of raised intrathoracic pressure, and consultation with an ear, nose, and throat (ENT) surgeon who might elect to explore the middle ear and to re- pair the rupture surgically. If the symptoms appear after a dive, they can mimic vestibular DCI. If there is any doubt, a diving medicine specialist should be consulted. Barotrauma of descent can also affect a blocked external auditory meatus, gas spaces in carious teeth and under fillings or, in the event of loss of breathing gas pressure, the whole body. Problems while at increased pressure For prolonged dives, compressed gas is delivered to the diver at the same pressure as the surrounding water. This can be via a hose from the surface. A continuous flow through the helmet or face mask is easily engineered but is wasteful of gas. Most divers now use valves that provide gas only on demand. Self-​contained underwater breathing apparatus (scuba) allows the diver to carry an on-​demand supply of gas independent from the surface. Basic configurations of this equipment rarely last for more than 1 h. Rebreather equipment achieves greater endurance by replacing oxygen and removing CO2 from exhaled gas so that it can be recirculated. Inert gas narcosis At raised partial pressure, nitrogen and several other inert gases with high solubility in lipids act like anaesthetics. Effects develop within minutes and reverse rapidly because they depend on passive solu- tion. Air is often breathed down to depths of 50 m, although sophis- ticated tests of cerebral function show impairment starting at 20 m. When deeper than 50 m, effects become more obvious. Narcosis is completely reversed on ascent. Using a less narcotic gas such as helium allows divers to reach the lowermost parts of the contin- ental shelves without narcosis. Divers can complete routine tasks while narcosed if they have repeatedly rehearsed them at increasing depths. Cognition and problem-​solving, however, remain impaired. Hypercapnia Work of breathing and physiological dead space increase as gas be- comes denser at pressure. Hyperventilation is difficult at depth but can still occur. Breathing a dense gas mixture such as air at great depth will cause hypercapnia. Some divers hypoventilate involuntarily and become hypercapnic even in favourable conditions at depths as shallow as 30–​40 m. Although these divers enjoy good gas economy, hypercapnia increases risk of inert gas narcosis, cerebral oxygen tox- icity, and DCI. Use of a less dense mixture, such as oxygen-​in-​helium, reduces this effect. Hypercapnia can also result from equipment mal- function, contaminated gas, or voluntary hypoventilation. Oxygen toxicity Oxygen toxicity is due to complex biochemical interactions and takes time to develop and to reverse. There is a wide range of inter-​ and intraindividual sensitivity. Inspired oxygen partial pressure exceeding 50 kPa is toxic to the lungs. Irritation of lung endothelium and epithelium causes a spreading tracheobronchitis and reduction in lung volumes, flows, and gas transfer. Symptoms appear after about 6 h at partial pres- sures around 79–​89 kPa and after 3 h at around 200 kPa. Advanced pulmonary changes can be irreversible, but symptoms typically di- minish rapidly in 2–​4 h with complete recovery in 1–​3 days. Lung function similarly recovers rapidly, although small decrements can persist for more than a week. Although pulmonary damage continues central nervous system toxicity becomes the primary limit to diving (Fig. 10.2.4.5) as in- spired partial pressure of oxygen rises further. This is unlikely to occur if the inspired partial pressure of oxygen does not exceed 200 kPa when at rest in a dry, comfortable environment but exercise, shivering, hypercapnia, anxiety, immersion, and pyrexia potentiate cerebral oxygen toxicity. As a result, inspired oxygen is usually maintained between maxima of 130 and 160 kPa when in water, depending on work levels. Manifestations of oxygen toxicity include visual disturbances, tinnitus, irritability, and dizziness. A general- ized seizure will usually follow if the oxygen partial pressure is not reduced promptly. Toxicity while immersed can be very dangerous

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1668 but, in the safety of a hyperbaric chamber, partial pressures up to 300 kPa are used for maximum therapeutic effect. Multiple therapeutic hyperbaric exposures can cause myopic lens changes. They have also been reported in divers breathing oxygen for many hours. Most are reversible within 12 weeks of last exposure. High-​pressure nervous syndrome Breathing oxygen-​in-​helium at depths in excess of 100 m causes tremor. Impaired higher function and level of consciousness and, in more severe cases, convulsions, occur at greater depths. It is pre- sumed that extreme pressures directly compress nerve components and affect their function. Depth and rate of pressurization influence the nature and incidence of symptoms. Slower compressions reduce both incidence and severity. Some habituation occurs during pro- longed exposure. Adding a small amount of narcotic gas such as ni- trogen to the breathing mixture reduces some of the manifestations but individual variation in side effects and optimum dose makes this technique unsuitable for commercial diving. Problems of ascent On ascent, partial pressures of gases fall as they expand, and less gas can remain in solution. Hypoxia of ascent Hypoxia will occur if breath is held for long enough in any circum- stance. Ordinarily, the rise in CO2 stimulates the breath-​hold diver to take a breath but predive hyperventilation will delay this, some- times to a dangerous extent. A frankly hypoxic mixture can be sup- plied by mistake during a compressed gas dive. In some rebreathers, a hard-​working diver can consume oxygen faster than the design delivers it to the mixture (dilution hypoxia). Although the oxygen partial pressure might be sufficient to sustain consciousness at depth in any of these situations, ascent will cause it to fall further. Barotrauma of ascent Middle-​ear barotrauma can occur on ascent as well as descent. The mechanism of injury is due to expansion of trapped gas. If one ear ventilates via its Eustachian Tube before the other, uneven vestibular stimulation can cause a transient ‘alternobaric vertigo’. If gas in sinuses cannot escape, it might eventually burst them. Rupture of the ethmoid sinus is rare but feared because of the risk of deep infec- tion. The gut is quite resilient, but in some cases, especially on rapid ascent, ruptures have occurred. Dental barotrauma is also reported and rapid ascents can fracture teeth. In a minority of individuals, the facial nerve or maxillary branch of the trigeminal nerve are exposed to pressure changes in the middle ear and maxillary sinus, respectively. If gas is not vented from the space on ascent, overpressure can impair the nerve’s blood supply, and hence cause a cranial nerve deficit a few minutes after as- cent. This could be misdiagnosed as DCI. Release of pressure brings about resolution, which is usually within minutes and unlikely to exceed 2 h. Compressed gas diving allows the diver to fill the lungs at depth and, therefore, to burst them on ascent unless they are adequately vented. About 9.3 kPa of sustained overpressure (barely 1 m of sea- water) bursts a lung. Ascent at a controlled rate, breathing nor- mally or exhaling, allows the lungs, which have a time constant of emptying close to 0.3 s, to empty and minimizes the risk of rupture. Ascent at too rapid a rate, or while breath-​holding, risks lung rupture. Central tears lead to mediastinal emphysema. Peripheral tears cause pneumothorax. Gas can embolize into the systemic cir- culation, the most significant targets being central nervous system and myocardium. Escaped gas expands as the ascent continues, making matters worse. The victim might lose consciousness or de- velop neurological deficit almost immediately. Otherwise dyspnoea, cough, haemoptysis, voice change, or discomfort in the throat or retrosternal region develop a few minutes later. There might be sur- gical emphysema of the neck and upper chest, increased cardiac dullness or crepitus, and/​or evidence of a pneumothorax. Patients with neurological signs should be recompressed as soon as possible. In the meantime, first aid management is oxygen (as close to 100% as possible) and careful rehydration. Recompression will also reduce the volume of escaped gas if severe pneumomediastinum or subcutaneous emphysema threatens the airway. Decompression illness (DCI) More inert gas dissolves in tissues as dive depth or duration increases. On a safe ascent, this gas comes out of solution, often forming bub- bles in the venous circulation, slowly enough for it to diffuse out harmlessly via the lungs. Tissues are said to be ‘supersaturated’ until all of the excess gas is eliminated. If bubbles are too large or too nu- merous, they can block blood vessels, damage vascular endothelium, and induce ‘foreign body’ reactions. More severe decompression can generate extravascular bubbles within solid tissues, causing distor- tion and even rupturing cells. The term decompression sickness de- scribes disease caused by gas coming out of solution. DCI includes both decompression sickness and disease caused by bubbles escaped from a ruptured lung. The lungs can filter out moderate numbers of venous gas emboli before they reach the systemic circulation. It is possible that this ‘filter’ might be circumvented by right–​left shunts such as patent for- amen ovale or pulmonary arteriovenous anastomosis. One in four healthy people has a patent foramen ovale, but many are only ‘probe-​ patent’ with little chance of shunting. Foramina exceeding 10 mm in maximum dimension are found in less than 1%. Over half of the Convulsions Time (h) PIO2 (atm) Reduction in vital capacity 0 10 20 30 5 4 3 2 1 0 Fig. 10.2.4.5  Commonly observed pulmonary and central nervous O2 toxicity versus time curves related to inspired Po2 (PIo2) (constructed from the data of many workers).

10.2.4  Diving medicine 1669 victims of neurological DCI in one study had medium or large patent foramina. Some have questioned whether a patent foramen ovale is a primary cause of decompression illness and have proposed that, for instance, a significant pulmonary bubble load could trigger vasocon- striction, raise right heart pressures, and open a previously ‘closed’ foramen. Increasing intrathoracic pressure by heavy lifting and Valsalva-​like manoeuvres to clear ears, could have a similar effect. It has been estimated that the odds ratio of serious decompres- sion sickness in divers with a patent foramen ovale is around 2:5. This estimate did not consider foramen ovale size and larger defects are likely to be associated with higher risk. Absolute risk of decom- pression sickness for the whole diving population is low, however, at a little over 2 per 10 000 dives, and primary screening for foramen ovale is not advocated. ‘Undeserved’ DCI justifies screening with bubble-​contrast echocardiography. If a large foramen ovale is found, the usual approach is to advise less provocative diving or percutan- eous closure of the defect. Migraine with aura is associated with an increased risk of patent foramen ovale and is accepted as an indica- tion for screening. The discovery of a large shunt in a diving candi- date who has not suffered from a diving-​related illness would usually be considered to be a significant risk factor for DCI. After breathing an unchanging gas mixture for 24–​48 h at a constant pressure, no more gas accumulates in tissues. Decompression from this ‘saturated’ state takes as long as several days, but it does not lengthen if dive duration is extended. This is the basis of saturation diving. A vast amount of experimental work has been done to determine the safe limits to ‘no-​stop’ diving (Fig. 10.2.4.6) and the depth–​time profiles that must be followed on returning to the surface after longer dives. DCI occurs in about 1% of dives conducted within ‘safe’ schedules, in some 2–​3% of dives at the limits of these schedules, and in many badly conducted dives. Signs of arterial gas embolism following pul- monary rupture will usually present within the first 10 min after sur- facing; 50% of all DCI cases will develop symptoms within 1 h of surfacing and 90% within 6 h. The most common presentation in military and commercial diving is limb pain, commonly of the shoulders or elbows in divers, and of the knees and hips in tunnel workers. Pain might present a few minutes or as much as 24 h after a dive, often as a dull and poorly localized ache of gradual onset. It is not usually made worse by moving the joint, although weight bearing might make knee pain worse. Signs of inflammation are uncommon. Left untreated, the pain will regress and disappear over 2 or 3 days. Recompression commonly improves the pain quickly. Although recreational divers also experience limb pain, neuro- logical symptoms are more likely. Sensory disturbance is common, with numbness and paraesthesiae being frequent manifestations. One fulminant form starts with pain distributed along a thoracol- umbar dermatome (girdle pain) followed by loss of sensation and power in the lower limbs. Involvement of the brain is common and can be subtle. This can impair insight and delay a diver’s decision to seek assistance. Denial is also a frequent feature. Any of the higher functions can be in- volved, including loss of short-​term memory, altered affect, visual disturbance, and loss of consciousness. Inner-​ear DCI can be con- fused with inner-​ear barotrauma. Bubbles do not necessarily respect normal anatomical boundaries, and patchy or multisystem presen- tations are common. Cardiopulmonary symptoms and signs are unusual but, if present, usually indicate a severe case. Cutaneous manifestations range from itching, sometimes with a papular rash, through to patches of skin ‘marbling’ characterized by a reticular cyanosis on a pallid back- ground with an erythematous periphery. Blockage of the lymphatic system by bubbles can cause tender nodes and oedema which typic- ally affects face, neck, or breast. It is not unusual to exhibit several manifestations, or for them to appear at different times and to evolve in different ways. Less specific constitutional symptoms, such as fatigue, malaise, headache, and anorexia can be difficult to distinguish from transient self-​limiting illnesses, but they are usually of no concern unless other manifest- ations are present or they are severe enough to affect function. Divers developing any manifestation of DCI within 24 h of a dive should be managed as if they have the condition unless an alternative diagnosis is more likely. First aid management is supplemental in- spired oxygen (as close to 100% as possible) and rehydration. All but trivial cases of DCI should be recompressed as soon as possible; it is an effective treatment and reduces the size and promotes resorption of existing bubbles before irreversible infarction and oedema occur while preventing formation of new bubbles. High inspired partial pressures of oxygen facilitate removal of excess inert gas, relieve is- chaemia, and reduce oedema, inflammation, and reperfusion injury. The goal is as complete a resolution of symptoms as possible at depth and to avoid recurrence on surfacing. Relapse or residual symptoms require retreatment, so detailed postrecompression examination is necessary. DCI may fail to resolve completely. Miscellaneous related problems A diver who ascends more rapidly than the planned decompres- sion schedule has ‘omitted decompression’. Risk of DCI is increased. Treatment is oxygen and, in more extreme cases, recompression. New exposure to an inert gas when saturated with another can increase overall gas burden if there is a mismatch in the rate at which the gases diffuse into and out of a tissue. This can cause a bubble-​related disease, for which decompression is not the imme- diate provocation and is known as ‘isobaric counterdiffusion’. The site affected depends on the location of the interface between the Time (h) 10 60 50 40 30 20 Above the line: OK to make an ascent without decompression stops Below the line: must make an ascent with decompression stops Depth (m) 0 0 1 2 3 4 5 6 Fig. 10.2.4.6  The ‘no-​stop’ diving curve that determines whether a dive has been shallow and brief enough for the diver to make an ascent to the surface without decompression stops.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1670 two gases. It can cause inner-​ear or skin symptoms in saturation divers. Progression can be halted by altering the gas mixture. It can be treated by recompression and prevented by increasing environ- mental pressure slightly in order to reduce super-​saturation of the tissue before changing gas mixtures. Several hours after breathing high fractions of oxygen, a diver can develop an exudate in the middle ear, yet remain able to ventilate the ears. This might be due to consumption of oxygen causing an insidious volume reduction, or due to a direct toxic effect of oxygen upon the middle-​ear epithelium. The problem resolves spontan- eously within hours. Differential pressure across a restricted aperture can generate large forces with serious, and often fatal, consequences. Examples include the inflow to a culvert or a sudden breach in a pipe con- taining gas at lower pressure than ambient. Potential mechanisms of injury include entrapment, compromised inspiratory effort, pri- mary trauma, and critical damage to equipment. Problems after the dive Autopsies on some asymptomatic divers with no history of DCI have revealed that their brains and spinal cords contain consid- erably more micro-​infarcts than those of non​diving controls. Although the consequences of such damage are considered slight or subclinical, subjective reports of forgetfulness and poor concen- tration have been correlated with diving experience. Subjectively forgetful divers, as a group, performed worse than controls in tests of cognitive function, especially memory. They also had struc- tural differences on brain MRI. A history of welding increases the probability of a diver reporting these problems and the respirable heavy metal particles arising from preparation of surfaces and the welding fumes might be responsible for at least some of these findings. Imaging of long bones of divers and caisson workers show aseptic infarcts (dysbaric osteonecrosis) in a sizeable minority (up to 11%). The incidence is higher in those with a history of overt DCI. Lesions can occur after a single decompression, but their in- cidence rises with age, depth, and diving intensity. Those in the head, neck, or shaft are asymptomatic, but those at juxta-​articular surfaces can be disabling. They are more common in caisson workers than divers, but are even seen in professional breath-​hold divers, such as the Ama of Japan, in whom the dissolved gas burden must be light. The aetiology is unknown, but gas embolism is the favoured explanation. Commercial diving, especially saturation diving, enlarges total lung capacity and forced vital capacity (FVC). This is attributed to training effects of prolonged breathing of compressed gases. The FEV1/​FVC ratio falls, due partly to the rise in FVC, but there are also hints of additional small-​airway damage. Pulmonary capillary blood volume, as judged by carbon monoxide transfer, also falls. This appears to be due to transient episodes of hyperoxia during saturation-​diving procedures, but might also be associated with venous gas emboli released during decompression. The effects are slight but definite and can be cumulative. There are no obvious clin- ical consequences. Mild high-​tone deafness is found in commercial divers and is at- tributed to the noise of gas flows within their helmets. Fitness to dive Fitness assessment balances real and theoretical hazards against em- ployer and physician liability and duty of care, legislation, and the candidate’s livelihood or desire to dive. Some organizations adopt didactic standards. Others use guidelines, which leave room for judgement by the physician and, sometimes, for informed risk to be carried by the candidate. Military and commercial diving is physically demanding and often remote from medical aid. These divers undergo periodic med- ical examinations. Periodicity and extent of examination depend on local regulations. Many recreational divers simply complete regular health declarations, undergoing examination only if a question is answered ‘positively’. Assessments aim to determine whether candidates: • are sufficiently physically fit to rescue a fellow diver, to swim in swift currents and rough waters, and to undertake any related non​diving tasks • are medically fit and have no problems that might incapacitate, impair, distract, predispose to decompression illness or baro- trauma, or otherwise make them a liability to themselves or others • have an acceptable risk of long-​term health consequences from diving • require any restrictions or adjustments We must avoid understatement of the dangers of diving, especially at its most extreme, but must also assess hazard and risk realistically, enabling imaginative solutions for, and greater acceptability of, dis- abled divers who can often dive usefully without jeopardizing health or safety of those involved. An individual who is bodily fit, mentally stable, free of conditions such as epilepsy, obstructive lung disease, ill-​controlled diabetes or asthma, alcohol or drug addiction, and has no history of ruptured eardrums or aural surgery is likely to be medically fit to dive. An acute chest, upper airway, or ear infection would be grounds for temporary unfitness. Diving should be avoided while taking medi- cation that could impair exercise capacity, ability to think clearly, or ability to orientate in space. Medical conditions that can mimic DCI deserve careful assessment. Women are advised not to dive during pregnancy as evidence is suggestive, though not yet con- clusive, that the unborn child is at increased risk of developmental defect. Compromised gas flow or gas exchange could predispose to injury or an inability to cope with the respiratory demands of diving. Risk factors for spontaneous lung rupture such as dis- tortion of lung tissue must be carefully assessed. Cross-​sectional imaging can identify many more bullae than are visible on plain films but, in the absence of a validated quantification of risk of barotraumatic rupture, interpretation in terms of fitness to dive remains subjective. If a candidate runs several kilometres a day, is a good swimmer, was always good at games at school, and has no history of recent respiratory disease, they are very likely to be fit to dive. FEV1 multi- plied by 35 measures the maximal voluntary ventilation, an indi- cator of respiratory fitness. FEV1 should, therefore, be more than 75% predicted. FVC should also be more than 75% predicted be- cause there is good evidence that subjects with a low FVC and, by

10.2.5 Noise 1671

10.2.5 Noise 1671

10.2.5  Noise 1671 inference, lungs that are too small for their bodies, are more liable to lung rupture on rapid ascents. Possible pre-​existing blebs, bullae, or air-​trapping on postincident CT of chest in a small case series of pulmonary barotrauma after free ascents in candidates with unusually large lungs (volumes greater than 120% of predicted) has raised concerns for some diving clin- icians. Until more is known, it would be prudent to consider care- fully all candidates with ‘outlier’ spirometry values, whether too large or small. The FEV1/​FVC ratio has not been found to be especially helpful. What matters most in diving is the time constant of emptying of the full lungs. The peak expiratory flow (PEF)/​FVC ratio is a reasonable measure of this, and PEF should be at least 1.5 times predicted FVC per second. Past or current asthma used to be an absolute contraindication to diving on theoretical grounds of increased risk of lung rupture on fast ascents. Childhood asthma often disappears, however, and very many people with current mild asthma are known to dive fre- quently without ill effect. The British Thoracic Society’s guidelines permit diving in asymptomatic asthma with normal spirometry, negative exercise test, PEF no more than 10% below best values, and requiring no more than regular inhaled anti-​inflammatory agents. Alternative causes of osteonecrotic lesions must be excluded be- fore a dysbaric aetiology is accepted. If lesions are asymptomatic, the medical examiner might consider restriction from experimental or provocative deep diving. Symptomatic lesions will be more re- strictive and, if the articular surface collapses, a joint replacement might be required. MRI has been shown to be more sensitive than traditional plain X-​ray screening. MRI can detect potential lesions within days of a dsybaric insult, whereas X-​ray evidence may take several months to appear. Some lesions visible on MRI, including those in juxta-​articular locations, have remained asymptomatic and have resolved spontaneously. As a result, such lesions require careful serial monitoring before any decision is made that will limit a diver’s employability in the long-​term. Conclusion More effective therapies for DCI are sought, either instead of, or to supplement recompression. Surface oxygen is used as a first aid measure but might have a role as a definitive treatment in selected groups. Intravenous perfluorocarbons and lidocaine have both attracted interest, but more evidence is required. An intervention as simple as an oral non​steroidal anti-​inflammatory drug showed promise in reducing compression requirements in a recent randomized controlled study. Further meticulous data collection will help to identify subgroups who need minimal or more aggressive treatment from the outset. It will also help to clarify issues of safety of drugs, and of medical conditions, while diving. A local hyperbaric facility can advise on management of diving disorders. In Britain, if the nearest suitable facility is not known, the British Hyperbaric Association provides a 24-​h advice line (07831 151523)  for England, Wales, and Northern Ireland and Aberdeen Royal Infirmary (0345 408 6008) provides the service for Scotland. There are many helpful organizations around the world. The Divers’ Alert Network, for instance, has international coverage, and contact details can be obtained from http://​www. diversalertnetwork.org/​contact/​international.asp FURTHER READING Bennett M, Mitchell S, Dominguez A (2003). Adjunctive treatment of decompression illness with a non-​steroidal anti-​inflammatory drug (tenoxicam) reduces compression requirements. Undersea Hyperb Med, 30, 195–​204. Bove AA (2004). Bove and Davis’ diving medicine, 4th edition. W.B. Saunders, Philadelphia, PA. British Thoracic Society Fitness to Dive Group (2003). British Thoracic Society guidelines on respiratory aspects of fitness for diving. Thorax, 58, 3–​13. Brubakk A, Neuman T (eds) (2003). Bennett and Elliott’s physiology and medicine of diving, 5th edition. W.B. Saunders, London. Edmonds C, et al. (eds) (2015). Diving and subaquatic medicine, 5th edition. CRC Press, Boca Raton, FL. Lundgren CEG, Miller JN (eds) (1999). The lung at depth. Dekker, New York, NY. Macdiarmid JI, et  al. (2004). Co-​ordinated investigation into the possible long-​term health effects of diving at work. In: Examination of the long-​term health impact of diving: the ELTHI diving study. HSE Books, HMSO, Norwich. Naval Sea Systems Command, U.S. Department of the Navy (2018). U.S. Navy Diving Manual (Revision 7, Change A, April 2018). https://www.navsea.navy.mil/Portals/103/Documents/SUPSALV/ Diving/US%20DIVING%20MANUAL_REV7_ChangeA-6.6.18. pdf?ver=2018-06-15-102549-030 Slade JB, et  al. (2001). Pulmonary edema associated with scuba diving: case reports and review. Chest, 120, 1686–​94. Smart D, et  al. (2015). Joint position statement on persistent for- amen ovale (PFO) and diving. South Pacific Underwater Medicine Society (SPUMS) and the United Kingdom Sports Diving Medical Committee (UKSDMC). Diving Hyperb Med, 45, 129–​31. UK Health and Safety Executive (2009). Research report RR761—​ Differential pressure hazards in diving. http://​www.hse.gov.uk/​re- search/​rrpdf/​rr761.pdf Wilmshurst P, Bryson P (2000). Relationship between the clinical fea- tures of neurological decompression illness and its causes. Clin Sci, 99, 65–​75. Wilmshurst PT, et  al. (1989). Cold-​induced pulmonary oedema in scuba divers and swimmers and subsequent development of hyper- tension. Lancet, i, 62–​5. 10.2.5  Noise David Koh and Tar-​Ching Aw† ESSENTIALS Noise can affect hearing in the occupational setting but can have other effects where exposures are non​occupational. For clinical † It is with great regret that we report that Tar-Ching Aw died on 18 July, 2017.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1672 purposes, noise is measured in decibels weighted according to the sensitivity of the human ear (dB(A)). Regardless of source, the effects of overexposure to noise are similar. Initially there is a temporary threshold shift, where reversibility of hearing loss is possible with re- moval away from further noise. Noise-​induced hearing loss occurs following prolonged or intense exposure, with poor prospects for improvement of hearing. The classical audiogram for noise-​induced hearing loss shows a 4 kHz dip. Non​auditory effects of prolonged noise exposure include annoyance, sleep disturbance, hypertension, and cardiovascular disease, stress, and impaired cognitive perform- ance. Prevention of noise-​induced hearing loss is by reducing ex- posure to noise at source, minimizing exposure time, using hearing protection, and participating in surveillance. Introduction Noise is any unwanted sound. Excessive noise damages the coch- lear hair cells, breaking and disrupting the cilia, which act as local electromechanical amplifiers. This can result in physical and psychological harm. Exposure The two important characteristics of sound are its intensity and fre- quency. The human audible sound intensity range is 0–​120 decibels. The decibel (dB) scale is logarithmic rather than linear, therefore every increase in sound intensity of 3 dB is equivalent to a doubling of sound intensity. In young adults, the ear sound frequency ranges from 20 Hz to 20 kHz, but its sensitivity is not equal across this range. To mimic the response of the human ear and to allow for the variation in ear sensitivity to different frequencies, sound level meters apply a weighting to the sound intensities, and express the readings as dB(A), that is, decibels weighted by the A scale (as de- fined by international standards). Typical sound levels are 65 dB(A) for normal conversation at a distance of 1 m; 140 dB(A) for a jet aircraft taking off 25 m away; and 160 dB(A) for a rivet gun near the ear. Noisy industries include manufacturing, construction, engineer­ ing, metalworking, motor sports, the military, and entertainment industries. Instantaneous noise levels can be assessed using a noise meter. For cumulative noise exposure, a personal noise dosimeter provides an ‘equivalent noise dose’ by averaging the frequencies and intensities over an 8-​h shift. In the United Kingdom, the Control of Noise at Work Regulations 2005 stipulate an exposure limit of 87 dB(A) averaged over 8 h/​day or 140 dB(A) for any instantaneous impulse noise. Besides occupational exposure, lower intensity community noise (e.g. from airports or urban traffic) is recognized to be associated with adverse health effects. Several studies have shown increased mortality from cardiovascular diseases among people living near noisy airports. The use of personal music players at excessive vol- umes for prolonged periods can also result in significant and haz- ardous noise exposure. There is also evidence that background noise can impair learning in schools. Clinical effects Exposure to loud noise can cause auditory and non​auditory effects. There is wide variation in individual susceptibility. Massive impulse pressures (e.g. from bomb blasts), can cause acute acoustic trauma. There can be a perforation of the tympanic membrane or disruption of the ossicular chain with associated pain and hearing loss, or occa- sionally hyperacusis. An early response to loud noise exposure (e.g. among those who work in noisy environments or those attending loud mu- sical events) is temporarily increased hearing threshold. This temporary threshold shift might be accompanied by tinnitus. The transient dullness of hearing typically lasts up to 24 h, after which hearing thresholds return to normal. With continuing exposure, the magnitude of this temporary sensorineural hearing loss and the recovery time increase until, after months or years, there is a permanent shift in threshold, which might be accompanied by tin- nitus. Hearing damage due to chronic or variable noise exposure might not become apparent until early or middle age, depending on when exposure commenced, and on duration and intensity of the noise. On audiograms, noise-​induced hearing loss is detected as a dip at 4 kHz (Fig. 10.2.5.1). Affected people find it difficult to distin- guish between similar sounds, particularly consonants, in the pres- ence of moderate background noise. With severe hearing loss, the listener may experience ‘loudness recruitment’, a rapid, uncomfort- able increase in sound perceived when intensity increases beyond the already abnormal hearing thresholds. With continued noise exposure, the 4-​kHz dip on audiograms extends to lower frequen- cies and hearing thresholds worsen. This might be combined with presbyacusis (age-​related hearing loss) in later years. Worldwide, about 16% of hearing loss is estimated to be associ- ated with occupational noise exposure. In addition to hearing loss, noise exposure can lead to non-​ auditory effects such as sleep disturbance, hypertension, and car- diovascular disease, stress, and impaired cognitive performance. 125 110 100 90 80 70 60 50 40 30 20 10 0 Frequency (Hz) Hearing level (dB) 250 tfe L th giR 500 8000 4000 1000 2000 Fig. 10.2.5.1  Audiogram typical of noise-​induced hearing loss with a dip at 4 kHz frequency. Red circles, right ear; blue crosses, left ear.

10.2.6 Vibration 1673

10.2.6 Vibration 1673

10.2.6  Vibration 1673 There might be symptoms of annoyance, distraction, fatigue, sleep disturbance, and feelings of isolation. These can combine to re- duce work output and efficiency. Accidents in noisy workplaces have been attributed partly to inability to hear verbal warnings or instructions clearly. Prolonged exposure to community noise might be associated with increased mortality from cardiovascular disease. Diagnosis A diagnosis of noise-​induced hearing loss is established by noise exposure assessment, a history of hearing difficulty that might be accompanied by tinnitus, and an audiogram showing the classical sensorineural 4-​kHz dip. Where abnormalities are detected, it is important to establish the history of occupational, leisure, and community noise exposure, exposure to ototoxic drugs and chemicals, previous ear pathology or surgery, other relevant medical history, and the compliance with use of hearing protection. Management and prevention Otoscopic examination, tuning fork tests, and bone-​conduction audiometry should be carried out to exclude conductive hearing loss. Unusual asymmetrical audiograms with vertigo or unilat- eral tinnitus require otorhinolaryngologist referral to exclude cerebellopontine angle pathology (e.g. acoustic neuroma). People with hearing loss and tinnitus might benefit from using hearing aids, counselling, and a tinnitus-​masking device. The emphasis in dealing with noise-​induced hearing loss must be on prevention. Employers whose workplaces are noisy should es- tablish a hearing conservation programme, with commitment to a robust ‘noise policy’, at the highest level of management. To control noise at source, engineering controls can be used to make equip- ment less noisy, or processes can be redesigned to reduce noise output. Exposure to noise can be limited by separating its source from the workers using soundproof enclosures or shelters. In the United Kingdom a limit for noise exposure has been set at 87 dB(A) averaged over an 8-​h day or 140 dB(A) for any instantaneous im- pulse noise. Hearing protection using ear plugs or ear muffs can reduce noise exposure at the ear by 3 dB(A) to 15 dB(A), but it must be fitted and used correctly to be effective. The diagnosis of workplace noise-​induced hearing loss in a worker should be treated as a sentinel event indicating that other workers were at similar risk and that prompt preventive measures should be implemented. After reduction of noise at source, any re- sidual noise exposure of workers (exceeding 85 dB(A) as defined by the United Kingdom Control of Noise at Work Regulations 2005) warrants health surveillance. This involves symptom review and annual audiometry. In some cases, it may be necessary to consider changing jobs. Clinicians should also advise affected people about benefits available from state compensation schemes. For control of community noise and exposure to personal music players, other preventive measures are required. These include urban planning, restriction of flight paths, and flight timings for air- ports, and innovative health education to reduce listening time and volume of personal music players. FURTHER READING Basner M, et al. (2014). Auditory and non-​auditory effects of noise on health. Lancet, 383, 1325–​32. Hammer MS, Swinburn TK, Neitzel RL (2014). Environmental noise pollution in the United States: developing an effective public health response. Environ Health Perspect, 122, 115. 10.2.6  Vibration Tar-​Ching Aw† ESSENTIALS Various occupations can lead to exposure to vibration, which can be transmitted to the whole body or localized to the hands. The main clinical effect of whole-​body vibration exposure is low back pain. Effects from hand-​transmitted vibration can be (1)  vascular, with manifestations of secondary Raynaud’s phenomenon; (2)  neuro- logical, often presenting as paraesthesia and reduced sensory percep- tion; and (3) musculoskeletal, including reduced grip strength and loss of manual dexterity. In tropical countries, as opposed to temperate climates, the clinical manifestations are predominantly neurological and musculoskeletal, instead of vascular. Management requires ex- clusion of differential diagnoses, and the identification and reduction of exposure to vibration at source. Diagnosis of an index case should prompt further investigation and where possible, modification of the system of work to prevent other cases from occurring. Introduction Vibration is ‘the mechanical oscillation of a surface around its refer- ence point’. Workplace exposure to vibration results in local effects, mainly on the hands when the vibration is transmitted to the upper limbs. The clinical syndrome used to be termed ‘vibration white finger’, highlighting the vascular features. The current term is ‘hand–​ arm vibration syndrome’, reflecting a combination of vascular, sen- sorineural, and musculoskeletal components. When vibration is transmitted to the whole body, systemic effects, mainly low back pain, result.† It is with great regret that we report that Tar-Ching Aw died on 18 July, 2017. Exposure It has been estimated that 5–​25% of the workforce in Europe is exposed to whole-​body vibration, while up to 11% are exposed to hand-​transmitted vibration. Occupational exposure to whole-​ body vibration occurs in helicopter pilots, and in drivers of heavy vehicles (e.g. tractors, forklift trucks, mobile cranes, and buses). † It is with great regret that we report that Tar-Ching Aw died on 18 July, 2017.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1674 The nature of the surface over which land vehicles are driven, as well as the characteristics of the vehicle cabs, contribute to the vibration. Hand–​arm vibration exposure occurs in factory workers involved in fettling, chipping, grinding (Fig. 10.2.6.1), riveting, swaging, and using handheld pneumatic hammers, drills, chisels, and rotary tools. Forestry, agricultural, and wood workers using chain saws, miners drilling rock surfaces, and construction and road workers using drills (Fig. 10.2.6.2), and compactors are also at risk. Clinical effects Whole-​body vibration Exposure to whole-​body vibration has been linked to physiological changes to the cardiovascular, respiratory, and musculoskeletal sys- tems. Clinical effects include headache, motion sickness, sleep and visual disturbances, and urinary and abdominal complaints. However, low back pain is the only effect reliably associated with whole-​body vi- bration. In vocational drivers, low back pain can occur as a result of vibration, poor posture within the vehicle cab, and from additional tasks, such as frequent handling or lifting of heavy loads. Hand–​arm transmitted vibration This causes secondary Raynaud’s phenomenon presenting as prom- inent episodic digital pallor, usually on exposure to cold or following contact with cold objects (Fig. 10.2.6.3). These symptoms often occur in the morning, or following outdoor activity such as fishing or gardening, especially in cold weather. The vascular changes might be accompanied by neurological and musculoskeletal effects that con- tribute to disability. Vascular and sensorineural effects might appear and progress independently. The latent period between initial exposure and development of symptoms is usually 5–​10 years, but can range from between 6 months and 20 years depending on intensity and dur- ation of exposure. The sequence of colour changes in the affected digits starts with pallor, followed by a bluish hue due to cyanosis, and then redness on reversal of the vascular spasm. Reversal can be spontaneous or can follow warming of the hands. Each episode usually lasts several minutes, and there can be several episodes per day or week. Neurological effects include paraesthesia, reduced temperature perception, loss of manual dexterity, and pain. Severe tingling and discomfort often follow rapid warming of the hands. Loss of the ability to distinguish and hold small objects, such as coins, or to button up clothing causes physical and social disability. Musculoskeletal effects are not as well established, although effects such as muscle weakness, bony exostoses, carpal tunnel syndrome, and Dupuytren’s contrac- ture have been associated with exposure to vibration. Diagnosis The criteria for a diagnosis of hand–​arm vibration syndrome are: • evidence of sufficient exposure to vibration; guides to the amount of exposure to vibration from various tools are available (e.g. on the National Institute of Working Life website, http://​umetech.niwl.se) Fig. 10.2.6.1  Exposure to vibration from grinding a metal component against a rotating wheel with an abrasive surface. Fig. 10.2.6.2  Exposure to vibration in a road worker from use of a handheld drill. Fig. 10.2.6.3  Patient with hand–​arm vibration syndrome showing prominent digital pallor.

10.2.6  Vibration 1675 • confirmed episodic pallor of the digits and/​or sensorineural effects • documented latent period between initial exposure to vibration and onset of symptoms 5–​10 years • assessing the likelihood of other causes of Raynaud’s phenomenon or sensory abnormalities The presence of associated musculoskeletal features supports the diagnosis. Physical examination might show callosities on the hands, loss of light touch sensation or two-​point discrimination in the affected digits, and poor grip strength, although there might be no obvious abnormalities, especially in the early stages of the disease. Various clinical and special tests have been used in the evalu- ation of patients with hand–​arm vibration syndrome. These in- clude digital blood pressure measurements, vibrotactile thresholds, sensory aesthesiometry, and cold provocation tests. However, the clinical and occupational history is of greater importance than the results of any of these tests in the diagnosis of hand–​arm vibration syndrome. The differential diagnosis should consider other causes of Raynaud’s phenomenon. The cause might be constitutional or it might be secondary to rheumatoid arthritis, systemic lupus erythematosus, scleroderma, and other autoimmune disorders, cryoglobulinaemia, frostbite, or thoracic outlet syndrome. Use of ergot, clonidine, and β-​blockers, occupational exposure to vinyl chloride monomer, and heavy cigarette smoking are other recog- nized causal factors. Management, treatment, and prevention The severity of hand–​arm vibration syndrome can be staged using the Stockholm Workshop Scale (Table 10.2.6.1). This scale provides separate staging for the vascular and the sensorineural effects. For example, stage ‘2L(3)/​1R(3)’ for the vascular component refers to three digits at stage 2 in the left hand; and three digits at stage 1 in the right hand (i.e. stage/​hand/​number of digits). Further subdivision of stage 2 into early and late effects has been suggested as a basis for deciding on a change of job duties. An alternative scoring system (the Griffin scale) summarizes severity of effect by giving a higher score for affected terminal phalanges and the thumbs. Such scoring systems provide a more objective basis for following up patients with hand–​arm vibration syndrome. Engineering controls can minimize the transmission of vibra- tion from machinery to the body or hands. The Health and Safety Executive in the United Kingdom has defined action values and limit values for exposure to vibration. Action values refer to steps that have to be put in place at these exposure levels, and limit values are legally binding standards that should not be exceeded. Patients with hand–​arm vibration syndrome might be able to con- tinue in the same job following reduction of exposure to vibration. Workers with continuing exposure to hand-​transmitted vibration should be under regular health surveillance. Where the condi- tion is severe and the source of vibration cannot be eliminated, redeployment should be considered. In early cases, redeployment might arrest or reverse the progression of symptoms. In severe cases, the disease might progress regardless of removal from fur- ther exposure to vibration. Advice to the patient includes avoidance or reduction of fur- ther exposure to vibration, use of appropriate gloves, keeping the body and hands warm especially in cold weather, and cessation of cigarette smoking. Vasodilatory drugs such as tolazoline, inositol, and cyclandelate, and calcium channel antagonists such as verapamil and nifedipine, angiotensin-​converting enzyme inhibitors, prosta- glandins, and stanazolol have been tried with varying success. In the United Kingdom, the patient should also be advised about entitle- ment to prescribed diseases benefits, which are awarded depending on severity. This UK scheme is administered by the Department for Work and Pensions through local social security offices. Other coun- tries have their own workmen’s compensation schemes for diseases of occupational origin. A diagnosis of Raynaud’s phenomenon should always include de- tailed inquiry into occupational exposure to vibration. Diagnosis of hand–​arm vibration syndrome should be viewed as a sentinel event warranting further investigation of the workplace to assess whether improvements in work practices can be implemented to prevent the occurrence of other cases. Table 10.2.6.1  The Stockholm Workshop Scale for hand–​arm vibration syndrome A. Vascular component Stage Grade Description 0 No attacks 1 Mild Occasional attacks affecting only tips of one or more fingers 2 Moderate Occasional attacks affecting distal and middle (rarely proximal) phalanges of one or more fingers 3 Severe Frequent attacks affecting all phalanges of most fingers 4 Very severe As in stage 3, with trophic changes in the fingertips B. Sensorineural component Stage Description 0SN Vibration-​exposed but no symptoms 1SN Intermittent numbness with or without tingling 2SN Intermittent or persistent numbness, reduced sensory perception 3SN Intermittent or persistent numbness, reduced tactile discrimination, and/​or manipulative dexterity

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1676 FURTHER READING European Agency for Safety and Health at Work (2008). Workplace exposure to vibration in Europe: an expert review. Office of Official Publications of the European Communities, Luxembourg. Health and Safety Executive (2005). Hand-​arm vibration: the control of vibration at work regulations 2005 & guidance on regulation. HSE Books, Sudbury. Health and Safety Executive (2016). Hand-​arm vibration exposure ­calculator.  http://​www.hse.gov.uk/​vibration/​hav/​vibrationcalc. htm Mason H, Poole K (2004). Clinical testing and management of indi- viduals exposed to hand-​transmitted vibration: an evidence review. Faculty of Occupational Medicine, London.

10.3 Environment and health 1677

10.3 Environment and health 1677

10.3.1 Air pollution and health 1677

10.3.1 Air pollution and health 1677

10.3 Environment and health CONTENTS 10.3.1 Air pollution and health  1677 Om P. Kurmi, Kin Bong Hubert Lam, and Jon G. Ayres 10.3.2 Heat  1687 Michael A. Stroud 10.3.3 Cold  1689 Michael A. Stroud 10.3.4 Drowning  1691 Peter J. Fenner 10.3.5 Lightning and electrical injuries  1696 Chris Andrews 10.3.6 Diseases of high terrestrial altitudes  1701 Tyler Albert, Eric R. Swenson, Andrew J. Pollard, Buddha Basnyat,
and David R. Murdoch 10.3.7 Radiation  1709 Jill Meara 10.3.8 Disasters: Earthquakes, hurricanes, floods,
and volcanic eruptions  1713 Peter J. Baxter 10.3.9 Bioterrorism  1718 Manfred S. Green 10.3.1  Air pollution and health Om P. Kurmi, Kin Bong Hubert Lam,
and Jon G. Ayres ESSENTIALS The term ‘occupational and environmental health’ includes any act of emission of any substance, likely to be hazardous in nature, which is either not originally present or is present in a higher concentration than normal in the natural atmosphere. Most air pollutants are gen- erated from human activities (e.g. energy, transportation, industry, agriculture), but natural events in the living (e.g. methane emissions in wetlands) and non​living environment (e.g. volcanic eruptions) also contribute to atmospheric air pollution, although their relative im- portance has declined since the Industrial Revolution and the advent of modern fossil fuel-​based economies. Pollutants may be classified as (1) primary (emitted directly into the atmosphere) or secondary (formed in the air through chemical reactions with other pollutants and gases); (2) indoor or outdoor; (3) gaseous or particulate. The main problem in determining health effects of individual air pollutants is that in real life they are never experienced in isolation, but many large studies have shown a link between air pollution and ad- verse cardiovascular and respiratory outcomes. Long-​term exposure to air pollution, especially fine particulate, could also adversely af- fect other outcomes such as diabetes, impaired cognitive function, preterm birth, and lower birthweight. Air pollution—​even at low levels—​possesses significant health risk, particularly among sensitive individuals. The latest Global Burden of Disease Study 2013 estimated 2.9 million premature deaths could be attributed to ambient air pol- lution, of which 88% occurred in low and middle-​income countries. Although by definition all air pollutants are hazardous, only the major ones (e.g. particulate matter, sulphur dioxide, nitrogen dioxide and ozone) in the atmosphere are monitored and regulated by legis- lation in most countries. Ambient air pollution in high-​income coun- tries has reduced significantly in the last few decades, but further reduction in pollutant levels will inevitably require very high cost. As low-​ and middle-​income countries are experiencing fast economic growth, air pollution is likely to continue to be a significant con- tributor of global morbidity and mortality in the future. Introduction Concerns about air quality and its effects on health have been ex- pressed for many hundreds of years. The lime burners of the 1300s were blamed for polluting the air and John Evelyn in his famous diary railed against the polluted air of the City of London. The do- mestic pollution of those years was later compounded by the ad- vent of the Industrial Revolution in the 18th century, when the sulphurous emissions from factories were regarded as a necessary evil meaning, as it did, jobs and payment for the working popula- tion. This grudging acceptance remained until the London smog of 4–​9 December 1952 when an increase in coal being burnt in do- mestic grates in a cold week together with a temperature inversion with a ‘blocking’ zone of high pressure accumulated pollutants at street level. Within a week, 4000 excess deaths occurred during

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1678 that six-​day period. In fact, a re-​assessment of the data suggested a ­further 13 500 deaths in the weeks and months that followed. As a direct consequence, the Clean Air Act of 1956 was passed, which re- sulted in a marked reduction in particulate pollution and to a lesser reduction in sulphur dioxide levels, such that in the late 1970s there was a general belief that air pollution and its potential health prob- lems had been conquered. However, in the 1980s a different type of air pollution—​nitrogen dioxide, fine particulates, and hydrocar- bons arising from vehicle emissions—​was realized. These pollutants are not only harmful in their own right but also act as a substrate for the formation of ozone, the combination of all these pollutants comprising the photochemical pollution so characteristic of Los Angeles, Athens, and Mexico City, and increasingly commonly in East and South Asia. As a result, urban dwellers worldwide are now facing long-​term exposures to air pollution. As we spend most of our lives indoors, indoor air quality is clearly equally important. In high-​income countries as well as in urban areas of low-​ and middle-​income countries (LMICs) most indoor levels of pollutants are driven by outdoor levels. In such settings tobacco smoke becomes the most important source of indoor pol- lutant, exposure to which has been shown extensively to impart ad- verse health effects. However, in the rural and less developed parts of the LMICs combustion products of solid fuels, derived from bio- mass (e.g. wood, animal dung, agricultural residues) and coal for heating and cooking purposes, dominate the scene. It has been esti- mated that nearly one-​third of the world’s population are relying on solid fuels, most of whom use traditional unventilated stoves of low efficiency, producing large quantities of pollutants, particularly par- ticulate matter and carbon monoxide. While there has been ample evidence to suggest the global burden of mortality and morbidity associated with household air pollution is substantial and compar- able to ambient air pollution, it has not received adequate attention until recently. At present the extent of the health effect(s) of these pollutants, alone or in combination, are a continuing matter for research and debate. Although acute effects have often been deemed to be small, as more studies of health effects of long-​term exposure are under- taken, the true picture of the burden due to air pollution is begin- ning to unravel. Sources and types of air pollutants Very often the term ‘air pollution’ is understood to be an act of emis- sion of any substance, likely to be hazardous in nature, which is either not originally present or is present in a higher concentration than normal in the natural atmosphere. While it is true that most air pol- lutants are generated from human activities (e.g. energy, transporta- tion, industry, agriculture), natural events in the living (e.g. methane emissions in wetlands), and non​living environment (e.g. volcanic eruptions, lightning strikes, natural radioactive decay) also con- tribute to atmospheric air pollution, but their relative importance has declined since the Industrial Revolution and the advent of modern fossil fuel-​based economy, at least in high-​income countries. Given the loose definition, it is not surprising that a long list of substances could be defined as air pollutants. They can be classified by number of ways, and Table 10.3.1.1 presents three commonly used criteria with further subclassifications. In the urban environment, the concentration of airborne pollu- tants is usually higher outdoors compared to indoors because of the high volume of vehicular emissions. However, improved thermal insulation and the use of air conditioning systems in modern dwell- ings may also facilitate the accumulation of indoor pollutants, par- ticularly when people smoke in their homes. On the other hand, in LMICs, especially in the rural areas, indoor pollutant level can be many times higher than the outdoor environment because of the use of solid fuels for cooking and heating, a common phenomenon in unventilated dwellings. Pollutants can also be grouped by their physical properties, often by their state and size. Under such a classification scheme particu- late matter (PM), sometimes known as total suspended particulate, is a mixture of solid particles and aerosols suspended in the air with Table 10.3.1.1  Major classifications of air pollutants A. Primary–​secondary pollutants a. Primary: Pollutants emitted directly into the atmosphere (e.g. SO2, some NOx species, CO, PM) b. Secondary: Pollutants that form in the air as a result of chemical reactions with other pollutants and gases (e.g. O3, NOx, and some PM) B. Indoor–​outdoor pollutants a. Indoor pollutants i.  Sources: Combustion, cooking, heating, smoking, particle resuspension, building materials, air conditioning, consumer products, biological agents ii. Products: Combustion products (e.g. tobacco and solid fuel), CO, CO2, sVOC (e.g. aldehydes, alcohols, alkanes, and ketones), microbial agents and organic dusts, radon, manmade vitreous fibres b. Outdoor pollutants i.  Sources: Industrial, commercial, transportation, agricultural, natural, transboundary ii. Products: SO2, O3, NOx, CO, PM, VOC, sVOC C. Gaseous–​particulate pollutants a. Gaseous: SO2, NOx, ozone, CO, VOC, sVOC (e.g. PAHs, dioxins, benzene, aldehydes, 1,3-​butadiene) b. Particulate: Coarse or thoracic particles (aerodynamic diameter 2.5–​10 µm; regulatory standard = PM10), fine PM, or respirable (0.1–​2.5 µm; regulatory standard = PM2.5); ultrafine PM (<0.1 µm; not regulated) SO2, sulphur dioxide; NOx, oxides of nitrogen; CO, carbon monoxide; CO2, carbon dioxide; sVOC, semi-​volatile organic compounds; PM, particulate matter; PAHs, polycyclic aromatic hydrocarbons. Pollutants such as particulate matter (PM), sulphur dioxide (SO2), and carbon monoxide (CO) are examples of primary pollutants which, as combustion product of fossil fuels, are released into the atmosphere directly from sources. Secondary pollutants are formed by primary pollutants in the atmosphere. One such example is ground (tropospheric) level ozone (as opposed to stratospheric ozone, which is naturally occurring), formed by chemical reactions of oxides of nitrogen and hydrocarbons in the presence of sunlight. Ozone is one of the main constituents of the photochemical smog, which is typical in cities with many motor vehicles on warm sunny days. From Bernstein JA, et al. (2004). Health effects of air pollution. Journal of Allergy and Clinical Immunology, 115(5), 1116–​23, with permission from Elsevier.

10.3.1  Air pollution and health 1679 variations in size, composition, and origin. The major components of PM are inorganic compounds such as sulphates, nitrates, am- monia, black carbon, sea salt, and mineral dust, and organic sub- stances including pollen and mould. Coarse or ‘inhalable particles’ have an aerodynamic diameter of ≤10 µm (known as PM10), are chiefly derived from attrition of larger particles particularly from abraded soil, road dust (from brake and tyre), construction debris, or also from aggregation of smaller combustion particles which can be inhaled to the lower respiratory system. Fine or ‘respirable par- ticles’ (aerodynamic diameter of ≤2.5 µm; PM2.5) are a subset of the PM10 fraction, which can penetrate deep into the lungs through the bronchioles to the alveoli where gas-​exchange takes place. Ultrafine particles are less than 100 nm in size, and may even penetrate into the systemic circulation. Measurement of a health effect The main problem in determining health effects of individual air pollutants is that in real life they are never experienced in isolation. There are multiple potential interactions, not only with other air pollutants, but with other factors such as weather conditions (as in the 1952 London smog), levels of airborne allergen, the presence of a respiratory tract infection, exercise, diet, socioeconomic status, and actively or passively inhaled cigarette smoke. Some health ef- fects may also be attributed to multiple risk factors at a time and hence making determination of specific, pollutant-​induced health effects difficult. There are three main ways whereby a health effect can be assessed: animal studies, challenge studies in humans, and by epidemiological studies (including cross-​sectional, time series, case-​crossover, and cohort studies). Each has their problems but when determining na- tional ambient air quality standards all available types of data may be taken into consideration. When determining national ambient air quality standards, the aim is to try and define a no observable effect level (NOEL), that is, the level of a specific pollutant where no health effect can be demonstrated. The rough rule of thumb that has been used is to take as the standard a level one-​tenth of the NOEL obtained from human chamber/​challenge studies or one-​hundredth of the NOEL determined by animal studies. Risk assessment is difficult and very largely arbitrary. Although individual risks are relatively easily measurable in the workplace, when considering am- bient airborne pollution individual risk and public health load has not been very well characterized to date. This is partly because the health effects of outdoor air pollution are considerably less than the risks of active cigarette smoking or the risks of other well-​recognized aetiological factors (e.g. high blood pressure). Moreover, potential sources of confounding, primarily in terms of socioeconomic status and cigarette smoking might have affected the interpretation of the health effects of air pollution. Ambient (outdoor) air pollution and health effects Large-​scale multicity and indeed multinational studies, such as the early Harvard Six Cities, American Cancer Society cohort, and Adventist Health Air Pollution (AHSMOG) studies in the United States, and the more recent NMMAPS (National Morbidity Mortality Air Pollution Study), APHENA (Air pollution and health: a European and North American approach), PAPA (Public health and Air Pollution in Asia), and ESCAPE (European Study of Cohorts for Air Pollution Effects) studies have provided evidence showing the link between air pollution and adverse cardiovascular and respiratory outcomes (including ischaemic heart disease, stroke, heart failure, asthma, and chronic bronchitis), particularly in terms of excess mortality associated with short-​ and long-​term exposures. Recent evidence suggests long-​term exposure to air pollution, es- pecially fine particulate, could also adversely affect other outcomes such as diabetes, impaired cognitive function, preterm birth, and lower birthweight. Indeed, the latest Global Burden of Disease Study 2017 estimated 4.2 million premature deaths could be attributed to ambient air pol- lution in 2015, of which more than 80% occurred in LMICs; the ma- jority of them in the World Health Organization (WHO) Western Pacific and Southeast Asia regions. However, as most of the studies investigating health effects to date have been done in high-​income countries where the current air pollutant level is 15–​20-​fold lower than that in LMICs, it raises the question whether concentration-​response functions, primarily derived from high-​income countries, could be extrapolated to estimate the burden in LMICs. More importantly, it is extremely difficult to ascertain and quantify the level and duration of exposure to airborne pollutant at both population and individual levels with high accuracy and precision. Coexposure at the same time to another pollutant will likely affect any physiological response as will the presence of cofactors such as age, individual susceptibility (pres- ence of comorbidities, concurrent respiratory tract infection, degree of bronchial responsiveness), smoking, physical activity (which in- creases effective lung dose due to enhanced ventilation) or diet. Such variations within and between individuals have added the complexity in interpreting the findings from previous literature. Although health effects of air pollution could be at times difficult to assess and to ascertain from observational studies, interventions (often associated with pollution abatement legislations) have pro- vided indirect evidence to support the link between air pollution and adverse health outcomes. A ban on ‘smoky’ coal sale and burning in Ireland, first introduced in Dublin in 1990 and subsequently ex- tended to cities and towns nationwide, has resulted in substantial reductions in cardiorespiratory hospital admissions and respiratory mortality. Associations between improved air quality and reduc- tion in mortality have also been reported following the reduction of sulphur content of fuel in Hong Kong in 1990. On the other hand, conflicting findings have been reported in studies investigating the effect of low emission/​congestion charging zones on pollutant levels. Nevertheless, a study in Rome estimated that residents living near main roads have gained 3.4 days per person following the intro- duction of low emission zones in 2006. Transient changes in air quality have also been demonstrated to have measurable population health benefits. During the 2008 Olympic and Paralympic Games (July–​September) the Chinese government implemented a series of aggressive pollution control measures in Beijing and surrounding areas, including restriction or complete shutdown of industrial and construction operations and an alternative-​day driving scheme. As a result, large reductions in gaseous and particulate pollutant levels were achieved during this period (-​13% to -​60%) compared with the pre-​Olympic period. During this time, there was a significant reduc- tion in levels of inflammatory and thrombotic biomarkers in healthy

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1680 young adults living in Beijing, but reverting to pre-​Olympic period levels as the control measures were relaxed after the Paralympic Games. A reduction in cardiovascular mortality was also observed in the same period. However, it is important to note that such major athletic events inevitably introduce disturbances to the normal city life, such that changes in health events might not have been entirely explained by changes in air quality. In the following paragraphs health effects of the major air pollu- tants will be described in some detail. Sulphur dioxide (SO2) SO2 comes from the combustion of sulphur-​containing fossil fuel, usually in power stations and ocean-​going vessels. Emissions in high-​income countries have been reduced drastically thanks to the availability of low-​sulphur fuel and improvement in flue-​gas desul- phurization. However, in countries with high reliance on coal, SO2 continues to be a major pollutant and precursor to acid rain. Controlled human challenge studies in both asthmatic and healthy individuals have reported that around 5 minutes’ exposure to inhaled SO2 induces rapid onset bronchoconstriction resulting in decrease in FEV1 or increase in airway resistance within 2 min- utes of exposure. SO2 causes tracheitis and reduces ciliary func- tion in vitro, while in vivo it can cause bronchoconstriction in exercising asthmatics at levels experienced in urban areas today. Non​asthmatic individuals, particularly those who are atopic, will also develop bronchoconstriction with SO2 on exercise but only at significantly with higher exposures (≥1300 µg/​m3). This sen- sitivity to the effects of SO2 could have a genetic component as a recent study among asthmatic patients spirometrically responsive (>12% decrease in forced expiratory volume in 1 s (FEV1)) to ex- posure to SO2 reported significant association between the TNF​α promotor polymorphism and asthma. It has been estimated that 25% of asthmatics are sensitive to less than 0.5 ppm of SO2 while exercising. Although among healthy adults the effect is usually short lived and there is a spontaneous recovery within 30 minutes of initial ex- posure, in asthmatic patients, recovery could take up to 4 hours. On the other hand, a meta-​analysis of time series studies (focusing on short-​term effect) has found positive association in terms of hospital admission as well as mortality for both cardiovascular and respira- tory health outcomes. Many studies worldwide have shown chronic effects of sulphur dioxide on absenteeism, respiratory symptoms, and on prevalence and mortality of chronic obstructive pulmonary disease (COPD). Particulate matter (PM10 and PM2.5) As described previously, PM comprises a wide range of particles or aerosols of different sizes and composition. Recent studies have shown PM to have specific health effects independent of SO2, which is often emitted simultaneously and the individual effects of which were previously thought to be difficult to disentangle. In high-​ income countries traffic emission, particularly diesel exhaust, is the main source of PM. It has been demonstrated that particle size is negatively related to the extent of adverse health effect, probably ex- plained by the high surface area to volume ratio in smaller particles, which encourages adsorption of more toxic compounds at a given mass. The small size also enables the particles to enter the systemic circulation and into tissues and organs. Human challenge and animal model studies have suggested that exposure to diesel exhaust has potent inflammatory effects involving lung epithelial cells and alveolar macrophages. Healthy volunteers ex- posed to 300 µg/​m3 diesel exhaust or particulates for one hour in a controlled chamber have been shown to have increased neutrophil counts in sputum and bronchial biopsy specimens and increases in IL-​ 6, IL-​8, and growth-​related oncogene α-​levels, with minimal change in lung function. In patients with mild asthma, exposure to diesel can increase airway hyperresponsiveness to methacholine and airway re- sistance. Longer-​term (3 months) of higher dose of diesel exhaust in animals increased airway hyperresponsiveness significantly but the ef- fect disappeared when the animals were relocated to cleaner environ- ment for further 3 months. On the other hand, the mechanism leading to adverse cardiovascular health outcome is relatively unclear. It has been reported that PM exposure is associated with up-​regulation of fibrinogen and platelet levels, while sequestration of red blood cells in the lung could increase the risk of cardiac arrhythmia. Epidemiological studies, particularly multicity studies, have provided substantial evidence to suggest short-​ and long-​term ex- posure to PM10 and PM2.5 results in increase in cardiorespiratory emergency room visits, hospital admissions and mortality (mainly ischaemic heart disease and stroke). Pooled analysis from 110 time series studies reported for a 10 µg/​m3 increase in PM2.5 concentra- tion there is an 1.5% increase in daily respiratory and 0.8% increase in cardiovascular mortality, respectively. With regard to the effects of long-​term exposure, the multicentre ESCAPE used data from 22 European cohorts and found 7% increase in all natural cause mor- tality per 5 µg/​m3 increase in PM2.5. There is also support for links between long-​term PM2.5 exposure and other endpoints, including adverse birth outcomes (low birth weight, preterm birth, and small for gestational age), childhood respiratory disease, and impaired neurodevelopment and cognitive function. Oxides of nitrogen (NOx) The major oxides of nitrogen, nitric oxide (NO), nitrous oxide (N2O), and nitrogen dioxide (NO2) are regarded together as NOx, but it is NO2 that receives most attention as the other two have no known deleterious health effects. While one-​third of ambient NO2 comes from diesel engines, the highest exposures to ambient NO2 for most individuals, however, are seen indoors in gas-​fired kitchens, where levels may reach 900 µg/​m3 for short periods. A more usual indoor level is 190 µg/​m3 (100 ppb) compared to an outdoor urban level of around 60 µg/​m3. The effects of ambient levels of NO2 is a matter of much debate as some question whether NO2 is merely an indicator of other pollu- tants (particularly PM). Oxides of nitrogen can react with ammonia, moisture, and other compounds to form small particles and is also a precursor to photochemical smog. An early study showed that 190 µg/​m3 NO2 increased airway responsiveness in exercising asth- matics, but subsequent studies, even up to exposures of 700 µg/​m3, showed no effect. In a controlled chamber study, healthy smokers exposed to high levels of NO2 had induced inflammatory response in the airway that was characterized by neutrophil influx and re- duced lymphocytes count. NO2 at very high levels are often only present in occupational settings, and can cause acute pulmonary oedema. Cross-​sectional and short-​term longitudinal epidemio- logical studies have shown NO2 to be associated with airway in- flammation in healthy subjects and increase in respiratory symptoms

10.3.1  Air pollution and health 1681 in asthma patients. Susceptible individuals living close to main roads are likely to have worsened respiratory symptoms leading to in- creased hospital admissions and premature mortality. Similar to PM, there is also evidence that long-​term exposure to NO2 is linked to increased cardiovascular emergency admissions and mortality, mainly due to stroke and myocardial infarction. A  pooled ana- lysis from 110 time series studies showed per unit µg/​m3 increase in NOx would increase the risk of all-​cause, cardiovascular and respiratory mortality by 0.17% (0.12–​0.23%), 0.11% (–​0.12–​0.35%), and 0.15% (–​0.29–​0.59), respectively. Ozone (O3) Ozone is a highly reactive molecule which is formed by the action of ultraviolet light on NOx and hydrocarbon fragments emitted by ve- hicles. It is formed mostly during the summer months and tends to build up over a prolonged warm spell. Levels are often higher in rural areas downwind from cities as urban nitric oxide neutralizes ozone while the formation of ozone takes a little time as polluted air is taken downwind. It is also probable that ozone enhances the formation of aerosol strong acid, thus significantly affecting the ambient pollutant mix. There is a diurnal variation with peak levels being achieved in late afternoon. In winter, ozone levels are almost unmeasurably low but in summer can exceed hourly values of 200 µg/​m3. Ozone causes inflammatory changes (mainly neutrophilic) in the bronchial mucosa at levels as low as 160 µg/​m3, at which level changes in forced vital capacity and FEV1 can be detected after an exposure of 1 hour and can persist for up to 24 hours. Individual responses to chal- lenge with ozone vary widely between normal and asthmatic individ- uals, but asthma patients appear not to be more sensitive to the effects than normal subjects. It is possible that women are more susceptible. Repetitive exposures seem to produce a lesser response suggesting a latency effect which is difficult to understand in view of the known inflammatory effect of this gas. Summer peaks of ozone have been associated with peaks in hospital admissions for asthma in several areas of North America. Ozone at levels which do not cause an effect on airway function can cause bronchoconstriction when patients are pre-​exposed to usual ambient SO2 which suggests a potentiating effect of air pollutants, which has great logical appeal. Ozone at 240 µg/​m3 has also been shown to enhance the bronchoconstrictor response to inhaled allergen which again reinforces the very likely possibility that current air pollutants exert their effect in a permissive way. The large multicentre APHENA study has provided data on short-​ term effects of ozone. It found statistically significant associations for 1-​hour ozone and all-​cause mortality and cardiovascular mortality. Recent cohort analyses suggest an effect of long-​term exposure to ozone on mortality, at least for respiratory or cardiorespiratory mor- tality, especially in people with potential predisposing conditions. Polycyclic aromatic hydrocarbons (PAHs) These form part of the total hydrocarbons, a wide range of com- pounds most of which have no known human health effect. When considering total hydrocarbons methane is usually excluded, the re- maining hydrocarbons largely comprising alkanes. Because of their association with PM, total hydrocarbons are usually expressed as parts per billion of carbon. PAHs are partitioned between the par- ticulate and gas phase, the lower molecular weight molecules being in the gas phase. The carcinogenicity of PAHs increases with the in- crease of their molecular weight, but reduction in acute toxicity. The main sources of airborne PAHs in the outdoor environment are from evaporation of solvents and fuels and from combustion of car- bonaceous materials at high temperature, predominantly in emissions from motor vehicles. Of the different PAH species, benzene and, to a lesser extent, benzo [a]‌anthracene, benzo[a]pyrene and dibenz[ah] anthracene are known to be carcinogenic. With introduction of stricter guidelines, current ambient B[a]P concentration has reduced from around 100 ng/​m3 in 1945 down to 0.3 ng/​m3 in high-​income countries. However, the major exposure to PAHs in human is tobacco smoking and burning of fuels, particularly in an indoor environment. Benzene is present in significant quantities in cigarette smoke and, on average, cigarette smokers take in about 2 mg/​day compared with less than 0.2 mg/​day for most non​smokers, although passively exposed non​smokers will be exposed to approximately 60% more benzene than non​passive non​smokers. Short-​term exposure to PAHs causes impaired lung function in asthmatics and thrombotic effects in people with major coronary heart diseases. Individuals in occupational set- tings exposed to air pollutants with a mixture of PAHs for long dur- ations have reported increases in incidence of skin, lung, bladder, and gastrointestinal cancers, as well as a raised risk of cell damage via gene mutation and cardiopulmonary mortality. Carbon monoxide (CO) The major source of CO for cigarette smokers is, and will remain, cig- arette smoke. For non​smokers, including children, vehicle emissions passively inhaled cigarette smoke, and indoor cooking are the main contributors. Exposures are far greater in tunnels, car parks, garages, and in dense, slow moving traffic. Kerbside levels in towns are of the order of 20 µg/​m3 which, with chronic exposure, would produce a carboxyhaemoglobin level of around 3%. WHO guidelines aim to keep blood levels of carboxyhaemoglobin to less than 2% in non-​ smokers. In non​smokers levels of carboxyhaemoglobin rarely exceed 3%. CO exerts its toxic effect not just by the formation of carboxy- haemoglobin but also because it shifts the oxygen dissociation curve to the left. As a result, blood levels of around 3.6% will reduce the time to onset of angina on exercise. The fetuses of smoking mothers might have carboxyhaemoglobin levels up to 2.5 times that of the mother. How the fetus is able to concentrate CO and what the health effects might be to the fetus are unknown. Recent studies have reported that exposure to CO is likely to reduce the maximal exercise capacity in healthy young individuals and hence reduces the time to angina and, in some cases, the time to ST-​segment depression in people with car- diovascular disease, albeit at a concentration that is lower than that needed to reduce exercise capacity in healthy individuals. Indoor air pollution and health effects While common air pollutants are often released to the atmosphere from traffic, power stations, and factories, this is not to say air pollu- tion exists only outdoors. It could easily be forgotten that even in in- door setting, where people could spend up to 90% of their time, there are outdoor–​indoor transfers, as well as indoor emission sources, and because they are confined environments concentration of pollutants could be much higher indoors than outdoors. Indoor air quality can be considered with respect to the domestic and occupational set- tings. In high-​income countries, domestic exposures are predomin- antly PM, CO, NO2, PAHs, and volatile organic compounds due to

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1682 combustion (tobacco smoke and fuel for cooking/​heating), domestic activities (cleaning and cooking), and building and furnishing ma- terials (adhesives and radon) and biological pollutants (allergens and moulds). From the occupational point of view, specific expos- ures which lead to occupational asthma are dealt with elsewhere but non​specific problems, such as the ‘sick building syndrome’ will be covered next in brief. In LMICs, however, poor indoor air quality is related to the burning of solid fuels using inefficient stoves, which produce very high concentrations of PM, CO, and other combustion products, often many folds higher than what is measured in the am- bient atmosphere. This household air pollution affects one-​third of the world’s population and is therefore a major cause of morbidity and mortality globally. It is estimated that in 2013, 2.9 million pre- mature deaths and 81.1 million disability adjusted life years (DALYs) can be attributed to household air pollution. India and China alone accounted for 60% of the total premature deaths. Control of indoor pollution is contentious. Most attempts are aimed at increasing room ventilation rates, which have shown sig- nificant improvements in odour levels and occupant satisfaction while air filtration, although attractive, has yet to be shown to be as effective. In LMICs where household air pollution dominates, a wide range of interventions, including use of cleaner alternative fuels (e.g. biogas), replacement of traditional cook stoves with more efficient ones, improving ventilation (e.g. retrofitting chimneys), and change in use behaviour (e.g. keeping children away from smoke), have been implemented with varying levels of success. Ascertainment and quantification of exposure indoors is more difficult than outdoors because there exist very large variations in pollutant levels across different microenvironments (e.g. rooms within the same dwelling) and exposure between and within indi- viduals, determined by several factors including proximity to emis- sion source, permeability and ventilation of the microenvironment, and time-​activity pattern of individuals. Consequently, personal ex- posures estimated from detailed diary of indoor and outdoor time-​ activity patterns will give a better idea of time exposure to specific pollutant(s) when trying to estimate a health risk. Selected indoor exposures in 
high-​income countries Tobacco smoke The most important indoor pollutant is tobacco smoke from the smouldering tobacco and exhaled by smokers. Known as environ- mental tobacco smoke or second-​hand smoke, it is a mixture of toxic and carcinogenic chemicals and particulates. A smoker of 20 cigar- ettes a day contributes about 20 µg/​m3 to 24-​h indoor particulate concentrations, and in a house containing several heavy smokers the 24-​h ambient air quality standard of 50 µg/​m3 for PM10 can easily be exceeded. Those smokers are exposed not only to mainstream smoke (inhaled through the mouthpiece) but also to the environmental tobacco smoke they generate (also known as sidestream smoke), which is more toxic on a weight for weight basis compared to main- stream smoke. As a consequence, the adverse health effects of ex- posure to environmental tobacco smoke are similar to those caused or exacerbated by direct smoking. In adults, cancers (particularly lung cancer), cardiovascular diseases (ischaemic heart disease and stroke), and respiratory diseases (COPD, onset and exacerbations of asthma, and respiratory tract infections) are the main causes of mortality and morbidity. Associations with other conditions such as cognitive impairment, degenerative eye disease, and mental ill health have been reported. Its irritant effect should not be ignored as this is likely to be the reason most people object to passive smoking. The effect of the irritation per se has no known long-​term physical effects, but the effect on quality of life at home (or work) can be con- siderable. In neonates and children, environmental tobacco smoke could have developmental effects (low birth weight, preterm birth, and sudden infant death syndrome) and respiratory effects (retarded lung function growth, lower respiratory tract infection, respiratory symptoms, and asthma onset and exacerbations). Recently a new concept of ‘third-​hand smoke’ has been postulated, which refers to tobacco smoke residue that settles onto surfaces and dust and re-​emitted and re-​suspended into the air. While currently there is no direct evidence of the harmful effect of this third-​hand smoke, tobacco-​specific lung carcinogen 4-​(methylnitroasmino)-​1-​ (3-​pyridyl)-​1-​butanone (NNK) has been found to be present on sur- faces in most homes occupied by smokers, providing rationale for further research in this field. An updated Cochrane review published in 2016 examined the ef- fects of implementing smoking bans and found consistent evidence of improving cardiovascular health outcomes and reducing mortality for associated smoking-​related illnesses. Several countries have ex- tended the smoking ban to private vehicles in the presence of chil- dren. The impact of such legislation is still too early to be assessed. Electronic cigarettes (e-​cigarettes) have gained popularity in re- cent years, either as a substitute of cigarettes or as an aid for smoking cessation. While these battery-​powered nicotine vaporizers do not emit combustion products of tobacco, the aerosols produced by e-​cigarettes not only significantly increase indoor PM2.5 concen- trations (up to 300 times), but can also contain toxicants including glycols, aldehydes, metals, volatile organic compounds, and PAHs, some of which do not appear in cigarette smoke. Data on the health effects of exposure to e-​cigarette aerosols have been sparse. In vitro studies have shown increases in oxidative stress and decreases in epi- thelial cell viability 24 h after e-​cigarette aerosol exposure compared to clean air controls. Results from small-​scale preclinical studies have been conflicting. Cooking emissions Natural gas cooking appliances are a source of indoor NO2 and CO and contribute to deteriorated air quality when not adequately venti- lated. The health effects of ambient NO2 and CO have been described earlier, and the use of gas for cooking have also been shown to be associated with respiratory symptoms among females. But cooking itself, which is the treatment of food with heat, would promote de- composition and volatilization of lipids and amino acids in food, leading to emissions. The types and levels of pollutants in cooking emissions are highly heterogeneous and depend on food ingredi- ents and methods of cooking (e.g. frying, grilling, and baking), but generally include PM, volatile organic compounds, PAHs, and het- erocyclic amines. Some of these compounds are known mutagens and carcinogens. A link between exposure to cooking emissions and lung cancer has been proposed, although causality has not been to- tally confirmed. It should be noted that all of the currently available data were derived exclusively from the Chinese population.

10.3.1  Air pollution and health 1683 Radon Radon-​222 is a noble gas, a natural decay product of radium-​226, with a half-​life of 3.8 days. It decays through four short-​lived ‘radon daughters’ to lead-​210. Two of the radon daughters, polonium-​218 and polonium-​214, are α-​emitters. These cause bronchial mucosal damage when they decay within the lung and inhalation can thus lead to lung cancer. It has been estimated that 3–​14% of all lung cancer cases could be attributable to radon. In the 1970s it became clear that radon is invariably present in in- door air. Indoor levels vary considerably and in some dwellings levels are unacceptably high. The worldwide average indoor radon level has been estimated at 39 Bq/​m3, compared to 5–​15 Bq/​m3 outdoors. The main sources of indoor radon are the rock or soil on which the house is built, building materials used in the construction of the dwelling, natural gas, and water usage. The pooling of studies from Europe, North America, and China has suggested an increase in risk of lung cancer of 8–​13% per 100 Bq/​m3 increase in indoor radon level. Biological pollutants Allergens and moulds are the most important biological pollutants found in indoor air. Excrements of pests (house dust mites, cock- roaches, rodents) and dander of pets (particularly cats) are powerful allergens that may lead to the development of asthma and other al- lergic diseases. Several dust mite control measures (e.g. washing of bed linens) have been shown to be highly effective in reducing mite allergen, but the efficacy of these measures in preventing asthma is still under debate. Removal or relocation of pets has been recom- mended as the best measure but the reluctance of owners to give up their pets makes this difficult to implement. Air filtration as an alter- native has been proposed but the results have not been encouraging. Dampness is common in dwellings around the world, especially in overcrowded and poorly ventilated environments. Mould is only but one group of microbes that thrives in damp conditions. Its spores and fragments are potent allergens that are largely responsible for the health effects. There is sufficient evidence to support a causal link between dampness (and mould) and asthma development and exacerbations. However, a systematic review of trials and controlled before–​after studies of the effects of repairing mould-​damaged buildings did not find convincing beneficial effect (reduction) in asthma-​related symptoms and respiratory infections. Sick building syndrome In the early 1970s, illness or symptom complexes were recognized as being related to occupancy of certain buildings. Sick building syn- drome is characterized by an increased prevalence in a particular building of a range of non​specific symptoms typical of mucosal ir- ritation (e.g. sore/​dry eyes, sore nose, dry mouth, sore throat) often with lethargy and headaches. Symptoms usually disappear as soon as an affected individual leaves the building, but a 12-​year follow-​ up on 239 diagnosed Swedish patients found that symptoms could be long-​lasting with significant impact on their social life. Indoor environmental parameters such as ventilation, temperature, relative humidity, and indoor chemical (e.g. formaldehyde, volatile organic compounds) levels have been proposed as risk factors but there does not appear to be a consistent relationship of presence of severity of symptoms. Symptoms are usually reported more often by women than men. It was initially believed that the problem of sick building syndrome was minor, but it is now recognized that the financial costs of the condition are substantial. The acceptance of sick building syn- drome as an entity has led to the condition playing an important role in new building design. In buildings already affected, modification of the ventilation very often produces significant improvements in symptom severity and frequency. Household air pollution in LMICs In LMICs, particularly sub-​Saharan Africa and South Asia, indoor air quality is dominated by emissions from solid fuel (coal and bio- mass) for cooking and heating. Although the proportion of the household relying on solid fuel has decreased substantially from 62% in 1980, 53% in 1990, 46% in 2005 to 41% in 2010, the total number of people using solid fuel has remained approximately at 2.8 billion in the last three decades. The types of fuel used depend upon local availability and seasonality, for example, coal is predom- inantly used in China, whereas biomass such as wood, rice husk, twigs, charcoal, and dried animal dungs is used in India. Solid fuels are often burnt in traditional cookstoves of low energy conversion efficiency without chimneys such that combustion is usually incomplete and releases large amounts of carbon (as par- ticulates) and CO. Women, due to their traditional role in domestic purposes in LMICs and young children, who often stay with their mothers, are exposed to very high concentrations of air pollutants. The 24-​hour concentrations of PM2.5 measured in kitchen environ- ment from different settings ranges from 100 to 5000 µg/​m3 which is many folds higher compared to the WHO recommended guide- lines of 25 µg/​m3. Similarly, concentration of CO in enclosed kit- chen with little ventilation can be many folds higher than the WHO recommended level of 7 mg/​m3 for 24-​hour measurement. In rural communities where solid fuels are predominantly used for domestic purposes, the air pollution from kitchen exfiltrates to outdoor creating higher ambient air pollutant concentrations. In a cross-​ sectional study of 250 Nepalese rural homes the mean 24-​hour in- door PM2.5 concentration in the kitchen where cooking is carried out was 455 µg/​m3, the concentration in veranda where the other family members spend most of their time was 129 µg/​m3 and nearly 100 m away from house was 7.4 µg/​m3 (Fig. 10.3.1.1). This suggests indi- viduals using cleaner fuels but living in a community where solid fuels are used are also likely to be exposed higher air pollution. Like cigarette smoke, biomass smoke is a concentrated cocktail of chemicals, the toxicity of which is dependent on the type of fuel. Animal studies suggest short-​term wood smoke exposure could af- fect the physiology of the larynx, airway, and alveoli. Biomass smoke also contains high levels of reactive oxygen species, causing inflam- matory response and DNA damage in vitro. Emissions arising from coal burning have been classified as definite carcinogens and those from biomass fuels as probable carcinogens. There have been very few controlled biomass smoke-​exposure studies in humans, but acti- vation of circulating platelets, neutrophils, and monocytes has been reported with high levels of leukocyte-​platelet aggregates in chron- ically biomass smoke-​exposed individuals. Long-​term exposure to coal smoke has been associated with over 130% increased risk of lung cancer and over threefold increase in nasopharyngeal cancer compared to unexposed individuals. PAHs, particularly benzo[a]pyrene, appear to be directly involved in car- cinogenesis, as they are readily absorbed through the respiratory tract, gastrointestinal tract, and skin and circulated systemically.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1684 There is good evidence to suggest that exposure to wood smoke is associated with an increased risk of COPD even among never smokers. A meta-​analysis reported 80% of excess risk of COPD and 132% of excess risk for chronic bronchitis among those ex- posed to household air pollution compared to those who used clean fuel for cooking. It has been suggested that the mechanism of how biomass smoke develop COPD might be similar to that of cigarette smoking. Both are generated from the combustion of plant materials, which generates complex carbon-​based par- ticles coated with PAHs and irritant gases such as formaldehyde and acrolein and this could increase the expression of some of the matrix metalloproteinases as in the case of tobacco smoke. In children, exposure to biomass smoke (average PM10 level up to 1000–​2000 µg/​m3) increases acute lower respiratory infections. The underlying mechanism by which biomass smoke predisposes individuals to respiratory infections is unknown, although evi- dence from controlled chamber study suggested biomass com- bustion may enhance susceptibility to bacterial pneumonia. Early life exposure to household air pollution might be associated with reduced lung function and increased respiratory symptoms in childhood and ultimately in adulthood, but this needs to be con- firmed in long-​term prospective cohorts. Although the association between cardiovascular diseases and ambient air pollution has been well established, there is no direct evidence to conclude the same with household air pollution. Few short-​term observational studies have reported increased sys- tolic and diastolic blood pressure, elevated oxidized low-​density lipoprotein, significant increase in platelet aggregation, platelet P-​selectin expression, raised anticardiolipin IgG and IgM anti- bodies, increased mean carotid intima-​media thickness and in- creased prevalence of atherosclerotic plaques among those exposed to biomass smoke. A randomized controlled trial reported reduced ST-​segment depression among those who received improved cook stoves, hinting exposure to wood smoke could affect ventricular repolarization. Other endpoints such as adverse pregnancy outcomes and peri- natal risks (birth defects, low birthweight, and sepsis), and impaired physical and cognitive development in children have been reported. Biomass users, especially in rural communities, often spend con- siderable amount of time collecting firewood and are likely to have injuries from falls and muscloskeletal problems. A significant in- creased risk of cataract and teary eyes among biomass smoke-​ex- posed individuals have also been reported. Switching to cleaner fuels, such as electricity, is the most effective intervention but this option is not always feasible. Adoption of im- proved cookstoves that are more efficient is a viable alternative. At present, 27% of the total population using solid fuels have access to improved cookstoves, most being in China (70%). Major pro- jects have been launched to produce and disseminate improved cookstoves on a global scale. A recent systematic review examined the effect of stove interventions and concluded while most studies reported a reduction of household air pollution particularly PM and CO levels, they were not enough to meet WHO air quality recom- mendations. Nevertheless, reductions in self-​reported respiratory and non​respiratory symptoms among those who received im- proved cookstoves have been found. On the other hand, there was no conclusive evidence of improvement in objective measures of lung function and birth weight. The results from these intervention studies highlighted importance of user behaviour and compliance (e.g. exclusive use of improved cookstoves above traditional op- tions) and the long-​term proper maintenance of the stoves. Air quality guidelines and standards Although by definition all air pollutants are hazardous, only the major ones (e.g. PM, SO2, NO2, and ozone) in the atmosphere are monitored and regulated by legislation in most countries. While no threshold concentration has been identified below which air pollu- tion has no effect on population health, different limits and target values have been proposed by various international organizations and national governments, driven by a basket of factors including scientific evidence of adverse health effect, cost, feasibility, and pol- itical will. In most cases, the national guidelines in LMICs tend to be more relaxed compared to those in the high-​income coun- tries. Table 10.3.1.2 shows three most widely used contemporary guidelines and standards from the WHO, the US Environmental Protection Agency (US EPA), and the European Union (EU). (a) (b) Fig. 10.3.1.1  Household air pollution in kitchen. (a) A man cooking food on traditional stove without chimney using wood; (b) a cot hanging from the kitchen ceiling where an infant is usually kept when the mother is performing the cooking activities. Photos taken by Om P Kurmi.

10.3.1  Air pollution and health 1685 In contrast, indoor air quality is not normally regulated, except for airborne pollutants generated in the workplace, the standards of which are enforced separately by occupational safety and health regulations beyond the scope of this Section. Certain countries have set up emission standards of volatile compounds (e.g. formal- dehyde) from building and interior furnishing materials, primarily aiming to control the associated sick building syndrome. WHO has issued non​legal binding guidelines based on health risk of damp- ness and mould (WHO 2009) and common chemicals found in indoor environment (WHO 2010), but the adoption of such guide- lines has not been widespread in its member states. WHO has also developed air quality guidelines on household air pollution, par- ticularly to those using solid fuel for cooking. The guideline is sum- marized in Table 10.3.1.3. Table 10.3.1.2  Various major national and international guidelines on major air pollutants Pollutants Mean concentration allowed when averaged over timea  10-​min 1-​hr 8-​hr 24-​hr Annual PM10 (µg/​m3) WHO –​ –​ –​ 50 20 US EPA –​ –​ –​ 150a EU –​ –​ –​ 50 40 PM2.5 (µg/​m3) WHO –​ –​ –​ 25 10 US EPA –​ –​ –​ 35Ϯ 12† or 15‡ EU –​ –​ –​ –​ 25 Nitrogen dioxide (µg/​m3) WHO –​ 200 –​ –​ 40 US EPA 190# –​ –​ –​ 99.7 EU –​ 200 –​ –​ 40 Ozone (µg/​m3) WHO –​ –​ 100 –​ –​ US EPA –​ –​ 140## –​ –​ EU 120 Sulphur dioxide (µg/​m3) WHO 500 –​ –​ –​ 20 US EPA –​ 200 b –​ –​ –​ EU –​ 350 –​ 125 –​ a Not to be exceeded more than once per year over three years; Ϯ 98th percentile averaged over 3 years; †Primary pollutant: Annual mean averaged over 3 years; ‡ Secondary pollutant: Annual mean averaged over 3 years; # 98th percentile of 1-​hour daily maximum concentration, averaged over 3 years; ## Annual fourth-​highest daily maximum 8-​hour concentration, averaged over 3 years; b Primary pollutant: 99th percentile of 1-​hr daily maximum concentration, averaged over 3 years. The latest guidelines used in tables can be obtained from WHO (http://​apps.who.int/​iris/​bitstream/​10665/​69477/​1/​WHO_​SDE_​PHE_​OEH_​06.02_​eng.pdf); US EPA (http://​www3.epa.gov/​ttn/​naaqs/​criteria.html) and EU (http://​ec.europa.eu/​environment/​air/​quality/​standards.htm) webpages. Table 10.3.1.3  Summary of published WHO air quality guideline values for household air pollution Pollutant (unit for guidelines) Mean concentration over averaging time Unit risk Comments 15 min 30 min 1 hour 8 hours 24 hours 1 year PM2.5 (µg/​m3) –​ –​ –​ –​ 25a 10 –​ 24–​hour guideline not to exceed max. 3 days/​year PM10 (µg/​m3) –​ –​ –​ –​ 50a 20 –​ 24–​hour guideline not to exceed max. 3 days/​year Benzene (risk of leukaemia per 1 µg/​m3) –​ –​ –​ –​ –​ –​ 6.0 × 10–​6 No safe level Carbon monoxide (mg/​m3) 100 –​ 35 10 7 –​ –​ –​ Formaldehyde (mg/​m3) –​ 0.1 –​ –​ –​ –​ –​ –​ Naphthalene (mg/​m3) –​ –​ –​ –​ –​ 0.01 –​ –​ Nitrogen dioxide (µg/​m3) –​ –​ 200 –​ –​ 40 –​ –​ PAHsb (risk of lung cancer per 1 ng/​m3) B[a]‌P –​ –​ –​ –​ –​ –​ 8.7 × 10–​5 No safe level a The 24-​hour average values for PM10 and PM2.5 refer to the 99th percentile of the distribution of daily values (i.e. the fourth next highest value of the year). b PAHs, polycyclic aromatic hydrocarbons, B[a]‌P, benzo[a]pyrene: In view of the difficulties in developing guideline for PAH mixtures, B[a]P was considered to represent the best single indicator compound.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1686 Table 10.3.1.4  The Daily Air Quality Index, showing exposure concentration and accompanied health messages (accessed from http://​uk-​air.defra.gov.uk/​) Air pollution index bands (value) → Low (1) Low (2) Low (3) Moderate
(4) Moderate
(5) Moderate
(6) High (7) High (8) High (9) Very high (10) Pollutants (µg/​m3) Sampling time Concentration of specific pollutants at different air pollution index bands PM10 24-​hr mean 0–​16 17–​33 34–​50 51–​58 59–​66 67–​75 76–​83 84–​91 92–​100 ≥101 PM2.5 24-​hr mean 0–​11 12–​23 24–​35

36–​41 42–​47 47–​53 54–​58 59–​64 65–​70 ≥71 SO2 15-​min mean 0–​88 89–​177 178–​266 267–​354 355–​443 444–​532 533–​710 711–​887 888–​1064 ≥1065 NO2 1-​hr mean 0–​67 68–​134 135–​200 201–​267 268–​334 335–​400 401–​467 468–​534 535–​600 ≥601 O3 8-​hr mean 0–​33 34–​66 67–​100 101–​120 121–​140 141–​160 161–​187 188–​213 214–​240 ≥241 Accompanying health messages for individuals
at risk Enjoy your usual outdoor activities. Adults and children with lung problems, and adults with heart problems, who experience symptoms, should consider reducing strenuous physical activity, particularly outdoors. Adults and children with lung problems, and adults with heart problems, should reduce strenuous physical exertion, particularly outdoors, and particularly if they experience symptoms. People with asthma may find they need to use their reliever inhaler more often. Older people should also reduce physical exertion. Adults and children with lung problems, adults with heart problems, and older people, should avoid strenuous physical activity. People with asthma may find they need to use their reliever inhaler more often. Accompanying health messages for the general population Enjoy your usual outdoor activities. Enjoy your usual outdoor activities. Anyone experiencing discomfort such as sore eyes, cough, or sore throat should consider reducing activity, particularly outdoors. Anyone experiencing discomfort such as sore eyes, cough, or sore throat should consider reducing activity, particularly outdoors. PM10, particulate matter less than 10 µm aerodynamic diameter; PM2.5, particulate matter less than 2.5 µm aerodynamic diameter; SO2, sulphur dioxide; NO2, nitrogen dioxide; O3, ozone.

10.3.2 Heat 1687

10.3.2 Heat 1687

10.3.2  Heat 1687 Air quality index Based on the air quality guidelines, government agencies around the world have drawn up variations of air quality index to commu- nicate information about real-​time and short-​term forecast levels of air pollution and its possible associated health effects to the public. In the United Kingdom, on the advice of the Committee on Medical Effects of Air Pollutants (COMEAP), the Department for Environment Food and Rural Affairs (Defra) uses pollutant (PM10, PM2.5, SO2, NO2, and ozone) concentration data obtained from its national monitoring network and computer based models to derive a health-​based Daily Air Quality Index, which provides recommendations on the actions to be taken by the public based on the current and forecasted (next day) air pollution levels. The Daily Air Quality Index is a 10-​point scale divided into four colour-​coded bands: low (1–​3), moderate (4–​6), high (7–​9), and very high (10) (Table 10.3.1.4). Members of the public will first de- termine whether they are at risk from air pollution (from the add- itional information accompanying the Daily Air Quality Index), and read the health messages corresponding to the pollution band for recommended actions (e.g. reducing physical exertion). The main objective behind this is the prevention of adverse health ef- fects from short-​term elevations in air pollution. It is anticipated that such a forewarning system could help sensitive individuals to modify their behaviour to reduce their individual exposure to the pollution, or reduce the severity of their symptoms. FURTHER READING Atkinson RW, et al. (2015). Fine particle components and health—​ a systematic review and meta-​analysis of epidemiological time series studies of daily mortality and hospital admissions. J Expo Sci Environ Epidemiol, 25, 208–​14. Brunekreef B, Holgate ST (2002). Air pollution and health. Lancet, 360, 1233–​41. Cohen AJ, et al. (2017). Estimates and 25-year trends of the global burden of disease attributable to ambient air pollution: an analysis of data from the Global Burden of Diseases study 2015. Lancet, 389, 1907–18. Gauderman WJ, et al. (2015). Association of improved air quality with lung development in children. N Engl J Med, 372, 905–​13. GBD Risk Factors Collaborators (2015). Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 coun- tries, 1990–​2013:  a systematic analysis for the Global Burden of Disease Study 2013. Lancet, 386, 2287–​323. Gordon SB, et al. (2014). Respiratory risks from household air pollution in low and middle income countries. Lancet Resp Med, 2, 823–​60. Kurmi OP, Lam KBH, Ayres JG (2012). Indoor air pollution and the lung in low-​ and medium-​income countries. Eur Resp J, 40, 239–​54. Pope CAI, Ezzati M, Dockery DW (2009). Fine-​particulate air pollution and life expectancy in the United States. N Engl J Med, 360, 376–​86. Soule EK, et al. (2017). Electronic cigarette use and indoor air quality in a natural setting. Tob Control, 26, 109–​12. Thomas E, et  al. (2015). Improved stove interventions to reduce household air pollution in low and middle income countries: a de- scriptive systematic review. BMC Public Health, 15, 1–​15. 10.3.2  Heat Michael A. Stroud ESSENTIALS Rising body temperature triggers behavioural and physiological responses including reduction in physical activity, alterations of clothing, skin vasodilatation, and sweating. Heat-​related illness is relatively common, especially with high humidity or prolonged physical activity. Risk can be reduced by acclimatization with re- peated heat exposure, but some individuals seem to be particularly susceptible. Clinical presentations of heat-​related illness include (1) ‘heat ex- haustion’—​the commonest manifestation, with symptoms including nausea, weakness, headache, and thirst. Patients appear dehydrated, complain of being hot, and are flushed and sweaty. Treatment re- quires rest and fluids, given orally or (in severe cases) intravenously. (2) ‘Heat stroke’ victims often complain of headache, may be drowsy or irritable, and may claim to feel cold. Core temperature is usually 38–​41°C, but the patient is shivering with dry, vasoconstricted skin. Treatment requires (a) aggressive rapid cooling—​tepid water and fan-​ assisted evaporation in the first instance, with more invasive meas- ures (e.g. intraperitoneal fluids, if required); (b) close biochemical monitoring; (c) supportive care for organ failure. There is significant mortality. Thermoregulation in the heat Most of human evolution took place in Africa and hence all hu- mans are heat tolerant. We try to maintain a near-​tropical micro- climate against our skin, by using clothing to reduce heat loss to our surroundings. Thermal balance is regulated by the hypothalamus, which integrates information from skin temperature sensors with core temperature data from receptors within walls of large blood vessels and the brain. Rising temperatures trigger both behavioural and physiological responses. Behavioural changes include reducing physical activity, altering clothing, and seeking shade or cool shelter. Cold drinks are also helpful. Although these responses seem simplistic, decisions may not be straightforward. If physical activity is low and water is in short supply, it is better to increase clothing cover and protect your- self from high radiant heat inputs. If activity must be continued and water is freely available, minimal clothing to permit maximal sweat evaporation is preferable. Immediate physiological responses involve vasodilatation of skin and subcutaneous blood vessels to enhance surface heat loss from radiation, conduction, and convec- tion. The vasodilatation is triggered by a sympathetic cholinergic reflex in response to skin warming, with additional direct effects of heat on arteriolar tone. In a resting person, skin vasodilatation can maintain thermal equilibrium in environmental temperatures up to 32°C, but with higher temperatures or heat production from activity, core temperatures will rise. This will trigger sweating to promote evaporative cooling.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1688 Heat acclimatization Repeated heat exposure can increase our capacity to lose heat by about 20-​fold. This is partly due to greater skin blood flow from in- creases in circulating volume and improved vasodilatory responses, but changes in sweating responses are more important. In the non​acclimatized, sweating is triggered by a rise in core tempera- ture of about 1°C and maximum rates are limited to about 0.5 litre/​ h. Following acclimatization, a 0.5°C core rise will trigger the re- sponse and sweat rates may exceed 2.0 litre/​h. Acclimatization also leads to aldosterone-​mediated reductions in sodium loss in both sweat and urine. The acclimatized individual therefore requires no sodium supplementation and giving supplements can delay the ac- climatization process. Avoiding them altogether, however, risks salt depletion in non​acclimatized persons during prolonged heat stress. Acclimatization develops swiftly and around 90% of maximum heat tolerance is present after 7–​10 days, on which core temperature has risen by more than 1°C for more than 1 h. Physical exertion com- bined with heat makes the changes even more rapid. After returning to cool environments, adaptation is lost in 20–​40 days. Susceptibility to heat-​related illness Although we are generally heat tolerant, heat-​related illness is rela- tively common, and several factors increase vulnerability. Above an environmental temperature of about 35°C, we tend to gain heat from our surroundings, and this, along with metabolic heat production, can only be lost via evaporation of sweat. Hot envir- onments with high humidity are therefore the greatest threat. Acclimatization status has a marked influence on heat-​related risks, the unacclimatized being prone to hyperthermia and salt depletion, while the fully acclimatized are vulnerable to dehydration from high sweat rates. Dehydration in itself limits sweating capacity and skin blood flow and hence increases risks. It can occur easily since thirst is a poor trigger for adequate drinking. Sweat rates in the acclima- tized can also exceed gut capacity for water absorption. Prolonged physical activity can cause heat illness under quite modest environmental conditions. This is particularly common when individuals are obliged to wear clothing that is insulative or vapour-​impermeable. Military heat casualties are sometimes due to these factors, but there have also been fatalities in soldiers who have been susceptible to heat for no obvious cause. Such genetic or constitutional vulnerability should be suspected whenever a heat-​ related problem occurs following relatively modest heat stress. These people should be strongly advised to avoid similar circumstances in future. Obesity and poor physical fitness are further risk factors in the heat, as is diabetic autonomic dysfunction. Older people are generally heat sensitive and, in addition, are prone to problems from the increased circulatory demands of vasodilatation. Drugs can also induce heat illness (see following paragraphs). Heat exhaustion Most casualties in hot environments suffer from heat exhaustion. There is usually a history of prolonged heat stress followed by nausea, weakness, headache, thirst, and sometimes collapse. Patients appear dehydrated with a tachycardia and low blood pressure. If hyper- thermic, the casualty should be complaining of feeling hot and will appear flushed and sweaty. The absence of these symptoms and signs, especially with a very high core temperature, suggests heat stroke. Heat exhaustion is ascribable to sodium and/​or water depletion, but discriminating between these can be difficult. Sodium depletion tends to be greater if the casualty was poorly acclimatized and hence sweated relatively more sodium than water. Conversely, water depletion is more common in acclimatized individuals. Muscle cramps or whole-​body dehydration without marked changes in haematocrit or serum proteins are suggestive of excessive sodium loss, but serum sodium tends to be normal in such cases unless enthusiastic fluid replacement without salt has led to hyponatraemia. This sometimes occurs in runners after completing marathons in hot environments. In predominantly water-​ depleted heat exhaustion, haematocrit, serum proteins, and serum so- dium tend to be high. Renal impairment occurs in either form of heat exhaustion and the treatment of both types often requires 5–​10 litres of oral or intravenous fluids in the first 24 h. Sodium supplementation is given as appropriate, but if sodium status is uncertain, it is usually safer to provide some than to precipitate acute hyponatraemia. Heat stroke Mild heat stroke has occurred when a hot environment or high ac- tivity levels have led to pyrexia with cerebral disturbance. Core tem- perature is usually 38–​41°C. The condition frequently follows heat exhaustion but temperature may have risen rapidly allowing no time for salt or water depletion. Sufferers have headaches and may be ei- ther drowsy or irritable. They often hyperventilate. The great danger is progression to more severe heat stroke, in which core tempera- ture reaches levels that cause irreversible denaturing of proteins. This usually occurs at above 41.5°C. Damage is widespread and par- ticularly affects brain, liver, kidney, and muscle. Furthermore, the hypothalamic thermoregulatory centre may fail, switching off vaso- dilatation and sweating, and switching on cold defences inappropri- ately. Patients may therefore claim to feel cold and on examination may be shivering with a dry, vasoconstricted skin. A disastrous vi- cious cycle of increasing temperatures can then ensue. Treatment for all heat stroke requires early recognition and rapid cooling. Tepid water and fan-​assisted evaporation may be more ef- fective than immersion in cold water, which can limit heat loss by stimulating intense peripheral vasoconstriction. Intraperitoneal fluids, paralysis, and ventilation may be needed and, in extreme circiumstances, cooling by cardiac bypass should be considered. Hyperkalaemia, hypocalcaemia, acidosis, rhabdomyolysis, dissem- inated intravascular coagulation, and hepatic or renal failure are common complications. Ventricular fibrillation is a frequent ter- minal event. Even if apparently resuscitated and cooled success- fully, a 12-​ to 24-​h ‘lucid interval’ may precede major deterioration. Permanent neurological damage is common. Drug-​induced heat illness Many drugs can cause mild degrees of pyrexia by inducing local or systemic inflammation or hypersensitivity. Some also increase sus- ceptibility to environmental heat by inhibiting central thermoregu- lation (e.g. barbiturates and phenothiazines) or reducing sweating

10.3.3 Cold 1689

10.3.3 Cold 1689

10.3.3  Cold 1689 capacity (e.g. anticholinergics). Salicylate overdose can generate heat stroke by increasing metabolic heat production while impairing hypothalamic regulation. There are two types of heat-​related drug reactions, however, which are particularly dangerous. Malignant hyperpyrexia This is usually a dominantly inherited condition, although dif- ferent gene defects may affect families. Administration of a variety of anaesthetic agents, including halothane and suxameth- onium, leads to rapid, massive heat production from generalized increases in skeletal muscle tone. Contraction is triggered at the muscle cell membrane and hence neuromuscular blocking agents are ineffective. Intravenous dantrolene, an inhibitor of muscle calcium flux, is helpful and can be used along with ventilation and cooling/​supportive measures. Fatalities are common, and it is therefore important to avoid risks whenever possible. In patients with a relevant personal or family history, in whom an anaesthetic is unavoidable, oral dantrolene should be given prior to the use of low-​risk agents. Neuroleptic malignant syndrome This condition has similarities to malignant hyperpyrexia but is in- duced by idiosyncratic reactions to normal doses of antidopaminergic drugs, including phenothiazines and butyrophenones. The onset is less rapid than malignant hyperpyrexia, occurring over a few days. The increased muscle tone is also induced presynaptically and hence neuromuscular blocking agents help. Some recreational drugs, such as ecstasy, can induce this type of response, although most cases of ecstasy-​induced hyperthermia are probably cases of heat stroke induced by enthusiastic dancing with limited fluid intake in hot, humid environments. FURTHER READING Bouchama A, Knochel JP (2002). Heat stroke. N Engl J Med, 346, 1978–​88. Hodgson P (1991). Malignant hyperthermia and the neuroleptic ma- lignant syndrome. In: Swash M, Oxbury J (eds) Clinical neurology, pp. 1344–​5. Churchill Livingstone, Edinburgh. Hubbard RW, Armstrong LE (1988). The heat illnesses: biochemical, ultrastructural, and fluid-​electrolyte considerations. In:  Pandolf KB, Sawka MN, Gonzalez R (eds) Human performance physiology and environmental medicine at terrestrial extremes, pp. 305–​59. Benchmark, Indianapolis, IN. 10.3.3  Cold Michael A. Stroud ESSENTIALS Humans are poorly adapted to cold, which can cause hypothermia, non​freezing cold injury, and frostbite. Hypothermia This occurs especially with wind and wetting, and is seen indoors in older people and those who are thin. At a core temperature of 35°C, victims complain of cold, act appropriately, shiver, and are peripherally vasoconstricted, but with further cooling they may be- come confused or drowsy and appropriate physiological responses disappear. Coma occurs at 26–​32°C, and death typically at 17–​26°C. General investigation and management is as for any comatose pa- tient, but specific issues include (1) accurate measurement of core temperature requires a low-​reading rectal thermometer; (2) meas- urement of serum amylase (risk of pancreatitis) and creatine kinase (risk of rhabdomyolysis); (3) rewarming—​if onset of cooling was pro- longed, rewarming should generally be slow; (4) diagnosis of death—​ apparently dead victims should be rewarmed whenever possible before resuscitation is abandoned. Non​freezing cold injury This occurs when skin temperatures below 12°C are maintained for prolonged periods, particularly in water (e.g. trench foot). This causes local tissue damage, particularly to nerves, which can be permanent. Frostbite Frozen tissues initially appear hard, white, and anaesthetic, but with rewarming become swollen, painful, and blistered. There may be ir- reversible necrosis, but initial appearances can be misleading and hence early amputation should be avoided. Once thawed, frostbite treatment is similar to that for burns. Thermoregulation in the cold It has only been 10 000 to 15 000 years since ancestral humans dwelt exclusively in warm or hot climates. Humans are therefore poorly adapted to cold, and hypothermia occurs quite frequently even in temperate regions. With water immersion it may occur even in the tropics. In truly cold areas, there is also the risk of non​freezing cold injury and frostbite. Nevertheless, behavioural changes allow us to operate safely even in the coldest environments. Core temperatures in the cold are usually maintained by ad- justments in clothing and physical activity. The latter can increase heat production from a resting 100 W to 1–​2 kW. This is very ef- fective. Although it takes highly specialized, multilayered clothing to keep warm while inactive in an environment of +5°C, clothing insulation equivalent to normal office dress (1 clo) will maintain core temperature even in an environment of –​20°C when working moderately hard. Our limited physiological cold protection is under hypothalamic control. Falling surface and, to a lesser extent, core temperatures lead to decreased blood flow in the skin due to increased sympathetic ad- renergic tone and direct cooling effects of cold on skin arterioles. This minimizes surface heat loss. Unfortunately, vasoconstriction also leads to severe cooling of the hands and feet with problems of temporary skin numbness, muscle weakness, and risks of more per- manent peripheral cold injury. It is often this peripheral cooling that limits our capacity to work in the cold. Falling skin temperatures will also lead to higher resting muscle tone and shivering, especially when declining core temperature

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1690 releases hypothalamic inhibition of shivering. These mechanisms can only increase resting heat production to around 500 W and, unlike newborn infants and some other mammals, adult humans cannot add significant non​shivering heat production to this figure. Effects of falling core temperature Falling core temperature leads to progressive decline in function. At 34–​36°C, hypothermic individuals are conscious of feeling cold and try to move around, add clothing, or seek shelter. Simultaneously, physiological defences are activated. With further falls of tempera- ture, mental and physical problems increase. Some people become withdrawn while others exhibit aggression or disinhibition. Once core temperatures reach 33–​34°C, victims often stagger and be- come confused or drowsy. It is also around this point that ‘paradox- ical undressing’ may occur. This phenomenon is well described and appears to be due to hypothalamic dysfunction with alteration of set-​point temperature. Victims therefore feel warm or even hot and appropriate behavioural and physiological responses disappear. At core temperatures varying between 26 and 32°C coma will ensue, and between 17 and 26°C cardiac output becomes inadequate to sustain life for prolonged periods. The risk of ventricular fibrilla- tion is also high. Nevertheless, successful resuscitations of victims with core temperatures below 15°C have been reported (see also Chapter 9.5.3). Causes of hypothermia Several factors increase hypothermic risk. Wetting of skin or clothing extracts enormous amounts of heat and reduces insulation of garments. Complete immersion is particularly hazardous and worldwide more than 100 000 people per year die of cold shock or inexorable hypothermia in the water. This far exceeds deaths from drowning without cold. Winds also increase environmental cooling and a still air temperature of +5 °C equates to –​50 °C if wind speed is 40 km/​h. Coupled with rain, these effects often contribute to hypo- thermic accidents among hill walkers and mountaineers, although in these cases fatigue may contribute. Prolonged exertion depletes muscle glycogen which reduces heat production capacity from both exercise and shivering. Low blood glucose also impairs hypothal- amic temperature control. Small, thin people cool easily because of their increased surface-​to-​volume ratios. They also have reduced subcutaneous insulation and low heat-​producing mass. A fat person can main- tain core temperature at rest, even if mean skin temperature is 12°C, whereas a thin person struggles to maintain thermal equi- librium with a skin temperature of 25°C. However, rapid cooling can sometimes have benefits. A small child in cold water may cool so rapidly that vagally triggered bradycardia and lowered brain metabolic demands may permit successful resuscitation after very prolonged immersion. Older people may also be small and thin and are at risk of so-​ called ‘urban hypothermia’. Poverty, illness, immobility, malnutri- tion, and a less sensitive regulatory system may contribute, but in many cases hypothermia on admission to hospital is secondary to other pathology (e.g. a stroke may have led to prolonged immobility in a cool environment). Drugs that impair consciousness or induce vasodilatation are risk factors, and alcohol is particularly hazardous. Alcoholics with no fixed abode and a tendency to hypoglycaemia are frequent urban cold casualties. Hypothermic illness General management of the hypothermic casualty is similar to that for any comatose or semicomatose person. Abnormalities in blood gases, pH, electrolytes, and glucose are common, and pancreatitis or rhabdomyolysis are recognized complications. Accurate meas- urement of core temperature is surprisingly difficult. Axillary, tym- panic, and oral temperatures can all be misleading. A low-​reading rectal thermometer is best. Hypothermia has one very specific risk. Pronouncement of death is fraught with difficulty since profound bradycardia, minimal stroke volume, and marked respiratory de- pression occur. The old adage that you are ‘never dead unless warm and dead’ must be taken seriously. A variety of rewarming methods are available. Warm blankets and hot drinks will suffice in many cases but, although they are widely used, metallized ‘space blankets’ are of no proven benefit. Warmed intravenous fluids are helpful and, in extreme cases, peritoneal warmed fluids or cardiac bypass can be used. Specialized equipment providing heated, humidified air also permits core rewarming. Hot baths are effective but difficult to use safely since a paradoxical fall in core temperature can occur as blood flow is rapidly restored to cold limbs. In general, if cooling was prolonged in onset or duration, rewarming must be undertaken with extreme caution. In critical cases, where rapid rewarming is needed, full resuscitation facilities must be available, although safe defibrillation in the presence of water is impossible. Careful monitoring during rewarming is vital. Blood volumes are often low due to early cold-​induced diuresis, followed by the in- ability of hypothermic kidneys to retain salt and water. In immer- sion casualties, hydrostatic effects on the limbs may have promoted additional fluid loss and, if possible, these people must be kept re- cumbent throughout rescue and rewarming to minimize risks from extreme postural hypotension. Warming cell membranes are ex- tremely unstable, and uncontrollable fluxes in potassium and other electrolytes may occur, although care must be taken in interpreting biochemical results from cold peripheral blood sampling. Non​freezing cold injury Local temperatures of less than 12°C prevent normal membrane pumping and paralyse nerve and muscle conduction. If such cooling is prolonged, permanent damage may ensue. Immersion in cold water is particularly likely to cause this type of damage and soldiers in military campaigns are frequent victims of ‘trench foot’. Long-​term damage is likely whenever an anaesthetic, paralysed, cold region becomes hot, red, painful, and swollen after rewarming, although this change may take several days. Degeneration of nerve and muscle can then follow, leading to prolonged anaesthesia, muscle contractures, or inappropriate peripheral vascular control with intolerance to local heat or cold. There may be slow improve- ment over months or years.

10.3.4 Drowning 1691

10.3.4 Drowning 1691

10.3.4  Drowning 1691 Frostbite Human tissues freeze at around –​2 °C. Ice forms outside cells but the remaining extracellular fluid becomes hyperosmolar and hence severe intracellular dehydration occurs. This denatures proteins. Vascular endothelial cells are particularly vulnerable, and following rewarming, small blood vessels may leak plasma and then become blocked by red cell sludge and clot. Additional ischaemic necrosis is then superimposed on the frost damage. Frozen tissues appear hard and white and are anaesthetic. Rewarming leads to pain and swelling, often accompanied by blis- tering. Deep-​freezing results in irreversible necrosis but appear- ances can be misleading, and early amputation of digits should be avoided. If still frozen, rewarming is best achieved rapidly by using immersion in water at 40–​42°C, although any thawing should be avoided if refreezing is likely. Once thawed, treatment is similar to that used for burns with prevention of infection paramount. Generous analgesia is required. FURTHER READING Dexter WW (1990). Hypothermia. Safe and efficient methods of rewarming the patient. Postgrad Med, 88, 55–​8, 61–​4. Giesbrecht GG (2000). Cold stress, near drowning and accidental hypothermia: a review. Aviat Space Environ Med, 71, 733–​52. Granberg PO (1997). Cold injury. In: Chant ADB, Barros D’Sa AAB (eds) Emergency vascular practice, pp. 119–​34. Hodder Arnold, London. Hamlet MP (1988). Human cold injuries. In: Pandolf KB, Sawka MN, Gonzalez R (eds) Human performance physiology and environ- mental medicine at terrestrial extremes, pp. 435–​66. Benchmark, Indianapolis, IN. Stroud MA (1993). Environmental temperature and physiological function. In: Ulijaszek SJ, Strickland SS (eds) Seasonality and human ecology, pp. 38–​53. Cambridge University Press, Cambridge. 10.3.4  Drowning Peter J. Fenner ESSENTIALS Drowning is a major preventable cause of death, most frequently in children and in developing countries. Aspiration (whether of salt or fresh water) is usual in drowning and near-​drowning (known as
non​fatal, or submersion injury) and leads to cardiac arrest within a few minutes. Death or severe neurological impairment occurs after submersion for more than 5–​10 min, but much longer durations may be tolerated in hypothermic conditions. Prevention Precautions include proper supervision of children in recreation areas such as swimming pools, beaches, and river banks, and of young children and epileptics in baths. Personal flotation devices (life jackets) are the best preventive strategy in boating activities. Prevention and rescue efforts of life-​savers are effective in swimming pools and on patrolled beaches. Clinical features Prognosis cannot reliably be predicted, but cardiovascular status is a better prognostic indicator than neurological presentation. Patients who are neurologically responsive at the scene of immersion, in sinus rhythm and with reactive pupils, have good outcomes. Those who are asystolic on arrival at hospital and remain comatose for more than 3 h have a poor prognosis unless they are hypothermic. Patients with a normal chest radiograph on admission usually survive. Management The factors that influence outcome are (1) immediate management—​ including rapid rescue; laying the victim on their side for assessment of the airway and breathing to assist drainage of any excess water from the airways and lungs; prompt and effective bystander cardiopulmonary resuscitation, using supplemental oxygen if available, preferably with oxygen of highest concentration possible (e.g. bag–​valve–​mask) and an oropharyngeal airway, endotracheal tube, or laryngeal mask airway in comatose victims (if suitably skilled personnel are present). (2) Hospital management—​important elements are (a) ventilatory support to main- tain adequate arterial oxygenation, which may involve the use of extra- corporeal membrane oxygenation and/​or cardiopulmonary bypass in refractory cases; (b) colloid resuscitation, (c) recognition and treatment of complications (e.g. secondary pneumonia). Definition Drowning has most recently been defined as ‘the process of experi- encing respiratory impairment from submersion/​immersion in liquid’. Outcomes of drowning should be classified as death, mor- bidity, and no morbidity. Recent guidelines suggest that the term ‘submersion injury’ or non​fatal drowning should replace ‘near-​ drowning’, although the latter is still commonly used. The lack of a universally agreed standard definition makes it difficult to evaluate the results of studies of drowning and submersion, particularly as drowning remains difficult to diagnose at autopsy. Mortality and morbidity Acute prolonged hypoxia causes haemodynamic effects, cerebral damage, and death. Neurological morbidity in survivors of near-​ drowning includes difficulties with learning, memory, attention and planning, and cerebral palsy. A large study of childhood immersions has shown that approximately 70% of survivors have no neuro- logical deficit, 30% have some deficit; 3% will live in a permanent vegetative state. Epidemiology The estimated incidence of drowning is 0.5  million per year, making it the fourth most common fatal injury worldwide in the

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1692 global burden of disease. It is the seventh leading cause of death from unintentional injury in all ages and the second leading cause in children aged 1–​14 years. Incidences of drowning are highest in children up to 4 years old. In infants and toddlers under 12 months, bathtub and bucket immersions are the highest cause of drowning. Ten per cent (10%) of fatal bucket or tub immersions are attrib- utable to child abuse. Smooth and slippery bathtubs are particu- larly dangerous and bathtub seats are unsafe, particularly if infants are left unattended. Worldwide, drowning rates in young children, many of whom are unsupervised, have decreased little, despite po- tentially effective preventive strategies such as fencing of swimming pools, providing education appropriate to the particular circum- stances, or increased surveillance. Although there are few such pre- ventive strategies for older children, drowning rates have declined dramatically in the last decade. Developing countries have the highest rates of drowning. Thirty-​ eight per cent (38%) of the world’s drownings occur in the Western Pacific region and Africa, where the drowning mortality rate is 13.1 per 100 000 population per year. Children aged 5–​14 years suffer the highest mortality rate. Children under the age of 5 years have the highest drowning mortality rate for both sexes. The mother’s age and literacy and family income are identified as risk factors. In Bangladesh, drowning has been shown to be a major cause of child- hood mortality: among 1-​ to 4-​year-​olds; there are 156 fatal drown- ings/​100 000 population per year. Younger males were at greatest risk of drowning in rural areas, mainly in ditches and ponds. In China, estimated drowning mortality rates for all age groups were 29.8/​100 000/​year for boys and 29.6 for girls. In the United States of America in 2000, more than 1400 children younger than 20 years drowned. It is the seventh leading cause of unintentional injury deaths for all ages, the second leading cause of all deaths from injury in children aged 1–​14 years and the third most common cause of fatality in people under 15 years (after car accidents and asphyxia). Many drownings occur during recre- ation in swimming pools, spas, hot tubs, lakes, rivers, or oceans. Approximately 53% of victims needed hospitalization or transfer for more specialized care. Drowning rates were highest among children up to 4 years old. Worldwide, ocean drownings are less common than freshwater drownings, probably because fewer children swim unsupervised in the ocean, and increasing numbers now swim on patrolled beaches in more-​developed countries, where prevention and the rescue efforts of life-​saving associations have proved effective. Rate of drowning varies with climate, availability of beaches, lakes, and other natural and artificial water sources, provision of life-​saving services, improvements in designs and rules for water craft, and the use of life jackets. Rock fishing carries a high risk of drowning and near-​drowning. A genetic basis has been suggested for unexplained drowning or near-​drowning. Ethnicity White American children aged 1–​4 years drown twice as often as African-​American children of the same age. These accidents usu- ally happen in residential swimming pools. Conversely, in the age group 5–​19 years, African-​Americans drown more often than white Americans. Australian aboriginal children drown more often than non​indigenous children. Worldwide, fatal drowning is generally more prevalent in indigenous races than in others. Alcohol Alcohol affects vision, balance, movement, and reasoning and is a major risk factor for drowning in adolescent and adult swimmers, water craft operators, and passengers, who fall overboard while in- toxicated. At the time of rescue, resuscitation, or death, 25–​50% of adult and adolescent victims of drowning had some exposure to alcohol. Pathophysiology Aspiration is usual in drowning. Earlier figures had suggested that approximately 10–​15% of victims of drowning had not aspirated water but recent figures show an incidence of only 2%. In these cases, death may result from laryngeal spasm and asphyxia during submersion. Early animal studies in anaesthetized dogs showed that spontaneous respiratory efforts continued for around 60 s after im- mersion. Complete cardiac arrest supervenes after 4.5 min (mean 262 s). Recent Chinese bronchoscopic studies in anaesthetized dogs whose lungs were filled with seawater showed that the bronchi fill with bronchoalveolar fluid, causing increasing blood lactate dehydrogenase-​L and alkaline phosphatase levels. Electron micros- copy shows injuries to type II alveolar epithelial cells, thickened re- spiratory mucosa, and platelet adherence. Haemodynamic effects following inspiration of liquid are similar. There is a rapid fall in cardiac output, while pulmonary capil- lary wedge pressure, central venous pressure, and pulmonary vas- cular resistance increase. Reduction in the dynamic compliance of the lungs is similar, following inspiration of all types of solutions. However, aspiration of large volumes of hypertonic seawater draws fluid into the lung from the circulation by osmosis, resulting in fluid-​ filled, non​ventilated, but perfused alveoli incapable of normal gas exchange while aspiration. Conversely, aspiration of large amounts of hypotonic freshwater may cause sufficient absorption of fluid into the circulation from the alveoli to cause both acute hypervolaemia and haemolysis. Within 1 h, pulmonary oedema develops, resulting in a decrease in circulating blood volume. Early studies suggested that 85% of human drowning victims aspirated 22 ml/​kg of water or less, but it has been estimated that about 10% of body weight of water may be absorbed from the lungs during freshwater drowning. Since the brain has a limited ability to maintain adenosine tri- phosphate levels anaerobically when cerebral blood flow is reduced, it suffers irreparable damage within 4–​6 min. Death or severe neuro- logical impairment occurs after submersion of more than 5–​10 min. However, in hypothermic conditions, brain activity may be restored after up to 60 min of submersion apnoea. Bystanders’ estimates of submersion time are usually inaccurate. Hypothermia (See Chapter 10.3.3.) A low body temperature generally indicates the severity of the drowning incident. Sudden immersion in ice water causes a reflex cardiorespiratory response, called ‘cold shock’, causing an initial gasp and hyperventilation despite hypocapnia and also hyper- tension. Continuous aspiration of cold water results in rapid core cooling, while the circulation is intact. Such victims may survive with little or no neurological deficit after long submersion with extreme hypoxia. After submersion for a maximum of 10 min in

10.3.4  Drowning 1693 water at 16°C, a good outcome can be predicted in 96.6% of victims. New evidence supports the use of mild hypothermia for periods of 12–​24 h in comatose drowned victims. A 6-​year-​old boy, who presented with a rectal temperature of 16.4°C after a 65-​min sub- mersion, survived, apparently neurologically intact when his blood was rewarmed in increments of 3°C over 96 min. However, later neuropsychological testing revealed cognitive difficulties, espe- cially global memory impairment, despite the fact that MRI and magnetoencephalography were normal. In adults, success is less common. A notable exception was a 31-​year-​old man with a core temperature of 23°C who had been asystolic for 80 min, but was warmed by cardiopulmonary bypass and recovered. Despite discouraging data from animal studies, recent reports suggest that in hypothermic submersion-​associated cardiac ar- rest, adrenaline and vasopressin may help to achieve the vaso- pressor response needed to restore spontaneous circulation prior to rewarming. This treatment could obviate prolonged mechanical cardiopulmonary resuscitation, or the use of extracorporeal circula- tion. It has proved effective in restoring spontaneous circulation, but one patient died of multiorgan failure 15 h later. In warm-​water drowning there appears to be no statistically significant correlation between duration of submersion and survival. Causes of drowning Drowning occurs in many different situations: after accidental im- mersion in people with little or no swimming ability, with head and neck injury, following cardiac and neurological emergencies (including epilepsy), as a result of alcohol and drugs, metabolic disease (including hypoglycaemia), and even child abuse and murder. In countries with long coastlines and many bathing beaches, drowning is common and is often caused by swimmers being caught in ‘rip currents’ (also known as ‘rips’) (large vol- umes of water returning back out to sea after onshore wave action) (Fig. 10.3.4.1); there is no such entity as the frequently suggested ‘undertow’. Swimmers in difficulty may be able to shout for help but, contrary to public opinion, those who are drowning do not. Most drowning victims adopt a characteristic vertical position in the water—​legs hanging vertically, head tilted back for quick ex- halation and inhalation before bobbing underwater, with no time or sufficient breath to call for help. After only 20–​60 s, victims may submerge permanently. Clinical features Prognostic indicators None of the recent developments in assessment, treatment, or equipment has improved survival rates among submersion victims. Prevention and rapid rescue remain the most effective means of re- ducing the toll. The key to a successful outcome and return to pro- ductive, full life is early bystander cardiopulmonary resuscitation, early and aggressive advanced life support methods (Fig. 10.3.4.2), induced hypothermia when appropriate, careful rewarming, including extracorporal membrane oxygenation, and extracorporal warming where needed. However, up to 25% of drowning victims presenting to the hospital emergency department will die and a further 6% suffer neurological sequelae. The prognosis cannot Fig. 10.3.4.1  Australians swimming in the sea at Petrel Cove near Victor Harbour despite notice warning of rip currents. Courtesy of D A Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1694 be predicted from the initial clinical presentation, laboratory, or electrophysiological examinations, but those with a normal chest radiograph on admission usually survive; Pao2 may not relate to radiographic appearances. Although the cause and pathophysio- logical changes of pulmonary insufficiency vary depending on the type and volume of fluid aspirated, serum electrolyte and haemo- globin concentrations (or haematocrit) do not predict survival. Cardiovascular status This is a better guide to outcome than neurological status. Mortality is high in victims with circulatory arrest on admission, but those in sinus rhythm with reactive pupils, who are neurologically re- sponsive at the scene of immersion, have good outcomes. Those who are asystolic on arrival at hospital and remain comatose for more than 3 h have a poor prognosis unless they are hypothermic. Rapid hypothermia from sudden submersion in cold water (see Chapter 10.3.3) carries a relatively good prognosis, compared to in- sidious hypothermia developed during prolonged submersion that results in cardiac arrest. Neurological status Victims who are alert when medical help arrives have a survival rate approaching 100%, whereas the prognosis in those who are coma- tose with fixed, dilated pupils is poor. Among victims with impaired consciousness, 87% will survive without neurological defects and 2% with minor defects, while 11% will die. Approximately 40–​50% of victims who are comatose on arrival have incapacitating brain damage. Those with no spontaneous limb movements and abnormal brain stem function 24 h after the accident have a poor neurological outcome. A modified Glasgow Coma Score is helpful in evaluating neuro- logical injury. A score of 5 or less predicts a mortality risk of over 80%. Pupil reactivity at the time of arrival differentiates survivors from fatalities but could not differentiate between those with minor or incapacitating neurological deficits. Fixed, dilated pupils or total flaccidity are associated with a high mortality. Victims with any motor activity, even posturing or seizures, in the immediate postresuscitation period had a higher incidence of intact survival, but abnormal posturing that persisted or recurred after 12–​24 h in- dicated a high probability of severe brain damage. An abnormal CT scan in the initial 36 h following an immersion incident is associated with a dismal prognosis. MRI with qualitative and quantitative MR spectroscopy data may allow a more accurate prognosis. The gravity of the early clinical state, the estimated duration of cardiorespiratory arrest, and the severity of the hypothermia, seiz- ures, and paroxysmal electroencephalogram (EEG) activity do not determine the severity of submersion injury encephalopathy. Early EEG patterns with moderate background activity, sleep patterns, response to auditory and painful stimulations, and numerous β- rhythms suggest a good outcome, whereas bad outcomes are sug- gested by high voltage, rhythmic δ-waves; biphasic sharp waves; monotonous EEG, ‘burst-​suppression’ pattern, and absence of β- rhythms. Children who show no spontaneous movements and have abnormal brainstem function 24 h after submersion injury are likely to suffer severe neurological deficits or death. Treatment Victims of submersion injury must be treated immediately for venti- latory insufficiency, hypoxia, and the resulting acidosis. A successful outcome depends on early effective resuscitation at the scene and on competent intensive life support. In-​water resuscitation is effective within 5 mins of the shore, or longer, if the victim shows signs of in- creased activity after the initial breaths of the shore. Immediate Laying victims on their side for assessment of the airway and breathing will assist drainage of any excess water from the airways and lungs (Fig. 10.3.4.2). If necessary, on-​site cardiopulmonary re- suscitation should be started as soon as possible using supplemental oxygen if available, preferably in the highest concentration (e.g. (a) (b) Fig. 10.3.4.2  (a, b) Cardiopulmonary resuscitation including defibrillation being carried out on the beach by Australian surf
life-​savers, in a man who suffered a cardiac arrest while swimming. Courtesy of P J Fenner.

10.3.4  Drowning 1695 bag–​valve–​mask). An oropharyngeal airway, endotracheal tube, or laryngeal mask airway should be inserted in comatose victims, if suitably skilled personnel are present. Pulse oximetry is helpful. Vomiting and regurgitation are significant risks during early resus- citation. Respiratory and cardiopulmonary arrest may occur after an apparently successful rescue, mandating close, uninterrupted monitoring, and the early administration of oxygen to all immer- sion victims. At the hospital On arrival at the hospital, after a clear airway and cardio-​circulatory support have been established, arterial blood gas tensions and pH should be measured. The pH of the blood will indicate whether there is a residual metabolic acidosis after a substantial period of hypoxia. Mechanical ventilation may be necessary with positive end-​ expiratory pressure, or continuous positive airway pressure to main- tain arterial oxygen pressure above 10 kPa with an inspired oxygen fraction below 0.6. After both freshwater and seawater aspiration, large volumes of intravenous colloid are usually needed while circulating blood volume and cardiac output are estimated. Freshwater aspiration is more likely to cause pulmonary oedema. A central venous catheter or pulmonary artery catheter helps to assess the effective circulating blood volume to guide fluid therapy. Failure of response to intra- vascular replacement with 20 ml/​kg of colloid is an indication for starting inotropes. Steroid and prophylactic antibiotic therapy do not appear to increase the chance of survival. Inpatient treatment Extracorporeal membrane oxygenation has been proved to be ef- fective after drowning. Patients with severe hypoxaemia may have irreversible cerebral ischaemia. A  3-​year-​old drowned girl in re- fractory cardiorespiratory arrest was successfully resuscitated using cardiopulmonary bypass, and then extracorporeal membrane oxy- genation for 4 days. Despite a prolonged period in a vegetative state, she later made an almost complete neurological recovery. If adult respiratory distress syndrome occurs, it is usually within 6 h of admission. There is evidence that alveolar epithelial barrier function is well preserved even after aspiration of large quantities of hypertonic salt water. Surfactant has been used with some suc- cess in refractory respiratory failure in near-​drowning, but it is expensive. The risk of secondary pneumonia is high, especially when mech- anical ventilation has been used. Although prophylactic antibiotics are not recommended, broad-​spectrum antibiotics may be required. Mild reversible renal impairment is rare. Initial serum creatinine, marked metabolic acidosis, abnormal urinalysis, or significant blood lymphocytosis are markers of impending acute renal failure. Prevention of drowning Swimming pools and natural bodies of water are the greatest risk to young children. Preventive measures include public media educa- tion and campaigns, parental education and supervision, training in cardiopulmonary resuscitation, better safety standards, and safety devices such as the fencing of swimming pools. The number of pool drownings in Brisbane, Australia, decreased after legislation made pool-​fencing compulsory. Strategies for the prevention of drowning should also consider hazards in rural areas. Multilingual notices on public beaches are important (Fig. 10.3.4.3) but are often ignored (Fig. 10.3.4.1). Swimming ability and safety skills of young children can be im- proved by training. Education of the public is essential. In Australian surf, only 17% of rescues and resuscitations, up to 95% of them suc- cessful, occurred within patrolled areas, while 55% (62% of them successfully) occurred outside patrolled areas. Resuscitation success rates fell with increasing distance from patrolled areas. Among non-​ boating drownings in Australia, 4.7% are among overseas tourists, 89% of whom drown in the ocean. An adult should supervise all epileptic children and infants aged under 3 while they are in the bath. Currently, up to 89% of children aged 35 to 59 months and 6% of those younger than 3 years of age are bathed without adult supervision. Drownings associated with boating and personalized water craft can be prevented by using life jackets (personal flotation devices), but as many as 50% of boaters do not use them. Efforts to increase their use should target adolescents, adults, and boating enthusiasts, especially those using motor boats. In Alaska’s commercial fishing industry, specific measures designed to prevent drowning after ves- sels have capsized and sunk have proved successful. In most age groups, more men drown than women. This probably reflects men’s overestimation of their abilities, and perhaps greater alcohol consumption. Middle-​aged men dominate the group who die of cardiac events (mostly on the surface) (Fig. 10.3.4.2). Fatalities from breath-​holding hypoxia during diving tend to occur in young males. Hyperventilation to increase breath-​hold time is a dangerous practice that should be discouraged. Drownings are rare at super- vised water parks, thanks to the large number of lifeguards on duty. Fig. 10.3.4.3  Multilingual talking sign warning of dangers on an Australian beach. Courtesy of P J Fenner.

10.3.5 Lightning and electrical injuries 1696

10.3.5 Lightning and electrical injuries 1696

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1696 FURTHER READING Bierens J, et  al. (eds) (2006). The handbook on drowning. Springer, Berlin. Hasibeder WR (2003). Drowning. Curr Opin Anaesthesiol, 16, 139–​45. Idris AH, et  al. (2003). Recommended guidelines for uniform re- porting of data from drowning:  the ‘Utstein style’. Resuscitation, 59, 45–​57. Papa L, Hoelle R, Idris A (2005). Systematic review of definitions for drowning incidents. Resuscitation, 65, 255–​64. Piette MH, De Letter EA (2006). Drowning: still a difficult autopsy diagnosis. Forensic Sci Int, 163, 1–​9. Salomez F, Vincent JL (2004). Drowning:  a review of epidemi- ology, pathophysiology, treatment and prevention. Resuscitation, 63, 261–​8. van Beeck EF, et al. (2005). A new definition of drowning: towards documentation and prevention of a global public health problem. Bull World Health Organ, 83, 853–​6. 10.3.5  Lightning and
electrical injuries Chris Andrews ESSENTIALS Lightning Lightning strikes are rare accidents but carry a 10% case fatality, killing 0.1–​0.3 per million population each year. During thunderstorms, the risk is increased by sheltering under trees or by being on open water, on tractors, or in open fields or in outdoor activity. Lightning is considered to cause instant asystole. It is suspected clinically if someone is found collapsed in the open with linear or feathered burns (see next), exploded clothing, and ruptured ear- drums. Victims are safe to handle, with most victims showing keraunoparalysis (cold, pulseless, mottled extremities, not unlike compartment syndromes). Immediate cardiopulmonary resuscita- tion is mandatory. Survivors might develop complications including pain syndromes and psychological sequelae. Burns are generally of minor consequence, unlike electrical injury. The most disabling con- sequences of the injury are generally the psychological sequelae. Electric shock The term electrocution implies death from electric shock. Survivors are said to have been exposed to or received an electric shock. Electrocution is the fifth most common cause of workplace death, mainly affecting utilities, mining, and construction labourers. Domestic electrical accidents are common, where contact with over- head power lines, faulty power tools, and particularly using extension cords, are the most common causes, with metal ladders and an- tennae being particularly dangerous. Prevention is by implementing codes of safe practice. Victims of electric shock might suffer prolonged attachment to the source of electric current and must be disconnected from the source before resuscitation. A victim still attached to a source of current is dangerous to touch. The most expedient method of disconnection is to turn the offending power off. Consequences of the shock include: (1)  Immediate scale—​ventricular fibrillation, which is the mechanism of fatality, sometimes leading to persistent cardiac dysfunction in survivors; (2)  Neurological and muscular manifestations, both early and late, including paraesthesiae, and pareses. In the early stages, gener- alized convulsions, respiratory embarrassment, due to tetanic spasm, and rhabdomyolysis may occur; (3)  Burns, which might be severe and require expert surgical atten- tion. Electroporation (a special form of cell membrane disrup- tion by electric fields) contributes to cell death; delineation using polaxamers can direct the extent of surgical debridement. (4)  In the longer term, persistent paraesthesiae and pareses occur, with a particular fatiguability of the musculature, evidenced
as loss of stamina. Visual and auditory dysfunction may occur. The most disabling consequence is the psychological syn- drome, which has its own unique character. Introduction Lightning is a powerful force; it provides spectacular displays and has evoked an extensive mythology. It is poorly understood in med- ical terms and it is only recently that its characterization, physically and psychologically, has been made more plain. The comparatively recent discovery and distribution of electricity have had an equally profound effect, and provide truth to the adage that ‘electricity is a good servant and a bad master’. Generation of electricity occurs in several ways, but in the final analysis electrons are imbued with energy by a generator of electric current. From any given generator notionally one conductor em- anates to be a source of current in the form of electrons, and this current is delivered by the conductor to a location where electrical work can be done. This conductor is generally termed the ‘active’ conductor. A second conductor is required to return the electrons in a less energetic state to the generator for re-​energization. This is the return path, or the neutral conductor. The electrical system is referenced to earth, and the second conductor might indeed be earth. Alternatively, a common method of injury occurs when an individual contacts the active conductor and earth simultaneously. While it could be argued that earth reference creates danger on the one hand, on the other technical considerations show that system stability and predictability is enhanced by earth reference. Epidemiology Lightning injury It is not appreciated that only a small fraction of lightning strike to individuals lead to fatality. The latest accepted case fatality of

10.3.5  Lightning and electrical injuries 1697 lightning shock is 10% and is around 0.3 per million population in the United States of America each year, but fewer than 0.1 per mil- lion in the United Kingdom. In the early part of the 20th century, most people struck by light- ning were outdoor workers (67%) and outdoor recreationalists (28%). Nowadays, the breakdown is 45% and 50%, respectively, explained by changes in social and work habits. Indoor strikes (e.g. by current conducted through communication or power ap- paratus) continue to account for about 5% of these accidents, but few fatalities. Men are more often injured than women (1.67 males to 0.33 females); the age group most at risk is 20–​29  years. Risky situ- ations include sheltering under trees (particularly), on open water, on tractors, in open fields, and playing outdoor sports, like golf. Regional differences correlate well with storm activity and popula- tion density in that area. Electrical injury Electrocution ranks fifth in the causes of workplace death, ac- counting for the death of 10 000 workers each year in the United States, with a further 10 million being injured. Most of the victims work for utility companies, followed by mining and construction workers. Contact with power lines causes 53% of fatal shocks, and contact with power tools accounts for a further 22%. The most dan- gerous times of day seem to be between 10.00 a.m. and 3.00 p.m. on Mondays, Tuesdays, and Thursdays. Most of the victims are trade and labouring staff; sales, clerical, and professional categories are at least risk. Metal ladders and antennae are particularly dangerous and can easily be hoisted into overhead power lines. Codes of safe practice are written accordingly. In domestic situations, contact with overhead lines by ladders and poles is again important. Faulty, including amateur, repair of equipment, and faulty apparatus, wiring, and especially power and extension cords account for large numbers of deaths and in- juries. Children are at particular risk. Death from domestic electric shock has shown a marked decrease with the introduction of re- sidual current devices. These are items of good practice, and sense if current is diverted from the active supply to earth rather than neutral, and then interrupt the supply in a matter of milliseconds. These will not ameliorate every accident, but are considered to act in 80% of cases, notably active conductor to earth shocks. While they ameliorate the fatal effects in these cases, victims still can have several of the stated consequences. Mechanisms of injury Lightning injury Lightning injury may be sustained in five separate ways:

  1. A person might be struck directly. This might represent the most common cause of fatality.
  2. A nearby object, such as a tree or a building, might be struck, and someone in direct contact with it might receive a shock.
  3. Without direct contact an arc might ‘jump’ to a nearby person from the struck object, thereby generating a ‘side flash’. This is particularly dangerous in the unwise event of sheltering under a tree.
  4. As current disperses away from the base of a strike to ground, an individual might divert current flowing in the ground through themselves. This is termed ‘shock due to increase in earth poten- tial’, or simply ‘earth potential rise’.
  5. A recently documented mechanism is the transient flow of cur- rent due to corona and streamer formation around the upper parts of an individual. The current to provide these streamers flows from ground through the individual to project upwards to reach a descending stroke from a cloud. It has been found that both cardiac and respiratory function cease instantaneously under lightning strike, the cardiac arrest being asys- tolic. Cardiac function restarts under local pacemaker control, but respiratory function does not recommence, and secondary hypoxic cardiac arrest supervenes. The major cranial orifices are portals of entry for lightning cur- rent, and from there the pathways to the brainstem are short. Respiratory function is thought to be affected in the brainstem. Fluid channels (cerebrospinal fluid and blood) might be channels to the myocardium. The QT prolongation resulting from lightning injury can predis- pose to episodic arrhythmias. There is no evidence at all for one dictum, viz., that lightning in- hibits body metabolism. Resuscitation is as urgent as with any other injury. There is no evidence for the notion sometimes quoted that resuscitation can be delayed in a lightning victim. There is similarly no evidence that metal on the head, or the presence of a mobile tele- phone (cellphone), predisposes to being struck. Electrical injury With electric shock, it is important to assess the points of entry and exit and the pathway of current through the body. Once the pathway has been determined, a locus for expected injury can be established, and the flow of current can be estimated from the applied voltage divided by the impedance of the proposed pathway. Most imped- ance is in the skin barriers, and is non​linear. There is initial (contact) impedance, which decreases as current flow continues. Impedance also varies with time since application, contact surface area, and frequency. Contacts can be with the active and neutral (return) conductor, or with the active conductor and earth, with earth providing the return path to the generator. Indeed, these considerations become blurred as most countries operate an MEN distribution system (Multiple Earthed Neutral). In this system, the earth referencing comes about as the neutral conductor is connected to earth regularly along its route back to the generator. Return paths are then shared between earth and neutral. For currents with a frequency of 15–​100 Hz, externally ap- plied from hand to hand, or hand to foot, relevant variables char- acterizing the current are the threshold of perception (0.5 mA) and ‘let go’ current (10 mA). The threshold of fibrillation (where threat to life can occur) is a higher threshold again. For example, there is a 50% chance of fibrillation when 2000 mA is conducted for 10 ms, or at the other extreme 100 mA conducted for 10 s, in a hand to foot path. Direct internal application of less than 200 µA to the heart muscle can induce fibrillation. Dangerous cur- rent levels as well as impedance parameters are documented in standards.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1698 Joule heating might account for tissue damage in the path of the current. It can be calculated from the power dissipation in the tissue—​the square of the tissue current (often hard to estimate) times its impedance. The complex phenomenon of electroporation, where cell membranes are breached by the electrical induction of unstable pores in the membrane, can also lead to cell death. The complex nature of internal electric fields leads to internal field damage which is difficult to quantify and predict. Presentation of the injured person The presentations and physical consequences of electrical and lightning injury are different, and hence the application of prin- ciples from one to the other is invalid, physically. On the other hand, the psychological consequences are similar, and they may be considered together. Lightning injury A witnessed strike offers the best chance of resuscitation. The victim is not dangerous to touch, and does not constitute a risk to the res- cuer. Immediate cardiopulmonary resuscitation is paramount. It has been stated that: Any person found with linear burns and clothing exploded off should be treated as the victim of a lightning strike. Feathering burns are pathognomic of lightning injury and occur in no other type of in- jury. . . . Another complex diagnostic of lightning injury includes linear or punctate burns, tympanic membrane rupture, confusion, and out- door location . . . . In assessing a lightning victim, the following features must be sought. Cardiovascular and pulmonary consequences Asystolic arrest is the main cardiac event in lightning injury. Electrocardiographic (ECG) findings can take many forms, with ischaemic and infarct forms. They almost invariably resolve com- pletely over time. Alterations in QT interval and arrhythmias of many kinds are seen. ECG changes might not occur until late in the course, and so are poor diagnostic tools. Respiratory arrest is common. A person not suffering cardiopulmonary arrest is highly unlikely to die from lightning strike. Neurological consequences Direct neural injury can occur both centrally and peripherally. All forms of intracranial bleeding have been reported. Direct cerebral damage particularly affects the basal ganglia, cerebellum, and brain- stem. Dural tears, scalp haematomata, and fractures are also seen. Seizures occur as a result of anoxia and injury. Peripheral nerve injury, including autonomic injury, can give pro- longed and long-​lasting disability, which often develops late. Other late features include spinal cord atrophic paralysis, cerebellar ataxia, incoordination, paraesthesiae, and aphasia. Continuing complex re- gional pain syndromes may be seen. Keraunoparalysis and burns More than 70% of victims demonstrate keraunoparalysis. This is a syndrome of cold, pulseless, mottled, and asensory extremities. The syndrome resembles a compartment syndrome and occurs in the line of passage of the strike current. It resolves spontan- eously within 24 h with no sequelae, and requires no surgical intervention. Burns are of minor consequence in lightning injury, and again require little intervention. Entry and exit burns might be full thickness, though small. Arborescent (feathering) burns re- semble fern-​like patterns on the skin (Fig. 10.3.5.1). Their aeti- ology has been convincingly shown to be due to an inflammatory response following field arcing across the skin. They fade within 24 h. Linear burns are due to the passage of hot plasma tongues over the skin. Eschar is simply allowed to separate without fur- ther treatment. Flash might be seen, like sunburn or welder’s flash, from the profound radiation of the strike. Sheet burns resulting from efflux of hot plasma can be a variant of linear burning, since both seem to follow moisture and sweat lines. There might be contact burns from metal such as buckles and coins. It is said that these are thermal, but doubt has been cast on this. Eye, ear, and explosive injuries The explosive force of the lightning insult blasts clothing apart (Fig. 10.3.5.2), and may cause percussive injury to the lungs and abdominal viscera. Tympanic membranes are usually ruptured, perhaps from the explosive force of the strike. Percussive eye in- jury, particularly retinal, has been reported. Cataracts can develop much later. Other injuries Renal and haematological damage have occasionally been re- ported. Several writers examining lightning strike during preg- nancy suggest that outcome for a fetus is poor, independent of Fig. 10.3.5.1  Example of keraunographic marking. Courtesy Dr Ajay Mahajan (Mahajan AL, Rajan R, et al. (2007). Lichtenberg figures: cutaneous manifestation of phone electrocution from lightning. J Plast Reconstr Aesthet Surg, 61(1),111–​13). Reprinted with permission from Elsevier.

10.3.5  Lightning and electrical injuries 1699 that for the mother. Menstrual and sexual difficulties have been reported, though the latter could be more of a psychological nature. Electrical injury In contrast with lightning injury, victims might suffer prolonged attachment to the source of electrical current, making them dangerous to touch. Before resuscitation, they must be separated from the current source, and this usually means interrupting the current flow at the supply point. Burns are far more serious, and might merit intense surgical treatment. The likelihood of internal burning (remembering the possibility of electroporation) might require further surgical inter- vention. Cardiac and respiratory burns can also occur. Cardiovascular consequences Ventricular fibrillation is the most common fatal cardiac ar- rhythmia following electrical injury. Cardiopulmonary resus- citation is urgently required. Electricity suppliers have standard first aid/​resuscitation procedures. Cardiac dysfunction can persist for long periods, and ECG changes may not resolve. Recent studies indicate the importance of vascular chan- nels for the passage of current. Alterations in vascular function have been documented, and further, similar alterations in ves- sels remote from the current passage have been demonstrated at the same time. These findings point to humoral factors in mediating electrical injury. The findings also cause us to ques- tion whether most of the damage is mediated by current passage through fluid channels, rather than previously claimed nerve conduction. Neurological and muscular consequences Neural injury can be categorized into early and late syndromes, at cerebral, cord, and peripheral levels. Early tetanic muscular contraction locks the victim on to the electrical conductors. This tetany can compromise respiratory function. Neurological injury might be hard to distinguish from hypoxic and vasospastic injury. Similarly, neural injury is often hard to separate from ischaemic injury due to vessel spasm. Early and late generalized convulsions can occur. Pareses and paraes- thesiae might develop, both early and late. There is a characteristic easy fatiguing of the musculature, which resembles a profound loss of stamina. In the long term, complex regional pain syndromes and other chronic pain syndromes must be considered. Burns Burns are often severe in electrical injury and merit much treatment effort. Arc and flame burns and contact burns from current entry and exit are seen. For example, tetanic gripping of the electrical con- ductor causes grasp burns to the hand. Severe internal thermal or electroporation damage can occur. The management is largely surgical. Joints, ligaments, and tendons might be severely damaged by the heat generated, and osteonecrosis might be seen. Amputations are relatively common. Other aspects Widespread muscle damage generates myoglobin that must be cleared by the kidney with a severe risk of renal damage. Other metabolic and biochemical disturbances secondary to hypoxia might develop. Massive hyperkalaemia has implications for the use of depolarizing muscle relaxants. Eye damage includes retinal damage, with punctation and detach- ment, and thermal damage to other media. During follow-​up the possibility of ocular pareses and cataracts must be recognized. After shock during pregnancy, the prognosis for the fetus is poor. Non​focal injury is more likely in survivors. Psychological consequences of electrical
and lightning injuries Although electrical and lightning injuries are fundamentally dif- ferent in nature and management, their psychological sequelae are similar. Sequelae (‘remote symptoms’, so-​called as they occur with no evidence of current passage through the brain) can be profoundly disabling. They most often come to attention via Worker’s Compensation and litigation, and if ignored, do a great disservice to a victim. They can persist for many years and might never resolve. The changes have the features of organic psycho- logical consequences. This is the overall picture, though it might be suggested that a psychological reaction to the loss of function and continuing pain might be functionally generated, although many believe that findings from neuropsychological testing strongly suggest an organic basis. Indeed, recent research in depression indicates that the hippocampus is found to diminish in volume, implicating that such volume changes, together with changes in cortisol and (a) (b) Fig. 10.3.5.2  Reconstruction of external result of a lightning strike. Courtesy Professor Mary Ann Cooper, University of Illinois, Chicago.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1700 brain-​derived neurotrophic factor, are important in causing de- pressive syndromes. Research in the sequelae of electric shock also shows reduction in hippocampal volume as a direct conse- quence of peripheral shock. Several of these symptoms develop over time postinjury, and might not be present for some weeks after the initial examin- ation where the emphasis is more on the physical symptoms. The state of the victim, in totality, deteriorates for 12–​18 months fol- lowing the injury, then improves to achieve stability 18 months to 3 years from the injury though falling short of original premorbid function. The remote symptoms follow two major categories—​emotional, personality, and behavioural consequences on the one hand, versus cognitive and higher function consequences on the other, with effects on cognitive agility and speed. Table 10.3.5.1 is a summary of psychological findings. An important part of evaluating a victim is to submit them to neuropsychological testing. The aim of such testing, in part, is to objectify the dysfunctions that are seen, and if possible, subdivide them more specifically. It is unfortunate that in evaluating an electric shock victim or a lightning victim, too many examiners regard the psycho- logical syndrome as either representing malingering, or of no consequence, to the victim’s severe detriment. It has been docu- mented that assessment by those unfamiliar with the injury over- looks or wrongly diagnoses over 90% of the resulting syndrome features. Treatment of the injuries First, urgent and life-​saving treatment must be administered. Secondly, there must be surveillance for delayed sequelae, and thirdly, long-​term monitoring for morbidity, including cataract formation and psychological problems. Lightning injury First, the casualty is resuscitated and evacuated. Cardiopulmonary resuscitation is continued until medical emergency help is obtained. Ventilation and cardiac support might be required. ECG monitoring must be used to detect subtle effects like QT prolongation. Associated trauma is treated. In the long term, patients are observed for development of pain syndromes. Ocular and auditory functions are monitored. Sensitivity to the psychological sequelae is required, and preventive interviewing might be useful. Carbamazepine, gabapentin, clonazepam, flecainide, and mexilitine are useful to control neurally derived pain and resulting weakness. An antidepressant is a useful adjunct to this. Electrical injury Urgent life support is indicated. Ventilatory and inotropic sup- port and correction of arrhythmias may be required together with correction of biochemical abnormalities, and attention to any myoglobinuria. For burns, progressive debridement and/​or ampu- tation might be needed. Specialized rehabilitation may be required. Associated trauma is treated. Ocular and auditory functions are monitored, and psycho- logical disturbances are reviewed. In the long term, surveillance is similar to lightning injury. Psychological elements In all cases, the management of the psychological syndrome is paramount and might be the greatest determinant of long-​term functional capability. Awareness of the impact of the injury on em- ployment and relationships and social networks is fundamental. Cognitive and computer aids are being developed, and cognitive support is important. Personality change, social withdrawal, sexual dysfunction, with loss of earning capacity, places a large strain on marital partnerships, and this requires special support. An antidepressant such as an SSRI/​SNRI (selective serotonin re- uptake inhibitor/​serotonin and norepinephrine reuptake inhibitor), or a tricyclic such as clomipramine, may be useful. Agomelatine is showing some promise in this application. Early and continuing neuropsychological assessment and support is desirable. Controversy The place of polaxamers in discovering the extent of electropor- ation and in delineating debridement levels is of great interest. A  polaxamer is a polymer with a central hydrophobic chain flanked by two hydrophilic chains. There are multiple vari- ants of each of these elements. In this structure, however, they Table 10.3.5.1  Proportions and enumeration of remote injuries Proportions Memory disturbance 71% Concentration disturbance 63% Aggression and irritation 67% Wariness and phobia 58% Loss of mental powers 50% Social isolation 38% Sleep disorder 38% and others including confusion, word finding disability, anxiety, depression, and learning disorders. Subdivisions of remote injuries Memory and learning deficits Globally more specifically Visual Visuospatial Auditory 19% 35% 38% 62% Verbal learning deficit 54% Verbal fluency deficit 46% Concentration and attention deficit 46%/​42% Executive function deficit 38% Reduced executive speed 62% and others including general and verbal IQ decrease, dynamic coordination decrease, slowed information processing, deficit in fine motor skills, phobia, and anxiety and depression.

10.3.6 Diseases of high terrestrial altitudes 1701

10.3.6 Diseases of high terrestrial altitudes 1701

10.3.6  Diseases of high terrestrial altitudes 1701 resemble the structure of cell membranes, and they can be patched into cell membrane defects including those caused by electroporation. On the one hand they might offer a therapeutic restorative for damaged membranes, and on the other a radio-​ labelled polaxamer might be used for imaging the extent of damage. The mechanisms of the psychological disability and remote in- jury are beginning to be elucidated. Victims are frequently written off as malingering or simply depressed, when a more extensive syndrome exists. Expert evaluation is highly desirable, especially if litigation or compensation is involved. The useful duration of monitoring of otherwise asymptomatic people has not been determined. A pressing need is the formulation of the psychological syndrome into a Diagnostic and Statistical Manual (DSM) formulation. FURTHER READING Andrews CJ (2006). Further documentation of remote effects of elec- trical injury, with comments on the place of neuropsychological testing and functional scanning. IEEE Trans Biomed Eng, 53, 2102–​13. Andrews CJ, et al. (1992). Lightning injuries: electrical, medical and legal aspects. CRC Press, Boca Raton, FL. Andrews CJ, Reisner AD (2017). The neuropsychological conse- quences of lightning and electrical injury: review and hypotheses for causation. Neural Regen Res, 12, 677–​86. Cherington M, Cooper MA (eds) (1995). Seminars in Neurology,
15 (3, 4). Cooper MA (1980). Lightning injuries: prognostic signs for death. Ann Emerg Med, 9, 134. Cooper MA, Andrews C, Holle R (2005). Lightning injuries. In: Auerbach P (ed) Wilderness medicine, 4th edition, pp. 73–​111. Mosby, St Louis, MO. Cooper MA, Andrews CJ (2005). Disability, not death, is the issue in lightning injury. Proc Int Conf On Lightn Stat Elec, Boeing, Seattle, WA. Hendler N (2005). Overlooked diagnoses in chronic pain: analysis of survivors of electric shock and lightning strike. J Occup Env Med, 47, 796–​805. Kurtulus, A., Acar, K., Adiguzel, E., Boz, B. (2008). Hippocampal neuron loss due to electric injury in rats: a stereological study. Legal Medicine (Tokyo), 22, 2671–​5. Lee RC, Capelli-​Schellpfeffer M, Kelley K (eds) (1994). Electrical injury: a multidisciplinary approach to prevention, therapy & re- habilitation. Ann N Y Acad Sci, 720. Lee RC, Cravalho EG, Burke JF (1992). Electric trauma. Cambridge University Press, Cambridge. Morse MS, Berg JS, TenWolde RL (2004). Diffuse electrical injury: a study of 89 subjects reporting long-​term symptomatology that is re- mote to the theoretical current Pathway. IEEE Trans Biomed Eng, 51, 1449–​59. Park K, et al. (2012). Alterations in arterial function after high-​voltage electrical injury. Critical Care, 16, R25. Reisner A (2013). Possible mechanisms for delayed neurological damage in lightning and electrical injury. Brain Injury, 27,
565–​9. 10.3.6  Diseases of high
terrestrial altitudes Tyler Albert, Erik R. Swenson, Andrew J. Pollard, Buddha Basnyat, and David R. Murdoch ESSENTIALS Ascent to altitudes above 2500 m leads to exposure to hypobaric hypoxia. This affects performance on first arrival at high altitude and disturbs sleep, but physiological changes occur over time to defend arterial and tissue oxygenation and allow the individual to adjust. This process of acclima- tization includes (1) an increase in the rate and depth of breathing; and (2) an increase in red cell mass, and in red cell 2,3-​diphosphoglycerate. Acclimatization is no longer fully possible at extreme altitude (>5800 m) and the exposed individual will gradually deteriorate. Altitude illness results from a failure to adjust to hypobaric hypoxia at altitude. Risk is increased by ascent to higher altitudes, by more rapid gain in altitude, and (in some people) genetic predisposition; the condition may be avoided in most cases by slow, graded ascent. Clinical presentation occurs soon after arriving at a new altitude, most often manifest as one of three conditions: Acute mountain sickness A common condition that presents with non​specific symptoms, including headache and anorexia. The victim is likely to be apathetic, but clinical examination is generally unremarkable. Mild cases usu- ally resolve with rest and avoidance of further ascent. Those whose symptoms fail to resolve (or worsen) should descend immediately. Treatment with acetazolamide (which can also be used as prophy- laxis) or dexamethasone is often given in severe cases. High-​altitude cerebral oedema An uncommon condition that typically presents with worsening symptoms of acute mountain sickness and ataxia, with progressive neurological symptoms including behavioural changes, confusion, and impairment of consciousness. Papilloedema and focal neuro- logical signs may be present. Treatment is urgent, with the most im- portant measure being descent. Oxygen or simulated descent using a portable hyperbaric chamber can be helpful. Dexamethasone is widely recommended (and can be used as prophylaxis). High-​altitude pulmonary oedema A relatively uncommon condition with significant mortality that typ- ically presents with dyspnoea and cough. Signs include low-​grade fever, tachycardia, tachypnoea, basal crepitations, and (in late dis- ease) cyanosis. Treatment is urgent, with the most important measure being descent. Oxygen should be given if available. Simulated des- cent using a portable hyperbaric chamber can be helpful. Nifedipine reduces pulmonary artery pressure, relieves symptoms, and is usually given (and can be used as prophylaxis). Chronic mountain sickness and other medical problems Chronic mountain sickness (Monge’s disease) is a disease of adults who reside for prolonged periods at high altitude and develop polycythaemia and eventually cor pulmonale. Symptoms appear to

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1702 resolve with descent, but treatment with venesection has been at- tempted in those who remain at altitude. High-​altitude pulmonary hypertension has been described in both infants and adults, pre- dominantly native lowlanders who ascend to and reside at high alti- tude: this also appears to resolve on descent. Pre-​existing medical conditions are mostly little affected by as- cent to altitude, but people particularly likely to be affected by hypoxia/​ altitude include those with (1) coronary ischaemia and a strongly positive exercise treadmill test; (2) sickle cell disease or trait; (3)  chronic pulmonary disease, especially pre-​existing pulmonary hypertension from any cause. Introduction High-​altitude regions of the world, once considered remote and ac- cessible to only a few individuals, are in fact home to over 200 million people living permanently above 2500 m in Asia, South America, and North America. In South America, miners and astronomers work for long durations at altitudes over 4500 m, while an equal number of tourists spend time at these altitudes in such activities as trekking, skiing, and pilgrimages. Anyone at these altitudes is sus- ceptible to high-​altitude illnesses which, if not recognized and ap- propriately treated, can spoil the experience of alpine environments, impair work ability, and in the worst scenarios cause death. The high-​altitude environment Although the percentage of oxygen in ambient air remains con- stant at 21%, barometric pressure decreases with increasing altitude, leading to a corresponding fall in partial pressure of oxygen (Po2) (Fig. 10.3.6.1). At 2500 m the barometric pressure and inspired Po2 are about 75% those of sea level values. At 5000 m, which is close to the maximum for permanent human habitation, Po2 is about half of the sea-​level value. On the summit of Mount Everest (8848 m), Po2 is about one-​third of the sea-​level value. In human physiology, the definitions of high altitude in Box 10.3.6.1 are commonly used. In addition to hypoxia, several other characteristics of the high-​ altitude environment are important to understand. Temperature decreases approximately 1°C for every 150 m rise in altitude, irrespective of latitude, so that high-​altitude areas are considerably colder. Ultraviolet penetration increases by approximately 12% for each 1000 m altitude gain, increasing the risk of sunburn, ultraviolet kera- titis, and other sun-​related problems. The low humidity at high altitude contributes greatly to fluid loss and dehydration, as does the increased solar radiation, which might be very much exaggerated by reflection from snow. While air quality is generally superior in mountainous areas, heavily populated areas in valleys that are prone to inversions can be associated with unhealthy concentrations of irritating gases and particulates. Effects of hypobaric hypoxia A reduction in exercise capacity is a major effect of ascent to high altitudes. Maximal oxygen consumption decreases by approxi- mately 10% for each 1000-​m gain in altitude above 1500 m, and this does not recover appreciably with acclimatization. Reduced max- imal oxygen consumption occurs by a reduction in mitochondrial Po2, interfering with the function of the electron transport chain and adenosine triphosphate synthesis, or through central inhibition in the brain to prevent injurious tissue hypoxia. Genetic factors are partly responsible for individual variations in exercise performance at high altitudes. For example, Tibetans, a population having resided at high altitude for more than 25 000 years, have a variant of the gene encoding a transcription factor (hypoxia inducible factor -​2, HIF-​2), and mountaineers who perform well at high altitudes have a higher expression of an angiotensin-​converting enzyme gene variant. Sleep can also be disturbed at high altitude. There is difficulty getting to sleep, frequent arousals, less rapid eye movement (REM) time, and a decrease in slow-​wave sleep. Periodic breathing, char- acterized by episodes of hyperpnoea followed by apnoea, is rela- tively common among travellers over 2500 m. It is thought to result from instability of the respiratory control system through enhanced hypoxic drive or response to CO2, and can be minimized by the use of acetazolamide. Hypoxaemia during apnoeic episodes during periodic breathing likely accounts for many of the arousals from sleep that are experienced at high altitude. 800 700 600 500 400 300 200 0 Barometric pressure (mmHg) Barometric pressure falls as altitude increases Altitude (m) Mont Blanc 4807 m Mount Everest 8848 m 8000 10000 6000 4000 2000 Fig. 10.3.6.1  Change in barometric pressure with altitude. © Pollard, Andrew J. and Murdoch, David R., The High Altitude Medicine Handbook (3e). Oxford: Radcliffe Medical Press Ltd; 2003. Reproduced with the permission of the copyright holder. Box 10.3.6.1  Altitude—​definitions Intermediate altitude (1500–​2500 m) Physiological changes due to hypobaric hypoxia (such as reduced exer- cise performance, increased ventilation, increased haematopoiesis) are detectable, but arterial oxygen saturation remains above 90%. Acute altitude illness is unlikely. High altitude (2500–​3500 m) Acute altitude illness is common following rapid ascent to this altitude. Very high altitude (3500–​5800 m) Arterial oxygen saturation falls below 90%. Acute altitude illness is common and marked hypoxaemia can occur during exercise and sleep. Extreme altitude (>5800 m) Further acclimatization cannot be achieved, progressive physiological deterioration occurs, and survival cannot be maintained permanently. Marked hypoxaemia occurs at rest.

10.3.6  Diseases of high terrestrial altitudes 1703 Neuropsychological changes at high altitude are often quite subtle, although various changes in mental performance have been docu- mented. Attention span, short-​term memory, arithmetic ability, and decision-​making can all be impaired at altitudes over 4000 m. Acclimatization Acclimatization is the process of gradual adjustment to high alti- tude hypoxia. In general, it is a physiological process involving a series of adjustments that occur in both the short term (minutes to hours) and long term (days to weeks). These changes enhance oxygen delivery to cells and efficiency of oxygen use. In contrast, ‘altitude adaptation’ refers to physiological changes that occur over longer time periods (decades and generations) and confer advan- tages for life at high altitude. Acclimatization reduces the impact of high-​altitude hypoxia, but is insufficient to fully return the body to its sea-​level normoxic capacities. The principal steps involved in high altitude acclimatization can be summarized as follows: Ventilation Hyperventilation is the most important feature of acclimatiza- tion and serves to defend alveolar Po2 and thus arterial and tissue Po2. Increases in the rate and depth of breathing, termed the hyp- oxic ventilatory response, are mediated by hypoxic stimulation of peripheral chemoreceptors located mainly in the carotid bodies. Hyperventilation increases alveolar Po2 in the face of decreased in- spired Po2, while also reducing alveolar Pco2 leading to a respira- tory alkalosis. Although initially the alkalosis limits the full hypoxic ventilatory response, eventually it is somewhat compensated for by increased urinary bicarbonate excretion over several days and subse- quently followed by a further increase in ventilation over the course of several weeks, reflecting an increase in intrinsic hypoxic sensitivity of the peripheral chemoreceptors. The degree of hyperventilation in response to high-​altitude hypoxia can be profound and, with exercise, it may be what subjectively causes a person to have to stop. Alveolar ventilation increases approximately fivefold on the summit of Mount Everest, where inspired Po2 is less than one-​third of its sea-​level value and arterial Pco2 values as low as 10 mmHg at rest have been recorded. Blood Although erythropoietin secretion is increased within 2 h of ascent to high altitude, it takes many days to weeks for an increase in red cell mass to occur. This ultimately increases the oxygen-​carrying capacity of the blood and permits greater oxygen transport to tis- sues. The shift in the oxyhaemoglobin dissociation curve to the right, which occurs on ascent and is due to an increase in red cell 2,3-​diphosphoglycerate, which favours unloading of oxygen in the tissues. However, this particular adjustment is offset by the shift to the left caused by the respiratory alkalosis mentioned previously, leaving the P50 essentially unchanged. Circulation Although there is an abrupt increase in cardiac output on ascent to high altitude, there follows a progressive decrease in stroke volume and maximal cardiac output is reduced at all levels of exercise, including maximal exercise. Although there is no evidence for insuf- ficient myocardial oxygenation, there is disagreement about whether the myocardium is depressed by hypoxia. There is an immediate redistribution of blood flow: coronary and cutaneous flow both fall, cerebral and retinal flow increase, and renal flow decreases initially although returns to normal with acclimatization. Fluid balance Central blood volume increases with ascent to high altitude due to peripheral venous constriction. This, in turn, can suppress anti- diuretic hormone and aldosterone to induce a diuresis along with an independent action of peripheral chemoreceptors stimulated by hypoxia to reduce renal sodium and water reabsorption. Extreme altitudes Acclimatization in adults seems to be possible up to about 5500 m. Above this height, there is a fine balance between adjustment to high altitude and deterioration as a result of chronic hypoxia. The term ‘high-​altitude deterioration’ refers to the general deterior- ation in physical condition that occurs after lengthy stays at extreme altitudes. Typical features include progressive weight loss (fat and muscle), worsening appetite, poor sleep, and increased lethargy. The most extreme altitudes, such as the summit of Mount Everest, are very close to the limit of human tolerance to hypoxia. Indeed, early estimates indicated that all available oxygen on the summit of Mount Everest would be required for basal oxygen uptake, with none left for physical exertion. Alveolar gas samples taken near the summit of Everest (8400 m; barometric pressure, 36.3 kPa) show an inspired Po2  = 6.27 kPa, and alveolar Po2  =  4.00 kPa. Mean arterial gas values at this altitude were: Po2 3.3 kPa; Pco2 1.8 kPa; pH 7.5; oxygen saturation 54%. Consequently, it is extraordinary that some humans are able to climb to this height without using supplemen- tary oxygen. Missing in our efforts to fully describe acclimatization and the profound altitudes that some individuals can attain are pro- found changes at the levels of the microcirculation and cellular me- tabolism. Many of these might involve the hundreds of genes that are up-​ and down-regulated by gene transcription factors called hypoxia inducible factors 1 and 2 (HIF). Several reports have suggested mild, possibly permanent, defects in cognition in climbers who have as- cended to extreme high altitudes. Illnesses due to altitude Until high-​altitude acclimatization has occurred, lack of physio- logical compensation for hypobaric hypoxia can manifest as altitude illness. Acute mountain sickness, high-​altitude pulmonary oedema, and high-​altitude cerebral oedema are recognized distinct clinical syndromes of altitude illness. Acute mountain sickness is both the most common and the quickest to develop, and while it often pre- cedes high-​altitude pulmonary oedema or high-​altitude cerebral oe- dema, either can occur in its absence. Development of altitude illness usually occurs after a rapid ascent, although there is considerable variation in susceptibility between in- dividuals. Genetic factors are likely to be important in determining susceptibility, but several other factors are contributory and are dis- cussed in the following paragraphs. Acute mountain sickness Incidence rates of acute mountain sickness vary with the absolute altitude gained and the speed of ascent. Some 30–​50% of those who ascend to 4500 m on a standard trek in the Himalayas develop acute mountain sickness. Its incidence is greater at higher altitudes and

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1704 with greater gains in altitude, and might be precipitated by physical exertion, although this remains controversial. Some people have a history of recurrent acute mountain sickness, suggesting individual susceptibility. Typically, symptoms of acute mountain sickness begin 6–​12 h after ascent to altitudes over 2500 m. The familiar features are non​specific symptoms that are readily confused with many other illnesses and include: • headache •  nausea • vomiting • fatigue • anorexia • dizziness • sleep disturbance For practical purposes, people ascending to altitude with unex- plained symptoms that include the above mentioned should be as- sumed to have acute mountain sickness. The headache is typically worse at night, upon lying down, and with a Valsalva manoeuvre. Anorexia is often pronounced. Clinical examination is typically un- remarkable, although it might reveal some peripheral oedema. There might be tachycardia and elevated core temperature. Typically, the person with acute mountain sickness is apathetic and withdrawn, often seeking solitude (see Fig. 10.3.6.2). The aetiology of acute mountain sickness is unknown. It has been argued that it is a mild form of cerebral oedema since it often precedes development of high-​altitude cerebral oedema, and the symptoms of acute mountain sickness include symptoms of head- ache and nausea consistent with a mild increase in intracranial pressure. Brain imaging studies have not found increases in global brain volume or oedema in the first 6–​10 h after exposure to hyp- oxia despite symptoms of acute mountain sickness, but it might be that these techniques lack sufficient sensitivity, particularly if the oedema is not global. However, brain volume does increase after longer exposure to hypoxia. Some recent evidence suggests that oxi- dative stress might be involved in the development of acute moun- tain sickness. Other theories include hypoxia-​mediated irritation of the trigeminal system from hypoxia-​mediated increase in radical oxygen species or nitrosative radicals. Mild acute mountain sickness usually resolves if the victim avoids further ascent and rests. Paracetamol (acetaminophen), non​steroidal anti-​inflammatory agents, and other analgesics might bring relief from headache. Those whose symptoms fail to resolve or worsen should descend immediately. More severe symp- toms will also resolve with descent, but some people will require treatment to facilitate descent. Supplementary oxygen might be beneficial if available. Treatment with acetazolamide (250 mg or- ally, three times daily), or dexamethasone (4 mg orally, four times daily) can be useful in severe cases. Acetazolamide is a carbonic anhydrase inhibitor, which increases renal excretion of bicar- bonate to induce a metabolic acidosis. The hyperventilation in- duced by the respiratory compensation improves oxygenation and helps relieve symptoms. Portable hyperbaric chambers are widely used on trekking routes and can be pressurized to simulate a 500–​700 m descent, tempor- arily relieving symptoms in order to facilitate a true descent. These chambers are inflated with a hand or foot pump to achieve the baro- metric pressure of a lower elevation. CO2 is removed by the airflow generated by the pumping action, and a CO2 scrubber is included in some models. Acute mountain sickness can be avoided or prevented in most cases by carefully graded ascent. Above 3000 m, a rate of ascent of 300–​600 m per day with a rest day every 1000 m will avoid symp- toms for most people. However, there is considerable individual variation. For some destinations, itineraries are rapid enough to in- duce symptoms of acute mountain sickness in a large proportion Fig. 10.3.6.2  A trekker in Kunde Clinic at 3840 m with symptoms of acute mountain sickness (headache, anorexia, lethargy, and malaise) en route to Everest Base Camp. Courtesy of T Albert.

10.3.6  Diseases of high terrestrial altitudes 1705 of travellers. For this reason, prophylaxis with acetazolamide, started on the day before ascent over 3000 m (125–​250 mg twice daily or 250–​500 mg daily of the slow-​release preparation) is fre- quently recommended for prevention and can be quite effective. Since the side effects induced by this drug may be serious (allergic reactions), mimic acute mountain sickness in some respects (head- ache, nausea, anorexia), or be intolerable (paresthesiae), several test doses should be tried before it is used for prophylaxis during ascent. There is evidence that acetazolamide doses of 125 mg twice daily can be effective in some people. Dexamethasone can also be useful for prophylaxis, although its mechanism of action is unknown. One study found that the inhaled corticosteroid budesonide, in doses unable to generate significant blood levels to act at the brain, was as effective as oral dexamethasone. However, two subsequent studies could not confirm this efficacy. In recent trials, ginkgo biloba has been found to be ineffective and thus is not advised, particularly since it is not a regulated pharmaceutical and preparations can vary in content and contamination with other substances. Theophylline reduces periodic breathing during sleep, but not oxygenation, and probably has little utility in prophylaxis. High-​altitude cerebral oedema Unlike acute mountain sickness, which is quite common among travellers to high altitude, high-​altitude cerebral oedema and high-​altitude pulmonary oedema are relatively uncommon. High-​ altitude cerebral oedema is more typical after ascent to altitudes over 4000 m, but cases have been described even at the modest ele- vation of 2100 m. As with acute mountain sickness, higher rates are found at the highest altitudes and after more rapid ascent. At 4000 to 5500 m, rates of 1% have been described amongst trek- kers. High-​altitude cerebral oedema is usually preceded by acute mountain sickness and is frequently associated with high-​altitude pulmonary oedema since the greater hypoxemia occurring with high-​altitude pulmonary oedema is the equivalent to suddenly being at an even higher altitude (see following paragraphs). Symptoms of acute mountain sickness have usually been present for 1 to 2 days before the onset of high-​altitude cerebral oedema. Risk factors for the development of high-​altitude cerebral oedema are probably similar to those recognized for other forms of altitude illness. high-​altitude cerebral oedema might be more common in the presence of intracranial space-​occupying lesions such as cysts or tumours. Worsening symptoms of acute mountain sickness and ataxia are typical early signs of development of high-​altitude cerebral oedema. Behavioural changes (being irrational, withdrawn, or exuberant), confusion, and a change in conscious level leading to coma might ensue. Papilloedema might be present. Both focal neurological signs and cranial nerve lesions can occur. Brain imaging studies show typical signs of cerebral oedema, particularly in the splenium of the corpus callosum, with changes in white matter signal, compression of sulci, and blunting of gyri. Lumbar puncture, if undertaken, re- veals elevated pressures but is otherwise normal. High-​altitude cere- bral oedema is indistinguishable clinically from many other causes of compromised cerebral function. There is a very high mortality among those who develop coma. It is likely that high-​altitude cerebral oedema is a vasogenic oe- dema resulting from injury to the blood–​brain barrier, following disturbances in cerebral autoregulation. Cytotoxic oedema from release of mediators in the central nervous system in response to hypoxia might also contribute. In view of the seriousness of high-​altitude cerebral oedema, treat- ment is urgently required, and the most important measure is imme- diate descent. Oxygen therapy or simulated descent using a portable hyperbaric chamber can improve oxygenation and symptoms and thus facilitate descent. Intravenous dexamethasone (8 mg followed by 4 mg, orally four times per day) might improve symptoms and is widely recommended. high-​altitude cerebral oedema tends to re- cover more slowly than other forms of altitude illness and ataxia is often the last sign to disappear. High-​altitude cerebral oedema is probably prevented by slow, graded ascent (as with acute mountain sickness described previ- ously) and prophylaxis with dexamethasone might be beneficial for those with a risk of the condition. High-​altitude pulmonary oedema High-​altitude pulmonary oedema typically occurs within 4  days of ascent to altitudes over 2500 m and might be accompanied by symptoms of acute mountain sickness or high-​altitude cerebral oe- dema. It presents more frequently with increasing altitude: 1–​2% of people may be affected at 4500 m, but rates as high as 10% have been reported after rapid ascent at this altitude. Previous episodes of high-​altitude pulmonary oedema for an individual repeating the same ascent rate and altitude gain confer a 60–​70% likelihood of re- currence. It can also occur in those who have become acclimatized at one altitude and then make a further ascent. A form of high-​altitude pulmonary oedema, known as re-​entry high-​altitude pulmonary oe- dema, can occur in adults and often more in children living at high altitude after returning home from a lowland vacation. It is a serious form of altitude illness and is associated with fatality when not man- aged urgently and appropriately. High-​altitude pulmonary oedema is slightly more common in men than women. The risk of it developing is increased by cold, rapid ascent, exertion, coexistent viral infection, and possibly by drugs or ingestions that cause respiratory depression, such as al- cohol. Individual susceptibility is well recognized and those who have previously suffered from it appear to be more susceptible in the future. There are various genetic associations described including pulmonary surfactant protein A, HLA DR6, HLA DQ4, epithe- lial sodium channel protein, and endothelial nitric oxide synthase. People with raised pulmonary blood flow or an exaggerated hypoxic pulmonary vascular response might also be more susceptible (i.e. those with atrial septal defect, unilateral absence of a pulmonary ar- tery or a chronic respiratory condition). People with high-​altitude pulmonary oedema typically present with dyspnoea out of proportion to others in the group and cough. The breathlessness is made worse with exertion and can present with blood-​tinged frothy sputum and frank haemoptysis. Other symp- toms include chest pain, orthopnoea, nausea, insomnia, headache, dizziness, and confusion. Low-​grade fever is a common finding to- gether with tachycardia, tachypnoea, crackles on auscultation of the chest, and cyanosis in late disease. Signs of right ventricular enlargement can be present with an accentuated pulmonary second heart sound and right ventricular heave. Oxygen saturations are decreased from the prevailing levels of those acclimatizing well, the electrocardiogram shows right axis de- viation, tachycardia, and peaked P-​waves, and the chest radiograph

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1706 shows pulmonary oedema (Fig 10.3.6.3), often more prominently on the right. In patients who have been studied with cardiac catheterization during high-​altitude pulmonary oedema, pulmonary arterial pres- sure is often quite elevated, but pulmonary wedge pressures are normal, ruling out heart failure. Most people who are susceptible to high-​altitude pulmonary oedema show an abnormal rise in their pulmonary arterial pressure at sea level during exposure to hyp- oxia or on normoxic exercise. Hypoxic pulmonary vasoconstriction varies almost fivefold among healthy persons, and at sea level or low altitudes poses no problems. The clinical syndrome is not unique, and similar findings occur in other respiratory diseases, including acute bacterial or viral pneu- monia and pulmonary embolism. High-​altitude cough (see fol- lowing paragraph) might also cause diagnostic confusion. As described earlier, high-​altitude pulmonary oedema appears to result from an exaggerated hypoxic pulmonary vasoconstrictor response. Because there is regional unevenness in the response, this leads to downstream pressure increase in over-​perfused areas that lead to capillary leak, either from pressure induced changes in permeability or at the extreme capillary stress failure and frank bleeding into the alveolar space. Once high-​altitude pulmonary oedema is recognized, the victim must descend. Even descent of several hundred metres can be enough to raise the barometric pressure sufficiently to increase inspired oxygen tensions and re- duce pulmonary artery pressure. Without appropriate manage- ment, it can be fatal, and further ascent should not be undertaken. In a mountain environment, immediate descent might be im- possible because of weather or other circumstances. The patient might be so breathless that they cannot move and, as mentioned earlier, at a very high risk to develop high-​altitude cerebral oe- dema. Adjunctive therapies might improve symptoms and allow descent. The patient should be encouraged to sit up to prevent orthopnoea. Oxygen should be given if available. As a strong inhibitor of hypoxic pulmonary vasoconstriction, nifedipine (20 mg slow release preparation, four times daily) reduces pul- monary artery pressure and relieves symptoms. Side effects of ni- fedipine include headache, dizziness, and postural hypotension, but the drug is generally well tolerated. Other pulmonary vaso- dilators such as hydralazine, phentolamine, inhaled nitric oxide, and sildenafil have been used and might be beneficial, but nifedi- pine is most widely used. Portable hyperbaric chambers are often available on commercial trekking routes (see Fig. 10.3.6.4). They can simulate descent, improve oxygenation, and relieve symptoms. Devices that help provide positive expiratory airway pressure might also improve oxygenation. The risk of high-​altitude pulmonary oedema is reduced by slow, graded ascent. Above 3000 m, a rate of ascent of 300–​600 m per day with a rest day every 1000 m is recommended. Nifedipine (20-​mg slow release preparation, three times daily) and other calcium channel blockers, dexamethasone (8 mg twice daily), inhaled salmeterol, a β2-​andrenoceptor agonist, and phosphodiesterase-​5 inhibitors are effective for prophylaxis in those known to be susceptible. Acetazolamide, which inhibits hypoxic pulmonary vasoconstriction by a mechanism unrelated to its action as a carbonic anhydrase in- hibitor, might also be effective in its prevention, although it has not been formally tested for this indication. Dexamethasone, while ef- fective in prevention for high-​altitude pulmonary oedema-​susceptible persons, does not appear to be efficacious in treatment. This might stem from the fact that its action largely is dependent on changes in gene transcription that require many hours to become effective. High-​altitude retinal haemorrhage Retinal haemorrhages occur frequently at altitudes of 5000 m or higher, even in those without acute mountain sickness or high-​ altitude cerebral oedema. Although usually asymptomatic, they can cause visual problems if the macula is involved (Fig. 10.3.6.5). The causes of high-​altitude retinal haemorrhage can include in- creased cerebral blood flow, Valsalva manoeuvres (during exer- tion or coughing), polycythaemia, and hypoxia-​mediated capillary endothelial permeability. In most instances of high-​altitude retinal haemorrhage without altitude illness, descent might not be neces- sary. The haemorrhages usually resolve within days to weeks. If vi- sion is compromised or there is concomitant altitude illness, descent is mandatory. Peripheral oedema Swelling of the hands, face, and ankles commonly occurs at high altitude and might not be related to acute mountain sick­ ness, high-​altitude cerebral oedema, or high-​altitude pulmonary Fig. 10.3.6.3  Chest X-​ray and CT scan of a patient with high altitude pulmonary oedema. Courtesy of E. Swenson.

10.3.6  Diseases of high terrestrial altitudes 1707 oedema. Anasarca is seldom seen. Descent or diuretics will treat the oedema. High-​altitude cough Dry hacking cough is a common, bothersome problem at high altitude, and has caused rib fractures in some severe cases. High-​ altitude cough is probably multifactorial in origin: water loss from the airways due to hyperventilation of cold-​dry air, post nasal drip, acute mountain sickness, high-​altitude pulmonary oedema, and bronchoconstriction have all been invoked as possible causes of alti- tude related cough. Breathing through a silk scarf, throat lozenges, and steam inhalation might be helpful. If there is nasal congestion, a decongestant nasal spray is useful. Effects of high altitude on pre-​existing
medical conditions Hypertensive patients should continue their medications at high altitude, and the vast majority of hypertensive skiers and trekkers do very well despite a transient rise in the blood pressure. Some pa- tients with labile hypertension might have a sudden, dangerous rise in their blood pressure at high altitude and, for them, blood pres- sure monitoring might be necessary. The exaggerated blood pres- sure response to high altitude is apparently mediated by increased α-​adrenergic activity. Hence an α-​blocker might be useful or, as shown recently in a well-​conducted randomized study, angiotensin receptor blockade with valsartan up to 3500 m is effective. People with stable coronary artery disease tolerate intermediate and high altitudes relatively well, even while exercising. This might be partly attributable to the marked reduction in maximal exer- cise at high altitude, which reduces myocardial oxygen demand and maximal heart rate. Animal experiments at high altitude have also demonstrated down-​regulation of the β-​receptors of the heart. However, travel to high altitude can precipitate new-​onset an- gina, although it is unclear whether this is related to exertion or to hypobaric hypoxia as such. People with cardiac risk factors or with previous myocardial ischaemia, coronary artery bypass surgery, or angioplasty are considered to be at high risk if they have a strongly positive exercise treadmill test. Although cold air and exercise are triggers for asthma, many asth- matics remain well at high altitude. This might be due to decreased density of the air, a lack of allergens, or the increase in steroid hor- mones produced under hypoxic stress. However, it is important for asthmatics going up to high altitude to carry their medicines (including an oral corticosteroid) with them and have a well-​defined action plan to deal with any exacerbation. Many people with well-​controlled epilepsy can venture safely to high altitudes, but there remain some causes for concern. Hyperventilation leading to hypocapnia and hypoxia are themselves triggers for seizure activity. People with sickle cell disease or trait are at high risk of sickle crises above 2000 m, and should avoid staying at altitude. Diabetics might find that increased energy expenditure at high altitude alters carbohydrate and insulin requirements. Consequently, rapidly acting insulin, close monitoring, availability of oral and intravenous glucose, and knowledgeable companions are im- portant. Loss of diabetic control due to intercurrent infections, like diarrhoea, is also possible. It is especially important that insulin pre- parations never freeze, and one option for prevention is by keeping them close to the body. Pre-​existing pulmonary hypertension from any cause can be a problem at high altitude and comes close to being an absolute contraindication to travel above very modest altitudes (<2500 m). Mitral stenosis, kyphoscoliosis, and congenital cardiac defects with pulmonary hypertension can predispose to high-​altitude pulmonary Fig. 10.3.6.4  An example of a portable hyperbaric chamber being set up in the Alps. Courtesy of E Swenson.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1708 oedema and are therefore hazardous at high altitude. Those with chronic obstructive pulmonary disease frequently report increased dyspnoea and reduced exercise tolerance when they ascend to high altitude. Other illnesses at altitude Focal neurological problems are occasionally encountered at high altitude. Transient ischaemic attacks and strokes, cerebral venous thrombosis, subarachnoid haemorrhage, high-​altitude syncope, delirium, transient global amnesia, cranial nerve palsies, cortical blindness, and amaurosis fugax have all been reported. However, it is unclear whether these deficits are related to hypoxia of high altitude and most need to be distinguished from acute mountain sickness and high-​altitude cerebral oedema. Venous thrombosis has been reported at high altitude, although its association with high-​altitude exposure is uncertain. Cases of cerebral venous thrombosis at high altitude have been reported in previously asymptomatic people with heterozygous protein C and S deficiency and antiphospholipid syndrome. Risk of thrombosis might have been increased by dehydration and polycythaemia. Immobility during inclement weather, coupled with dehydration and polycythaemia, might also predispose to deep vein thrombosis leading to pulmonary embolism. Unanswered questions about the effects of high altitude on blood coagulation include the use of oral contraceptives and whether prophylactic aspirin prevents throm- bosis. Subjects taking warfarin might experience changes in the drug’s metabolism and monitoring may be needed, thus restricting high altitude destinations to those places where international normalized ratio measurements are possible, such as large ski re- sorts with medical centres. Chronic mountain sickness (Monge’s disease) and high-​altitude pulmonary hypertension Excessive erythrocytosis, severe hypoxaemia, and, in some cases, moderate to severe pulmonary hypertension leading to cor pulmonale are the features of chronic mountain sickness. This dis- ease, also known as Monge’s disease, is a disease of long-​term resi- dents of altitudes above 2500 m. Besides South America, where this condition was originally described, chronic mountain sickness has also been documented in Colorado and in the Han Chinese population in Tibet. Migration to low altitude cures the problem. Venesection, acetazolamide and, possibly, angiotensin converting enzyme inhibitors have been shown to be helpful. It is important to distinguish chronic mountain sickness from chronic obstructive pulmonary disease. High-​altitude pulmonary hypertension is now the accepted term for diseases that include adult subacute mountain sickness and infantile subacute mountain sickness. Unlike chronic mountain sickness, which is characterized by erythrocytosis, the primary fea- ture of this condition is pulmonary hypertension leading to heart failure. The adult form has been described exclusively in Indian sol- diers living at extreme altitudes for prolonged periods. The infantile form has been seen mainly in Han Chinese immigrants in Tibet. These conditions bear a striking pathophysiological resemblance to brisket disease in cattle. Descent from high altitude completely cures the problem. High-​altitude pulmonary hypertension of chronic onset is also well described. FURTHER READING Bärtsch P, et al. (2004). Acute mountain sickness: controversies and advances. High Alt Med Biol, 5, 110–​24. Bärtsch P, et  al. (2005). Physiological aspects of high-​altitude pul- monary edema. J Appl Physiol, 98, 1101–​10. Bärtsch P, Swenson ER (2013). Acute high-​altitude illnesses. NEJM, 368, 2294–​302. Basnyat B, Murdoch DR (2003). High-​altitude illness. Lancet, 361, 1967–​74. Baumgartner RW, Siegel AM, Hackett PH (2007). Going high with preexisting neurological conditions. High Alt Med Biol, 8, 108–​16. Dehnert C, et al. (2005). Identification of individuals susceptible to high-altitude pulmonary oedema at low altitude. Eur Respir J, 25, 545–51. (a) (b) Fig. 10.3.6.5  Contrasting retinal appearances on Mount Everest. (a) Normal retina of a well-​acclimatized and well-​oxygenated climber shortly after reaching the summit (8848 m). (b) Retinal haemorrhages in a poorly acclimatized climber at the North Col (7100 m). Copyright Daniel Morris, Newcastle.

10.3.7 Radiation 1709

10.3.7 Radiation 1709

10.3.7  Radiation 1709 Imray C, et al. (2011). Acute altitude illness. BMJ, 343, d4943. Lipman GS, et  al. (2012). Ibuprofen prevents altitude illness:  a
randomized controlled trial for prevention of altitude illness with nonsteroidal anti-​inflammatories. Ann Emerg Med, 59, 484–​90. Luks AM (2012). Clinician’s corner:  what do we know about safe
ascent rates at high altitude? High Alt Med Biol, 13, 147–​52. Luks AM, Swenson ER (2007). Travel to high altitude with pre-​ existing lung disease. Eur Respir J, 29, 770–​792. Maggiorini M, et al. (2006). Both tadalafil and dexamethasone may reduce the incidence of high-​altitude pulmonary edema: a random- ized trial. Ann Intern Med, 145, 497–​506. Rimoldi SF, et al. (2010). High-​altitude exposure in patients with car- diovascular disease:  risk assessment and practical recommenda- tions. Prog Cardiovasc Dis, 52, 512–​24. Roach RC, Hackett PH (2001). Association of polymorphisms in
pulmonary surfactant protein A1 and A2 genes with high-​altitude pulmonary edema. Chest 128, 1611–​20. Schoene RB (2004). Unraveling the mechanism of high altitude pul- monary edema. High Alt Med Biol, 5, 125–​35. Swenson ER, Schoene RB (2014). High altitude pulmonary edema.
In: Swenson ER, Bartsch P (eds) High altitude: human adaptation to hypoxia, pp. 405–28. Springer Publications, New York, NY. 10.3.7  Radiation Jill Meara ESSENTIALS Ionizing radiation Ionizing radiation has sufficient energy to break chemical bonds and produce charged ions in living tissue. These changes might cause cell death, but breaks of both strands of a DNA molecule that do not kill a cell can be a precursor of cancer. Excluding medical exposures, natural radiation accounts for most human exposure, which produces health effects that might be (1) stochastic, where the probability of manifesting the effect depends on the radiation dose, including carcinogenesis and in- duction of heritable defects; (2) psychological, especially following accidental exposures; and (3) tissue reactions, occurring when suf- ficient cells are killed after exposure to radiation doses above a cer- tain threshold, including the acute radiation syndrome (radiation sickness) and radiation burns. Prevention—​legislative dose-​limits prevent tissue reactions and re- duce the risk of stochastic effects, although all doses should be kept as low as reasonably achievable. Non​ionizing radiation Ultraviolet radiation affects primarily (1) the skin, causing sunburn in the short term and skin cancer in the long term; and (2) the eye, causing photokeratitis and photoconjunctivitis (arc eye, snow blind- ness) in the short term, and conjunctival and corneal disorders and cataracts in the long term. Information on the health effects of other types of non​ionizing radiation (e.g. radiofrequency microwaves, and power-​frequency electric and magnetic fields) is less robust, but controls are recom- mended to prevent those health effects that are established. Introduction The term radiation applies to emissions in the electromagnetic spectrum. Only ionizing radiation is energetic enough to cause ionization of matter. There are natural sources of ionizing radi- ation, such as radon gas or cosmic rays, and manufactured sources, such as X-​rays and radioactive isotopes produced in nuclear re- actors. Excluding medical exposures, natural radiation accounts for most human exposure. Some types of non​ionizing radiation are also health hazards. These include radiant heat, ultraviolet ra- diation, radio waves, microwaves, and power-​frequency electro- magnetic fields. Historical perspective The dangers of ionizing radiation became apparent almost as soon as experiments with radioactive materials began. In the early 20th century, radiologists often calibrated their machines by the dose causing erythema on their hands. Many, including Marie Skłodowska-​Curie and her daughter Irène Joliot-​Curie, died of radiation-​induced cancers. Despite universal exposure to natural background radiation, there is a general fear of ionizing radiation, especially that associated with nuclear power and nuclear weapons. The hazards of certain types of non​ionizing radiation, such as sunburn and electrical discharge in thunderstorms, are well known, but the health of pioneers in non​ionizing radiation re- search was not affected. Recently, the safety of power-​frequency and radiofrequency fields—​at the levels to which the public are exposed—​has been questioned. However, hypotheses about possible long-​term health effects, such as induction of cancer, lack biologically plausible mechanisms or confirmation by high-​quality epidemiological studies. Ionizing radiation: Mechanism of harm Atoms, radioactivity, and radiation Isotopes of some elements are unstable and undergo radioactive decay. The time taken for half of a given quantity to decay is called the half-​life, which ranges from fractions of a second to thousands of years, depending on the particular isotope. The unit of radio- activity is the becquerel (Bq). 1 Bq equals 1 atomic disintegration per second. The natural radionuclide potassium-​40 (40K), with a half-​life of 1.250 × 109 years, makes up 120 parts per million of all potassium on Earth. Since there is about 4000 Bq of 40K in an average person, there are about 14 million radioactive disintegra- tions per hour from 40K inside the average human body. Unstable isotopes (radionuclides) decay and release energy as subatomic particles (α-​ or β-​particles) or γ-​rays. X-​rays are pro- duced by bombarding a metal target with electrons in a vacuum.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1710 Neutrons are produced during nuclear fission reactions. These vary in the extent to which they can penetrate the body and can interact with tissues and cells. α-​Particles are densely ionizing and are stopped by the dead layer of the skin, but constitute a hazard if taken into the body. β-​Particles can penetrate the body to the depth of a few centimetres; X-​ and γ-​rays penetrate the body and, if not absorbed, pass through it. Lead shielding is needed to pro- tect against X-​rays and γ-​rays. These properties of radiation affect the location and extent of cellular damage following exposure and dictate the protective methods required. Ionizing radiation has sufficient energy to break chemical bonds and produce charged ions in living tissue. Most of these changes are inconsequential, others can be repaired, but there is a finite prob- ability that damage might cause cell death. Breaks of both strands of a DNA molecule might not kill a cell, but they are known to be a precursor of cancer. Measuring radiation risk Acute cell damage depends on the energy imparted by the radiation. The mean energy absorbed per unit mass of tissue (absorbed dose) is measured in gray (Gy). 1 Gy is equal to 1 joule (J) deposited per kilo- gram of tissue. Radiation and tissue-​weighting factors are used to convert the absorbed dose in Gy to an effective dose in sieverts (Sv). This allows external and internal exposures from all types of ion- izing radiation to be integrated into one dose, on the basis of equality of stochastic risk. The United Kingdom average annual individual natural background radiation dose is 2.3 mSv. The typical dose from an anteroposterior chest radiograph is 0.02 mSv and that from an abdominal CT scan is 10 mSv. Health effects of exposure to ionizing radiation There are three types of health effects associated with exposure to ionizing radiation:  stochastic effects, psychological effects, and tissue reactions. • In stochastic effects, the probability of manifesting the effect depends on the radiation dose and include carcinogenesis and induction of heritable defects. Radiation-​induced cancer is clinically and pathologically indistinguishable from idiopathic cases. Risks at low-​radiation doses are extrapolated from animal, experimental, and epidemiological studies at higher doses as- suming a linear no-​threshold model. This implies that there is no ‘safe’ radiation dose, but very small exposures convey very small risks. The absolute cancer risk per unit of radiation dose (risk coefficient) is estimated to be 5.5%/​Sv. Recent data suggest that cardiovascular system damage might also be a stochastic effect of radiation, with a similar risk coefficient as for cancer induction. • Psychological effects are found especially following acci- dental exposures. Readers are referred to the literature on risk communication. • Tissue reactions (also called deterministic effects) occur after ex- posure to radiation doses above a certain threshold, when suffi- cient cells are killed. These include the acute radiation syndrome (radiation sickness) and radiation burns (Fig. 10.3.7.1). Radiation accidents are rare and the initial symptoms of radiation sickness are non​specific, resembling influenza or food poisoning, so phys- icians might be involved in diagnosis and treatment before the true cause is appreciated. Patients might present to a range of dif- ferent medical settings. For example, the theft of a caesium-​137 (137Cs) radiotherapy source in Goiânia, Brazil, led to 50 people being overexposed, and resulted in four deaths. Many people and large areas of land and property were contaminated before the true cause of the incident was appreciated. Clinical features of radiation-​induced
tissue reactions External exposures, either whole body or partial, do not render patients radioactive and thus pose no radiation risk to medical at- tendants. If the patient has ingested or inhaled radioactive mater- ials, or has wounds containing them (internal exposure), they and their waste products can pose a persisting radiation or contamin- ation hazard to other people. Decontamination of radioactive ma- terial on skin or clothing is often straightforward, but should not take precedence over life-​saving procedures. If contamination is suspected, contact a radiation-​protection expert for monitoring and avoid spread of material. Stable iodine can be used to block uptake of radioactive isotopes of iodine. Chelating agents, such as ethylene- diamine tetraacetic acid, and ion-​exchange resins, such as Prussian blue, can be used to enhance excretion of certain internal radio- nuclides, such as 137Cs and actinides. Partial-​body exposures, especially of the extremities, might not be accompanied by systemic disease if the equivalent whole-​ body dose does not reach the symptom threshold. Symptoms of Fig. 10.3.7.1  Mature radiation burn showing central necrosis and annular epilation. Reproduced with permission from The Radiological Accident in Lilo, International Atomic Energy Agency, Vienna (2000). © IAEA, 2000.

10.3.7  Radiation 1711 radiation burns include erythema, oedema, dry and wet desquam- ation, blistering, pain, necrosis, and gangrene. There are no path- ognomonic features, but margins of ulcers might show epilation. Radiation burns can extend deep into the soft tissue, increasing fluid loss and risk of infection. Skin injuries evolve slowly, usually over weeks to months, can become very painful, and are resistant to treatment. Acute radiation syndrome The acute radiation syndrome is a rare (handfuls of cases per year worldwide), multiphasic illness. The prodrome of high exposure to external ionizing radiation is sudden anorexia, nausea, and vomiting, headache, fatigue, fever, and diarrhoea, sometimes with erythema and itching, usually lasting 24–​48 h. The timing of onset, severity, and duration of prodromal symptoms depend on the radiation dose. After a latent period of apparent recovery, effects of the killing of cells—​especially stem cells—​appear. Severity depends on the radi- ation dose. The main clinical features are: • haematopoietic syndrome, at whole-​body radiation doses ex- ceeding 1 Gy—​significant reductions in blood cell counts, infec- tion, haemorrhage, and anaemia • gastrointestinal syndrome at whole-​body radiation doses around 6 Gy—​breakdown of the integrity of the gut wall leading to mas- sive fluid and electrolyte loss and ingression of pathogens • radiation pneumonitis and the cerebrovascular syndrome (at doses exceeding 20 Gy)—​respiratory failure, hypotension, and major im- pairments of cognitive function • radiation burns if the skin dose exceeds 20 Gy If the patient survives this phase, recovery is likely. High radiation doses can also lead to permanent sterility. Several triage categories have been published, relating the severity and time-​course of symptoms and signs to prognosis. Although the threshold radiation dose for symptoms is approximately 1 Gy, lymphocyte dosimetry can detect acute doses down to about 100 mGy. Patients who also have conventional injuries have a worse prognosis. Without medical treatment, an acute dose of approximately 4 Gy is likely to be fatal within 60 days in 50% of those exposed. Doses over 10 Gy are likely to be fatal sooner, despite treatment. Similar doses over longer periods (days, weeks, and so on) might cause less severe symptoms as the body has time to repair the damage and the main concern in such patients may be the stochastic risks. Clinical investigation This includes full history, examination, cytogenetic and regular blood tests. The estimated radiation dose is needed to predict the clinical course of the patient and plan treatment. This dose should be revised as treatment progresses because the heterogeneous na- ture of accidental exposures makes the scale of radiation damage difficult to estimate. Vomiting less than 2 h after acute exposure indicates a dose of at least 3 Gy. However, there is considerable individual variation and vomiting is not invariable, even at high doses. Prodromal symp- toms last for more than 24 h with doses exceeding 6 Gy. The pattern of fall in blood levels of lymphocytes, granulocytes, platelets, and red cells depends on radiation dose. For pure γ-​field exposures, the dose (between 0.1 and 10 Gy), follows first-​order kinetics and can be estimated by multiplying the lymphocyte depletion rate con- stant by 8.6. Chromosome aberration assays, mainly dicentrics (chromosomes with two centromeres) in lymphocytes or other chromosomal ab- normalities detected by fluorescence in situ hybridization, can be used to give a more precise estimate of whole-​body dose. These as- says can be used for several years after exposure. Treatment of acute radiation syndrome Good clinical care ensures the best chance of recovery, provided that some stem cells have survived the radiation exposure. Early treatment of associated conventional injuries is important. Routine monitoring should include daily full blood counts, and blood cul- tures and other infection screens, especially in febrile patients. As a rule of thumb, patients with an estimated dose of 2 Gy or more should be observed in hospital and monitored for onset of acute radiation syndrome, but not all will require intensive treat- ment. Patients with doses of more than 4 Gy should be presumed to be developing acute radiation syndrome. Early arrangements should be made for specialist treatment. The mainstays of treatment are: • symptomatic treatment (e.g. early wound closure, antiemetics, analgesics, and fluid replacement) • early cytokine (colony stimulating factor) therapy • avoiding infection by barrier isolation (or reverse isolation) with strict environmental control, oral feeding with cooked food only and meticulous hand and nail hygiene, skin, and hair disinfection and minimization of invasive procedures • supporting affected organs until surviving stem cells multiply and repopulate the relevant organ/​tissue (e.g. consider gastrointestinal decontamination, antibiotics, blood, and platelet transfusions). Avoid antacids, proton pump inhibitors, and H2 blockers to maintain gastric acidity; use sucralfate to avoid stress ulcers Bone marrow transplants have not been proven to be beneficial. There is weak evidence for erythropoiesis stimulating agents and haematopoietic stem cells having benefit. Haematopoietic syndrome Reverse barrier nursing and topical treatments to decrease bacterial/​ fungal colonization should be used. Intravenous lines should be kept to a minimum and sited to decrease infection risk. Febrile neutro- penic patients should be given broad-​spectrum antimicrobials. Established infections should be treated as for other patients with neutropenic sepsis. Early use of antifungal agents or antiviral drugs might be required to prevent late mortality. Gastrointestinal syndrome Use supportive therapy to prevent infection and dehydration. 5-​hydroxytryptamine-​3 (5HT3) receptor antagonists should be used prophylactically if whole-​body dose exceeds 2 Gy. Diarrhoea should be treated with antidiarrhoeals, fluids, and electrolytes. Prophylactic antibiotics should be considered. Food with a low microbial content might minimize infection risks. Enteral feeding should be used if possible.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1712 Treatment of radiation burns Wound contamination is treated by gentle wound cleansing and de- bridement. Wastes arising should be treated as contaminated. Care should be taken not to break intact skin and introduce internal con- tamination. Systemic corticosteroids are now not recommended without a specific indication. Radiation burn treatments include: topical steroids, hyperbaric oxygen, pentoxifylline with oral vitamin E, wet dressings, alginates, hydrocolloids, and anti-​inflammatory agents. Growth factors have been used to foster granulation and epithelialization. Wide exci- sion, surgical repair, and skin grafting might be necessary by sur- geons experienced in the management of chronic vascular injury. Systemic mesenchymal stem cells have been used, but need further evaluation. Combined injury Surgical correction of life-​threatening and other major injuries should be carried out as soon as possible (within 36–​48 h); elective procedures should be postponed until late in the convalescent period (45–​60 days), following haematopoietic recovery. Surgical wounds and traumatic lacerations tend to heal more slowly in ir- radiated tissues. Health effects of exposures to 
non​ionizing radiation Ultraviolet radiation Ultraviolet radiation primarily affects the skin and the eye. The short-​term skin effect is sunburn, with erythema and oedema. In some people, sunburn is followed by increased production of mel- anin (suntan) but this offers only minimal protection against further exposure. Acute ocular exposure to ultraviolet radiation can lead to photokeratitis and photoconjunctivitis (arc eye, snow blindness, and so on). The most serious long-​term effect of ultraviolet radiation is induc- tion of skin cancer. Non​melanoma skin cancers, mainly basal cell carcinomas and squamous cell carcinomas, are common in white populations but are rarely fatal. The overall incidence is difficult to assess because of underreporting, but is likely to exceed 100 000 cases per year in the United Kingdom. The incidence of malignant melanoma, which is much more likely to be fatal, has increased substantially in white populations for several decades causing about 2150 deaths/​year in the United Kingdom. Chronic exposure to solar radiation causes photo-​ageing of the skin, characterized by a leathery, wrinkled appearance and loss of elasticity. Suberythemal quantities of ultraviolet radiation are beneficial in stimulating vitamin D synthesis in the skin. Vitamin D has been associated with several musculoskeletal and non​musculoskeletal health outcomes, including hypotheses about protection from cancer. Overall, there is insufficient evidence on vitamin D and non​musculoskeletal health outcomes to set adequate vitamin D intakes from dietary or UV sources over and above those established to prevent musculoskeletal disease. Repeated ocular exposure is a major factor in corneal and con- junctival diseases, such as climatic droplet keratopathy, pterygium, and, probably, pinguecula. Cumulative exposure to ultraviolet radi- ation is a major cause of cortical cataracts, but its importance in the general population remains uncertain. Immune responses Exposure to ultraviolet radiation can suppress immune responses by complex mechanisms, but the significance for human health and response to vaccinations is uncertain. Radiofrequency electromagnetic waves The widespread adoption of radiofrequency microwaves in wireless technology, including mobile phones and wi-​fi, has led to concerns about adverse health effects. High exposure to radio frequencies can cause thermal burns. There is no evidence that there is signifi- cant risk to the general public from exposure to radiofrequency radiation or from use of micro/​radiowave appliances. However, these are new technologies and a cautious approach is appropriate because of the lack of scientific evidence. Public Health England (PHE) recommends that children should use mobile phones only for important calls. Power-​frequency electric and magnetic fields There are concerns that power-​frequency electric and magnetic fields might have adverse effects on health even at levels below those required to interfere with nerves through induced fields and currents. The evidence is controversial. However, epidemiological studies had shown a consistent statistical association—​not neces- sarily indicating causation—​between unusually high background magnetic fields in homes and/​or residential proximity to power lines and increased risk of childhood leukaemia (possibly 2–​5 at- tributable cases per year in the United Kingdom). This prompted the International Agency for Research on Cancer to classify power-​ frequency electric and magnetic fields as ‘possibly carcinogenic’. In March 2004, the United Kingdom Health Protection Agency (now PHE) recommended that the government should consider precau- tionary protection from power-​frequency electric and magnetic fields. More recent studies have failed to confirm this association. Static magnetic fields Head movements in static magnetic fields stronger than 2 T can cause symptoms such as vertigo, nausea, a metallic taste, and phosphenes (seeing light without light entering the eye). Humans undergoing MRI (magnetic resonance imaging) are exposed to static magnetic fields exceeding 2 T. There are insufficient data to indicate long-​term health effects of exposures to static electric and magnetic fields. Stronger fields should be used with care in controlled or experi- mental situations with more rigorous patient monitoring. Limits have also been advised for switched gradient and radiofrequency exposures from MRI. FURTHER READING Bennett P, et al. (eds) (2010). Risk communication and public health, 2nd edition. Oxford University Press, Oxford. Dainiak N, et al. (2011). First global consensus for evidence-​based management of the hematopoetic syndrome resulting from ex- posure to ionising radiation. Disaster Med Publ Hlth Preparedness, 5, 202–​12.

10.3.8 Disasters Earthquakes, hurricanes, floods,

10.3.8 Disasters: Earthquakes, hurricanes, floods, and volcanic eruptions 1713

10.3.8  Disasters 1713 Dainiak N, et al. (2011). Literature review and global consensus on man- agement of acute radiation syndrome affecting non-​hematopoetic organ systems. Disaster Med Publ Hlth Preparedness, 5, 183–​201. Gourmelon P, et al. (2010). European consensus on the medical man- agement of acute radiation syndrome and analysis of the radiation accidents in Belgium and Senegal. Health Physics, 98, 825–​32. Mettler F, Upton A (2008). Medical effects of ionizing radiation, 3rd edition. Saunders Elsevier, Philadelphia, PA. Radiation Emergency Assistance Center/​Training Site (2013). The medical aspects of radiation incidents. Oak Ridge USA, Publications & Radiation Resources. https://​orise.orau.gov/​reacts/​documents/​ medical-​aspects-​of-​radiation-​incidents.pdf Rojas-​Palma C, et al. (2009). TMT Handbook: handbook for the man- agement of the public in the event of a malevolent use of ionising radi- ation. http://​www.tmthandbook.org/​ Waselenko JK, et  al. (2004). Medical management of the acute ra- diation syndrome:  recommendations of the Strategic National Stockpile Radiation Working Group. Ann Int Med, 140, 1037–​51. Websites Advisory Group on Non-​Ionising Radiation (2002). Health effects from ultraviolet radiation. Documents of the NRPB. Vol. 13, No. 1. https://​www.gov.uk/​government/​uploads/​system/​uploads/​attach- ment_​data/​file/​414185/​NRPB_​Doc_​series_​vol13_​no1.pdf Advisory Group on Ionizing Radiation (2009). High dose radiation effects and tissue injury. Documents of the Health Protection Agency RCE-​10. http://​webarchive.nationalarchives.gov.uk/​20140714084352/​ http://​ www.hpa.org.uk/​webc/​HPAwebFile/​HPAweb_​C/​1237362785677 AGNIR (2012). Health effects from radiofrequency electromagnetic fields: report of the advisory group on non-​ionizing radiation, Doc HPA, RCE-​20. https://​www.gov.uk/​government/​publications/​ radiofrequency-​electromagnetic-​fields-​health-​effects Armed Forces Radiobiology Research Institute (2003). Medical man- agement of radiological casualties handbook, 2nd edition. Military Medical Operations, Armed Forces Radiobiology Research Institute, Bethesda, MD. http://​ehs.columbia.edu/​2edmmrchandbook.pdf Health Protection Agency (2008). Protection of patients and volunteers undergoing MRI procedures. Documents of the Health Protection Agency RCE-​7. https://​www.gov.uk/​government/​uploads/​system/​ uploads/​attachment_​data/​file/​329364/​Protection_​of_​patients_​ and_​volunteers_​undergoing_​MRI_​procedures.pdf International Atomic Energy Agency and World Health Organization (2000). How to recognize and initially respond to an accidental ra- diation injury. http://​www-​pub.iaea.org/​MTCD/​publications/​PDF/​ IAEA-​WHO-​L-​Eng.pdf National Radiological Protection Board (2004). Advice on limiting exposure to electromagnetic fields (0–​300GHz). Documents of the NRPB, Vol. 15, No. 2, http://​webarchive.nationalarchives.gov.uk/​ 20140629102627/​ http:/​www.hpa.org.uk/​Publications/​Radiation/​ NPRBArchive/​DocumentsOfTheNRPB/​Absd1502/​ National Radiological Protection Board (2004). Review of the scien- tific evidence for limiting exposure to electromagnetic fields (0–​300 GHz). Documents of the NRPB, Vol. 15, No. 3. http://​webarchive. nationalarchives.gov.uk/​20140722091854/​ http:/​www.hpa.org.uk/​ Publications/​Radiation/​NPRBArchive/​DocumentsOfTheNRPB/​ Absd1503/​ National Radiological Protection Board (2005). Mobile phones and health 2004: report by the board of NRPB. Documents of the NRPB, Vol. 15, No. 5. http://​webarchive.nationalarchives.gov. uk/​20140714084352/​ http://​www.hpa.org.uk/​webc/​HPAwebFile/​ HPAweb_​C/​1194947333240 10.3.8  Disasters: Earthquakes, hurricanes, floods, and
volcanic eruptions Peter J. Baxter ESSENTIALS Natural disasters (including earthquakes, volcanic eruptions, hur- ricanes, floods) cause tens of thousands of deaths and adversely affect the lives of hundreds of millions of people every year. The trend is for the impacts to increase alongside the continuing expan- sion of human populations into regions at risk and with environ- mental degradation making human settlements more vulnerable, especially in heavily urbanized areas and megacities. This reckless development is going on in most countries of the world, even in places prone to natural disasters. Recent remarkable advances in forecasting weather-​related disasters (hurricanes and floods) have to be matched with adequate disaster preparedness in those communities at high risk if they are to be translated into effective warnings, especially in low-​income countries. Earthquakes remain notoriously unpredictable and have the greatest mortality toll of all natural disasters. For those natural disasters that are weather related, climate fore- casters say that, on current trends with an increase in the world’s average temperature of 2–​3°C and an associated warming of the oceans, there is likely to be growing potential for more intense and more frequent hurricanes in regions that already suffer from these extreme events and, in temperate regions, more severe winter storms and wider fluctuations in rainfall with flooding and more heat waves. Rising sea levels will increase the severity and the fre- quency of coastal floods everywhere. On the positive side, climate change fears have been an important catalyst in making some coun- tries pay more attention to addressing the mitigation of their present weather-​related disasters. The scale of some recent catastrophes events has added a fur- ther sense of urgency. The three mega-​disasters (so-​called be- cause mortality exceeded 100 000 people in each) —​the Southeast Asian tsunami (2004), Hurricane Nargis, Burma (2008), and the Haiti earthquake (2010)  —​were followed by the Tohoku earth- quake (2011), Japan, in which tsunami waves caused over 18 000 deaths and severely damaged the supposedly earthquake-​resistant Fukushima nuclear plant that had been built along the same coast with tsunami protection that proved to be inadequate. This ‘domino’ or cascading effect triggering a technological disaster in its wake demonstrated the problems that can be faced even in ad- vanced industrial nations prone to natural cataclysms. Introduction A natural disaster is the massive ecological breakdown in the relation between humans and their environment, a serious and sudden (or insidious, as in drought) event on such a scale that

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1714 the stricken community needs extraordinary efforts to cope with it, often with outside help or international aid. Very high mor- tality is not necessary in this definition, which is not specific- ally health based. Between 300 and 400 natural disasters occur throughout the world every year. The occurrence rates of geo- physical hazards like earthquakes and volcanic eruptions have not varied much since the time of Neanderthals over 40 000 years ago, but there is evidence in recent years for a rise in the number of weather-​related disasters, such as severe floods and windstorms. Furthermore, the potential for increasing losses of life and prop- erty as populations expand in regions of high natural risk was exemplified by the Southeast Asian tsunami that devastated the Indian Ocean region on 26th of December 2004 leaving more than 250 000 people dead and at least 1.7 million displaced from their homes in 10 countries. Natural disaster risk reduction is one of the targets embedded in the UN Sustainable Development Goals with 187 member states adopting the Sendai Framework for Disaster Risk Reduction 2015 to 2030. By 2030, the Framework calls for substantial global re- ductions in the following areas: mortality, the number of people affected, direct economic losses, and damage to critical infrastruc- ture. Countries will have to make their own plans to support the implementation of the Framework, with greater international co- operation to support developing countries. For access to early warning systems, disaster risk information and assessment are pri- orities, while health features more strongly than in the previous UN decadal initiatives. Natural disasters are by definition chaotic, but communities in disaster-​prone countries can plan and prepare against them. Scientists are working on improving forecasting and on the mod- elling of their impacts. Most deaths in sudden-​onset disasters happen before outside aid arrives, hence strengthening local response capacity in the first hours is crucial. However, inter- national disaster relief can be rapidly and effectively dispatched to needy countries that are politically willing to accept it:  al- though medical relief teams might arrive too late to treat most of the critically injured, teams with engineering and disaster relief skills will have an important role in restoring roads and bridges, bringing in potable water, ensuring solid waste management, food protection, vector control, and sanitation. Even in disasters that have major human impacts, attendances at medical facilities can return to normal levels even within a few days of the acute phase, once the injured have been cared for, when the priority becomes the restoration of primary healthcare and the needs of survivors. Rehabilitation in the post-​disaster phase should be an essential consideration in the emergency response phase, and ultimately dir- ected towards measures for reducing the pre-​disaster vulnerability. Yet many financial donors view disasters as mere temporary inter- ruptions in development. In April and May 2015 two earthquakes in Nepal killed some 8000 people and reduced 300 000 houses to rubble. On 25 April, a 7.8 magnitude earthquake struck with its epi- centre 77 km northwest of the capital Kathmandu which suffered widespread destruction along with the collapse of mud and stone housing in rural areas. The impact was so widespread that an ava- lanche was triggered on Mount Everest. For a poor country it will take many years to recover—​blowing away the myth that life gets back to normal in a few weeks after a disaster. History should not have to repeat itself—​with the limited progress in the past over strengthening building codes, reinforcing old buildings, protecting health facilities, and educating the population about risks in a country where seismologists had warned for years about the severe earthquake risk. Pre-​disaster measures Accurately forecasting the timing and size of these sudden-​onset natural disasters is rarely possible. This fact constrains efforts to prevent loss of life by timely evacuation of people from the areas at risk before disaster strikes. Disasters leave a trail of devastation and are quite different from the major incidents hospitals usually plan for, in that normal lifelines and infrastructure break down in the devastation. Thus, essentials such as transport, communica- tions, and power will be the first to fail or will be severely curtailed, thereby crippling the immediate emergency response. In the worst examples hospitals can be severely damaged and the staff being among the victims. But despite their chaotic aspects, disasters are amenable to sci- entific study and a growing body of physical scientists in various fields are directing their energies towards disaster mitigation, par- ticularly in devising hazard warning systems, engineering solu- tions, and disaster risk information and assessment. There is a growing need for social scientist involvement in implementing risk reduction interventions and developing community resilience in disaster-​prone areas. As well as implementing forecasts and warnings, the traditional approach to disaster risk reduction, depending upon the hazard, relies on a platform of engineering measures such as constructing river and coastal flood defences, ensuring regulations are fol- lowed to build seismic resistant buildings in earthquake zones, cyclone shelters, and land-​use planning to minimize the occupa- tion of risky zones—​the commonest example being floodplains. Community preparedness and emergency planning should in- clude the full involvement of the health sector. We should add to this traditional list improving resilience and communicating risk, as well as effectively translating into practice the findings of the latest scientific research. Poverty and social marginalization in mid-​ to low-​income coun- tries, especially, remain potent sources of global vulnerability to natural disasters. Less well-​publicized are disasters occurring in regions of conflict and humanitarian crisis, or complex emergen- cies, for example, in the Democratic Republic of Congo, Darfur (Sudan), and Eritrea (Ethiopia). International relief organiza- tions might not be able to safely or freely move in their response. Hence, in the Southeast Asian tsunami, access to some regions of Indonesia and Thailand was prevented by security issues. In 2008, Burmese people living in the Irrawaddy Delta had no warning from the government of the approach of Cyclone Nargis and hurricane preparedness measures were non​existent. Over 138 000 people died, most from drowning, but the crisis was made worse when, in the immediate aftermath, 1.5 million homeless survivors were left without food, water, or shelter while the Burmese government vacillated for weeks over accepting international aid. International disaster relief is nowadays capable of being rapidly dispatched to needy countries and is on such a global scale that epidemics and

10.3.8  Disasters 1715 famine are no longer the feared Horsemen of the Apocalypse they once were. Earthquakes Over time more deaths are caused by earthquakes than by any of the other causes of natural disaster. Many parts of the world lie along fault lines and are known to be vulnerable to devastating earth- quakes, but it remains impossible to predict precisely where and when a quake will strike. Most deaths and injuries are caused by collapsing buildings, but secondary causes such as fires can take their toll. When timber, masonry, reinforced concrete, and other types of buildings collapse, they inflict injuries to occupants in dif- ferent ways and with different degrees of severity. In the collapse of masonry buildings, an important cause of death is often suffocation from the weight and dust shaken from the wall or roof material which may also bury the victims. Falling masonry causes crush in- juries to the head and chest, external or internal haemorrhage, and chest compression (traumatic asphyxia). Little is known about the survival times of people when they get trapped in collapsed build- ings, but most victims will die immediately or within 24 hours from their injuries if they are not rescued, depending upon such factors as the severity of after-​shocks, fire outbreaks, cold, and rain. Rapid extrication of survivors and application of first-​aid by the uninjured immediately after the event could potentially save up to 25–​50% of injured victims. Greatest demand for emergency medical care is within the first 24 hours and the need for emergency treatment quickly fades, though search and rescue teams might continue trying to find survivors for three to five days. Multiple trauma is the main feature, with the risk of doctors missing internal injuries in the stress of the emergency. Causes of delayed death include de- hydration, hypothermia, crush syndrome, and postoperative sepsis. Most of those requiring medical assistance suffer minor injuries such as lacerations and contusions. In the Southeast Asian tsunami, an earthquake of magnitude 9 on the Richter scale off the coast of the island of Sumatra on 26 December 2004 suddenly forced the seafloor upwards by some 10 m, creating a wave that surged through the Indian Ocean. The surface perturbation was initially small, but when the water grew shallow, near the coast, the tsunami waves formed. Without warning, the waves hit Indonesia and Thailand within an hour, and then Sri Lanka and India, ultimately reaching as far as East Africa. The province at the north-​western end of Sumatra, Aceh, suffered overwhelming devastation. More than 20 000 homes were destroyed, over a 100 000 people were killed, and some 700 000 people were displaced. Many victims were health service staff, which hampered the emergency response. In all countries affected by the tsunami the main public health infrastructure remained intact as the devastation was limited to coastlines, so the feared epidemics of vector-​borne diseases, such as malaria and dengue, as well as cholera and dysentery, were able to be prevented. Large numbers of dead and small numbers of major injuries in survivors in comparison are typical of flood disasters in general, as the severely injured quickly succumb in the water; the injured survivors were mainly treated by local health teams. Many of the patients requiring surgery had infected wounds following contamination by sand and mud. Respiratory tract infections and pneumonia were common among patients who had come close to drowning. Psychosocial needs were identified on a massive scale, but the appropriateness and effectiveness of specific interventions in such disasters to prevent post-​traumatic stress disorder remains a controversial issue. The power of tsunamis was demonstrated again in the Tohuko earthquake in 2011, which had its epicentre off the south-​east coast of Japan and triggered a tsunami that swept the Japanese coastline leaving total devastation and 15 891 dead with 2579 missing. Many more people would have died but for the tsunami warnings that are routinely practised in the shoreline towns. But the tsunami waves also overcame the engineering defences of the Fukushima nuclear power plant, which in retrospect had not been built to withstand tsunami of this size. The resultant severe damage to the plant’s nu- clear cores led to the release of substantial amounts of radioactive material into the environment. The evacuation of 170 000 residents within a radius of 30 km of the power plant was urgently undertaken, but concern grew that the possible spread of contamination in the atmosphere could have gone as far afield as Tokyo and beyond. These fears were later shown to be groundless, but they raised anxiety in Japan and neighbouring countries. Studies have shown that although no discernible physical effects are expected from the radiation leak, psychological and social problems largely stemming from heightened risk perceptions have had a devastating effect on people’s lives. One immediate economic casualty was the fishing industry along the coast due to market fears over radioactive contamination of the sea. The evacuation itself was not without risk, however, with 50 deaths attributed in hospital pa- tients; in the first three months of the evacuation mortality rates in older people needing nursing or hospital care rose significantly. A rare example of the importance of disaster preparedness was the earthquake in Bam, Iran, on 26 December 2003, which resulted in 26 271 deaths and the nearly complete destruction of the city of 80 000 inhabitants. The loss of about one-​third of the inhabitants, including 200 out of 500 doctors, was attributed to the weak, mud brick construction. The health infrastructure was destroyed, but within 48 hours some 11 972 of the 15 000 injured survivors had been air evacuated by the military to hospitals in the rest of the country, and others were transported to treatment facilities by re- latives. By the time foreign medical teams arrived, their main task was to provide routine healthcare to the residual population living in shelters. By contrast, the Pakistan earthquake on 8 October 2005 hit the impoverished mountainous north of the country where ac- cess to hundreds of remote villages was hindered by damaged and blocked roads. Over 73 000 people died and 69400 people had ser- ious injuries; over 3 million people were left homeless. As houses were mostly constructed of weak, rubble masonry walls supporting concrete slabs for roofs, the violent shaking easily razed buildings to the ground or triggered landslides. Roof slopes then fell on top of the occupants (Fig. 10.3.8.1) and caused multiple trauma, such as spinal and pelvic fractures. Significant numbers of amputations were performed, and post-​disaster reconstructive plastic surgery was frequently needed to treat the often severe and localized soft tissue damage caused by entrapment (Fig. 10.3.8.2). In 2008, the recent rapid economic development and accom- panying building boom in China lay behind the destruction caused by the largest earthquake to strike the country in recent times (7.9 on the Richter scale), when entire towns collapsed in the mountainous

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1716 Sichuan province, leaving 80 000 people dead and at least 5 million homeless. Poor building quality has been blamed for the catastrophic failure of homes and schools, the latter being a major psychological issue for parents with one child families. Despite the rapid mobil- ization of thousands of troops to the area, only a few survivors were retrieved from the rubble, sadly emphasizing the country’s failure to incorporate seismic resistance in the new community developments. Poverty profoundly influenced the impacts of the January 2010 earthquake in Haiti, one of the poorest countries in the world. The earthquake ranked as the deadliest to have occurred in western hemisphere history. Despite the island’s position along a known major earthquake fault line, there was a total lack of preparedness. Devastation was immense with 220 570 people killed and 3.7  million people directly affected. Haiti was left helpless and massive international help was desperately needed. Casualties with multiple trauma were so numerous that they were sent for treatment to hospitals all over the Caribbean re- gion. Tragically, UN soldiers sent to provide logistical help were evidently the source of the cholera epidemic that struck the island in October 2010, leading to 8000 further deaths and another 650 000 islanders infected at a time when half the popu- lation still had no access to clean water. Volcanoes About 500 to 600 volcanoes around the world are known to be cap- able of eruptive activity and several major eruptions occur every year. The vast majority of dangerous volcanoes are explosive and unpredictable in their behaviour, whereas the less common lava flow eruptions normally allow people to escape from their path. Most deaths and injuries in explosive eruptions (such as the one that engulfed ancient Pompeii and Herculaneum in AD79) are caused by pyroclastic flows and surges, which are clouds of intensely hot gas and ash that can travel at hurricane speeds. Survivors are un- common, but will have severe, extensive skin burns and inhalation thermal injuries. The scope of the destruction was shown in the worst volcanic disaster in the 20th century at St Pierre, Martinique, in 1902, when 28 000 people were killed in a laterally directed pyro- clastic surge, leaving only two survivors in the totally devastated city. Vesuvius in Italy is an example of one of the world’s most dangerous volcanoes, and uncontrolled development has left over one million people living in an area which could be devastated by pyroclastic flows. Another major cause of death is the lahar (wet, debris flows) formed by newly erupted ash mixed with heavy rain or unstable masses that are mobilized by meltwater from glaciers or by crater lakes. The eruption of the Nevado del Ruiz volcano, Colombia, in 1984 triggered a huge lahar through the rapid melting of ice in the summit glacier: no warning was received in the towns below where some 24 000 people were buried by mud. By contrast, in one of the largest eruptions of the 20th century, at Mount Pinatubo in the Philippines in 1991, 50 000 people were successfully evacuated from the threat of pyroclastic flows, but over 300 died from the collapsing of roofs burdened with accumulated rain and ash, while sheltering inside their homes. The eruption of the Soufrière Hills volcano on the tiny Caribbean island of Montserrat, a UK Overseas Territory, began in July 1995 and was to last over 15 years. After a slow start it gradually escalated, forcing the evacuation of thousands of people from their homes be- cause of the threat of pyroclastic flows and surges. By 1997, these currents had devastated the southern part of the island, evicting three-​quarters of the total island population of 12 000 people. Air pollution from volcanic gases and ash emissions was a major con- sideration because of the close proximity of the population to the volcano and the frequent eruption of fine, respirable-​sized ash Fig. 10.3.8.1  Pakistan earthquake, 2005. Collapsed concrete roof slab, typical example of vulnerable building construction in the impacted area (see text). Source: Emily So. Fig. 10.3.8.2  Pakistan earthquake, 2005. Treating severe soft tissue crush injuries in a patient rescued from a collapsed building. Source: Emily So.

10.3.8  Disasters 1717 containing hazardous amounts of crystalline silica, a mineral that causes silicosis. Hurricanes Hurricanes are one of a broad class of extreme weather phenomena that include winter storms (snow, sleet, freezing rain), thunder- storms (e.g. tornadoes, heavy rains, lightning, wind, and hail), ex- treme precipitation (e.g. flood and flash floods), and windstorms. Hurricanes (or typhoons as they are called in the Western Pacific) are tropical cyclones that form over warm oceans with ocean sur- face temperatures over 26°C. Once overland they soon run out of energy and rapidly abate, but can still cause severe flooding from heavy rain. Very high wind speeds, up to 250 km/​h, are restricted to a relatively narrow track, usually no more than 150 km wide, within which localized gusts may even achieve tornadoes speeds and be extremely destructive. Most deaths and injuries, however, are not from the effects of wind on people (who normally shelter indoors for protection) or from building damage (building collapse or being struck by flying debris). Instead, deaths and injuries are commonly the result of flooding by the sea surge as the hurricane strikes land, or concurrent heavy rainfall (typically up to 60 cm over a larger area and extending further inland than high winds) triggering landslides. Hurricanes lift the sea, forming a sea surge that typically rises 3–​4 m above existing heights, and the wind generates waves on top of these. Some storm surges can hit coasts well ahead of the landfall of the actual storm and can travel with nearly the same rapidity, and de- structiveness, as tsunami waves. Over 90% of fatalities in hurricanes are drownings associated with storm surges or floods. Other causes of death include burial beneath houses collapsed by wind, penetrating trauma from broken glass or wood, blunt trauma from floating objects or debris, or entrap- ment in mudslides. The greatest need in the post-​impact phase is the provision of adequate shelter, water, food, and clothing, and sanita- tion. Most victims suffer from lacerations caused by flying glass or other debris, or minor trauma such as closed fractures and puncture wounds. Katrina was the third most powerful storm ever to make land- fall in the United States of America, attaining hurricane category five status before it struck the Louisiana coast on the morning of 29 August 2005. It left breaches in the levee system of New Orleans that created catastrophic flooding of an area of more than 400 km², submerging half a million homes and trapping tens of thousands of people. Critically, the city’s mayor did not issue a mandatory evacu- ation order until the day before the hurricane hit, which was too late for many to evacuate in time, including the poor, who had no means of transport. In Louisiana and Mississippi 1700 people died, most by drowning. The emergency response was woeful. Up to 20 000 evac- uees were abandoned in the city’s Superdome sports stadium for five days before being evacuated to other shelters. One crowded public hospital left cut off for five days without electrical power, clean water, and medical supplies was rendered helpless by the engulfing floodwater. The victims were predominantly black and poor. In the aftermath, nearby states were able to absorb several hundred thousand evacuees from the city in a few days. Despite forebodings, epidemics of diar- rhoeal diseases, respiratory tract infections, and mosquito-​borne disease, in particular West Nile virus, did not occur. Increases in suicide and psychiatric morbidity were found in follow-​up studies of evacuees, and although psychological distress was common, it re- solved in most of the people over time. Recent advances in meteorology on tracking the paths of hurri- canes in time and space are now making inroads into reducing loss of life by forecasting their landfall far enough ahead for thousands of people to be directed to safety, provided the warnings are effect- ively communicated by government officials. A dramatic example was in the Indian state of Odisha (Orissa). In 1999 one of the largest cyclones to strike the coast left 10 000 dead in its wake: although it was forecast, no warnings were disseminated, and no preparedness to move the population to safety was in place. In October 2013 an al- most identical cyclone hit the same coast, but with the loss of only 17 people—​the difference was that the warning was followed by 1 mil- lion people being temporarily evacuated from the coastal area by the Indian army. Floods In addition to the major losses of life that can be caused by hurri- canes and their associated sea surges, floods mostly result from mod- erate to large events (heavy rainfall, snowmelt, high tides) occurring within the expected range of streamflow or tidal conditions. In the United Kingdom, as in many countries with low-​lying coastal land, the hazards of coastal flooding from sea surges and high tides dom- inates over river flooding, although the latter is becoming more fre- quent, particularly in the last few years when there has been a trend of abnormally wet winter months and localized repeated flooding, though it is not yet possible to attribute this to climate change. Flood warning and forecasting, combined with effective land manage- ment, community preparedness, and evacuation planning, are as essential as engineered river and coastal defences. The primary cause of death from floods is drowning, but trauma from impact with floating debris and by hypothermia due to cold exposure are also important. The proportion of survivors requiring emergency medical care is small as most injuries are minor, such as lacerations. This absence of survivors with severe or multiple trauma is likely to reflect the delay in search and rescue through the flood waters, and victims drown or die from their injuries and the effects of exposure before help arrives. Increased morbidity and mortality in survivors of floods was reported in the year after the UK East Coast flood in 1953, in which over 300 people drowned. An increase in suicides and mental health problems arose after the severe flooding caused by heavy rains in central Europe in July 1997. The mental health impacts of floods as a cause of enduring disaster morbidity re- quires further epidemiological study, especially as flooding is likely to become more disruptive and more widespread as a consequence of climate change. Bangladesh is one of the most flood-​prone countries in the world. A quarter of its 151 million population are at the mercy of cyc- lones and almost three-​quarters live in regions prone to a variety of flooding types: coastal floods, riverbank flooding, flash floods and monsoon rains, and meltwater from the Himalayas. The probability is that under climate change the flood hazards will get worse. Over the last 50 years, targeted improvements in disaster management, re- ducing poverty, and stepping up civic resilience measures have made

10.3.9 Bioterrorism 1718

10.3.9 Bioterrorism 1718

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1718 a substantial difference to the health and economic impacts of these severe events. Post-​disaster relief Most traumatic deaths in sudden-​onset disasters will happen before outside aid arrives, and so building local resilience and the capacity to react to disaster is essential. However, relief teams from outside a disrupted region or country may find that they have an important role in restoring roads and bridges, bringing in potable water, water purification, ensuring solid waste management, food protection, vector control, and sanitation. Attendances at medical facilities may return to normal within a few days of the disaster onset, and restor- ation of primary care then becomes the priority rather than emer- gency treatment. However, the earthquakes in Pakistan (2005) and Nepal (2015) already mentioned in this chapter caused widespread destruction in remote, mountainous regions of the countries, and the damage to roads reduced accessibility even further, so the treat- ment of many casualties was inevitably delayed. In Nepal, WHO working with the Ministry of Health and Population coordinated the national and international humanitarian aid, sending foreign medical teams that had registered with them to deliver healthcare in the affected districts. Epidemiology has an important role in post-​disaster assessment and health surveillance, particularly when large populations have been relocated, as well as investigating the causes of mortality and morbidity in disasters, such as mental ill health and long-​term phys- ical health consequences. The evidence base for the effectiveness of health-​directed interventions in humanitarian crises, including natural disasters, needs strengthening and this is at last being ac- knowledged by research grant awarding bodies and non​government organizations. Similarly, scientific evaluation of disaster relief efforts, in par- ticular the knee-​jerk response of sending aid—​any type of aid, in- stead of aid matched to need, in the mistaken and even patronizing belief that the affected population is too shocked and helpless to take responsibility—​has incentivized governmental relief organizations to review their practices. Indiscriminate aid can actually hinder the efforts of countries already fully under strain—​the examples are le- gion. Many donated medicines are out of date or simply inappro- priate, food supplies are not tailored to requirements, international relief teams being sent without the right expertise or capable of being self-​sufficient, including uninvited individuals or medical doctors inexperienced in disaster response, and who might con- sequently do more harm than good. Infectious disease epidemics are still mistakenly and widely held to inevitably follow in the wake of mass fatalities and the disposal of corpses, with the result that local healthcare workers, already in short supply, might be diverted into undertaking mass immunization campaigns and mass burials without good reason. Inevitably, the pressing priorities in most disaster settings are not conducive to undertaking well-​conducted evaluations of relief prac- tice. The medical literature on natural disasters is relatively small and in recent decades the global challenge has focused on improving preparedness and risk reduction, but progress has been lamentably slow. This global overview of progress over the last decade may not be as encouraging as many of us would like to read, but the need for effective health sector participation in both planning for and re- sponding to natural disasters has never been greater. FURTHER READING Cash RA, et al. (2013)Reducing the health effect of natural hazards in Bangladesh. Lancet, 382, 2094–​103. Fink S (2013). Five days at memorial: life and death in a storm ravaged hospital. Atlantic Books, London. Hasegawa A, et al. (2015). Health effects of radiation and other health problems in the aftermath of nuclear accidents, with an emphasis on Fukushima. Lancet, 386, 479–​88. Katz JM (2013). The big truck that went by: how the world came to save Haiti and left behind a disaster. Palgrave Macmillan, New York, NY. Koenig KL, Schultz CH (2016). Disaster medicine. Cambridge University Press, Cambridge. Websites Centre for Research on the Epidemiology of Disasters (CRED). http://​ www.cred.be Pan American Health Organization (WHO) Emergency Preparedness and Disaster Relief. http://​www.paho.org/​disasters UN International Strategy for Disaster Risk Reduction. https://​www. unisdr.org/​ 10.3.9  Bioterrorism Manfred S. Green ESSENTIALS Bioterrorism is the deliberate use of biological agents to cause illness, death, and fear for ideological or personal purposes. Most poten- tial bioterrorism agents occur naturally as known pathogens and are classified by the United States Centers for Disease Control as follows: (1) Category A, with greatest risk to the public and national security This comprises (a) infectious and contagious diseases—​smallpox, plague, and viral haemorrhagic fevers; (b) infectious but not con- tagious diseases—​anthrax, and tularaemia and (c) toxins—​botulism; (2) Category B, with intermediate risk This comprises those causative agents that are relatively easy to spread and produce diseases with moderately high death rates; (3) Category C—​emerging infectious diseases that could be en- gineered to spread and cause high rates of morbidity and mortality. The agents might be disseminated through aerosolization, food, human carriers, infected insects, or water. The incubation periods of these agents can vary from hours to weeks, with early symptoms mimicking many other infectious diseases. The diagnosis might not be suspected unless cases occur in clusters. Early identification of outbreaks will depend largely on the ability of primary care and emergency room physicians to identify and promptly report cases to the public health authorities. A major concern is that diagnosis of these extremely un- common diseases might not be considered by physicians who have

10.3.9  Bioterrorism 1719 rarely, if ever, seen such cases. Specific treatment (if available) of affected individuals will depend on the pathogen, and for contagious diseases such as smallpox and plague, isolation of patients and their contacts, barrier nursing, quarantine, and restriction of the movements and social interactions of people are important control measures. Decontamination is relevant mainly for anthrax and smallpox, in the environment of an aerosol attack and at places where patients were treated. Public education and effective risk communication are essen- tial in managing a bioterrorism attack:  (1) clinicians and public health personnel need access to up-​to-​date information; (2)  the general public requires nontechnical descriptions of the diseases and simple instructions on how to act in an emergency situation. Primary prevention should include addressing the root causes of terrorism, developing comprehensive preparedness programmes, and educating health professionals to deal with an outbreak. Introduction The potential public health threat posed by bioterrorism could make exceptional demands on clinicians. Rapid diagnosis will have im- plications far beyond the individual patient. It will initiate a process of preventive actions, which could impact on the lives of thou- sands. Clinicians might have to treat infectious disease casualties en masse under emergency situations, while ensuring the protection of healthcare workers and other patients. Clinical presentations might be atypical because of the nature of the exposure and the possibility that the organism might have been genetically mutated. Antibiotic resistance and vaccine failure could be encountered and laboratories are likely to be overburdened. Public panic could exacerbate ethical dilemmas in the triage for specialized care in limited facilities. Historical perspective The use of biological agents as weapons inspires a special abhor- rence and dread. International agreements, such as the Geneva Protocol in 1925 and the Biological Weapons Convention in 1972, banned their use and production. However, in the early 1990s, it was revealed that anthrax spores were accidentally released from a military facility in Russia in 1979, causing an outbreak of respira- tory anthrax. Evidence emerged that the former Soviet Union had continued a bioweapons programme, generating concerns that bioweapon agents and the expertise for their production might reach terrorist groups. Biological weapons Almost all potential bioterrorism agents occur naturally as known pathogens, although many are zoonoses, not normally affecting humans. The United States Centers for Disease Control and Prevention (CDC) classified potential bioterrorism agents into three categories (Box 10.3.9.1). Category A agents have the highest priority since they are considered the greatest risk to the public and national security. These can be subclassified into agents that are in- fectious and contagious, those that are infectious but not usually contagious, and toxins. Category B includes diseases that are con- sidered an intermediate risk to the public since the causative agents are relatively easy to spread and the diseases result in moderately high death rates. Category C agents include emerging pathogens, which could be engineered to spread and cause high rates of mor- bidity and mortality. Since the category A biological agents have been weaponized in past programmes, they are currently of greatest concern. Here they are briefly described but more details about their clinical aspects are provided in Section 8. Diseases that are both infectious and contagious Smallpox (Chapter 8.5.4) is the prototype of potential bioterrorism agents that are both infectious and contagious. Although eradicated in 1978, it is believed to have been weaponized by the Soviet Union. Universal vaccination was phased out in the 1970s and since the case fatality in unvaccinated subjects is around 30%, smallpox is one of the most feared bioterrorism threats. Secondary cases can occur through droplet spread, direct contact with skin lesions or body fluids, and rarely through airborne transmission. The plague bacillus (Yersinia pestis) (Chapter  8.6.16) was in- cluded in the bioweapons programmes of both the United States of America and the Soviet Union. Untreated pneumonic plague has a case fatality approaching 100%. The organism can spread from person to person through droplets, causing several generations of the disease. Box 10.3.9.1  Examples of bioterrorism agents by category Category A Infectious and contagious diseases • Smallpox (Variola major) • Plague (Yersinia pestis) • Viral haemorrhagic fevers (filoviruses, e.g. Ebola, Marburg, and arena- viruses, e.g. Lassa, Machupo) Infectious but not contagious diseases • Anthrax (Bacillus anthracis) • Tularemia (Francisella tularensis) Toxins • Botulism (Clostridium botulinum toxin) Category B • Brucellosis (Brucella spp.) • Epsilon toxin of Clostridium perfringens • Food safety threats (Salmonella, Escherichia coli 0157, Shigella) • Glanders (Burkholderia mallei) • Meliodosis (Burkholderia pseudomallei) • Psittacosis (Chlamidia psittaci) • Q fever (Coxiella burnetii) • Ricin from Ricinus communis (castor bean) • Staphylococcal enterotoxin b • Typhus fever (Rickettsia prowazekii) • Viral encephalitis (alphavirus, e.g. Venezuelan equine encephalitis, eastern equine encephalitis, western equine encephalitis) • Water-​safety threats (e.g. Vibrio cholerae, Cryptosporidium parvum) Category C • Emerging infectious diseases such as Nipah virus and hantavirus Source: http://​www.bt.cdc.gov/​agent/​agentlist-​category.asp

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1720 Viral haemorrhagic fevers caused by the filoviruses (Chapter 8.5.17) and arenaviruses (Chapter 8.5.18) may have been weaponized by the former Soviet Union, Russia, and the United States. The Soviet Union is reported to have produced quantities of Marburg, Lassa, Ebola, Junin, and Machupo viruses. Second and later generations of disease can occur through direct contact with body fluids of the patients. Healthcare workers are at greatest risk. Infectious but not contagious diseases Anthrax spores (Chapter 8.6.21) were among the leading agents in biological weapons programmes, since they are highly stable, viru- lent, resistant to drying, and easily disseminated. Aerosolized spores cause inhalation anthrax, which has an untreated case fatality ap- proaching 100%. The spores can survive in the environment for many years, although once on the ground, they will tend to produce cutaneous anthrax. The spore-​forming coccobacillus Francisella tularensis (Chapter 8.6.20), has been weaponized in biowarfare programmes. The un- treated case fatality could be 30–​60%. There is no secondary person-​ to-​person spread. Toxins Botulinum toxin, produced by Clostridium botulinum (Chapter 8.6.25), is one of the most potent neurotoxins known and has been weapon- ized. In a bioterrorist incident, it could be disseminated either through food or by aerosol. The untreated case fatality approaches 100%. Ricin is a protein cytotoxin produced from the castor bean Ricinus communis. There is no antidote. Patients affected by toxins are not contagious at any stage of the disease. Dissemination of bioweapons Biological agents might be disseminated through aerosols, food, human carriers, infected insects, or water. Aerosolization maximizes the number of people exposed, causing the most damage. Release of contagious agents at different sites could greatly amplify the out- break. Since most potential agents are not normally aerosol trans- mitted, the resulting illnesses could occur with shorter incubation periods and atypical clinical manifestations. Clinical effects are likely to depend on the dose. Epidemiology Documented contemporary attempts at planning or occasionally attempting bioterrorism have employed Salmonella typhimurium, botulinum toxin, anthrax spores, Q fever bacteria, Ebola virus, and ricin. In 1978, a Bulgarian dissident was assassinated in London by a pellet, probably of ricin, that was implanted into his leg. In 2001, six envelopes contaminated with powdered anthrax spores were mailed in the United States and infected 22 people. Half suffered from inhal- ation anthrax and the others from cutaneous anthrax. Thousands of workers received prophylactic therapy, and a large-​scale decontamin- ation programme was implemented (Fig. 10.3.9.1.) Radiological and chemical terrorism are also potential threats. The only documented incident of radiological terrorism occurred in 2006, when a former officer in the Russian security services was assassinated by exposure to α-​emitting polonium-​210 (210Po), in a public place in London. Although no other cases were detected, others could have been exposed through ingestion of the material from contamination of their hands. The initial symptoms could be confused with an infectious disease. In 2018 in the UK a nerve agent, novichok, was used in an attempt to assassinate a former Russian agent and his daughter. Some weeks later a UK citizen died from ex- posure to the same agent following contact with the presumed liquid containing novichok that had been left near the scene. Prevention Prevention of bioterrorism includes addressing the causes of ter- rorism and developing appropriate preparedness strategies. In addition to international condemnation of the development of bioweapons, access to production capabilities must be controlled. Effective preparedness is in itself a deterrent and requires coordin- ation between agencies and specialists from multiple disciplines. Food supplies must be protected from deliberate contamination. 1 0 0 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 30 31 32 33 2 3 4 5 6 7 8 9 10 Day Per cent Fig. 10.3.9.1  Simulated epidemic curve for a point-​source outbreak of inhalation anthrax without intervention. Reproduced from Scheulen J, Latimer C, Brown J (2006). Electronic Mass Casualty Assessment and Planning Scenarios (EMCAPS). Johns Hopkins University. Internet http://​www.hopkins-​cepar.org/​EMCAPS/​EMCAPS accessed July 14, 2007.

10.3.9  Bioterrorism 1721 Although water is an unlikely vehicle for bioterrorism, drinking-​ water sources require special security measures. Preparedness programmes include training and, where indicated, pre-​exposure vaccination of ‘first responders’. The infrastructure to deal with the impact of different biological agents will require increased clinical surge capacity and patient isolation facilities. Children, pregnant women, and the immunocompromised might have special needs. Dead patients need to be handled using the same barrier precautions as for live patients. Antivirals and immunoglobulins are currently considered only for treatment and not for prophylaxis. At present, vaccines are rele- vant only for smallpox and anthrax. In most countries, more than 50% of the population has never been vaccinated against smallpox. Antibody titres have been shown to decline markedly 5–​10 years following vaccination, although residual immunity might persist for many years. However, previously vaccinated, milder cases in the community could increase the risk of spread. Several countries have carried out vaccination programmes against smallpox for military personnel and first responders. Anthrax vaccine is given routinely in some military populations. Some countries have established national stockpiles of pharmaceuticals and vaccines for use in the event of biological or chemical attacks. The global inventory of smallpox vaccine, together with the possibility of diluting vaccine, exceeds three billion doses. Preparedness for a potential bioterrorism incident remains a high public health priority in most countries, but there are important logistic issues regarding stockpiling and mass ad- ministration of vaccines against potential bioterrorism agents, both prior to and during incidents. A dual vaccine has recently been de- veloped against both smallpox and anthrax by integrating immune-​ enhancing cytokine IL-​15 and the protective antigen of B. anthracis into the Wyeth vaccinia virus. This has been proven to be efficacious against both smallpox and anthrax in laboratory animals. If human trials of safety and efficacy are successful, such a vaccine will have sev- eral important advantages. First, it will be a combined vaccine against two diseases. Second, as a vaccinia-​based vaccine it can be lyoph- ilized without loss of potency and so will not be dependent on the cold-​chain, which will greatly simplify storage, stockpiling, and field delivery. Thirdly, it is likely to reduce the number of doses necessary to achieve protection against anthrax. Hence, in practice it could be an extremely important advance for mass vaccination in general and for vaccinating first responders in particular. The recent Ebola epidemic in West Africa has stimulated the development of effective vaccines and treatment for Ebola virus disease. Secondary prevention depends on comprehensive surveillance and clinical awareness, both for detecting and characterizing the event. This will facilitate prompt implementation of treatment and, where appropriate, postexposure prophylaxis. Rapid implemen- tation of measures such as vaccination, isolation of patients, and quarantine of contacts can ameliorate the spread. Tertiary preven- tion includes early treatment and rehabilitation of those people who contract the disease and public information campaigns to reduce the long-​term psychological impact of the incident. Clinical features The incubation periods of potential bioterrorism agents can vary from as little as several hours to weeks. The incubation period for smallpox is between 7 and 14 days, but could be less following ex- posure to aerosol. Pneumonic plague is likely to develop within 24 h to 2 days after aerosol exposure. Inhalation anthrax has an incuba- tion period of 1–​6 days, but is probably dose-​related, and could be longer than 40 days. For inhaled botulinum toxin, the incubation period is estimated to be between 12 and 80 h and for ricin, perhaps even less. Diseases such as anthrax, smallpox, and tularemia usually present early on with influenza-​like illnesses, but if exposure is by aerosol, the symptoms might differ from the naturally occurring diseases. Diseases like plague and tularemia can present as pneumonia. Agents such as smallpox will subsequently develop a typical rash. In the later stages, both anthrax and smallpox commonly develop neurological symptoms. The haemorrhagic fevers are characterized initially by high fever and bleeding tendencies. Inhaled botulinum toxin causes acute, afebrile, descending flaccid paralysis starting with ptosis and muscles innervated by cranial nerves. Inhaled ricin causes fever, chest tightness, dyspnoea, nausea, and arthralgia, within 4–​8 h, followed by acute respiratory distress syndrome and death within 18–​24 h. Differential diagnosis The early symptoms of diseases caused by potential bioterrorism agents can mimic a large spectrum of diseases since influenza-​like illness is a common presentation for many. Since diseases such as plague and tularemia can present with pneumonia, cases might not be suspected unless they occur in clusters. Even with the classical sign of widened mediastinum which frequently characterizes in- halation anthrax, it might not be simple to distinguish from other severe pneumonias. Early identification of deliberately caused out- breaks will depend largely on the ability of primary care and emer- gency room physicians to identify and promptly report cases to the public health authorities. A major concern is that diagnosis of these uncommon diseases will not be considered by physicians who have rarely, if ever, seen such cases. Many of the biological agents can be identified by hospital labora- tories. However, some require more specialized laboratories and international collaboration. New techniques, especially those based on polymerase chain reaction, are being developed to accelerate spe- cific diagnosis. The safety of laboratory workers must be protected. Surveillance and early detection The objectives of surveillance for bioterrorism incidents are twofold. Firstly, early detection of cases can facilitate prompt treatment, iden- tification of the exposure source, rapid introduction of prophylaxis and, where necessary, isolation of cases and imposition of quaran- tine. Secondly, surveillance systems have a major role in monitoring the progress of an outbreak to support decisions on upgrading and redistributing health services and provide reliable and timely infor- mation to the media and the public. Traditional surveillance, based on routine physicians’ reports, could have serious limitations in a bioterrorism incident. Early cases might be missed due to a failure to suspect unusual diseases. Thus, there might be considerable delays in alerting public health

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1722 authorities due to the lag time between the initial symptoms and definitive diagnosis. Furthermore, access to timely, processed infor- mation during the epidemic might be seriously limited. Recognizing these limitations, surveillance for symptoms and signs, known as ‘syndromic surveillance’, has been proposed as a more sensitive method for early detection of an outbreak. Although theoretically appealing, in practice, syndrome surveillance is likely to be most useful to complement early detection and reporting of the index cases by alert physicians. Sources of data for syndromic surveillance are usually recorded visits to primary care physicians and emergency rooms and use of prescription and non​prescription medication. Computer analysis of the data allows temporal and geo- graphical trends to be identified. Clusters in families or in age groups will be useful in locating the exposure source. Surveillance systems must include clear procedures to be followed when a suspected in- cident is reported. Although syndromic surveillance systems are highly sensitive, they tend to have both low specificity and positive predictive value. Abundant false positive reports could desensitize and paralyse the system. Despite the limitations of syndromic surveillance, once an out- break has been confirmed, such systems will provide timely data on the location, nature, and evolution of the outbreak. Electronic data systems will, however, be important for confirming and tracking the outbreak, and they can reduce delays in reporting and reliance on reports from individual physicians. They can play a critical role in the control of the epidemic by providing information for ‘situational awareness’. Epidemiological investigation The main objectives of the investigation are to identify and charac- terize the outbreak and predict its course. For bioterrorism incidents, the investigators should have specialized knowledge of the possible biological agents and the natural history of the diseases. Close col- laboration with the police, public health authorities, and the media is essential. Patient details should include the date and time when symptoms started, signs and symptoms and, when smallpox is sus- pected, the vaccination history. It is important to establish which public places patients have visited in the incubation period of the suspected agent. Those patients reported to have similar symptoms and contacts should be interviewed. It is important to document the natural history of the disease for each patient. When investigating po- tential bioterrorism outbreaks, there is a need to identify the source of the implicated pathogen. This was typified in the investigation of the anthrax letters incident in the United States in 2001. Recently the investigators in that outbreak used comparative genome analysis and demonstrated that the genotypes detected in the B. anthracis morphotypes isolated for the letters were different from the Ames strain common in the environment. This study has provided support for the added value of whole-​genome sequencing, and comparative genomics for potential bioterrorism outbreak investigations. Postexposure prophylaxis Postexposure prophylaxis is appropriate for some, but not all, bio- terrorism agents. Vaccination against smallpox, within 3–​4  days of exposure, appears to provide protection against clinical disease. However, since the incubation period is usually longer than 4 days, the lag time for recognizing index cases might render postexposure vaccination effective only for contacts of those initially exposed. ‘Ring vaccination’ involves intensive tracing and vaccination of all primary contacts, followed by vaccination of the secondary contacts as opposed to mass vaccination immediately following diagnosis of the first cases. Ring vaccination accompanied by vaccination in af- fected regions, followed by countrywide mass vaccination, is likely to be the most effective strategy. This approach has now also been proposed for the control of Ebola virus disease epidemics. For some agents, postexposure prophylaxis with antimicrobials has a role. Ciprofloxacin, doxycycline, and ampicillin are used for postexposure prophylaxis against anthrax and plague. In the case of anthrax, it should be combined with vaccination. Results of animal studies suggest that postexposure antivirals could be ef- fective in a smallpox outbreak. Ribavirin might have some efficacy in postexposure prophylaxis of RNA viral haemorrhagic fevers such as arenaviruses. Isolation and quarantine For contagious diseases, such as smallpox and plague, isolation of patients, barrier nursing, quarantine of contacts, and restriction of the movements and social interactions of people are important control measures. Results of modelling studies suggest that closing schools and reducing crowding and the use of public transport would be effective in limiting the spread of an epidemic. For contagious diseases, there are specific guidelines for the use of masks by healthcare personnel and emergency workers. Surgical masks might be adequate for droplet spread, whereas N95-​type masks would be necessary to protect against aerosols. However, they are more expensive, require special fitting, and cannot be worn for long periods. The efficacy and practicability of the use of masks by the general public are less clear. Public education and risk communication The novel and largely unpredictable effects of biological weapons are likely to increase the uncertainty surrounding a bioterrorism incident. Public education and effective risk communication are essential in order to bolster public confidence and improve co- operation with the authorities. Clinicians and public health per- sonnel should have access to up-​to-​date information. The general public should be provided with nontechnical descriptions of the diseases and simple instructions on how to act in an emergency situation. Risk communication associated with a bioterrorist event can be divided into five stages: prior to the event, on suspicion of an event, on confirmation of the event, during the event, and following the event. At each stage, the public is likely to ask questions relevant to that stage. Since the authorities may possess very little factual information, the public might suspect that information is being withheld, resulting in hostility. Thus, it is important that the public messages be reassuring while sharing uncertainties. Overreaction or panic should be anticipated. This can be exacerbated by rumours

10.3.9  Bioterrorism 1723 or unsubstantiated statements by professionals or laypeople not in- volved in managing the outbreak. A variety of problems should be anticipated during an outbreak, including atypical presentations of cases and varying responses to treatment and prophylaxis. Side effects of the medications and vaccines might be reported. Discovery of new exposure foci and reports of disease in apparently unexposed people could cause dis- quiet and mistrust. There might be inadequate isolation of patients and a breakdown of the implementation of quarantine. Untried new treatments might be proposed by unauthorized professionals or lay people. Following a bioterrorist incident, residual public fear and anx- iety is likely to persist. Inevitably, there will be questions about the extent to which the authorities were able to control the incident, criticism of actions taken or not taken, and general recriminations. Public messages should be broadcast about the lessons learned from the incident and actions that will be taken to address deficiencies. It is essential that the authorities maintain transparency in order to strengthen public trust. Decontamination Decontamination is relevant, mainly for anthrax and smallpox, in the environment of an aerosol attack and at places where patients were treated. Sodium hypochlorite solution is effective in most set- tings. Bedding and clothing of patients should be sterilized or dis- posed of where indicated. Low humidity and temperature prolong survival of the smallpox virus in the environment, and on scab ma- terial; it can remain viable for months. Legal and ethical aspects Bioterrorism preparedness requires the necessary legislation to en- able the public health authorities to carry out measures with ad- equate legal backing. Laws that are of particular importance relate to closing buildings, taking over hospitals, ordering isolation and quarantine, and active surveillance of presumed infected individ- uals and their contacts. Ethical issues might arise in the triage of patients for admission to overburdened hospital wards and inten- sive care units. Areas of uncertainty or controversy Bioterrorism incidents have so far been very rare, and preparedeness is based on an assumption that the potential risk is both real and severe. There are some concerns that the investment of large re- sources in bioterrorism preparedness could come at the expense of other essential public health activities. Research should be en- couraged to assess the risks, costs, and benefits of the preparedness activities, in order to strike a reasonable balance. New surveillance systems, particular those based on syndromic surveillance, might be insufficiently specific and too much of a burden on the health services to be sustainable for long. Uncertainty remains about the efficacy of vaccines and antimicrobial therapy in the event of an out- break. Renewed assessment of conventional approaches is particular appropriate now, in the light of the experience gained during the recent Ebola virus disease epidemic in West Africa. Likely future developments The threat of bioterrorism is likely to increase, demanding greater resources to deter attacks and improve surveillance, vaccines, and medications. Conclusions Bioterrorism is a low-​risk but high-​impact public health emergency. Deterrence remains the prime goal. Reducing the motivation for terror and banning internationally the use of biological weapons should be promoted at all levels. Sensible preparedness for bioter- rorist incidents is a deterrent in itself and ensures that public health systems and society will deal effectively with an incident. Risk com- munication needs to be strengthened. Such measures will also im- prove general emergency preparedness and the control of infectious diseases. FURTHER READING Adalja AA, Toner E, Inglesby TV (2015). Clinical management of potential bioterrorism-related conditions. New England Journal of Medicine, 372, 954–62. Armengaud J (2017). Striking against bioterrorism with advanced proteomics and reference methods. Proteomics, 17, (1–2). Arnon SS, Schechter R, Ingelsby TV, et al (2001). Botulism toxin as a biological weapon. Medical and public health management. Journal of the American Medical Association, 285, 1059–70. Banada PP, Deshpande S, Chakravorty S, et al (2016). Sensitive
detection of Francisella tularensis directly from whole blood by use of the GeneXpert System. Journal of Clinical Microbiology 55, 291–301. Banada PP, Deshpande S, Russo R, et al (2017). Rapid Detection of Bacillus anthracis bloodstream infections by use of a novel assay in the GeneXpert System. Journal of Clinical Microbiology 55, 2964–71. Bonin S (2007). In A Wegner, V Mauer, M Dunn, series ed. International Biodefense Handbook. An inventory of national and international biodefense practices and policies; Center for Security Studies at ETH Zurich. Borio L, Ingelsby T, Peters CJ, et al (2002). Hemorrhagic fever viruses as biological weapons. Journal of the American Medical Association, 287, 2391–405. Bozzette SA, Boer R, Bhatnagar V, et al (2003). A model for a smallpox vaccination policy. New England Journal Medicine, 348, 416–25. Covello VT, Peters RG, Wojtecki JG, Hyde RC (2001). Risk com- munication, the West Nile virus epidemic, and bioterrorism: responding to the communication challenges posed by the inten- tional or unintentional release of a pathogen in an urban setting. Journal of Urban Health, 78, 382–91. Darsow U, Sbornik M, Rombold S, et al (2016). Long-term safety of replication-defective smallpox vaccine (MVA-BN) in atopic eczema and allergic rhinitis. Journal of European Academy of Dermatology and Venereology, 30, 1971–7.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1724 Dennis DT, Ingelsby TV, Henderson DA, et al (2001). Tularemia as a biological weapon. Medical and public health management. Journal of the American Association, 285, 2763–73. Green MS, LeDuc J, Cohen D, Franz DR. Confronting the threat of bioterrorism – realities, challenges and defensive strategies (2018). Lancet Infectious Diseases, (In press). Henderson DA, Ingelsby TV, Bartlett JG, et al (1999). Smallpox as a biological weapon. Medical and public health management. Journal of the American Medical Association, 281, 2127–37. Kupferschmidt K (2017). Labmade smallpox is possible, study shows. Science, 357, 115–6. Mechaly A, Vitner E, Levy H, et al (2018). Simultaneous immuno­ detection of anthrax, plague, and tularemia from blood cul- tures by use of multiplexed suspension arrays. Journal of Clinical Microbiology, 26, 56(4). pii: e01479–17. doi: 10.1128/JCM.01479-17. Print 2018 Apr. Meselson M, Guillemin J, Hugh-Jones M et al (1994). The Sverdlovsk anthrax outbreak of 1979. Science, 266, 1202–8. Miller SL, Clements N, Elliott SA, et al (2017). Implementing a negative-pressure isolation ward for a surge in airborne infectious patients. American Journal of Infection Control, 45, 652–9. Mortimer PP (2003). Can postexposure vaccination against smallpox succeed? Clinical Infectious Diseases, 36, 622–8. Pennisi E (2016). Pocket DNA sequencers make real-time diagnostics a reality. Science, 351, 800–1. Quaglio G, Goerens C, Putoto G, et al (2016). Ebola: lessons learned and future challenges for Europe. Lancet Infectious Diseases, 16, 259–63. Rotz LD, Khan AS, Lillibridge SR, Ostroff SM, Highes JM (2002). Public health assessment of potential biological terrorism agents. Emerging Infectious Diseases, 8, 225–30. Sandman PM. Bioterrorism risk communication policy (2003). Journal of Health Communication, 8 Suppl 1, 146–7; discussion 148–51. Satterly NG, Voorhees MA, Ames AD, Schoepp RJ (2016). Comparison of MagPix assays and Enzyme-Linked Immunosorbent Assay for detection of hemorrhagic fever viruses. Journal of Clinical Microbiology, 55, 68–78. Schoch-Spana M, Cicero A, Adalja A, et al (2017). Global catastrophic biological risks: toward a working definition. Health Security, 15, 1–6 Sharp NJ, Molineux IJ, Page MA, Schofield DA (2016). Rapid de- tection of viable Bacillus anthracis spores in environmental sam- ples by using engineered reporter phages. Applied Environmental Microbiology, 82, 2380–7. Stern D, Olson VA, Smith SK, et al (2016). Rapid and sensitive point- of-care detection of Orthopoxviruses by ABICAP immunofiltra­ tion. Virology Journal, 13, 207. Sun W (2016). Plague Vaccines: Status and Future. Advances in Experimental Medicine and Biology, 918, 313–36. Suwantarat N, Apisarnthanarak A (2015). Risks to healthcare workers with emerging diseases: lessons from MERS-CoV, Ebola, SARS, and avian flu. Current Opinion in Infectious Diseases, 28, 349–61. Tao P, Mahalingam M, Zhu J, et al (2017). A bivalent Anthrax-Plague vaccine that can protect against two Tier-1 bioterror pathogens, Bacillus anthracis and Yersinia pestis. Frontiers in Immunology, 8, 687. Titball RW, Burtnick MN, Bancroft GJ, Brett P (2017). Burkholderia pseudomallei and Burkholderia mallei vaccines: Are we close to clinical trials? Vaccine, 35, 5981–9. Watson AK, Ellington S, Nelson C, et al (2017). Preparing for bio- logical threats: Addressing the needs of pregnant women. Birth Defects Research, 109, 391–8. Webb RP, Smith TJ, Smith LA, et al (2017). Recombinant Botulinum neurotoxin Hc Subunit (BoNT Hc) and catalytically inactive Clostridium botulinum Holoproteins (ciBoNT HPs) as vaccine candidates for the prevention of botulism. Toxins (Basel), 9, 269, pii E269.

10.4 Poisoning 1725

10.4 Poisoning 1725

10.4.1 Poisoning by drugs and chemicals 1725

10.4.1 Poisoning by drugs and chemicals 1725

10.4 Poisoning CONTENTS 10.4.1 Poisoning by drugs and chemicals  1725 John A. Vale, Sally M. Bradberry, and D. Nicholas Bateman 10.4.2 Injuries, envenoming, poisoning, and allergic
reactions caused by animals  1778 David A. Warrell 10.4.3 Poisonous fungi  1817 Hans Persson and David A. Warrell 10.4.4 Poisonous plants  1828 Michael Eddleston and Hans Persson 10.4.1  Poisoning by drugs
and chemicals John A. Vale, Sally M. Bradberry, and D. Nicholas Bateman ESSENTIALS Poisoning is usually an acute, short-​lived event which necessitates immediate care, though complications such as rhabdomyolysis may persist for a few days. Less commonly, symptoms may arise only after prolonged exposure, as occurs with many heavy metals. Rarely, sequelae may not occur until many years after exposure (e.g. with vinyl chloride). It must be stressed that exposure does not necessarily equate with poisoning as uptake of the agent involved is required but, even if this occurs, poisoning does not necessarily result as the amount absorbed may be too small. Poisoning may be accidental or deliberate; it is usually accidental in small children, but in adults it is almost invariably deliberate. Less commonly, it may be iatrogenic. Occupational poisoning is frequent in developing countries. Clinical assessment Assessment of a poisoned patient involves taking an appropriate history and performing a physical examination (including an as- sessment of the level of consciousness, ventilation, and circulation). Diagnosis is based on the history, circumstantial evidence (if avail- able), the presence of typical features, and, occasionally, on the re- sults of toxicological and other investigations. The medical approach should never be confined to the poison and its effects (e.g. deliberate self-​harm may be indicative of a significant psychiatric disorder that requires diagnosis and treatment). Biochemical abnormalities due to disturbed metabolic pro- cesses are common in severely poisoned patients. These may be of diagnostic value, but mostly their recognition and treatment are important in management. Acid–​base disturbances, particularly re- spiratory acidosis (due to central nervous system depression or pul- monary toxicity), and metabolic acidosis (due to lactic acidaemia or derangements of intermediary metabolism), are common. Plasma electrolyte abnormalities, particularly hypo-​ or hyperkalaemia, are observed and are most often due to redistribution of potassium across cell membranes. Hypoglycaemia and, less commonly, hyper- glycaemia may also occur. Management Initial management involves the treatment of any potentially life-​ threatening conditions, such as airway compromise, breathing difficul- ties, haemodynamic instability, and clinically significant arrhythmias. Thereafter, convulsions and temperature disturbances should be treated and fluid, acid–​base, and electrolyte abnormalities corrected. There is no evidence that the use of methods to reduce absorption from the gastrointestinal tract—​such as activated charcoal, gastric lavage, syrup of ipecacuanha, cathartics, or whole-​bowel irrigation—​ improves the clinical outcome in poisoned patients. However, ac- tivated charcoal and gastric lavage may be considered in patients who have ingested life-​threatening amounts of a toxic agent up to 1 h previously. Antidotes exert their beneficial effects by a variety of mechan- isms, including forming an inert complex with the poison, acceler- ating detoxification of the poison, reducing the rate of conversion of the poison to a more toxic compound, competing with the poison for essential receptor sites, blocking essential receptors through which the toxic effects are mediated, and bypassing the effect of the poison. There are, however, only a small number of poisons for which there is a specific antidote, and few antidotes are employed regularly in clinical practice; these include acetylcysteine, naloxone, and flumazenil. To increase poison elimination, treatment with multiple-​dose ac- tivated charcoal (in patients who have ingested carbamazepine, dap- sone, phenobarbitol, quinine, or theophylline), urine alkalinization

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1726 (in patients with moderately severe salicylate poisoning) or haemo- dialysis (which significantly increases the elimination of ethanol, ethylene glycol, isopropanol, lithium, methanol, and salicylate) should be considered, although there is no conclusive evidence that these treatments improve outcome. Most countries have a poisons information service, which pro- vides advice to medical staff (e.g. in the United Kingdom healthcare professionals may obtain online advice from https://​www. TOXBASE.org and by telephone), and in most cases, to the general public as well. Advice should always be sought if unfamiliar poisons are encountered or if there is clinical uncertainty about optimal management. Introduction Poisoning is usually an acute event demanding immediate care and attention, but the consequences of exposure sometimes per- sist. Distinctive sequelae may not appear until many years have elapsed (e.g. with carcinoma of the oesophagus following ingestion of corrosives or hepatic haemangiosarcoma from vinyl chloride exposure). Symptoms may arise only after prolonged exposure, as with many metals. Exposure by oral, inhalational, dermal, or other routes on their own does not necessarily indicate poisoning. Uptake is required for there to be a toxic effect, but even if this occurs, poi- soning does not necessarily result as the amount absorbed may be too small. If poisoning does occur, the ensuing clinical syndrome may be distinctive; for example, fixed dilated pupils, exaggerated tendon re- flexes, extensor plantar responses, depressed respiration, and car- diac tachyarrhythmias suggest tricyclic antidepressant poisoning; anaemia, constipation, colic, and motor nerve palsies are indicative of lead poisoning. However, with many psychotropic medicines there may only be non​specific central nervous depression, respira- tory impairment, and hypotension. Poisoning may be accidental or deliberate. It is usually accidental in small children, but in adults it is almost invariably deliberate (de- liberate self-​harm) or, rarely, it may be with homicidal intent. It may also be iatrogenic in those aged under six months (e.g. involving overtreatment with paracetamol). Occupational poisoning is common in developing countries and continues to occur in the developed world. The medical approach to poisoning should never be confined to the poison and its effects. All the circumstances surrounding the episode must be considered, especially in cases where litiga- tion may follow (e.g. in the event of an occupational mishap with a chemical). It is therefore important that the doctor concerned, having instituted any necessary life-​saving measures, should take a careful history, retain all pertinent evidence such as a sui- cide note and biological specimens, make a meticulous record of symptoms, signs, progress, and outcome, and remember issues of confidentiality. Most countries have a poisons information service, which pro- vides advice to medical staff (e.g. in the United Kingdom healthcare professionals may obtain online advice from https://​www. TOXBASE.org and by telephone), and in most cases, to the general public as well. Advice should always be sought if unfamiliar poisons are encountered or if there is clinical uncertainty about optimal management. Epidemiology Poisoning, either accidental or deliberate, is a common presentation in all countries throughout the world. This phenomenon is, how- ever, a relatively new one. Before the 1950s, hospital admissions from self-​harm, now the most frequent cause of poisoning presen- tation to healthcare, was extremely rare worldwide. The reasons for this changing pattern of poisoning are poorly understood. Patients who have suffered toxic exposures present to healthcare facilities in a variety of ways, including to primary care physicians, hospital emergency departments, and hospital outpatients; rarely, patients are discovered dead. Collecting statistics on poisoning is, therefore, a complex issue and there is currently no universally agreed system for documenting and comparing rates of poisoning in different countries. Most statistics refer to hospital admissions (as opposed to hospital presentations in emergency departments) or poisoning-​related deaths. Health statistical data from developed countries are usually more sophisticated than those from the developing world, although local surveys suggest that the incidence of self-​harm is little different in these different types of society. Poisons information centres also col- lect information about the types of agent people are exposed to or ingest, many of which do not result in clinical ill health, further com- plicating health statistics. There are clear age differences in the frequencies and causes of poisoning. In children under the age of 10, accidental poisoning is extremely common, particularly in the very young who tend to place household objects into their mouths. From the age of 10 up- wards, self-​harm becomes predominant, peaking in the late teens to late twenties and then gradually declining in incidence in higher age groups. The health departments of most developed countries publish data on poisoning mortality on the internet; collection of hospital admission data is less routine and is best in countries which have centralized healthcare provision. Hospital admissions due to poisoning Poisoning causes 5–​10% of acute hospital medical presentations in developed countries. In the United Kingdom, there are currently 350 000 to 400 000 per annum. Since deliberate self-​harm is a risk factor for further such episodes, approximately 25% of these cases occur in the same patient group. Self-​harm in women and men is somewhat different, and other than young children, mortality data for out-​of-​hospital deaths show that men are more likely to succeed in killing themselves than women. The severity of poisoning depends on the quantities ingested, but hospital statistics do not provide adequate data to as- sess this. Mortality data provide information on the relative toxicity of different compounds if it can be expressed per head of popula- tion exposed. This is a technique that is best used for assessing the toxicity of prescription medicines; it is much less easily applied to chemicals and household products when measures of availability are not so readily obtained. Many cases of ‘poisoning’ in children are more accurately de- scribed as ‘exposures’, since symptomatic poisoning is uncommon,

10.4.1  Poisoning by drugs and chemicals 1727 particularly in developed countries. While drug errors can result in poisoning, data are very difficult to collect. Patients who harm themselves often do so because they are acutely stressed. Few have formal psychiatric diagnoses, such as depression or psychosis, or are truly suicidal, since the incident is impulsive rather than planned. Such differences in behaviour affect mortality rates. Mortality is often higher in older age groups where overdose planning has been more careful, and has involved prescrip- tion medicines which typically are more toxic than over-​the-​counter preparations taken by younger patients. Impulsive behaviour is often associated with ingestion of alcohol and as many as two-​thirds of men, and nearly one-​half of women, take alcohol in association with an overdose. In many cases of self-​harm more than one drug is in- cluded in the cocktail, making clinical management more complex, particularly if two or more agents acting on the same body system are involved. This applies particularly to drugs acting on the brain, kidney, and cardiovascular system. The increasing worldwide use of drugs of abuse has also in- fluenced patterns of poisoning, and many cases of poisoning in this population result from variations in quality of supply or experimentation. Prescription medicines are used in most self-​harm episodes in the United Kingdom, the rest of Europe, North America, and countries in the developed world and their availability, therefore, influences the numbers of patients seen. The diagnoses for which the drug is used will also affect how often it is taken as a self-​harm agent. Thus, self-​harm with drugs for peptic ulcer disease is very uncommon, whereas overdose with antidepressants is much more frequent. The type of agent taken in overdose is culturally determined. In the United Kingdom, paracetamol contributes approximately one-​ third of all poisonings seen in hospitals but, although common in North America and other parts of Western Europe, the proportion of cases is lower. In developing countries, such as Sri Lanka and India, drugs are much more expensive and the agents ingested are either plants, such as yellow oleander, or the widely available pes- ticides. Agrichemicals cause less than 0.05% of hospital admissions for poisoning in England and Wales, whereas in Sri Lanka they are associated with around 70% of all cases of self-​harm. Consequently, although the numbers of patients self-​harming per head of popula- tion are quite similar, the mortality rates in Sri Lanka are orders of magnitude higher than in Western Europe. Deaths from poisoning In developed countries, most deaths from poisoning occur before admission, and less than 1% of patients presenting to hospital with poisoning are likely to die. The risk of mortality is very dependent on agent, and heroin, cocaine, and other high-​risk recreational drugs are associated with higher death rates. In England and Wales, the mortality from drug-​related poisoning was falling but has recently increased and in 2015 was higher than any time since 1993 with 3674 deaths; 2479 (67%) involved illegal drugs. In previous decades, carbon monoxide (from coal gas) and barbit- urates were common causes of death. Substitution of natural gas for coal gas and changes in prescribing patterns have altered the agents most frequently associated with death. For example, the United Kingdom and European Medicines Agencies have enacted legisla- tion on co-​proxamol (paracetamol plus dextropropoxyphene) in re- sponse to concerns about its toxicity in overdose. Such changes have significantly reduced mortality rates from this cause in the United Kingdom. Childhood poisoning Accurate data on childhood poisoning are difficult to obtain. Many children with relatively mild features will be managed at home or in emergency departments, where national statistical data are not routinely collected, and so in this population national statistics are unreliable, except for agents that cause death. Deaths in children are usually attributable to inappropriate storage, including toxic pharmaceuticals, such as digoxin and quinine, household products and, increasingly, drugs of abuse. The pattern of child poisoning varies in different countries. Herbicide expos- ures are common in agricultural countries where these materials are stored in the home. In developed countries, exposures may occur when a young child finds and takes a relative’s medicines. Child-​resistant containers have reduced poisoning rates in chil- dren, but tragedies still occur. Diagnosis Diagnosis of acute poisoning requires that the doctor not only es- tablish that exposure to a poison has occurred, but also its chem- ical composition and the route and magnitude of exposure, so that the features likely to develop can be anticipated and risk assessed. As in any other branch of medicine, diagnosis of acute poisoning is based on the patient’s history and on a combination of circumstan- tial evidence, the findings on physical examination and appropriate investigations. However, in acute poisoning, there are many obstacles to establishing the information required. Young children may not be able to give a history; adults are often unreliable; physical signs are rarely diagnostic; circumstantial evidence may be unavailable, tenta- tive, or misleading; and laboratory diagnosis is rarely comprehensive. History Since accidental poisoning in childhood is most common between the ages of 9 months and 5 years, an unequivocal history is unlikely to be forthcoming from the victim but may be obtainable from older witnesses. However, statements about quantities must be interpreted cautiously since an accurate assessment of the amounts in original containers is rarely available. In contrast, since 90% or more of adults presenting with acute poi- soning are conscious or drowsy, it should be possible to obtain a his- tory of self-​poisoning. A few patients adamantly deny having taken poisons but most usually admit to it without hesitation, although problems arise in trying to establish precisely the nature and quan- tity of what has been taken. Comparison of patients’ statements with poisons detected by laboratory analysis of blood or urine consist- ently reveals major differences in about half the cases. Consequently, patients are often thought to be deliberately untruthful. However, self-​poisoning is commonly an impulsive act. The patient ingests the contents of the first bottle that comes to hand, often while under the influence of alcohol, and so inaccuracies in the history are not sur- prising. Although about 60% of episodes involve drugs prescribed for the victims or their relatives, like many other patients they are often ignorant of the names.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1728 Assessment of the amounts of drugs ingested are even more dif- ficult. Few patients count the number of tablets they consume and neither patient nor doctor can accurately interpret a ‘handful’, a ‘strip’, or similar arbitrary quantity. Circumstantial evidence In the diagnosis of acute poisoning, circumstantial evidence be- comes important when patients are unable to give a history (as is likely with young children), are confused, or are unconscious, or are unwilling to do so. However, although circumstantial evidence may strongly suggest poisoning, it is seldom incontrovertible. It takes several forms. Circumstances under which found In the case of infants, the mother may return to the kitchen or bath- room to find her child with some substance all over their hands, face, and clothing, or surrounded by pills, one of which the child may be eating. The assumption that some has been ingested may not be cor- rect, and the amount swallowed is a matter of speculation. Self-​poisoning is a common cause of coma in previously healthy young adults. Adults may be found unconscious with tablet particles around the mouth or on clothing as the only clue to diagnosis. More often, the presence of empty drug containers with occasional tablets or capsules near the patient suggests the diagnosis. Less commonly, they are found unconscious or dead in some remote location. The lack of personal effects to indicate who they are or where they live may suggest a desire not to be identified and should arouse suspicion of poisoning. Protestations by relatives that the patient would never take an overdose are often incorrect and should not prevent full investigation in appropriate circumstances. Suicide notes Suicide notes are reliable indicators of poisoning in the absence of physical violence as a cause of coma. The note may specify what has been taken in addition to expressing despair, futility, worthlessness, and remorse. Features There are few symptoms or physical signs that cannot be attributed to one poison or another. However, a clinical feature rarely arises in isolation and clusters of features are of much greater diagnostic value. Those most commonly encountered in present-​day practice are given in Table 10.4.1.1. Conscious patients with abnormal behaviour, perhaps in com- bination with auditory and visual hallucinations, may have ingested amfetamines or other psychoactive stimulants, such as phen- cyclidine, lysergic acid diethylamide (LSD), ‘magic’ (psilocybin-​ containing) mushrooms (see Chapter  10.4.3), or drugs such as the older antihistamines and tricyclic antidepressants, which have marked anticholinergic actions. Drowsiness, ataxia, dysarthria, and nystagmus are common after ingestion of benzodiazepines. Coma with hypotonia and hyporeflexia may follow, particularly if alcohol has also been taken. Hypotension, hypothermia, and respiratory depression are rare. Poisoning with tricyclic antidepressants causes hypertonia, hyperreflexia, extensor plantar responses, and dilated pupils. Sinus tachycardia and pro- longation of the QRS interval on the electrocardiogram support a diagnosis of intoxication with these drugs; hypotension and hypo- thermia are uncommon. Tricyclic antidepressants and mefenamic acid are common causes of seizures. Coma with pinpoint pupils and a reduced respiratory rate is virtually diagnostic of poisoning with opioid analgesics and is an indication for a therapeutic trial of naloxone. Many patients with opioid poisoning will be habitual drug abusers and have vene- puncture marks and evidence of venous tracking, particularly in the antecubital fossae. Alcohol may be smelt on the breath, as might solvents such as toluene, acetone, or xylene as the result of ‘sniffing’ glues, cleaning agents, or other preparations. Burns around the lips or in the buccal cavity or pharynx indicate inges- tion of corrosives. Skin blisters Skin blisters may be found after poisoning with a wide variety of drugs including barbiturates, tricyclic antidepressants, benzodi- azepines, and non​drug toxins. They often occur over bony prom- inences that have been subjected to pressure and, less frequently, at sites where two skin areas have been in contact (e.g. the inner aspects of the knees) and are not specific for any poison. Neurological signs Lateralizing neurological signs Since most serious poisonings are associated with impairment of consciousness, neurological signs are particularly important. Lateralizing signs (unless they are attributable to a known neuro- logical disease) virtually exclude a diagnosis of acute poisoning. Table 10.4.1.1  Common feature clusters in the poisoned patient Feature cluster Likely poisons Coma, hypertonia, hyperreflexia, extensor plantar responses, myoclonus, strabismus, mydriasis, sinus tachycardia Tricyclic antidepressants: less commonly antihistamines, orphenadrine, thioridazine Coma, hypotonia, hyporeflexia, flexor or non​elicitable plantar responses, hypotension Barbiturates, benzodiazepines, and alcohol combinations, severe tricyclic antidepressant poisoning Coma, miosis, reduced respiratory rate Opioid analgesics Nausea, vomiting, tinnitus, deafness, sweating, hyperventilation, vasodilatation, tachycardia Salicylates Hyperthermia, tachycardia, delirium, agitation, mydriasis MDMA (Ecstasy) Amfetamines Miosis, hypersalivation, rhinorrhoea, bronchorrhoea Organophosphorus and carbamate insecticides, nerve agents

10.4.1  Poisoning by drugs and chemicals 1729 Such findings have been recorded with barbiturate and phenytoin poisoning, but so rarely that the general rule is not compromised. Pyramidal tract signs The usual features of pyramidal tract involvement (hypertonia, hyperreflexia, and extensor plantar responses) are commonly found in tricyclic antidepressant poisoning and with other drugs with marked anticholinergic actions (e.g. the older antihistamines). However, these signs may be abolished in deep coma. Abnormal movements Unconscious patients may respond to painful stimuli with flexor and extensor limb movements of the type seen in decorticate and decere- brate states. However, in poisoning, these signs do not indicate irre- versible brain damage, and patients showing them can be expected to recover fully; hypoglycaemia must be excluded in these cases. Acute dystonic movements (including acute torticollis, orolingual dyskinesias, and oculogyric crises) are also produced; these are usually caused by first-​generation antipsychotics such as chlorpro- mazine, haloperidol, or prochlorperazine. Choreoathetosis has been reported as a rare presenting feature of poisoning with organophos- phorus insecticides. Pupillary changes in poisoning Inequality of the pupils is not uncommon in poisoned patients. Widely dilated pupils that react poorly to light may be caused by poisons with anticholinergic actions (e.g. tricyclic antidepressants) or sympathomimetic effects (e.g. amfetamines) or agents causing blindness (e.g. quinine, methanol). Miosis is usually caused by opioid analgesics or poisons with cholinergic or anticholinesterase actions (e.g. organophosphorus insecticides, nerve agents). The degree and speed of reaction of the pupils to light is of no clinical value. Ocular signs A variety of ocular signs including strabismus, internuclear oph- thalmoplegia, and total external ophthalmoplegia, may be found in acutely poisoned patients. Strabismus has been described in poi- soning with phenytoin, carbamazepine, and tricyclic antidepres- sants. Usually the optic axes diverge in the horizontal plane but, in some patients, there is additional vertical deviation. It is present transiently and only in patients who are unconscious. Dysconjugate, roving eye movements may also be seen if both eyes are observed for a period. It is important to know that such abnormalities occur so that they are not misattributed to intracranial vascular lesions or some other pathology requiring surgical intervention. Dysconjugate eye movements may become apparent only when oculovestibular reflexes are examined by caloric stimuli. Installation of ice-​cold water into the external auditory meatus should make both eyes turn to the side irrigated, and failure of one eye to de- viate is evidence of internuclear ophthalmoplegia and a lesion of the medial longitudinal fasciculus. This has been reported in poisoning with a variety of drugs including tricyclic antidepressants, pheno- thiazines, benzodiazepines barbiturates, and ethanol, and can be de- tected in 10% of cases if caloric tests are carried out. Both sides are usually affected, but internuclear ophthalmoplegia on testing one side only also occurs in acute poisoning. Loss of oculocephalic and oculovestibular reflexes It is widely accepted that absence of oculocephalic and oculovesti­ bular responses indicates severe brainstem damage and the likeli- hood that the patient will not survive. However, this is not the case in acute poisoning where these reflexes may be abolished in patients who subsequently make a full recovery. Visual impairment Visual impairment is associated most commonly with quinine and methanol poisoning. Investigations Haematological and biochemical Information about the nature of poisons ingested can occasionally be deduced from standard haematological and biochemical investi- gations, and from arterial blood gas analysis (Table 10.4.1.2). Toxicological screening Toxicological screening for poisons in an unconscious patient is often requested when the cause of coma is unknown. Although identification of a drug or other chemical may reassure the clin- ician, this alone is not a good reason for the request. The clinician should consider how the result of a screen will alter management. The pattern of drugs involved in poisoning in most developed countries is such that specific treatment (e.g. antidotes, techniques to enhance elimination of the poison) is unlikely to be available, and management will therefore be supportive. Screening is labour-​ intensive, time-​consuming, and expensive, and in most cases, cannot be justified on an emergency basis because it will not alter the management of the patient, though there are important excep- tions (Table 10.4.1.3). Table 10.4.1.2  Haematological and biochemical investigations that assist in management • Serum sodium concentration (e.g. hyponatraemia in ecstasy (MDMA) poisoning) • Serum potassium concentration (e.g. hypokalaemia in theophylline poisoning, hyperkalaemia in digoxin poisoning, rhabdomyolysis, haemolysis) • Plasma creatinine concentration (e.g. renal failure in ethylene glycol and diethylene glycol poisoning) • Blood sugar concentration (e.g. hypoglycaemia in insulin and severe untreated paracetamol poisoning, hypoglycaemia, and hyperglycaemia in salicylate poisoning) • Serum calcium concentration (e.g. hypocalcaemia in ethylene glycol poisoning) • Serum alanine aminotransferase/​aspartate aminotransferase activities (e.g. increased in paracetamol poisoning) • Acid–​base disturbances, including metabolic acidosis • Methaemoglobin concentration (e.g. in nitrite poisoning) • RBC cholinesterase activity (e.g. organophosphorus insecticide and nerve agent poisoning)

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1730 However, emergency measurement of the serum or plasma con- centration of the agents in Table 10.4.1.3 is important to ensure ap- propriate clinical management. ECG A routine ECG is of limited diagnostic value but is important in pa- tients who are unconscious or thought to have ingested a cardiotoxic drug. Sinus tachycardia with prolongation of the QRS interval in an unconscious patient suggests tricyclic antidepressant poisoning. With increasing cardiotoxicity, it may be impossible to detect P-​ waves, and the pattern then resembles ventricular tachycardia. Overdose with cardiac glycosides or potassium salts also induces characteristic ECG changes. Q–​T interval prolongation is a recog- nized adverse effect of several drugs in overdose (e.g. quetiapine, terfenadine, and quinine) and predisposes to ventricular arrhyth- mias, notably torsade de pointes. X-​rays Routine radiology is of little diagnostic value. It can be used to con- firm ingestion of metallic objects (e.g. coins, button batteries) or in- jection of globules of metallic mercury. Rarely, hydrocarbon solvents (e.g. carbon tetrachloride) may be seen as a slightly opaque layer floating on the top of the gastric contents with the patient upright, or outlining the small bowel. Ingested packets of illicit substances may be discernible on a plain radiograph, but CT or MRI is more reliable in detecting such objects. General management Antidotes and methods to enhance elimination are available for only very small number of poisons, and the management of the great ma- jority of poisoned patients is based on what has been called ‘an or- derly if unspectacular regimen of supportive therapy’. A small but significant number of poisoned patients arrive at hospital with respiratory obstruction, ventilatory failure, or in car- diorespiratory arrest. In these cases, conventional resuscitation takes precedence over detailed assessment of the patient and at- tempts to obtain a history. The opioid antagonist naloxone is safe and should be used whenever there is the slightest suspicion that an opioid is involved. Its use intravenously will resurrect a comatose, hypoventilating patient within seconds and, even if it is given in- appropriately, it is highly unlikely to have adverse effects. Unconscious patients need scrupulous attention to respiration, hypotension, hypothermia, and other complications, if they are to survive. Expert nursing is as important as medical measures. Airway Establishment and maintenance of an adequate airway is of para- mount importance in the management of unconscious poisoned patients. The airway may be obstructed by the tongue falling back, dental plates being dislodged, other foreign bodies, buccal secre- tions, vomitus, and flexion of the neck. In the first instance, the neck should be extended and the tongue and jaw held forwards. Secretions in the oropharynx must be removed, and an oropharyngeal airway should be inserted before turning the patient into a semi-​prone pos- ition. If the cough reflex is absent, an endotracheal tube should be inserted to prevent aspiration into the lungs and allow regular suc- tion of bronchial secretions. It is then important to ensure that the inspired air is adequately warmed and humidified. Ventilation Once a clear airway has been established, the adequacy of spontan- eous ventilation should be assessed. Pulse oximetry can be used to measure oxygen saturation. The displayed reading may be inaccurate when the saturation is below 70%, when peripheral perfusion is poor, and in the presence of carboxyhaemoglobin or methaemoglobin. Only measurement of arterial blood gases, however, indicates the presence both of hypercapnia and hypoxia. The presence of venti- latory insufficiency (as determined by arterial partial pressure of oxygen ≤9 kPa on air and/​or arterial partial pressure of CO2 ≥6 kPa) should lead to consideration of the need for intubation and assisted ventilation if the central respiratory depression cannot be reversed by administration of a specific antidote such as naloxone. Unconscious poisoned patients often have a mild, mixed respira- tory and metabolic acidosis with CO2 tensions at the upper limit of normal, and oxygen tensions that fall with increasing depth of coma. Increasing the oxygen contents of the inspired air is often sufficient to correct hypoxia. High-​inspired oxygen concentrations are impera- tive in patients with carbon monoxide and cyanide poisoning, and in pulmonary oedema resulting from inhalation of irritant gases. Cardiovascular function Cardiovascular function should be assessed by measuring pulse, blood pressure and temperature (core and peripheral). ECG should be monitored in moderately or severely poisoned patients, par- ticularly when a drug with a cardiotoxic action has been ingested. Echocardiography may occasionally be useful in such patients. Hypotension Although hypotension (systolic blood pressure <80 mm Hg) is a rec- ognized feature of acute poisoning, the classical features of shock (tachycardia and pale, cold skin) are seen only rarely because only a minority of patients are severely poisoned. Hypotension and shock may be caused by a direct cardiodepressant action of the poison (e.g. β-​blockers, calcium channel blockers, tri- cyclic antidepressants); vasodilatation and venous pooling in the lower limbs (e.g. angiotensin-​converting enzyme (ACE) inhibitors, phenothiazines); decrease in circulating blood volume because of Table 10.4.1.3  Poisons for which emergency measurement is important for management • Carboxyhaemoglobin • Digoxin • Ethanol (when monitoring treatment in ethylene glycol and methanol poisoning) • Ethylene and diethylene glycols • Iron • Lithium • Methanol • Paracetamol • Salicylate

10.4.1  Poisoning by drugs and chemicals 1731 gastrointestinal losses (e.g. theophylline), increased insensible losses (e.g. salicylates), increased renal losses (e.g. diuretics), and in- creased capillary permeability. Hypotension may be exacerbated by coexisting hypoxia, acidosis, and dysrhythmias. Correct management of individual cases depends on accurate identification of the cause. Young patients are generally not at risk of cerebral or renal damage unless the systolic blood pressure falls below 80 mm Hg but, in those over the age of 40 years, it is preferable to keep the systolic blood pressure above 90 mm Hg. The treatment of hypotension depends on the cause. In all cases restoration of perfusion and oxygenation is the aim. While it is rea- sonable initially to administer a bolus of intravenous crystalloid, care must be taken to avoid volume overload in cases of primary cardiac failure. Such patients may require inotropic support with a sympathomimetic inotrope, such as dobutamine 2.5–​10 micro- grams/​kg/​min or dopamine 2–​5 micrograms/​kg/​min. Hypotension caused by vasodilatation that does not respond to intravascular volume expansion may warrant a vasoconstrictor sympathomimetic drug, such as noradrenaline (norepinephrine) 40 micrograms (base)/​ml at an initial rate of 0.16–​0.33 ml/​min, or metaraminol (which has the potential advantage that it can be administered via a peripheral line) 15–​100 mg by intravenous infusion. It must be recognized, however, that blood pressure may be raised at the expense of perfusion of vital organs, such as the kidneys. Response to treatment should be monitored not only by blood pressure but also other markers of improved per- fusion and oxygenation including skin colour and temperature, urine output, cerebration, and resolution of metabolic (lactic) acidosis. Hypertension A few drugs (e.g. cocaine and amfetamines), when taken in over- dose, may produce systemic hypertension. If this is mild and as- sociated with agitation, a benzodiazepine may suffice. In more severe cases, there may be a risk of arterial rupture, particularly intracranially. To prevent this, intravenous isosorbide dinitrate 2–​ 10 mg/​h (up to 20 mg/​h if necessary), or glyceryl trinitrate 10–​200 micrograms/​min by intravenous infusion (paediatric dose 0.2–​0.5 micrograms/​kg/​min) should be administered until blood pressure elevation is controlled. Arrhythmias Although many poisons are potentially cardiotoxic, the incidence of serious cardiac arrhythmias in acute poisoning is very low. Tricyclic antidepressants, β-​adrenoceptor blocking drugs, calcium channel blockers, cardiac glycosides, amfetamines, cocaine, bronchodila- tors (particularly theophylline and its derivatives) and antimalarial drugs are the most likely causes. Cardiotoxicity usually occurs to- gether with other features of severe poisoning, including metabolic acidosis, hypoxia, convulsions, respiratory depression, and abnor- malities of electrolyte balance, which should be corrected before considering the use of antiarrhythmic drugs. The latter have narrow therapeutic ratios and their use may further impair myocardial function. In general, drug therapy should only be given for persistent, life-​threatening arrhythmias associated with peripheral circula- tory failure. The drug used must be selected from knowledge of the pharmacology and toxicology of the poison involved and in such a way that it will not further compromise cardiac function. For ex- ample, in tricyclic antidepressant poisoning, arrhythmias are due to sodium channel blockade exacerbated by acidosis and are best treated with hypertonic sodium bicarbonate, 50–​100 mmol. Convulsions Convulsions are potentially life-​threatening because they cause hypoxia and metabolic acidosis and may precipitate cardiac ar- rhythmias and arrest. Short, isolated convulsions do not require treatment but those which are recurrent or protracted should be suppressed with intravenous diazepam 10–​20 mg in an adult (lor- azepam 4 mg is an alternative). This drug is highly effective in ad- equate doses and alternatives are seldom needed. However, it is important to remember that giving benzodiazepines in this way may potentiate the respiratory depressant effects of the drugs inducing seizures. The combination of convulsions, coma, and vomiting, which may occur with theophylline poisoning, is particularly dangerous and, in these circumstances, it may be preferable to paralyse the patient, insert an endotracheal tube, and start assisted ventilation. However, although this ensures control of the airway and oxygenation, thus avoiding the risk of inhalation of gastric contents, it does not sup- press seizure activity; cerebral function must therefore be moni- tored, and parenteral anticonvulsants given as required. Temperature disturbances Hypothermia Any poison which depresses the central nervous system may impair temperature regulation and cause hypothermia, especially when dis- covery of the patient is delayed and environmental temperatures are low. This important complication may be missed unless tempera- ture is recorded rectally using a low-​reading thermometer. In se- vere cases, peripheral and core temperatures should be monitored. Treatment includes nursing the patient in a warm room (27–​29°C) and a heat-​conserving ‘space blanket’. Cold intravenous fluids should be avoided and fluid bags for use should be stored in the room, or the lines should pass through a heating device. Hyperthermia Rarely, body temperature may increase to potentially fatal levels after poisoning with central nervous system stimulants such as cocaine, amfetamines (including ecstasy (MDMA)), monoamine oxidase inhibitors, or theophylline. In such cases, muscle tone is often grossly increased, and convulsions and rhabdomyolysis are common. Cooling measures should be instituted, sedation with di- azepam should be given and, in severe cases, dantrolene 2–3 mg/kg, then 1 mg/​kg should be administered intravenously. Acid–​base disturbances Acid–​base disturbances commonly accompany coma due to drugs. Acute respiratory acidosis is less common than might be expected, but some elevation of arterial CO2 tensions towards the upper limit of normal is usual. This, in combination with mild hypoxia in the deeper grades of coma, produces overall acidaemia. In general, acidosis should be prevented and managed by ensuring adequate ventilation, oxygenation and tissue perfusion, and control of convul- sions rather than by giving bicarbonate. However, several poisons, particularly methanol and ethylene glycol, cause life-​threatening

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1732 metabolic acidosis which should be corrected by infusion of sodium bicarbonate (see individual poisons). Acute respiratory alkalosis, often in combination with a minor metabolic acidosis, is commonly found in acute salicylate poisoning. The metabolic component may require treatment if it is the dom- inant feature and is causing overall acidaemia. Respiratory alkalosis should not be treated. Electrolyte abnormalities Electrolyte abnormalities may result from acid–​base disturbances or the direct effects of poisons. Massive tissue damage, usually rhabdomyolysis, may allow potassium to leak from cells leading to potentially lethal hyperkalaemia. Cardiac glycosides cause hyperkalaemia, secondary to loss from cells due to inhibition of the membrane sodium–​potassium pump, while the reverse occurs with sympathomimetic drugs. Oxalic acid and ethylene glycol (which is metabolized to oxalic acid) may cause hypocalcaemia by leading to the formation of in- soluble calcium oxalate, which is deposited in tissues. Similarly, in- gestion of fluorides is also a possible cause of hypocalcaemia; but the amounts children tend to ingest in the form of tablets to pre- vent dental caries seldom cause serious problems. Ingestion of po- tassium salts, even in sustained release formulations, may lead to hyperkalaemia and fatal arrhythmias. Bladder care Urinary retention is a common complication of acute poisoning, par- ticularly with tricyclic antidepressants and other drugs which have marked anticholinergic actions. However, bladder catheterization is all too often an unconsidered measure in unconscious poisoned pa- tients. Coma itself is not an indication for bladder catheters in poi- soned patients, the great majority of whom regain consciousness within 12 h. The bladder can usually be induced to empty reflexively (provided it is not allowed to become grossly overdistended) by ap- plying gentle suprapubic pressure. Catheterization should be reserved for those patients in whom suprapubic pressure is insufficient to empty the bladder, and in those thought to be developing renal failure. Skin, muscle, and nerve lesions Bullous lesions should be left intact until they burst, to reduce the risk of infection. De-​roofing should be performed when the blister bursts; a non​adhesive dressing is then applied. Rhabdomyolysis is a further possible result of immobility and may occur in combination with skin lesions or independently. Poisoning is the most common non​traumatic cause of this condition and it may lead to acute renal failure and, rarely, to ischaemic muscle con- tractures and long-​term disability. Urgent orthopaedic referral is in- dicated if a compartment syndrome is suspected. Peripheral nerves such as the radial, ulnar, and common peroneal may also be dam- aged by direct pressure while the patient is unconscious. Unconscious patients should be turned from side to side at least every 2 h. Antidotes Antidotes may exert a beneficial effect by: • forming an inert complex with the poison (e.g. deferoxamine, D-​penicillamine, dicobalt edetate, dimercaprol, digoxin-​specific antibody fragments, HI-​6, hydroxocobalamin, obidoxime, pralidoxime, protamine, Prussian (Berlin) blue, sodium calcium edetate, succimer (DMSA), unithiol (DMPS)) • accelerating detoxification of the poison (e.g. acetylcysteine, so- dium thiosulfate) • reducing the rate of conversion of the poison to a more toxic com- pound (e.g. ethanol, fomepizole) • competing with the poison for essential receptor sites (e.g. oxygen, naloxone, phytomenadione) • blocking essential receptors through which the toxic effects are mediated (e.g. atropine) • bypassing the effect of the poison (e.g. oxygen, glucagon) The most frequently used antidote in the developed world is acetylcysteine for paracetamol poisoning. Naloxone for opioid an- algesics, oxygen for carbon monoxide and, possibly, flumazenil for benzodiazepines are the only antidotes commonly needed in the management of unconscious poisoned patients. Other antidotes of proven value are listed in Table 10.4.1.4. The reader is recom- mended to read the relevant section in the chapter to obtain further advice. Antivenoms for bites and stings by venomous animals are discussed in Chapter 10.4.2. Table 10.4.1.4  Poisons for which there are specific antidotes Poison Antidote Aluminium Deferoxamine (Desferrioxamine) Arsenic Dimercaprol (BAL), succimer (DMSA) Benzodiazepines Flumazenil β-​adrenoceptor-​blocking drugs Atropine, glucagon Calcium channel blockers Atropine Carbamate insecticides Atropine Carbon monoxide Oxygen Copper d-​Penicillamine, unthiol (DMPS) Cyanide Dicobalt edetate, hydroxocobalamin,
oxygen, sodium nitrite, sodium thiosulfate Diethylene glycol Fomepizole, ethanol Digoxin and digitoxin Digoxin-​specific antibody fragments Ethylene glycol Fomepizole, ethanol Hydrogen sulphide Oxygen Iron salts Deferoxamine (Desferrioxamine) Lead (inorganic) Succimer (DMSA), sodium calcium edetate Methaemoglobinaemia Methylthioninium chloride (methylene blue) Methanol Ethanol, fomepizole Mercury (inorganic) Unithiol (DMPS) Nerve agents Atropine, obidoxime, pralidoxime, HI-​6 Oleander Digoxin-​specific antibody fragments Opioids Naloxone Organophosphorus insecticides Atropine, obidoxime, pralidoxime Paracetamol Acetylcysteine Thallium Prussian (Berlin) blue Warfarin and other anticoagulants Phytomenadione (vitamin K1)

10.4.1  Poisoning by drugs and chemicals 1733 Reduction of poison absorption Prevention of absorption of volatile poisons through the lungs ob- viously requires removal from the toxic atmosphere and occasion- ally removal of soiled clothing as well. The latter is also necessary when absorption is thought to have been percutaneous. In addition, the contaminated skin should be thoroughly washed with soap and water. Although it appears logical to assume that removal of unabsorbed drug from the gastrointestinal tract (‘gut decontamination’) will be beneficial, the efficacy of current methods remains unproven, and efforts to remove small amounts of ‘safe’ drugs are clearly not worth- while or appropriate. Activated charcoal Activated charcoal adsorbs a wide variety of drugs and toxic agents; the exceptions are acids and alkalis, ethanol, ethylene glycol, iron, lithium, and methanol. In studies in volunteers given 50 g activated charcoal, the mean reduction in absorption was 40%, 16%, and 21% at 60 min, 120 min, and 180 min, respectively, after ingestion. Based on these studies, ac- tivated charcoal 50–100 g should be considered in those who have ingested a potentially toxic amount of a poison (known to be ad- sorbed by charcoal) up to 1 h previously. There are insufficient data to support or exclude its use after 1 h. There is no evidence that ad- ministration of activated charcoal improves the clinical outcome. Gastric aspiration and lavage Gastric emptying studies in volunteers provide no support for the use of gastric lavage. In the single clinical study in which benefit was claimed for lavage within 1 h of overdose, patients also received acti- vated charcoal. There was also selection bias, and hence conclusions based on these data are limited. Thus, gastric lavage should not be used routinely in the management of poisoned patients as there is no evidence that it improves outcome, and it may cause significant morbidity. The efficacy with which lavage removes gastric contents decreases with time; therefore, lavage should be considered only in patients who have ingested life-​threatening amounts of a toxic agent up to 1 h previously. Emesis with syrup of ipecacuanha Syrup of ipecacuanha contains the active alkaloids emetine and cephaeline. Although syrup of ipecacuanha is an effective emetic, there is no evidence that its use prevents significant absorption of toxic material and, moreover, its adverse effects (e.g. persistent vomiting, diarrhoea, lethargy, drowsiness) may complicate diag- nosis. It is not recommended. Whole-​bowel irrigation Theoretically, the more quickly a slowly absorbed poison passes through the gut, the less it is absorbed. The opposite may apply to rapidly absorbed drugs. Whole-​bowel irrigation using polyethylene glycol electrolyte solutions does not result in absorption of fluid and electrolytes, even though large volumes are administered rapidly via a nasogastric tube. Some volunteer studies have shown substan- tial decreases in the bioavailability of ingested drugs, but no con- trolled clinical trials have been conducted and there is no evidence that whole-​bowel irrigation improves outcome. Based on volunteer studies, whole-​bowel irrigation may be considered following poten- tially toxic ingestion of sustained release or enteric-​coated drugs and in body packers. Cathartics Cathartics have been used alone and with activated charcoal. Cathartics alone have no role in the management of poisoned pa- tients. Based on available data, routine use of a cathartic with acti- vated charcoal is not endorsed. Methods to increase poison elimination Once a poison has been absorbed and providing there is no antidote, it is reasonable to consider the use of treatments that might speed its elimination from the body. Multiple-​dose activated charcoal Use of multiple-​dose activated charcoal involves repeated admin- istration of oral activated charcoal to increase the elimination of a drug that has already been absorbed into the body. Elimination of a drug with a small volume of distribution (<1 litre/​kg), low pKa (which maximizes transport across membranes), low binding af- finity, and prolonged elimination half-​life following overdose is par- ticularly likely to be enhanced by multiple-​dose activated charcoal. Multiple-​dose activated charcoal also improves total body clearance of the drug when endogenous processes are compromised by liver and/​or renal failure. Activated charcoal adsorbs material in the gut, which may be rele- vant in cases of poisoning with slow-​release drug preparations. It also adsorbs drugs that are secreted in the bile, thereby preventing intestinal reabsorption, and binds any drug that diffuses from the circulation into the gut lumen. After absorption, drugs re-​enter the gut by passive diffusion if the concentration in the gut is lower than that in the blood. The rate of passive diffusion depends on the con- centration gradient and the intestinal surface area, permeability, and blood flow. Occasionally, drugs such as digoxin may be secreted ac- tively by the intestinal mucosa, though the contribution of active secretion to the effect of multiple-​dose activated charcoal on drug clearance is unlikely to be greater than that of passive diffusion. Although many studies have demonstrated that multiple-​dose ac- tivated charcoal significantly increases drug elimination, it has not been shown to reduce morbidity and mortality in controlled studies in poisoned patients. At present, use of multiple-​dose activated charcoal should be considered only in patients who have ingested a life-​threatening amount of carbamazepine, dapsone, phenobarbital, quinine, and theophylline. Clinical experience in adults suggests that charcoal should be ad- ministered in an initial dose of 50–​100 g and then at a rate of not less than 12.5 g/​h, preferably via a nasogastric tube. Smaller initial doses (10–​25 g) can be used in children because, generally, smaller overdoses have been ingested and the capacity of the gut lumen is smaller. If the patient has ingested a drug that induces protracted vomiting (e.g. theophylline), intravenous ondansetron is effective as an antiemetic and thus enables administration of multiple-​dose ac- tivated charcoal. A total dose of 200 g (in adults) is usually sufficient.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1734 Urine alkalinization Increasing the urine pH enhances elimination of salicylate, phenobar- bital, and chlorophenoxy herbicides (e.g. 2,4-​dichlorophenoxyacetic acid, mecoprop). However, except for salicylate poisoning, urine alkalinization is not recommended as first-​line therapy for poisoning with these agents, as multiple-​dose activated charcoal is superior for phenobarbital, and a substantial diuresis is required in addition to urine alkalinization to achieve clinically important elimination of chlorophenoxy herbicides. Urine alkalinization is a metabolically invasive procedure re- quiring frequent biochemical monitoring and medical and nursing expertise. Before commencing urine alkalinization, it is important to correct plasma volume depletion, electrolytes (administration of sodium bicarbonate exacerbates pre-​existing hypokalaemia), and metabolic abnormalities. Sodium bicarbonate is most con- veniently administered intravenously as an 8.4% solution (1 mmol bicarbonate/​ml). Sufficient bicarbonate should be administered (225 mmol was the mean amount required in one study) to ensure that the pH of the urine, which is measured by narrow-​range indi- cator paper or a pH meter, is more than 7.5 and preferably close to 8.5. As the administration of sodium bicarbonate forces potassium into cells, it is important that the patient has a normal serum po- tassium concentration before sodium bicarbonate is administered. Sodium bicarbonate 8.4% is highly irritant to veins and severe tissue damage can ensue if extravasation occurs. A secure, preferably wide-​ bore cannula (or central venous line), must therefore be used. Dialysis, haemodialfiltration, haemofiltration,
and haemoperfusion Haemodialysis, haemodialfiltration, and haemoperfusion are of no value in patients poisoned by drugs with large volumes of distribution (e.g. tricyclic antidepressants), because the plasma contains only a small proportion of the total amount of drug in the body. Haemodialysis, and to a lesser extent haemodialfiltration, sig- nificantly increases elimination of ethanol, ethylene glycol, isopro- panol, lithium, methanol, and salicylate, and is the treatment of choice in all cases of severe poisoning with these agents. Although haemofiltration is widely available, it is much less efficient than haemodialysis and haemodialfiltration and, therefore, should not be used unless the alternatives are unavailable. Charcoal haemoperfusion can significantly reduce the body burden of phenobarbital, carbamazepine, and theophylline, but multiple-​dose activated charcoal is as effective and simpler to use. Drugs Angiotensin-​converting enzyme inhibitors Clinical features Hypotension is the principal feature, occasionally accompanied by drowsiness. Angioedema, hyperkalaemia, and renal failure are rec- ognized. The fall in blood pressure is often much greater than from therapeutic doses and the suggestion that ACE inhibitors have a ‘ceiling’ effect on blood pressure is incorrect. Treatment ACE inhibitors are likely to bind to activated charcoal, which should be administered in early presentations. The principles of supportive care include volume expansion and subsequent use of inotropes. Since most patients who take these drugs in overdose are on treatment for hypertension or heart failure, careful manage- ment is necessary as they may already have impaired myocardial function. Antibacterial agents Most patients who take an antibiotic overdose are asymptom- atic and require no treatment. There have been single case re- ports of renal failure after overdose with co-​trimoxazole or aminoglycosides, pancreatitis with erythromycin, haemorrhagic cystitis with amoxicillin, and seizures with amoxicillin and other β-​lactam antibiotics. Rifampicin may induce metabolism of other drugs, an effect that takes several days. Erythromycin may cause QT prolongation and torsade de pointes. Isoniazid is dealt with next. Anticoagulants Warfarin is an antagonist of the synthesis of the vitamin K de- pendent clotting factors II, VII, IX, and X. Newer oral anticoagulant agents, such as dabigatran, a thrombin inhibitor, and rivaroxaban and apixaban, both direct factor Xa inhibitors, impact specific tar- gets in the coagulation cascade. Toxicity is more likely to occur in the setting of therapeutic anticoagulation, or because of a drug interaction, than as a consequence of acute overdose. Pre-​existing renal impairment is the major factor affecting the toxicity of newer agents. Clinical features Epistaxis, gingival bleeding, spontaneous bruising, haematomas, haematuria, bilateral flank pain, rectal bleeding, and haemor- rhage into any organ. Spontaneous haemoperitoneum has been re- ported. Severe blood loss may result in hypovolaemic shock, coma, and death. Treatment For warfarin if major bleeding occurs, give phytomenadione (vitamin K1) 5 mg by slow intravenous injection together with prothrombin complex dried 25–​50 units/​kg or if unavailable fresh frozen plasma 15 ml/​kg. If the INR is 8.0 or more and there is no active bleeding discon- tinue warfarin (restart when the INR <5.0), give phytomenadione 1–​3 mg by slow intravenous injection and repeat the dose if the INR is 8.0 or more 24 h later. If the INR is 6.0–​8.0, and there is no active bleeding or only minor bleeding, warfarin should be discontinued and restarted when the INR is less than 5.0. If continued anticoagulation is unnecessary and the INR is 4.0 or less and there is no active bleeding, treatment with phytomenadione is not required. If the INR is 4.0 or more, phytomenadione 5 mg by slow intravenous injection (100 µg/​kg body weight for a child) should be administered. A specific monoclonal antibody, idarucizumab 5–10g should be given, to reverse the effects of dabigatran. Rivaroxaban and apixaban

10.4.1  Poisoning by drugs and chemicals 1735 have no specific antidotes, but clotting factor concentrate or fresh frozen plasma should be tried in active bleeding. Anticonvulsants: Carbamazepine Clinical features Carbamazepine is structurally related to the tricyclic antidepressants and has similar anticholinergic actions. Overdose causes dry mouth, coma, convulsions, nystagmus, ataxia, and incoordination. The pu- pils are often dilated, divergent strabismus may be present and com- plete external ophthalmoplegia has been reported. Hallucinations may occur, particularly in the recovery phase. Treatment Multiple-​dose activated charcoal has been shown to increase elimin- ation of carbamazepine significantly. Anticonvulsants: Phenytoin Clinical features Acute overdose results in nausea, vomiting, headache, tremor, cere- bellar ataxia, nystagmus, and rarely, loss of consciousness. Treatment Multiple-​dose activated charcoal may increase phenytoin elimin- ation though this has not been confirmed. Anticonvulsants: Sodium valproate Clinical features Most frequently there is drowsiness, impairment of consciousness, and respiratory depression. In severe poisoning, myoclonic jerks and seizures may occur and cerebral oedema has been reported. Liver damage, pancreatitis, and metabolic acidosis, perhaps due to changes in fatty acid metabolism, are very unusual but potential complications. Treatment Treatment is symptomatic and supportive. Haemodialysis is ef- fective in removing sodium valproate and should be employed in severe poisoning, particularly if severe hyperammonaemia and electrolyte and acid–​base disturbances are present. Anticonvulsants: Gabapentin Clinical features and treatment Lethargy, ataxia, slurred speech, and gastrointestinal symptoms may develop. Management is supportive. Anticonvulsants: Lamotrigine Clinical features and treatment Lethargy, coma, ataxia, nystagmus, seizures, and cardiac conduction abnormalities have been reported. Management is supportive. Anticonvulsants: Levetiracetam Clinical features and treatment Lethargy, coma, and respiratory depression have been observed. Management is supportive. Anticonvulsants: Tiagabine Clinical features and treatment Lethargy, facial grimacing, nystagmus, posturing, agitation, coma, hallucinations, and seizures have been reported. Management is supportive. Anticonvulsants: Topiramate Clinical features and treatment Lethargy, ataxia, nystagmus, myoclonus, coma, seizures, and a normal anion gap metabolic acidosis have been observed; the latter may be due to inhibition of renal cortical carbonic anhydrase. Metabolic acidosis can appear within hours of ingestion and persist for days. Management is supportive. Antidepressants These come in a variety of pharmacological groups, but share the common effect of altering central monoamine function. Toxicity is largely dependent on other properties of these drugs. Antidepressants: Tricyclic antidepressants Several different pharmacological actions determine the features of overdose. Reuptake of monoamines (noradrenaline and serotonin) into central and peripheral neurones is blocked. Anticholinergic ac- tions cause reduced gut motility, dry mouth, and tachycardia; sodium channel blockade (e.g. amitriptyline) with class I antiarrhythmic ac- tion prolongs the QRS complex; α-​adrenergic and histamine antag- onism results in hypotension and sedation. Clinical features Clinical features evolve as the drug is absorbed, usually within 30–​ 60 min of ingestion. Patients who remain conscious 6 h after in- gestion are unlikely to have taken a large overdose. Early features include drowsiness, sinus tachycardia, dry mouth, and dilated pupils. Urinary retention, increased reflexes, extensor plantar responses, and gaze palsies may then develop. Patients who become uncon- scious, Glasgow Coma Score (GCS) less than 8, or are unresponsive to pain, are at increased risk of more serious complications, particu- larly seizures. The risk of ventricular arrhythmias may be predicted from the length of the QRS complex. Changes in repolarization pat- tern may also be seen with abnormal T-​waves and apparent changes in the ventricular axis. This pattern mimics the Brugada syndrome, the congenital abnormality associated with ventricular fibrillation. Features include ST elevation in leads V1–​3, with right bundle block often associated with serious ventricular arrhythmias. Treatment Patients with depressed consciousness and prolonged QRS interval are at risk of seizures and arrhythmias. Maintenance of acid–​base balance in these patients is crucial. Early and prompt treatment with sodium bicarbonate, even in patients who are not overtly acid- otic, ameliorates cardiac effects of tricyclics. Sodium bicarbonate 50–100 mmol (50–100 ml of 8.4%) should be administered. If given into a peripheral vein, there is a risk of necrosis if it extravasates. Indications for bicarbonate include QRS duration greater than 120 msec, existing arrhythmias or hypotension resistant to fluid

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1736 resuscitation. The aim is to maintain the arterial pH in the range 7.5–​7.55 without producing greater alkalaemia. Class 1a and class 1c antiarrhythmic drugs are contraindicated because they have the same sodium channel blocking activity as tricyclic antidepres- sants. Convulsions should be treated conventionally with diazepam 10–​20 mg intravenously, in an adult or lorazepam 4 mg. The α-​adrenoceptor blocking properties of tricyclics can cause severe hypotension. Noradrenaline is the most appropriate inotrope to use in this situation. In the past, physostigmine was advocated to counteract the anti- cholinergic action of tricyclic antidepressants, but most European toxicologists do not recommend this. During recovery from tricyclic poisoning, there may be a pro- longed period of delirium with auditory and visual hallucinations. Sedation with diazepam is appropriate until the patient recovers. All tricyclic antidepressants may cause these features but dosulepin (dothiepin) is the most toxic in overdose, followed by amitriptyline. Antidepressants: selective serotonin reuptake
inhibitors (SSRIs) Citalopram, fluoxetine, fluvoxamine, paroxetine, and sertraline are antidepressants that inhibit serotonin reuptake (SSRIs) and lack the anticholinergic actions of tricyclic antidepressants. Clinical features Clinical features of these agents are principally due to serotonin-​like effects, and include nausea and vomiting, agitation, and tachycardia. Convulsions may occur after larger ingestions. Hypertonia and marked clonus are common features of significant poisoning, and increased muscle activity results in a rise in serum creatine kinase activity. Citalopram is the most toxic of the group in overdose. All SSRIs occasionally cause arrhythmias. Treatment In patients who consume more than one drug affecting serotonin receptors (e.g. tricyclic antidepressants, monoamine oxidase in- hibitors, drugs of abuse, including, in particular, ecstasy), the serotonin syndrome may occur. Features include marked agi- tation and increased muscle activity resulting in hyperpyrexia. About half the patients have central nervous system features including delirium and hallucinations. Other features include autonomic instability with tachycardia and labile blood pressure. Specific serotonin antagonists such as cyproheptadine may be useful though cannot be administered parenterally. Alternatively, benzodiazepines (e.g. diazepam orally or parenterally) may help reduce agitation. Antidepressants: Venlafaxine Venlafaxine is a drug that inhibits the reuptake of serotonin and nor- adrenaline (SNRI). In overdose, it has features of both SSRIs and tricyclic antidepressants but it lacks anticholinergic activity. Clinical features Drowsiness and convulsions are the main central nervous system effects. Tachycardia, ventricular arrhythmias, and changes in blood pressure are the main cardiovascular effects. Treatment Management of metabolic acidosis is important to reduce the risk of arrhythmias, which are more common in patients who have had convulsions. Convulsions are managed conventionally with di- azepam 10–​20 mg intravenously or lorazepam ​4 mg intravenously. Activated charcoal should be considered if more than 12.5 mg/​kg was ingested within the previous hour. Venlafaxine prolongs the QT interval so that torsade de pointes is a risk which, should it occur, is treated conventionally by correcting acidosis and with intravenous magnesium 1–2 g over 30–60 sec, repeated in 5–15 min. Antidepressants: Monoamine oxidase inhibitors (MAOIs) These have well-​established adverse interactions with foods containing tyramine. The classical MAOIs such as phenelzine, isocarboxazid, and tranylcypromine are now rarely used, and the new more specific inhibitors of MAOI type A (moclobemide) and type B (selegiline) produce less serious adverse effects in overdose. Classical MAOIs prevent the breakdown of catecholamines within the nerve ending, and result in excess sympathomimetic effects per- ipherally, and excess adrenergic effects centrally. In patients who are naïve to the drugs, onset of inhibition of enzyme takes several hours, and clinical features may not be seen immediately. In patients on chronic therapy, the onset will be more rapid. Clinical features Principal effects are central nervous system stimulation with excite- ment, restlessness, hyperpyrexia, hyperreflexia, convulsions, and coma. These may go on to cause rhabdomyolysis. Cardiovascular ef- fects include tachycardia and changes in blood pressure, depending on whether the effects of epinephrine (vasodilation) or norepineph- rine (vasoconstriction) predominate. Treatment Treatment is supportive, with careful monitoring. Patients who develop central excitation should be treated with large doses of di- azepam. This will reduce centrally stimulated muscle contraction and hence pyrexia and muscle damage. Cardiovascular monitoring is essential. Changes in blood pressure should be managed where possible with drugs that are not sympathomimetic agonists. Use of β-​blockade can result in an unopposed α-​agonist effect causing large rises in blood pressure. Hypertension is best controlled with an intravenous nitrate, such as glyceryl trinitrate. Antidiabetic agents Intentional overdose with insulin and oral hypoglycaemic agents is uncommon. However, deaths from hypoglycaemia following poi- soning with insulin and sulfonyureas have been reported. Metformin rarely causes hypoglycaemia since its mode of action is to increase glucose utilization. Risk of hypoglycaemia in overdose is low from more recently introduced antidiabetics. These include subtype 2 sodium-​glucose transport protein (SGLT2) inhibitors, meglitinides, thiazolidinediones, α-​glucosidase inhibitors, GLP-​1 analogues, and DDP-​IV inhibitors. Clinical features Features of hypoglycaemia include drowsiness, coma, twitching, convulsions, depressed limb reflexes, extensor plantar responses,

10.4.1  Poisoning by drugs and chemicals 1737 tachypnoea, pulmonary oedema, tachycardia, and circulatory failure. Hypokalaemia, cerebral oedema, and metabolic acidosis might occur. Neurogenic diabetes insipidus and persistent vegeta- tive state are possible long-​term complications. Lactic acidosis is a potentially serious complication of metformin overdose. SGLT2 inhibitors increase renal glucose clearance and, in overdose, cause polyuria, hypovolaemia, hypotension, and acute renal injury. Thiazolidinediones may cause hepatic dysfunction and SGLT2 in- hibitors cause diabetic ketoacidosis and renal impairment. Treatment In all cases of poisoning with insulin or a sulfonylurea, prompt diag- nosis and treatment are essential if death or cerebral damage from neuroglycopenia are to be prevented. The blood or plasma glucose concentration should be measured urgently and intravenous glucose given. Glucagon may be ineffective in hypoglycaemia due to exhaus- tion of hepatic stores of glucose. Recurring hypoglycaemia is highly likely. A  continuous infu- sion of glucose, together with carbohydrate-​rich meals, is required in cases of severe insulin poisoning, though there may be difficulty in maintaining normoglycaemia. In the case of sulfonylurea poi- soning, however, further glucose (although its administration may be unavoidable) only serves to increase the already high-​circulating insulin concentrations. Octreotide 50 microgram IV, followed by an infusion of 25 microgram/h is preferred in severe sulfonylurea poisoning. Antihistamines First-​generation antihistamines include brompheniramine, chlorphenamine, cyclizine, diphenhydramine, promethazine, and trimeprazine. Second-​generation drugs include cetirizine, loratidine, and fexofenadine. Clinical features Older antihistamines have anticholinergic actions but less potent central nervous system toxicity than other anticholinergic drugs. Delirium may be a particular problem in very young children and older people following a substantial acute overdose. Rhabdomyolysis is a well-​recognized complication of severe antihistamine poisoning. The second-​generation drugs generally cause less sedation and less psychomotor impairment, but some have been associated with cardiotoxicity causing QTc interval prolongation and ventricular tachycardia, including torsade de pointes. Treatment A 12-​lead ECG and cardiac monitoring for at least 12 h is recom- mended after a substantial overdose. Management should otherwise follow the same principles as for tricyclic antidepressant poisoning (see earlier). Antimalarials: chloroquine Toxicity can result from doses greater than 1 g (c. 6 tablets) in adults. Clinical features Cardiac arrest is commonly the first clinical manifestation of poi- soning, but hypotension usually precedes it and may progress to cardiogenic shock with pulmonary oedema. Electrocardiographic abnormalities, bradyarrhythmias, and tachyarrhythmias are common and are similar to those seen in quinine poisoning. Visual disturbance, agitation, drowsiness, acute psychosis, dystonic reac- tions, seizures, and coma may ensue. Hypokalaemia is common and is due to potassium channel blockade. Treatment Supportive measures should be employed and hypokalaemia corrected. There is no specific antidote. Sodium bicarbonate 50–​200 mmol (50–​200 ml of 8.4%) is indicated if the ECG shows intraventricular block but will exacerbate hypokalaemia, which should be corrected first. Mechanical ventilation, the administra- tion of epinephrine 1–​10 µg/​kg per minute and high doses of di- azepam (1 mg/​kg as a loading dose and 0.25–​0.4 mg/​kg per hour maintenance) may reduce the mortality to 10% in severe poisoning. Multiple-​dose activated charcoal may enhance chloroquine elim- ination. Extracorporeal elimination techniques do not have a role. Extracorporeal life support has been utilized successfully in severely poisoned patients unresponsive to conventional measures. Antimalarials: quinine Quinine cardiotoxicity is due to sodium channel blockade. Clinical features Cinchonism (tinnitus, deafness, vertigo, nausea, headache, and diar- rhoea) is common at plasma concentrations greater than 5 mg/​litre. In more serious poisoning, collapse with impairment of conscious- ness (due to ventricular arrhythmias), convulsions, hypotension, pul- monary oedema, and cardiorespiratory arrest may be observed. The latter is often preceded by ECG conduction abnormalities, particu- larly QT prolongation. Hypoglycaemia, resulting from insulin release, occurs even with therapeutic doses and must be excluded in all cases. About 40% of patients develop ocular features, which may be unilat- eral, including blindness, contracted visual fields, scotomata, dilated pupils, blurred disc margins, macular oedema, arteriolar spasm, and late optic atrophy. Oculotoxicity is likely when plasma concentrations exceed 10 mg/​litre. Visual loss is permanent in about 50% of cases. Treatment Multiple-​dose activated charcoal increases quinine clearance. Extracorporeal elimination techniques and stellate ganglion block are of no value. Electrolyte and acid–​base disturbances and hypo- glycaemia should be corrected. Hypertonic sodium bicarbonate will correct acidosis that persists despite fluid resuscitation and adequate oxygenation and is recommended first-​line therapy for conduction abnormalities due to sodium channel blockade, including QRS and QT prolongation. Overdrive pacing may be required if torsade de pointes occurs and does not respond to magnesium sulfate 1–2 g over 30–60 sec, repeated in 5–15 min. Antimalarials: Primaquine Clinical features The main concern about primaquine is its ability to cause methaemo- globinaemia in overdose. Other adverse effects reported are head- ache, nausea, abdominal pain, haemolytic anaemia, particularly in patients with glucose-​6-​dehydrogenase deficiency, and leucopenia.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1738 Treatment Treatment is supportive. Clinically significant methaemoglobin- aemia (generally >30%) is treated conventionally with intravenous methylthioninium chloride (methylene blue), 1–​2 mg/​kg body weight. Antipsychotics Antipsychotic drugs are thought to act predominantly by effects on dopamine D2 receptors. Older antipsychotics were phenothiazines, such as chlorpromazine, and butyrophenones, such as haloperidol. Selective (‘atypical’) antipsychotic drugs include amisulpride, aripiprazole, clozapine, olanzapine, quetiapine, and risperidone. Conventional antipsychotics have many actions, including anti- histamine and anticholinergic activity. Chlorpromazine blocks α-​, β-​, and 5HT-​receptors in vitro. Features such as postural hypo- tension are likely to be due to the sum of these effects. Clinical features In overdose, the predominant clinical features of all antipsychotics are sedation, loss of consciousness, and hypotension. Respiratory depression may occur in more severe cases. Hypotension and vaso- dilatation are features of chlorpromazine poisoning. Some have caused QT prolongation in overdose. A 12-​lead ECG should be obtained to check QT duration. Occasionally seizures are reported. ECG abnormalities have been seen with some of the newer atyp- ical antipsychotics, but they cause less cardiovascular disturbance than the older drugs. Muscle contraction due to central extra- pyramidal effects may result in rhabdomyolysis in severe cases. Neuroleptic malignant syndrome, seen during therapeutic use of these compounds, is uncommon in acute poisoning and should be treated conventionally. Treatment Management is supportive. Dystonic reactions may occur, particu- larly in young adults. These should be treated conventionally with procyclidine 5–​10 mg intravenously, or diazepam 10–​20 mg intra- venously or orally. Benzodiazepines These are widely used as tranquillizers, hypnotics, sedatives, and for emergency management of convulsions and hyperthermia. They also have abuse potential. Clinical features Although many benzodiazepines have active metabolites ac- counting for their sometimes-​prolonged sedative effects, all are remarkably safe when taken in excess alone. However, there is in- dividual variation in response, influenced by habituation and tol- erance, which develop during chronic therapy. Otherwise healthy elderly people may respond to an overdose with prolonged toxicity. Benzodiazepines potentiate the effects of other central nervous system depressants, particularly alcohol, tricyclic antidepressants, and barbiturates. Dizziness, drowsiness, ataxia, and slurred speech are the usual features; coma, respiratory depression, and hypoten- sion are uncommon and usually mild. Flurazepam is most likely to cause significant central nervous system depression. Amnesia of events during the period of drug effect is also seen. Treatment The use of flumazenil is potentially hazardous in patients who have co-​ingested proconvulsant drugs, particularly tricyclics, or who are habituated to benzodiazepines from therapeutic use (risk of acute withdrawal and fits). Flumazenil should therefore not be used routinely in benzodiazepine poisoning, nor as a diagnostic test. It should be given to avoid assisted ventilation in a patient who is otherwise going to require intubation, particularly in those with ex- isting chronic airways obstruction. Flumazenil has a short half-​life (40–​80 mins) and therefore repeated doses of 0.5–2 mg IV (or an infusion) may be required. β-​Adrenoceptor blocking drugs (β-​blockers) β-​adrenoceptor blocking drugs (β-​blockers) exert their toxic effects in overdose not only by blocking the β1-​ and β2-​adrenoceptors, but also by virtue of their membrane stabilizing activity, which results in a quinidine-​like effect on the action potential as a result of sodium channel blockade; this produces QRS widening, which predisposes to ventricular arrhythmias. Clinical features Symptoms usually occur within 6 h of ingestion of non​sustained re- lease preparations. Sinus bradycardia may be the only feature after a small overdose, but if a substantial amount has been ingested, coma, convulsions (particularly with propranolol), profound bradycardia, and hypotension may occur. Other effects include drowsiness, de- lirium, hallucinations, low-​output cardiac failure and cardiorespira- tory arrest (asystole or ventricular fibrillation). Bronchospasm and hypoglycaemia occur rarely. First-​degree heart block, intraventricular conduction defects, right and left bundle branch block, ST segment elevation, ventricular extrasystoles, and disappearance of the P-​wave may be noted on the electrocardiogram. Sotalol has been reported to cause QT interval prolongation and ventricular arrhythmias and asystole may follow severe overdose from any β-​adrenoceptor blocking drug. Treatment A delay in treatment may be fatal in patients who are severely poi- soned. The blood pressure and cardiac rhythm of the patient should be monitored immediately in an intensive care area and supportive measures implemented. Glucagon is the drug of choice for severe hypotension; it bypasses the blocked β-​receptor, thus activating adenyl cyclase and pro- moting the formation of cAMP (which has a direct β-​stimulant ef- fect on the heart) from adenosine triphosphate (ATP). It should be given in a bolus dose of 50–​150 µg/​kg (typically 10 mg in an adult) over 1 min, followed by an infusion of 1–​5 mg/​h according to re- sponse. Conventional inotropes are less effective than glucagon in severe cases. If bradycardia is refractory to atropine 0.6–​1.2 mg intravenously, repeated as necessary, transcutaneous or transvenous pacing should be considered. Sodium bicarbonate may reverse the cardiotoxic ef- fects of β-​blockers with membrane stabilizing activity and should be considered for the treatment of ventricular dysrhythmias. Occasionally, diazepam 10–​20 mg intravenously may be needed for convulsions. If bronchospasm supervenes, salbutamol (albuterol) by nebulizer, should be employed. Hypoglycaemia should be corrected.

10.4.1  Poisoning by drugs and chemicals 1739 β2-​Adrenoceptor agonists Poisoning with β2-​adrenoceptor stimulants, including fenoterol, pirbuterol, reproterol, rimiterol, salbutamol (albuterol), and ter- butaline, has followed deliberate and accidental ingestion of these drugs and has also resulted from confusion over the difference be- tween oral and parenteral doses. β2-​agonists act on β2-​adrenergic receptors and increase intracel- lular cAMP. In addition to initiating relaxation of bronchial, vas- cular, and uterine smooth muscle, β2-​agonists cause glycogenolysis in skeletal muscle and hepatic glycogenolysis and gluconeogenesis. Hypokalaemia is caused by β2-​receptor-​mediated activation of Na+-​ K+-​ATPase, with extracellular potassium being shifted into the intracellular compartment; hypokalaemia may precipitate supraventricular and ventricular arrhythmias. Clinical features Tremor, sinus tachycardia, agitation, convulsions, supraventricular and ventricular arrhythmias, hypokalaemia, hyperglycaemia, and ketoacidosis are the typical features of severe poisoning with β2-​agonists. Psychosis and hallucinations are observed occasionally. Treatment Severe hypokalaemia should be corrected as soon as possible by the administration of an infusion of potassium at a rate of 40–​60 mmol/​h diluted in 5% dextrose. A  non​selective β-​blocker, such as propranolol 1–​5 mg by slow intravenous injection, will also reverse hypokalaemia and tachyarrhythmias, but its use may ex- acerbate pre-​existing obstructive airways disease. Supraventricular tachycardia has been treated successfully with adenosine 6 mg IV. Convulsions are usually single and short-​lived but, if necessary, diazepam, 10–​20 mg intravenously, may be given. Bismuth chelate (tripotassium dicitratobismuthate) Although bismuth absorption from bismuth chelate is low after a therapeutic dose, a significant quantity may be absorbed after overdose. Clinical features Self-​poisoning with large doses of bismuth chelate has caused re- versible renal failure. Bismuth encephalopathy has occurred fol- lowing chronic excess ingestion. Treatment Chelation with unithiol or succimer enhances urine bismuth clear- ance but there is no evidence that these agents prevent nephrotox- icity. Extracorporeal renal support may be required for renal failure. Calcium channel blockers Calcium channel blockers act by blocking voltage-​gated calcium channels at cardiac conducting and contractile tissue and vascular smooth muscle. Clinical features In overdose, calcium channel blockers cause nausea, vomiting, dizziness, slurred speech, confusion, sinus bradycardia and tachy- cardia, prolonged atrioventricular conduction, atrioventricular dis- sociation, hypotension, pulmonary oedema, convulsions, coma, hyperglycaemia, and metabolic acidosis. When a sustained release preparation has been ingested, the onset of severe features may be delayed for more than 12 h. Cardiac complications are usually more serious following overdose with verapamil or diltiazem than with the dihydropyridines, such as nifedipine and amlodipine. Large overdoses carry a poor prognosis, particularly in patients with is- chaemic heart disease and in those taking β-​blockers. Treatment Calcium chloride (10%, 5–​10 ml at 1–​2 ml/​min) or calcium gluconate (10% solution) 10–​20 ml intravenously may reverse pro- longed intracardiac conduction times. If significant hypotension persists despite volume replacement, intravenous glucagon 10 mg (150 µg/​kg) should be given to an adult and can be followed by an infusion 5–​10 mg/​h depending on response. If hypotension persists, administer a sympathomimetic amine intravenously. Insulin–​dextrose euglycaemia has been shown to improve myo- cardial contractility and systemic perfusion and may be used as an adjuvant to a sympathomimetic amine. There is some evidence that intravenous intralipid is useful in patients who do not respond to other measures. Cardiac pacing may have a role if there is evidence of atrioventricular conduction delay, but there may be failure to capture. Successful use of intra-​aortic balloon pumping, cardiac bypass, and extracorporeal membrane oxygenation (ECMO) have been reported in extremely severe cases. Dapsone Dapsone is a sulfone antibiotic used in the management of leprosy and dermatitis herpetiformis. Clinical features Dapsone poisoning is potentially very severe, resulting in meth- aemoglobinaemia, haemolysis, hepatitis, and central nervous system effects including drowsiness, coma, and seizures. Treatment Multiple-​dose activated charcoal increases dapsone elimination. Methaemoglobin concentrations above 30% should be treated with methylthioninium chloride (methylene blue) 1–​2 mg/​kg intraven- ously. Dapsone-​induced methaemoglobinaemia may persist for sev- eral days necessitating repeated doses of methylthioninium chloride or an infusion. Haemodialysis enhances dapsone elimination and has been used successfully in the management of life-​threatening ingestions. Digoxin and digitoxin Digoxin and digitoxin toxicity occurs in: • patients on regular therapy who gradually accumulate drug due to excess dosing, or development of incipient renal impairment • patients who take a single, large overdose, both in those on chronic therapy and those naïve to the drug Interpretation of the clinical and biochemical features differs be- tween these situations. In acute poisoning, the most significant feature normally seen is bradycardia. Since digoxin acts on a Na+-​K+-​ATPase, and subse- quent changes in the myocardium develop following this, onset of the effects of digoxin in overdose may take up to 12 h. In very large overdoses, however, severe features may develop sooner than this,

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1740 although in clinical practice very large overdoses are uncommon. Because of the action on Na+-​K+-​ATPase, the serum potassium concentration increases, and a very high serum potassium concen- tration is therefore a useful, rapidly measurable marker for severe di- goxin poisoning. Measurement of the plasma digoxin concentration will confirm the diagnosis. This is particularly the case in patients on chronic therapy who may have less dramatic changes in serum po- tassium, perhaps because of coexistent diuretic therapy, and where clinical features may be more predominantly tachycardias. Patients require treatment for cardiovascular compromise, not for blood concentrations. In acute poisoning, blood concentrations may rise to quite high values (above 5 µg/​litre) without necessarily causing severe clinical features. These rises may be transient as the drug redistributes into fatty stores after absorption. In chronic therapy, plasma concentrations give a better indication of the quan- tities of digoxin present in the body, and in acute overdose several plasma concentration measurements may need to be taken over a short period to assess the dose absorbed. Clinical features Nausea, vomiting, and bradycardia may occur. Sinus bradycardia is the most important and earliest feature in acute poisoning. Malignant ventricular arrhythmias are seen in patients with severe poisoning. At high doses, central nervous system features including drowsiness and hallucinations may be present. Treatment In patients who are vomiting, the airway needs to be protected, and consideration should be given to administering charcoal later than 1 h in patients who have ingested significant quantities, as this is such a toxic compound. The temptation to treat moderate hyperkalaemia should be resisted, as this will interfere with moni- toring clinical response. Patients should be treated on the basis of their cardiovascular status, not the plasma concentrations of digoxin alone. Patients with bradycardia who are symptomatic should receive atropine and have any acid–​base disturbance corrected. In patients with significant bradycardia or malignant ventricular arrhythmias, the most effective therapy is neutralization of digoxin with digoxin antibody. Doses of antibody recommended by the manufacturers are designed to completely neutralize all digoxin present in the pa- tient. Such an approach is unwarranted, particularly in patients on chronic therapy with digoxin in whom complete reversal of digoxin will unmask the disorder for which they are being treated. Initially, half the calculated total neutralizing dose should be given. Further doses can be given subsequently, if necessary. In patients who receive the antibody, clinical improvement will occur rapidly, usually within 20 min. Failure to respond indicates either an incorrect diagnosis or continued absorption of digoxin. Measurement of serum digoxin concentration is not possible once the digoxin antibody has been administered, since currently avail- able assays measure both bound and free compound. Extracorporeal elimination techniques are ineffective in removing digoxin though multiple-​dose activated charcoal may increase elimination. Diuretics Most diuretic overdoses are minor, although inevitably some dis- turbance of fluid and electrolyte balance will result. When combined diuretic and potassium formulations are ingested, the potassium content is likely to pose the greater risk. More serious consequences are likely if a potassium-​sparing diuretic has been ingested. Clinical features Symptoms and signs of toxicity include anorexia, nausea, vomiting, diarrhoea, profound diuresis, dehydration, and hypotension. In addition, dizziness, weakness, muscle cramps, tetany and, occasion- ally, gastrointestinal bleeding may be seen. The electrolyte and meta- bolic disturbances that may be observed include hyponatraemia, hypoglycaemia or hyperglycaemia, hyperuricaemia, hypokalaemia, and metabolic alkalosis. Hyperkalaemia develops following the ingestion of combined diuretic and potassium preparations and potassium-​sparing diuretics, such as amiloride, spironolactone, or triamterene. Small-​bowel ulceration and stricture formation have followed poisoning due to diuretics with an enteric-​coated core of potassium chloride. Treatment Symptomatic and supportive therapy should be employed with correction of fluid and electrolyte imbalance. Patients with severe hyperkalaemia may need a glucose and insulin infusion followed by oral or rectal administration of an ion-​exchange resin. Iron Most medicinal preparations of iron are as the ferrous salt. Ferrous iron is oxidized to the ferric state before being absorbed. It is im- portant to differentiate vitamin preparations that contain iron from medicinal preparations, since the former generally do not cause significant clinical problems unless very large amounts are taken. Since iron toxicity is quite closely related to dose per kilo- gram ingested, serious poisoning is more likely to occur in young children than in adults. The anticipated toxicity of iron is nor- mally estimated by calculating the dose of elemental iron present in the preparation, which varies from salt to salt. Ingestions above 150 mg/​kg of elemental iron are generally extremely severe and may be fatal. Iron salts are both locally corrosive within the gastrointestinal tract and in the cell act as cellular toxins, probably by altering the function of mitochondria. In severe poisoning, patients are uncon- scious and suffer from circulatory collapse and hepatic injury. Clinical features Depending on the severity of poisoning features may vary, and in severe cases, features would be expected within the first 6 h and include nausea, vomiting, and abdominal pain. Iron will stain the vomit and faeces (diarrhoea) and may also cause intestinal ulcer- ation and result in haemorrhage. Large amounts of iron may be vis- ible on a straight abdominal radiograph, but this should not be done routinely to confirm iron ingestion in children. Following absorption of iron there is often a period of relative calm during which iron is taken into cells before its toxic effects mani- fest. In severe poisoning, profound hypotension, metabolic acidosis, coma, and features of hepatic necrosis and renal failure ensue. Such patients require intensive supportive care and mortality rates are high. In patients who recover from severe poisoning, gut strictures following scarring from ulceration may be problematic. The com- monest site is around the pylorus, particularly in young children.

10.4.1  Poisoning by drugs and chemicals 1741 Treatment Although dose is related to toxicity, patients may be sometimes in- accurate in their history, and since vomiting is a frequent early fea- ture it may be difficult to assess exactly how much iron has been absorbed. Plasma concentration measurements on more than one occasion may assist this process. High serum iron concentrations (>90 micromol/litre or 5 mg/​litre) in the first 4–​6 h after overdose are more likely to indicate severe poisoning. In this situation, iron will be circulating free in plasma and may result in toxicity. As iron does not bind to charcoal, patients who present early with suspected large iron ingestions should be considered for gastric aspiration or lavage, though in practice this is rarely per- formed since vomiting is a prominent early feature in these cases. Whole-​bowel irrigation has been advocated following ingestion of slow-​release iron preparations, though data on its efficacy is an- ecdotal. In patients with significant elevated iron concentrations (>90 micromol/​litre or 5 mg/litre) and features suggestive of sig- nificant poisoning, the specific iron-​chelating agent deferoxamine (desferrioxamine) should be administered intravenously. There are few human data to support the usual dose regimen of deferoxamine 15 mg/​kg/​h up to a maximum of 80 mg/​kg, which binds relatively little elemental iron. Toxicity with reported during its use in the management of chronic disorders, such as haemachromatosis and haemoglobinopathies, is not a feature of its use in the management of iron poisoning and should not, therefore, be used as a guide to limit dosing in severe cases. Deferoxamine may cause hypotension and there are occasional reports of anaphylactoid reactions. Once deferoxamine has been administered, interpretation of iron concentrations becomes impossible because the iron bound to deferoxamine is detected in the laboratory assay. Iron-deferoxamine complex col- ours urine red. Patients who have not developed features of poisoning within 6 h have probably not ingested very large quantities of iron, unless they have taken a slow-​release product. Most patients merely re- quire treatment for their gastrointestinal disturbance. Since iron preparations are more commonly given to women who are preg- nant than other groups of the population, iron overdose may be seen more frequently in pregnant women. There is currently no evidence to suggest these patients should be treated differently because of pregnancy, and deferoxamine, should certainly not be withheld in patients who are deemed to require it. Isoniazid Poisoning with isoniazid is potentially very serious, but uncommon. Isoniazid depresses brain concentrations of γ-​aminobutyric acid (GABA), thus leading to seizures. Clinical features The ingestion of isoniazid 80–​150 mg/​kg body weight is likely to cause severe poisoning. Nausea, vomiting, slurred speech, dizzi- ness, and visual hallucinations may develop. Stupor, coma, and con- vulsions follow rapidly and may be associated with hyperthermia, hyperreflexia, extensor plantar responses and, later, rhabdo- myolysis. In addition, dilated pupils, sinus tachycardia, and urinary retention may be observed. In severe cases, hypotension, acute renal failure, and respiratory failure may ensure. Marked metabolic (lactic) acidosis is common. Less commonly, hyperglycaemia, keto- acidosis, glycosuria, and ketonuria are found. Treatment Supportive measures, including the correction of metabolic acid- osis, should be instituted immediately if the patient is unconscious. Pyridoxine 1 g for 1 g of isoniazid ingested should be given intra- venously to control convulsions. When the ingested dose of iso- niazid is unknown, an initial intravenous dose of pyridoxine 5 g should be given. Due to the mechanism of toxicity, diazepam alone may be ineffective but the use of diazepam and pyridoxine is syner- gistic and both should be used for convulsions. Pyridoxine 5 g may be repeated if convulsions persist (in one case 52 g pyridoxine was given intravenously without ill effects). Lithium carbonate Lithium carbonate remains the drug of choice for the treatment of recurrent bipolar illness. It has a low therapeutic index and toxicity is usually the result of therapeutic overdosage (chronic toxicity) ra- ther than deliberate self-​poisoning (acute toxicity). Chronic tox- icity is usually explained by a reduction in lithium renal clearance without a reduction in dose. Single large doses are occasionally in- gested by individuals on long-​term treatment with the drug (acute on therapeutic toxicity). Clinical features Features of intoxication include thirst, polyuria, diarrhoea, and vomiting, and, in more serious cases, tremor, impairment of con- sciousness, hypertonia, and convulsions; irreversible neurological damage may occur. Measurement of the serum lithium concentra- tion confirms the diagnosis. Chronic toxicity is usually associated with concentrations above 1.5 mmol/​litre. However, acute massive overdosage may produce much higher concentrations without causing toxic features, at least initially. This is explained by plasma lithium concentrations that are substantially higher than cen- tral nervous system lithium concentrations before distribution is complete. Treatment Activated charcoal does not adsorb lithium. Treatment is sup- portive together with measures to enhance the rate of lithium elimination. Haemodialysis should be considered if neurological features are present, if renal function is impaired and if chronic toxicity or acute on therapeutic toxicity are the modes of presen- tation. The efficacy of haemodialysis is limited by the relatively slow movement of lithium ions across cell membranes. It is easy to reduce serum lithium concentrations, but they frequently re- bound when treatment is stopped and clinical improvement is much slower. Repeated haemodialysis sessions are usually re- quired. Continuous haemodialfiltration can be used if conventional haemodialysis is not available, though clearance of lithium is less efficient. Nitrates Organic nitrates such as isosorbide mononitrate and isosorbide dinitrate are vasodilators that act by relaxing vascular smooth muscle. These drugs are essentially nitric oxide donors, which in- crease nitric oxide-​induced activation of guanylate cyclase with

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1742 subsequent elevation of cGMP concentrations. Their effects in over- dose are directly related to their therapeutic actions. These drugs undergo extensive first pass metabolism in the liver. Exposure to in- organic nitrates is principally via drinking water. Clinical features The symptoms and signs caused by pharmaceutical nitrates in overdose are due primarily to excessive arteriolar and venous dilatation. Headache and vomiting are common, accompanied by flushing of the skin and dizziness. Sinus tachycardia, severe orthostatic hypotension, and syncope may develop. Convulsions and coma may be seen in severely poisoned patients. In contrast to poisoning by inorganic nitrates, methaemoglobinaemia is seen very rarely with organic nitrates. Methaemoglobinaemia caused by inorganic nitrates is in fact due to conversion of nitrate to nitrite by gastrointestinal bacteria, following ingestion. This is encoun- tered primarily among infants bottle-​fed with formula milk that has been made up with water with high nitrate content. Treatment Treatment is supportive. Methaemoglobin concentrations exceed­ ing 30% should be treated with intravenous methylthioninium chloride (methylene blue) 1–​2 mg/​kg. Non​steroidal anti-​inflammatory drugs These include a variety of different groups of drugs, all acting by inhibition of cycloxygenase enzymes. The newer, so-​called selec­tive, agents are thought to inhibit the inducible form of cycloxygenase (COX-​2) more than the other forms of the enzyme. These drugs, therefore, all tend to inhibit prostaglandin synthesis, and the main toxicity seen in overdose is on the kidney. In very large doses, central nervous system effects may be seen but these are uncommon. Non​steroidal anti-​inflammatory drugs (NSAIDs) come in ­different chemical groupings:  oxicams (meloxicam, piroxicam, tenoxicam) and phenylpropionic (arylpropionic) acid derivatives (e.g. fenbufen, ibuprofen, naproxen, tiaprofenic acid, mefenamic acid). The COX-​2 selective agents are also potentially nephrotoxic in overdose due to inhibition of renal prostaglandin synthesis. Clinical features Most NSAIDs cause only nausea except in severe cases. Mefenamic acid is an exception and produces nausea, vomiting and, occasion- ally, bloody diarrhoea; drowsiness, dizziness and headaches are common; and hyperreflexia, muscle twitching, convulsions, cardio- respiratory arrest, hypoprothrombinaemia, and acute renal failure have also been reported. Treatment Treatment of poisoning with NSAIDs is generally supportive. It is important to check renal function at an interval after ingestion in patients who have ingested large doses. Changes in serum potas- sium and elevations in serum creatinine concentrations are to be expected in patients ingesting toxic doses. Treatment of renal im- pairment is conventional. Dialysis may be required in very severe cases. Activated charcoal should only be considered in patients who have ingested very large doses of non​steroidals (generally >20 tab- lets). Convulsions with mefenamic acid are unlikely to be persistent but, if they are, should be managed by diazepam 10–​20 mg intra- venously or lorazepam ​4 mg intravenously. Opiates and opioids See section later on in this chapter, ‘Opiates and opioids’. Paracetamol (acetaminophen) Mechanisms of toxicity The toxicity of paracetamol is related to its metabolism (Fig. 10.4.1.1). In therapeutic doses, 60–​90% is metabolized by conjugation to form paracetamol glucuronide and sulphate. A much smaller amount (5–​10%) is oxidized by mixed function oxidase enzymes to form a highly reactive compound (N-​acetyl-​ p-​benzoquinoneimine, NAPQI), which is then immediately con- jugated with glutathione and subsequently excreted as cysteine and mercapturate conjugates. Only 1–​4% of a therapeutic dose of the drug is excreted unchanged in urine. In overdose, larger amounts of paracetamol are metabolized by oxidation because of saturation of the sulphate conjugation pathway. As a result, liver glutathione stores become depleted so NHCOCH3 NHCOCH3 NHCOCH3 NHCOCH3 NHCOCH3 NHCOCH3 NHCOCH3 NHCOCH3 SCH2CHCO2H SCH2CHCO2H OSO3H OC6H9O6 OH N-Acetyl-p-benzoquinone- imine (NAPQI) Cysteine conjugate Glutathione conjugate Paracetamol mercapturic acid Paracetamol sulphate Paracetamol glucuronide Paracetamol O O GSH GSH OH OH δ+ Fig. 10.4.1.1  Metabolism of paracetamol.

10.4.1  Poisoning by drugs and chemicals 1743 that the liver is unable to deactivate the toxic metabolite. NAPQI is believed to have two separate but complementary effects. Firstly, it reacts with glutathione, thereby depleting the cell of its normal defence against oxidizing damage. Secondly, it is a potent oxidizing as well as arylating agent; it inactivates key sulphydryl groups in certain enzymes, particularly those controlling calcium homeostasis. Paracetamol-​induced renal damage probably results from a mechanism similar to that causing for hepatotoxicity (i.e. by for- mation of NAPQI). As would be expected from the mechanism of toxicity, the severity of paracetamol poisoning is dose related. An absorbed dose of 15 g (200 mg/​kg) or more is potentially serious in most patients. There is, however, some variation in individual susceptibility to paracetamol-​ induced hepatotoxicity. Those with a high alcohol intake and poor nu- trition, and those suffering from anorexia nervosa or acute starvation have glutathione depletion and are at higher risk. Individuals with HIV-​related disease also appear to be more susceptible to paracetamol-​induced hepatic damage. Those receiving enzyme-​ inducing drugs are also at greater risk. Clinical features The features of paracetamol poisoning are summarized in Table 10.4.1.5. Biochemical and haematological abnormalities may also occur (Table 10.4.1.6). Following the ingestion of an overdose of paracetamol, patients usually remain asymptomatic for the first 24 h, or at most develop anorexia, nausea, and vomiting. Paracetamol-​induced kaluresis may cause hypokalaemia. Liver damage is not usually detectable by routine liver function tests until at least 12 h after ingestion of the drug, and hepatic tenderness and abdominal pain are seldom exhibited before the second day. Liver damage reaches a peak, as assessed by plasma alanine or aspartate aminotransferase (ALT, AST) activity or prothrombin time (international normalized ratio, INR), 72–​96 h after ingestion. More often there is prolongation of the prothrombin time and a marked rise in aminotransferase ac- tivity (activities of several thousand are not uncommon) without the development of fulminant hepatic failure. Renal failure due to acute tubular necrosis develops in about 25% of patients with severe hepatic damage and in a few without evidence of serious disturb- ance of liver function. Other features, including hypoglycaemia and hyperglycaemia, cardiac arrhythmias, pancreatitis, gastrointes- tinal haemorrhage, and cerebral oedema may all occur with hepatic failure due to any cause and are not direct consequences of para- cetamol toxicity. Paracetamol can cause metabolic acidosis at two distinct periods after overdosage. Transient hyperlactataemia is frequently found within the first 15 h in all but minor overdoses and appears to be due to inhibition of mitochondrial respiration at the level of ubi- quinone and increased lactate production. It is rarely of clinical con- sequence, although in very severe paracetamol poisoning (plasma paracetamol concentration >500 mg/​litre at 4 h after ingestion) the acidosis may be very rarely associated with coma. The second phase of hyperlactataemia and acidosis occurs in those patients who present late and go on to develop hepatic damage; in this instance, decreased hepatic lactate clearance appears to be the major cause, compounded by poor peripheral perfusion and increased lactate production. Hypophosphataemia is a recognized complication of acute liver failure, including that due to paracetamol, and may contribute to morbidity and mortality by inducing mental confusion, irrit- ability, coma, and abnormalities of platelet, white cell, and erythro- cyte functions. Phosphaturia appears to be the principal cause of hypophosphataemia in paracetamol poisoning; it may occur in the absence of fulminant hepatic failure and indicates paracetamol-​ induced renal tubular damage. Prediction of liver damage In the early stages following ingestion of a paracetamol overdose, most patients have few symptoms and no physical signs. There is thus a need for some form of assessment that estimates the risk of liver damage at a time when the liver function tests are still normal. Details of the dose ingested may be used but, in many cases, the his- tory is unreliable and, even when the dose is known for certain, it does not take account of early vomiting and individual variation in response to the drug. However, a single measurement of the plasma paracetamol con- centration is an accurate predictor of liver damage provided that it is taken not earlier than 4 h after ingestion of the overdose. Information gained from several studies has enabled the production of graphs which may be used for prediction of liver damage and which serve as Table 10.4.1.5  Clinical features of untreated paracetamol poisoning (>200 mg/​kg) Day 1 Day 2 Day 3 Asymptomatic May become asymptomatic or develop symptoms de novo (in severe untreated poisoning) Nausea Vomiting Jaundice → liver failure → hepatic encephalopathy Vomiting Hepatic tenderness ± generalized abdominal tenderness Back pain + renal angle tenderness → renal failure Abdominal pain Occasionally, mild jaundice Cardiac arrhythmias Anorexia Disseminated intravascular coagulation Pancreatitis Table 10.4.1.6  Biochemical and haematological abnormalities in paracetamol poisoning Biochemical abnormalities Haematological abnormalities AST/​ALT ↑↑ PT ↑ Bilirubin ↑ Platelets ↓ Blood sugar ↓ Clotting factors II ↓ V ↓ VII ↓ Creatinine ↑ Lactate ↑ Phosphate ↓ Amylase ↑ Potassium ↓ early due to kaluresis
↑ later in renal failure

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1744 a guide to the need for specific treatment (Fig. 10.4.1.2). These may not be accurate with slow-​release products. In patients who have taken several overdoses of paracetamol over a short period of time, the plasma paracetamol concentration will be meaningless in relation to the treatment graph. Such patients should be considered at risk and treated. Patients who regularly consume al- cohol in excess of currently recommended limits (particularly those who are malnourished), those who regularly take enzyme-​inducing drugs (e.g. carbamazepine, phenytoin, phenobarbital, and rifam- picin) and those with conditions causing glutathione depletion (e.g. malnutrition and HIV infection) may be at risk of liver damage. Studies in the early 1970s showed that 60% of patients whose plasma paracetamol concentration was above the line drawn between 200 mg/​litre (1.32 mmol/​litre) at 4 h and 50 mg/​litre (0.33 mmol/​ litre) at 12 h after the ingestion of the overdose were likely to sustain liver damage (ALT or AST >1000 u/​litre), unless specific protective treatment were given. There are now two approaches to risk management used world- wide. In the United Kingdom, following a decision by the MHRA in 2012 to abandon a detailed risk assessment in the decision to treat, it is deemed that patients with concentrations above a ‘treatment line’ starting at 100  mg/​litre (0.66  mmol/​litre 4 h after ingestion (Fig. 10.4.1.2) require therapy with antidote. In North America and Australasia, a parallel line starting at 150 mg/​litre 4 h after ingestion is used. The MHRA decision aims to prevent one death in the United Kingdom approximately every 2.1 years, and has been criticized as being too cautious and resulting in many unnecessary treatments. Patients who ingest multiple overdoses, or take repeated thera- peutic excess, are at greater risk of liver damage and decisions to treat them are generally based on dose ingested. Current UK advice is very conservative with a treatment cut-​off of paracetamol dose above 75 mg per kilogram in 24 h. Prognostic factors The overall mortality of paracetamol poisoning in untreated pa- tients is only of the order of 5%. A rise in transaminase (ALT/​AST) activity is usually the first liver function test to become abnormal, but a rise in INR is of particular value in assessing the prognosis of an individual patient. The more rapid the increase in ALT and INR, the worse the prognosis of the patient. A prothrombin time of more than 20 s at 24 h after ingestion indicates that significant hepatic damage has been sustained, and a peak prothrombin time of more than 180 s is associated with a chance of survival of less than 8%. Acid–​base disturbances are also a good guide to prognosis. Systemic acidosis developing more than 24 h after overdose in- dicates a poor prognosis; patients with a blood pH below 7.30 at this time have only a 15% chance of survival. In addition, a rise in the serum creatinine concentration is associated with poor survival; patients with a serum creatinine concentration above 300 µmol/​litre have only a 23% chance of survival. A study of prognostic indicators in patients who died of paracetamol-​ induced fulminant hepatic failure treated conventionally com- pared measurement of factors V and VIII with conventional tests. An admission pH below 7.30 with a serum creatinine concentra- tion above 300 µmol/​litre and a prothrombin time above 100 s in patients with grade III–​IV encephalopathy had a sensitivity, predictive accuracy, positive prediction value, and specificity of 91, 86, 83, and 91%, respectively. However, a factor VIII/​V ratio above 30 had comparable values of 91, 95, 100, and 100%. Novel biomarkers (e.g. microRNAs) are now being studied which may give an earlier more accurate risk assessment than is possible with conventional approaches. Treatment Consider administering activated charcoal for patients presenting within 1 h of overdose. Parenteral fluid replacement should be given if nausea persists or vomiting occurs. Patients who have taken staggered overdoses should be treated with an antidote irrespective of the plasma paracetamol concentra- tions. They can be discharged after antidotal treatment, provided they are asymptomatic and the INR, plasma creatinine concentra- tion, and ALT activity are normal. Patients who present 15 h or more after overdose tend to be more severely poisoned and at greater risk of developing serious liver damage and should receive antidotal treatment as the plasma con- centration alone may not be an accurate guide of severity; it may be non​detectable at the time of late presentation. The INR, venous pH, plasma creatinine concentration, and liver function tests are helpful in determining prognosis. Acetylcysteine Acetylcysteine acts by replenishing cellular glutathione stores and may also repair oxidation damage caused by NAPQI either directly or, more probably, through the generation of cysteine and/​ or gluta- thione. It may also act as a source of sulphate and so ‘unsaturate’ sulphate conjugation. Two principal regimens for acetylcysteine have been em- ployed. The most widely utilized worldwide is a 21-​h protocol (Table 10.4.1.7); the oral protocol previously used in the United 240 220 200 Plasma paracetamol concentration (mg/litre) 180 160 140 120 100 80 60 40 20 0 0 2 4 6 8 10 Paracetamol nomograms 12 14 16 18 Time (hours) 20 22 24 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 Original ‘200 mg’ Line UK ‘100 mg’ Line International ‘150 mg’ Line 1.6 Plasma paracetamol concentration (mmol/litre) Fig. 10.4.1.2  Paracetamol treatment nomograms. Those above the treatment lines are regarded at risk and treated with acetylcysteine. Thresholds differ in different countries, the ‘100 mg’ line being used in the United Kingdom. There is a scarcity of data after 15 h and the dotted lines show extrapolations used in clinical practice.

10.4.1  Poisoning by drugs and chemicals 1745 States being replaced by IV therapy. An alternative 12 h IV protocol (100 mg/​kg over 2h, 200 mg/​kg over 10 h) that is associ- ated with far fewer antidote-​induced adverse events is also being employed. Provided that acetylcysteine is administered within 8–​10 h of overdose, the development of hepatic damage is normally pre- vented; thereafter, the protective effects decline rapidly. Up to 10% of patients treated with intravenous acetylcysteine (20–​21 ​h regimen) develop rash, angioedema, hypotension, and broncho- spasm. Far higher proportions (up to 70%) develop nausea and vomiting, with treatment required in up to 30%. These reactions, which are due to the initial bolus, cause very few fatalities but cause treatment interruption and patient distress. Anaphylactoid reac- tions are far more common at lower paracetamol concentrations, so more aggressive use of the antidote increases the incidence of these reactions. Antihistamines, such as chlorpheniramine and broncho- dilators (e.g. salbutamol), may be given if such anaphylactoid reac- tions do occur, but discontinuing the infusion temporarily may be all that is required. Management of severe liver damage A 10% glucose solution should be administered to prevent the onset of hypoglycaemia. If fulminant hepatic failure supervenes, the use of a continued intravenous acetylcysteine (the 16-​h infu- sion is continued until recovery or death) will reduce morbidity and mortality. In one prospective study, the survival rate in 25 patients with paracetamol-​induced fulminant hepatic failure was 20%, with an in- cidence of cerebral oedema and of hypotension requiring inotropic support of 68 and 80%, respectively. With acetylcysteine, the com- parable figures in 25 matched patients were 48% (survival rate), 40% (cerebral oedema), and 48% (hypotension). A proton pump inhibitor will reduce the risk of gastrointestinal bleeding from ‘stress’ ulceration/​erosion. There is no evidence that fresh frozen plasma prevents gastrointestinal haemorrhage in pa- tients with severe coagulation abnormalities (prothrombin time

100 s). If acute renal failure supervenes, then this should be man- aged conventionally. Liver transplantation has been performed successfully in patients with paracetamol-​induced fulminant hepatic failure. Salicylates Salicylate poisoning may result from overdose of aspirin tablets, per- cutaneous absorption of salicylic acid (used in keratolytic agents), and ingestion of methyl salicylate (‘oil of wintergreen’). In therapeutic doses, aspirin is absorbed rapidly from the stomach and small intestine, but in overdose, absorption may occur more slowly, and plasma salicylate concentrations may continue to rise for up to 24 h. The pharmacokinetics of elimination of aspirin are important determinants of salicylate toxicity. Biotransformation to both salicyluric acid and salicylphenolic glucuronide (Fig. 10.4.1.3) is saturable with the following clinical consequences:  (1) the time needed to eliminate a given fraction of a dose increases with increasing dose; (2) the steady state plasma concentration of sa- licylate, particularly that of the pharmacologically active non-​ protein-​bound fraction, increases more than proportionately with increasing dose; and (3) renal excretion of salicylic acid becomes increasingly important; a pathway which is extremely sensitive to changes in urinary pH. When ingested in overdose, salicylates directly stimulate the re- spiratory centre to produce both increased depth and rate of res- piration, thereby causing a respiratory alkalosis (Fig. 10.4.1.4). At least part of this effect is due to local uncoupling of oxidative phos- phorylation within the brainstem. In an attempt to compensate, bicarbonate, accompanied by sodium, potassium, and water, is excreted in the urine resulting in dehydration and hypokalaemia. More importantly, the loss of bicarbonate diminishes the buffering capacity of the body and allows an acidosis to develop more easily. Very high salicylate concentrations in the brain depress the re- spiratory centre and may further contribute to the development of acidaemia. Simultaneously, a variable degree of metabolic acidosis develops, not only because of the presence of salicylic acid itself, but also be- cause of interference with carbohydrate, lipid, protein, and amino acid metabolism by salicylate ions (Fig. 10.4.1.4). Inhibition of citric acid cycle enzymes causes an increase in circulating lactic and pyruvic acids. Salicylates stimulate fat metabolism and cause increased production of the ketone bodies, β-​hydroxybutyric acid, acetoacetic acid, and acetone. Dehydration and lack of food intake, because of vomiting, further contribute to the development of ke- tosis. Protein catabolism is accelerated and synthesis diminished. Aminotransferases (responsible for the interconversion of amino acids) are inhibited. Increased circulating blood concentrations of amino acids result, together with aminoaciduria; inhibition of active tubular reabsorption of amino acids also contributes. Aminoaciduria increases the solute load on the kidneys and, thereby, increases water loss from the body. A primary toxic effect of salicylates in overdose is uncoupling of oxidative phosphorylation (Fig. 10.4.1.4). ATP-​dependent re- actions are inhibited, and oxygen utilization and CO2 production are increased. Energy normally used for the conversion of inorganic phosphate to ATP is dissipated as heat. Hyperpyrexia and sweating result, causing further dehydration. Fluid loss is enhanced because Table 10.4.1.7  Dosing regimen for acetylcysteine (21 h regimen) • 150 mg/​kg over 60 min, then 50 mg/​kg over the next 4 h and 100 mg/​kg over the next 16 h • Total dose, 300 mg/​kg over 21 h Conjugation with glycine Conjugation with glucuronic acid Salicyluric acid Salicylacyl glucuronide Salicylphenolic glucuronide Aspirin Salicylic acid Gentisic acid Hydrolysis Hydroxylation Fig. 10.4.1.3  Metabolism of aspirin.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1746 salicylates stimulate the chemoreceptor trigger zone and induce nausea and vomiting and, thereby, diminish oral fluid intake. If de- hydration is sufficiently marked, low cardiac output and oliguria will aggravate the metabolic acidosis already present which, if severe, can itself diminish cardiac output. Glucose metabolism also suffers as a result of uncoupled oxi- dative phosphorylation because of increased tissue glycolysis and peripheral demand for glucose (Fig. 10.4.1.4). This is seen principally in skeletal muscle and may cause hypoglycaemia. The brain appears to be particularly sensitive to this effect and neuroglycopenia can occur in the presence of a normal blood sugar level when the rate of utilization exceeds the rate at which glucose can be supplied from the blood. Increased metabolism and peripheral demand for glucose activates hypothalamic centres, resulting in increased adrenocortical stimulation and re- lease of adrenaline. Increased glucose 6-​phosphatase activity and hepatic glycogenolysis contribute to the hyperglycaemia, which is sometimes seen following ingestion of large amounts of salicylate. Increased circulating adrenocorticosteroids exacerbate fluid and electrolyte imbalance. Although this is rarely a practical problem, salicylate intoxi- cation may be accompanied by hypoprothrombinaemia due to a warfarin-​like action of salicylates on the physiologically important vitamin K epoxide cycle. Vitamin K is converted to vitamin K 2,3-​epoxide and then reconverted to vitamin K by a liver mem- brane reductase enzyme, which is competitively inhibited by war- farin and salicylates. Clinical features and assessment of severity of salicylate intoxication The dose of salicylate ingested and the age of the patient are the principal determinants of the severity of an overdose. The plasma salicylate concentration should be determined on admission, but it is important to repeat it 2 h later to ensure that the concentration is not rising. If the concentration has risen, the level should be re- peated after a further 2 h. Generally speaking, plasma salicylate concentrations that lie between 300–​500 mg/​litre some 6 h after in- gestion of an overdose are associated with only mild toxicity, concen- trations between 500 and 700 mg/​litre are associated with moderate toxicity, and concentrations in excess of 700 mg/​litre confirm severe poisoning. Salicylate poisoning of any severity is associated with sweating, vomiting, epigastric pain, tinnitus, and deafness (Table 10.4.1.8). Young children quickly develop metabolic acidosis following the ingestion of aspirin in overdose, but by the age of 12 years the usual adult picture of a combined dominant respiratory alkalosis and mild metabolic acidosis is seen. To some extent, the presence of an alkalaemia protects against serious salicylate toxicity because salicylate remains ionized and unable to penetrate cell membranes easily. Development of acidaemia allows salicylates to penetrate tis- sues more readily and leads, in particular, to central nervous system toxicity characterized by excitement, tremor, delirium, convulsions, stupor, and coma. Very high plasma salicylate concentrations cause Hepatic glycogenolysis Respiratory alkalosis Stimulation of respiratory centre Stimulation of chemoreceptor trigger zone HCO3 −(and Na+ K+and H2O) excretion Vomiting and decreased oral fluid intake Dehydration and electrolyte imbalance Glucocorticoids Catecholamines Tissue glycolysis Hyperglycaemia Respiratory rate Hyperpyrexia and sweating Uncoupling of oxidative phosphorylation Neuroglycopenia Hypoglycaemia Metabolic acidosis Ketone body formation Stimulation of lipid metabolism Respiratory acidosis Depression of respiratory centre Inhibition of Krebs cycle enzymes Inhibition of amino acid metabolism Cardiac output Renal clearance of sulphuric and phosphoric acids Circulating pyruvic and lactic acid levels Circulating amino-acid levels Buffering capacity Aminoaciduria CO2 production O2 consumption Fig. 10.4.1.4  Pathophysiology of salicylate poisoning. Table 10.4.1.8  Clinical features of salicylate poisoning • Nausea, vomiting, and epigastric discomfort • Irritability, tremor, tinnitus, deafness, blurring of vision • Hyperpyrexia, sweating, dehydration • Tachypnoea and hyperpnoea • Non​cardiogenic pulmonary oedema • Acute renal failure • Mixed respiratory alkalosis and metabolic acidosis (except in children who usually develop metabolic acidosis alone) • Hypokalaemia, hypernatraemia, or hyponatraemia • Hyperglycaemia or hypoglycaemia • Hypoprothrombinaemia (rare) • Confusion, delirium, stupor, and coma (in severe cases)

10.4.1  Poisoning by drugs and chemicals 1747 paralysis of the respiratory centre and cardiovascular collapse due to vasomotor depression. Pulmonary oedema is seen occasionally in salicylate poisoning and, although this is often due to fluid overload as a result of treat- ment, it may be non​cardiac and occur in the presence of hypovol- aemia. In these circumstances, the pulmonary oedema fluid has the same protein and electrolyte composition as plasma, suggesting in- creased pulmonary vascular permeability. Although aspirin overdose may be complicated by inhibition of platelet aggregation and hypoprothrombinaemia, gastric erosions, and gastrointestinal bleeding are rare following acute salicylate overdose. Oliguria is sometimes seen in patients following the ingestion of salicylates in overdose. The most common cause is dehydration but, rarely, acute renal failure or inappropriate secretion of antidiuretic hormone may occur. Although the urine pH may be alkaline in the early stages of salicylate overdose, it soon becomes acidic. Measurement of ar- terial blood gases, pH, and standard bicarbonate may show a re- spiratory alkalosis in the early stages of salicylate intoxication accompanied by the development of a metabolic acidosis. The plasma potassium concentration is often low; rarely, the blood sugar may be high. Treatment The plasma salicylate concentration should be re-​measured 2–​3 h after the first measurement. Dehydration, electrolyte imbalance and, most importantly, metabolic acidosis should be corrected. The role of multiple-​dose activated charcoal in increasing salicylate elimination is controversial and it cannot be recom- mended on current evidence. As the relationship between renal clearance of salicylates and urine pH is logarithmic, urine alkalinization should be undertaken in patients with a plasma salicylate concentration greater than 500 mg/​litre, particularly if an acidosis is present. The therapeutic aim is to make the urine alkaline (ideally, pH 7.5–​8.5), and in adults this may be achieved by administration of sodium bicarbonate, 225 mmol (225 ml of 8.4%); further doses of bicarbonate are given as re- quired. Hypokalaemia should be corrected before administration of sodium bicarbonate, because this lowers the serum potassium concentration further. In patients with severe poisoning (plasma salicylate concentration >700 mg/​litre or >5.1 mmol/​litre), haemo- dialysis should be considered, particularly when severe acid–​base abnormalities are present. Pulmonary oedema occasionally complicates salicylate toxicity. Fluid overload should be excluded as far as possible but, if increased pulmonary vascular permeability is suspected, measurement of the pulmonary artery wedge pressure may be needed both for confirm- ation of the diagnosis and to monitor subsequent fluid adminis- tration. Positive end-​expiratory pressure ventilation appears to be beneficial. Theophylline Poisoning may complicate therapeutic use, as well as being the re- sult of deliberate self-​poisoning. If a sustained-​released formula- tion has been ingested, peak plasma concentrations of the drug are frequently not attained until 6–​12 h after overdose and the onset of toxic features is correspondingly delayed. Clinical features Symptoms include nausea, vomiting, hyperventilation, haematem- esis, abdominal pain, diarrhoea, sinus tachycardia, supraventricular, and ventricular arrhythmias, hypotension, restlessness, irritability, headache, hyperreflexia, tremor, and convulsions. Hypokalaemia results from Na+-​K+-​ATPase activation. A mixed respiratory alkal- osis and metabolic acidosis is common. Most symptomatic patients have plasma theophylline concentrations in excess of 25 mg/​litre. Convulsions are seen more commonly when concentrations are greater than 50 mg/​litre. Treatment Multiple-​dose activated charcoal (e.g. 50 g 4-​hourly) enhances the systemic elimination of theophylline. Intractable vomiting may be alleviated by ondansetron, 8 mg intravenously in an adult. Gastrointestinal haemorrhage may require blood transfusion and the administration of a proton pump inhibitor intravenously. Tachyarrhythmias may be induced by the rapid flux of potassium across cell membranes and early correction of hypokalaemia may prevent their development. The plasma potassium concentration should therefore be measured on admission and at regular intervals thereafter while the patient is symptomatic. Potassium supplements will be needed in almost all cases and doses of up to 60 mmol/​h may be required at the outset in severe cases. Non​selective β-​adrenoceptor blocking drugs, such as propranolol, may also be useful in the treat- ment of tachyarrhythmias secondary to hypokalaemia. There may be a role for extracorporeal elimination techniques in very severe poisoning (plasma theophylline concentration >100 mg/​litre). Thyroxine Clinical features Only a small percentage of patients who ingest large amounts of thyroid hormones develop features of toxicity. Symptoms develop within a few hours with triiodothyronine (T3) and after 3–​6 days with thyroxine (T4). They tend to resolve in about the same time as they take to develop. Sinus tachycardia, tremor, anxiety, irrit- ability, insomnia, hyperactivity, sweating, diarrhoea, and fever are most common. Atrial fibrillation and convulsions have also been re- ported. Myocardial necrosis occurs rarely. Treatment Serum T4 and T3 concentrations should be measured approximately 12 h after ingestion (this need not be measured as an emergency). Those with high T4 concentrations should be reviewed for evidence of toxicity on the fourth or fifth day after ingestion. Patients who develop toxicity should be given propranolol 10–40 mg, 3–4 times a day for 5 days. Drugs of abuse Amfetamines and MDMA (ecstasy) Amfetamines, particularly methamfetamine (‘crystal meth’, ‘ice’) and MDMA, are abused widely. Features of poisoning are related predominantly to stimulation of central and peripheral adren- ergic receptors and, in addition, hyperthermia and hyponatraemia (secondary to inappropriate anti-diuretic hormone) may

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1748 develop in severe MDMA toxicity. Poisoning is usually the result of recreational use. Clinical features These drugs cause increased alertness and self-​confidence, euphoria, extrovert behaviour, increased talkativeness with rapid speech, lack of desire to eat or sleep, tremor, dilated pupils, tachycardia, and hypertension. More severe intoxication is associated with excit- ability, agitation, paranoid delusions, hallucinations with violent behaviour, hypertonia, and hyperreflexia. Convulsions, rhabdo- myolysis, hyperthermia, and cardiac arrhythmias can develop in the most severe cases. Rarely, intracerebral and subarachnoid haemor- rhage and acute cardiomyopathy occur and may be fatal. In the case of MDMA, hyperthermia, disseminated intra- vascular coagulation, rhabdomyolysis, acute renal failure, and hyponatraemia are observed commonly in severe cases, in addition to those features described earlier. Death occurred in 2 of 17 pa- tients with serum sodium concentrations of 107–​128 mmol/​litre. Their clinical course was remarkably similar; initial vomiting and disturbed behaviour was followed by seizures, drowsiness, a mute state, and disorientation. Severe hepatic damage, including ful- minant hepatic failure, has also been reported. The serotonin syn- drome has been described. Treatment Intravenous fluids should be given for dehydration. Diazepam 10–​20 mg intravenously or haloperidol 2.5–​5.0 mg intramuscu- larly are effective in controlling agitation. The peripheral sym- pathomimetic actions of amfetamines may be antagonized by β-​adrenergic blocking drugs. Although acidification of the urine increases renal elimination of amfetamines, sedation is usu- ally all that is required. Dantrolene 1 mg/​kg intravenous should be administered for hyperthermia that does not respond to conventional cooling methods. In most cases, hyponatraemia responds to fluid restriction alone. Transplantation may be indicated in patients who develop MDMA-​induced fulminant hepatic failure. Cannabis Cannabis is obtained from the plant Cannabis sativa which con- tains over 400 compounds including over 60 cannabinoids. The most potent cannabinoid is Δ9-​tetrahydrocannabinol (THC), which is responsible for the psychoactive effects seen with use; other cannabinoids include Δ8-​tetrahydrocannabinol, cannabinol, and cannabidiol. Smoking is the usual route of use, but cannabis is occasionally ingested as a ‘cake’, made into a ‘tea’ or injected intravenously. Clinical features Acute use Features include euphoria, distorted and heightened images, col- ours, and sounds; altered tactile sensations, sinus tachycardia, hypotension, and ataxia. Visual and auditory hallucinations, de- personalization, and acute psychosis are particularly likely to occur after substantial ingestion in naïve cannabis users. Cannabis im- pairs all stages of memory including encoding, consolidation, and retrieval. Memory impairment following acute use may persist for months following abstinence. Cannabis infusions injected intravenously may cause nausea, vomiting, and chills within minutes; after about 1 h, profuse watery diarrhoea, tachycardia, hypotension, and arthralgia may develop. Marked neutrophil leucocytosis is often present, and hypoglycaemia has been reported occasionally. Chronic use Heavy users suffer impairment of memory and attention and poor academic performance. There is an increased risk of anxiety and de- pression. Regular users are at risk of dependence. Cannabis use re- sults in an overall increase in the relative risk for later schizophrenia and psychotic episodes. Cannabis smoke is probably carcinogenic. Treatment Most acutely intoxicated patients require no more than reassurance and supportive care. Sedation with diazepam, 10 mg intravenously, repeated as necessary, should be administered to patients who are disruptive or distressed. Haloperidol, 2.5–​5 mg intramuscularly re- peated as necessary, is occasionally required. Synthetic cannabinoid receptor agonists Synthetic cannabinoid receptor agonists (SCRAs) are full canna- binoid type 1 (CB1) receptor agonists and bind to these receptors with a higher affinity than Δ9-​tetrahydrocannabinol. Furthermore, unlike the metabolites of Δ9-​tetrahydrocannabinol, the metabol- ites of several synthetic cannabinoids retain high affinity for, and exhibit a range of intrinsic activities at, CB1 and CB2 receptors. Cannabinoids also bind non​specifically to cellular membranes and act on opioid and benzodiazepine receptors, prostaglandin syn- thetic pathways, and protein metabolism. These interactions have the potential for complex effects and are likely to contribute to toxicity. Clinical features Current third and fourth generation synthetic cannabinoid receptor agonists (SCRAs) produce more severe clinical features than earlier SCRAs. A reduced level of consciousness, tonic–​clonic convulsions, transient respiratory failure, and severe agitation, particularly on re- covery, are typical. Treatment Treatment is symptomatic and supportive. As impaired ventilation is transient, supported ventilation is not usually necessary. Cathinones, benzofurans, and related compounds Cathinones are derivates of cathionine, which occurs naturally in the herb Catha edulis (Khat). Structurally these are phenylethyla­ mines, similar in structure to catecholamines and amfetamines. Mephedrone (4-​methylmethcathinone) is one of the most widely abused cathinones, others include mexedrone, methylone, butylone, and fluoromethcathinone. Purity of these street drugs varies widely. These are usually insufflated (snorted) or swallowed. Other phenylethylamines, such as bromofurans (‘Benzofury’), have many similar effects to amfetamines, but some have more hallucinogenic effects due to the receptor specificity of the individual compounds. Methylenedioxypyrovalerone (MDPV, ‘ivory wave’) inhibits the re- uptake of dopamine and norepinephrine centrally, and is a potent cause of psychiatric features.

10.4.1  Poisoning by drugs and chemicals 1749 Clinical features These agents act as stimulants, causing agitation, hallucin- ations, increased muscle activity with bruxism (teeth grinding), hyperpyrexia, sweating, dilated pupils, tachycardia, and arrhyth- mias. Toxicity is increased if co-​ingested with other stimulants, or drugs affecting central amine mechanisms (e.g. antidepressants, tramadol). Metabolic complications include hypokalaemia, hyper- glycaemia, and metabolic acidosis. Complications include seizures and rhabdomyolysis. Treatment Intravenous fluids should be given for dehydration. Diazepam, ini- tially 10–​20 mg intravenously, is the drug of choice for agitation: this is also appropriate to control excess muscle activity. Large doses may be required. ECG and cardiovascular monitoring is necessary. Management of other features is symptomatic. Cocaine In recent decades, there has been a considerable increase in the rec- reational use of cocaine. It is a powerful local anaesthetic and vaso- constrictor and may be abused by smoking, ingestion, injection or by ‘snorting’ it intranasally. Users, body packers, and those who swallow the drug to avoid being found in possession of it (‘stuffers’), are at risk of overdose. ‘Street’ cocaine is cocaine hydrochloride, which is water soluble, so can be injected or snorted. It may be dis- solved in an alkaline solution from which the cocaine is extracted into ether, which is then evaporated to leave relatively pure (‘free- base’) cocaine. ‘Crack’ (cocaine also without the hydrochloride moiety) is extracted by using baking soda (sodium bicarbonate). Other drugs, such as ethanol, cannabis, and conventional hypnotics and sedatives, are frequently taken with cocaine to reduce the inten- sity of its less pleasant effects. Clinical features The features of cocaine poisoning are similar to those of amfetamine. In addition to euphoria, it also has sympathomimetic effects including agitation, tachycardia, hypertension, sweating, and hallucinations. Prolonged convulsions with metabolic acid- osis, hyperthermia, rhabdomyolysis, ventricular arrhythmias, and cardiorespiratory arrest may follow in the most severe cases. Less common features include dissection of the aorta, myocarditis, myo- cardial infarction, dilated cardiomyopathy, subarachnoid haemor- rhage, cerebral haemorrhage, and cerebral vasculitis. Several rare complications of the method of use of cocaine have been reported. These include pulmonary oedema after intra- venous injection of freebase cocaine and pneumomediastinum and pneumothorax after sniffing it. In addition, chronic ‘snorting’ has caused perforation of the nasal septum, rhinorrhoea of cerebro- spinal fluid due to thinning of the cribriform plate, and pulmonary granulomata. Treatment Users who are intoxicated may require sedation with diazepam to control agitation or convulsions; very large doses of diazepam may be required. Measures to prevent further absorption are not usually relevant. Hypertension and severe tachycardia may be controlled with a β-​blocker but, in one case at least, the use of propranolol caused paradoxical hypertension. Accelerated idioventricular rhythm should not normally require treatment but ventricular fib- rillation and asystole should be managed in the usual way. Ethanol Ethanol is commonly ingested in beverages before, or concomitantly with, the deliberate ingestion of other substances in overdose. It is also used as a solvent and is found in many cosmetic and antiseptic preparations. It is rapidly absorbed through the gastric and intestinal mucosae. Gastric alcohol dehydrogenase isoenzyme has a role in metabolizing ethanol before absorption, thereby preventing ethanol entering the systemic circulation, particularly following ingestion of moderate amounts of alcohol. Absorbed ethanol is initially and principally converted to acetaldehyde by an NAD-​dependent hep- atic alcohol dehydrogenase. A small proportion is oxidized by the microsomal ethanol oxidizing system and the catalase pathway. Acetaldehyde is removed by oxidation via the NAD-​dependent en- zyme aldehyde dehydrogenase, to yield acetate and, subsequently, CO2 and water. About 95% of ingested ethanol is oxidized to acet- aldehyde and acetate; the remainder is excreted unchanged in the urine, and, to a lesser extent, in the breath and through the skin. Ethanol is a central nervous system depressant that interferes with cortical processes in small doses and may depress medullary function in large doses. The effects of ethanol on the central nervous system are generally proportional to the blood ethanol concentration. Ethanol is also a peripheral vasodilator. In the severely intoxicated, it may cause hypothermia and hypotension. Ethanol metabolism results in accumulation of free NADH, with resulting increase in the NADH:NAD ratio and inhibition of hepatic gluconeogenesis, which may cause hypoglycaemia, particularly in children or when poisoning follows fasting, exercise, or chronic malnutrition. An in- crease in the lactate:pyruvate ratio may also ensue, with develop- ment of hyperlactataemia. Clinical features Ethanol exacerbates the effects of other central nervous system de- pressants, in particular, hypnotic agents. In those not tolerant, the fatal dose of ethanol alone is between 300 and 500 ml absolute al- cohol, if this is ingested in less than 1 h. The features of ethanol poi- soning are summarized in Table 10.4.1.9. Table 10.4.1.9  Clinical features of ethanol poisoning Mild intoxication (500–​1500 mg/​litre) Emotional lability, and slight impairment of visual acuity, muscular coordination, and reaction time Moderate intoxication (1500–​3000 mg/​litre) Visual impairment, sensory loss, muscular incoordination, slowed reaction time, slurred speech Severe intoxication (3000–​5000 mg/​litre) Marked muscular incoordination, blurred or double vision, sometimes stupor and hypothermia, occasionally hypoglycaemia and convulsions Coma (>5000 mg/​litre) Depressed reflexes, respiratory depression, hypotension, and hypothermia. Death may occur from respiratory or circulatory failure or as the result of aspiration of stomach contents in the absence of a gag reflex

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1750 Severe hypoglycaemia typically occurs within 6–​36 h of inges- tion of a moderate to large amount of alcohol by either a previously malnourished individual or one who has fasted for the previous 24 h; it is common in children 5 years of age or less. The patient is often comatose, hypothermic, and convulsing, with conjugate de- viation of the eyes, trismus, and extensor plantar reflexes; the usual features of hypoglycaemia (e.g. flushing, sweating, tachycardia) are often absent. Convulsions are the most common presenting sign in children with hypoglycaemia. Lactic acidosis (usually only mild) is an uncommon but potentially serious complication of acute ethanol intoxication and occurs particularly in patients with severe liver disease, pancreatitis, or sepsis. Hypovolaemia, which may accompany severe intoxication, predisposes to lactic acidosis. Treatment Supportive measures are all that are required for most patients with acute ethanol poisoning, even if the blood ethanol concentration is very high. Particular care should be taken to protect the airway. In more severe cases, acid–​base status should be determined. Lactic acidosis requires correction of hypoglycaemia, hypovolaemia, and circulatory insufficiency, if present. An infusion of sodium bicar- bonate may be necessary in severely poisoned patients in whom a lactic acidosis persists. Blood sugar should be determined hourly in severe cases and the rate of intravenous glucose adjusted accordingly. If blood sugar concentrations decrease despite an infusion of 5–​10% dextrose, a 50% glucose solution, 50 ml intravenously, should be given because hypoglycaemia is usually unresponsive to glucagon. Haemodialysis may be considered if the blood ethanol concentra- tion exceeds 7500 mg/​litre and if a severe metabolic acidosis is pre- sent, which has not been corrected by the measures outlined earlier. Fructose is of negligible clinical benefit in accelerating ethanol oxi- dation and may cause acidosis; it should not be used. γ-hydroxybutyric acid and analogues γ-hydroxybutyric acid (GHB) is a liquid that is abused as a body-​ building agent (it stimulates growth hormone release) and as a seda- tive drug of abuse. It is a precursor of gamma-​aminobutyric acid (GABA) and acts as an agonist at GABAB receptors as well as at a GHB-​specific receptor in the brain. γ-​butyrolactone (GBL) and 1,4-​ butane-​diol (1,4-​BD) are GHB precursors, converted into it after in- gestion. They are organic industrial solvents found in products such as acetone-​free nail polish removers and paint strippers. 1,4-​BD is also marketed as a dietary supplement. Clinical features Low doses cause mild agitation, excitement, nausea, and vomiting, with euphoria and hallucinations at higher doses. Coma, brady- cardia, and respiratory depression occur in the most severely poi- soned. The most unique aspect of GHB poisoning is its very brief duration. Patients may progress from deep coma, requiring intub- ation, to self-​extubation and full alertness over only a few hours. A GHB withdrawal syndrome can occur in chronic abusers with clinical features occurring within 6–​12 h of the last dose. Features include insomnia, tremor, and confusion, which may progress to de- lirium not dissimilar to the alcohol withdrawal syndrome. Treatment Supportive measures to maintain adequate ventilation and circula- tion should be employed, and this is often all that is required. GHB withdrawal should be managed as for acute alcohol withdrawal. Baclofen, a specific GABAB receptor agonist, has been used success- fully in conjunction with benzodiazepines, to control withdrawal from GHB and GBL. Ketamine Ketamine is a dissociative anaesthetic acting on NMDA-​R recep- tors. It is used as a drug of abuse for its hallucinogenic effects. In repeated doses it is toxic to the bladder causing irreversible bladder wall thickening and infiltration, with smaller capacity, muscle in- stability, and urothelial ulceration. Urinary frequency, incontinence, and hydronephrosis are complications. Several synthetic derivates are also abused in the hope they are ‘bladder safe’. Clinical features The main features are psychological with dissociative agitation, ag- gression, and paranoia. There is a risk of physical harm both to pa- tient and carers. Hallucinations may result in injury from ‘flying’ or walking into traffic. Chronic use may result in dependency, GI symptoms and, most importantly, urological tract damage. Treatment Agitation should be managed supportively. Diazepam 10–​20 mg IV initially may be used for severe agitation. Patients should be coun- selled about the risks of abuse. Lysergic acid diethylamide Lysergic acid diethylamide (LSD) acts as an antagonist at periph- eral 5-​HT receptor subtypes, but as a 5-​HT2A receptor agonist in the central nervous system. LSD and MDMA (ecstasy) are sometimes combined (‘XL’; ‘candyflipping’) to increase the response to MDMA. Clinical features The ability of LSD to distort reality is well known. Visual hallucin- ations, distortion of images, agitation, excitement, dilated pupils, tachycardia, hypertension, hyperreflexia, tremor, and hyperthermia are common; auditory hallucinations are rare. Time seems to pass very slowly, and behaviour may become disturbed with paranoid delusions. Panic attacks are relatively common, but frank psychotic episodes (which may result in homicide) are not. The psychoactive effects can last for 48 h. Episodic visual disturbances (‘flashbacks’; hallucinogen per- sisting perception disorder) occur in which the effects of LSD are re-​experienced without further exposure to the drug. The symp- toms include false fleeting perceptions in the peripheral fields, flashes of colour, geometric pseudohallucinations, and positive afterimages. These disturbances may persist for several years but are often treatable with benzodiazepines and exacerbated by phenothiazines. Treatment Most patients will require little more than reassurance and sedation. Supportive measures are all that can be offered to those who are seriously ill.

10.4.1  Poisoning by drugs and chemicals 1751 Opiates and opioids Opioids are a large group of drugs, which act on opioid receptors and are usually used as analgesics. Abuse of opiates, particularly heroin, causes many patients to present with unintentional overdose, which is normally from intravenous injection (needle marks visible) but may occur from inhalation (‘chasing the dragon’). Oral ingestion in addicts is less common. Many addicts abuse other drugs in addition to opioids, and the combination of benzodiazepines and opioids are particularly hazardous. Some opioids have other effects not medi- ated through opioid receptors. Methadone has been shown to in- hibit potassium channels at high doses and is also associated with sudden death in susceptible patients due to QT prolongation and torsade de pointes. Buprenorphine, a partial agonist opioid, is now used as an alternative to methadone in replacement programmes. Fentanyl is a very potent opioid available in a range of formula- tions, particularly transdermal. Illicit extraction into an IV prepar- ation has been reported. Tramadol is an opioid with serotonergic metabolites. It causes both convulsions and respiratory depression in overdose. Clinical features Cardinal signs of opiate overdose are pinpoint pupils, reduced re- spiratory rate, and coma. Vomiting may also occur, particularly after intravenous injection in naïve users, and complicates the clin- ical pattern due to aspiration pneumonia. Methadone acts slowly (peak effects usually 4–​6 h after ingestion), though its onset may be more rapid when given intravenously. Non​cardiogenic pulmonary oedema is seen in a proportion of severe opioid overdoses and is treated by positive pressure ventilation. Hypothermia may occur in patients lying outside. Rhabdomyolysis has also been associated with opioid ingestion. Buprenorphine is potentially seriously toxic if given intraven- ously, and in some countries, has been combined with naloxone to reduce the acute hazard. Treatment Naloxone is a pure opioid antagonist. It will reverse the central ef- fects of all opioids if given in sufficient dose. In the event of veins not being accessible, intramuscular use is an alternative, but the onset will be slower. Use of naloxone by nebulizer has also been used in methadone poisoning. Failure of a suspected opioid poisoning to respond to an adequate dose of naloxone (at least 2.4 mg IV in an adult) should prompt reassessment of the diagnosis. It may indicate co-​ingestion of other central nervous system depressants, or inges- tion of γ-​hydroxybutyrate, which also causes small pupils and loss of consciousness. Naloxone has a half-​life of approximately 45–​90 min so its duration of action is, therefore, much shorter than that of the opioids for which the patient is being treated. Naloxone may there- fore be given by infusion; the normal advised dose is approximately two-​thirds of that required to fully wake a patient, every hour. This dose can be reassessed at regular intervals depending on the ex- pected half-​life of the ingested product. Morphine has active metabolites (morphine 6-​glucoronide), which may become relevant in large overdoses. This metabolite is renally excreted and more potent than the parent compound, thus poisoning may be prolonged in older people or in patients with renal impairment or renal damage following rhabdomyolysis. Other supportive care should be administered as necessary, including re- spiratory support. Significant hypotension due to pure opioid effects will usually re- spond to naloxone; patients who are managed just by ventilation may therefore be treated unnecessarily aggressively with fluid replace- ment. In some patients, high concentrations of opioids, such as co- deine, cause histamine release and whealing and itching of the skin, effects that should be treated conventionally with antihistamines. Tryptamines—​synthetic and natural This group of drugs are hallucinogens and include the naturally occurring mushroom hallucinogenic alkaloids psilocin and psilo- cybin, and synthetic compounds such as dimethyltryptamine and α-​methyltryptamine; there are also 4-​ and 5-​ substituted deriv- ates (e.g. 4-​hydroxy-​N,N,-​diethyltryptamine (4-​HO-​DET) and 5-​ methoxy-​N,N-​diisopropyltryptamine (5-​MeO-​DIPT)). Toxicity is related to stimulation of serotonin 2A receptors (5HT2A). Clinical features The features of toxicity are almost universally related to the hallucino- genic potential and psychosis induced by these agents. Additional features including dizziness, weakness, and tremor are common. In severe toxicity seizures, tachycardia, arrhythmias, abdominal symp- toms, and renal injury are seen. Rarely vasospasm may occur. Treatment This is generally supportive, with reassurance the main approach. Agitation should be managed by diazepam (10–​20 mg IV). Severe vasospasm should be managed aggressively by intra-​arterial α- adrenoceptor antagonists or nitrates. Metals Aluminium Aluminium hydroxide is used as an antacid and occasionally as a phosphate binder in the management of chronic renal failure. Aluminium sulphate is employed in water purification and paper manufacture. Aluminium may be absorbed orally and by inhalation. More than 90% of absorbed aluminium is bound to transferrin. Though some accumulates in brain tissue, most body aluminium is stored in bone and the liver. It is excreted mainly via the kidneys so accumulation may occur in the presence of renal failure. Clinical features Acute poisoning Ingestion of a significant quantity of a soluble aluminium salt such as aluminium sulphate causes burning in the mouth and throat, nausea, vomiting, diarrhoea, abdominal pain, hypotension, seizures, haemolysis, haematuria and, rarely, hepatorenal failure. Topical alu- minium sulphate may be irritant to the skin and eyes. By contrast, insoluble aluminium salts, such as aluminium oxide, do not produce an acute toxic response. Chronic poisoning Inhalation of ‘stamped aluminium powder’ can cause a persistent cough and breathlessness due to lung fibrosis or occupational asthma.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1752 Increased death rates from some types of cancer have been observed in aluminium production, but these effects are believed to be the result of exposure to other substances, such as benzopyrene, rather than exposure to aluminium. Aluminium may cause contact allergy. Aluminium encephalopathy is a potential, though now very un- usual, complication in patients with chronic renal failure adminis- tered aluminium-​containing phosphate binders or dialysed using aluminium-​contaminated water. The latter is fortunately now very rare as a result of advancements in dialysis water filtration. The ac- cumulation of aluminium in the brain produces cognitive decline, ataxia, dysarthria, myoclonic jerks, and seizures. Aluminium intoxi- cation may also contribute to renal osteodystrophy and anaemia in patients with chronic renal impairment. Aluminium has been im- plicated in Alzheimer’s disease, but a definitive causative association has not been established. Treatment Deferoxamine forms a stable complex with aluminium which it mo- bilizes primarily from bone with subsequent urinary elimination of the chelate. Deferoxamine is absorbed poorly from the gastrointes- tinal tract and must be administered parenterally. The deferoxamine chelate is dialysable and all published clin- ical studies of aluminium chelation using deferoxamine have involved patients in renal failure undergoing either dialysis or haemofiltration. There is evidence that deferoxamine can improve aluminium-​induced encephalopathy, bone disease, and anaemia in dialysis patients. Specialist advice should be sought. Arsenic Arsenic forms organic and inorganic compounds in trivalent and pentavalent states. Inorganic arsenical compounds may generate arsine gas (see ‘Arsine’, later in this chapter) when in contact with acids, re- ducing metals, sodium hydroxide, and aluminium. Common sources of exposure include fish consumption, traditional medicines, and groundwater contamination in Asian countries. In addition, arsenic trioxide has been used in the treatment of acute promyelocytic leu- kaemia. Most inorganic arsenicals are well absorbed following inges- tion and skin absorption may occur from prolonged exposure. Soluble arsenic compounds can also be absorbed by inhalation. Arsenic crosses the placental and blood–​brain barriers rapidly. Following distribution arsenic accumulates in bone, hair, and nails. The half-​life is generally 2–​10 days and excretion is predominantly in the urine. Clinical features Acute poisoning This can follow accidental, or deliberate ingestion, the toxicity being largely dependent on the water solubility of the ingested compound. Within 2 h of substantial ingestion of a soluble arsenical compound, severe haemorrhagic gastroenteritis may ensue with collapse and death usually within four days. A metallic taste, salivation, mus- cular cramps, facial oedema, difficulty in swallowing, hepatorenal dysfunction, convulsions, and encephalopathy are reported. A per- ipheral neuropathy (predominantly sensory), bone marrow de- pression, striate leukonychia (Mee’s lines), and hyperkeratotic, hyperpigmented skin lesions are common in those surviving a substantial ingestion. In moderate or severe arsenic poisoning, in- vestigations may show anaemia, leucopenia, thrombocytopenia and disseminated intravascular coagulation. ECG abnormalities have been reported and include QT prolongation and ventricular arrhythmias. Chronic poisoning The ingestion of arsenic in contaminated drinking water or ‘tonics’ leads to progressive weakness, anorexia, nausea, vomiting, stoma- titis, colitis, increased salivation, epistaxis, bleeding gums, conjunc- tivitis, weight loss, and low-​grade fever. Characteristically, there is hyperkeratosis of the palms and soles of the feet, ‘raindrop’ pigmen- tation of the skin, and Mee’s lines on the nails. A symmetrical periph- eral neuropathy is typical. Hearing loss, psychological impairment, and EEG changes have been reported. Other chronic effects include disturbances of liver function and ulceration and perforation of the nasal septum. In Taiwan, chronic arsenic exposure has been shown to cause blackfoot disease, a severe form of peripheral vascular dis- ease, which leads to gangrenous changes. Arsenic is classified by the International Agency for Research on Cancer (IARC) as a class 1 (confirmed) carcinogen. Chronic ex- posure to arsenic in drinking water has been causally linked to lung, skin, kidney, and bladder cancer, while occupational exposure to ar- senic is associated with lung cancer. Treatment The traditional chelator dimercaprol has been superseded by succimer (DMSA) and unithiol (DMPS). Both chelators are effective, but unithiol is thought to be superior. The intravenous dose of unithiol is 30  mg/​kg/​day for 5  days. Alternatively, unithiol 77  mg/​kg/​day orally may be administered for 5 days if the IV formulation is un- available. However, nausea and vomiting may limit oral unithiol ad- ministration. Convulsions, cardiovascular effects, and respiratory symptoms should be treated conventionally. Haemodialysis may be required to increase elimination if renal failure develops. Cadmium If hygiene is poor, workers can be exposed to cadmium from the smelting and refining of metals, from soldering or welding metal that contains cadmium, or in plants that make cadmium products such as batteries, coatings, or plastics. Itai-​itai disease (literally ‘ouch-​ouch’ disease, so named because of the effects of severe pain in the joints), occurred in Toyama Prefecture, Japan, in 1950 and was due to mass cadmium poisoning as a result of mining. Clinical features Cadmium compounds are poorly absorbed orally but are well ab- sorbed through the lungs. Cadmium is deposited in the liver and kidneys and very slowly excreted in the urine (half-​life 10–​30 years). Acute poisoning The ingestion of cadmium salts (>3 mg/​kg body weight) may lead to gastrointestinal disturbance which, in severe cases, may progress to cir- culatory collapse, acute renal failure, pulmonary oedema, and death. Inhalation of cadmium oxide fumes produced in welding or cut- ting has led to the development of severe lung damage and death. Often, there are no initial symptoms but after some 4–​10 h, there is increasing respiratory distress. Dyspnoea, cough, and chest pain are accompanied by chills and tremor. Severe pulmonary oedema may develop, or chemical pneumonitis in less severe cases. Recovery may be complicated by progressive pulmonary fibrosis.

10.4.1  Poisoning by drugs and chemicals 1753 Chronic poisoning Repeated exposure to cadmium, such as occupationally, leads to renal tubular dysfunction with glycosuria, aminoaciduria, and hypercalciuria, an increased incidence of renal stones and osteomal- acia. Less common features include anosmia, anaemia, teeth dis- coloration, and neuropsychological impairment. Later, emphysema may develop. Workers repeatedly exposed to high concentrations of cadmium have developed carcinoma of the prostate or lung. Treatment There is no clinical evidence that a substantial body burden of cad- mium may be chelated by any currently available antidote. Chromium Chromium exists mainly in two oxidation states: trivalent (Cr3+) and hexavalent (Cr6+). Cr3+ exposure has limited toxicological relevance, due to its low absorption and inability to cross cell membranes. In contrast, Cr6+ compounds are highly reactive, powerful oxidizing agents that inflict severe local damage. They are also well absorbed through most routes of exposure and cross cell membranes readily. Somewhat paradoxically, Cr6+ induces its devastating systemic toxicity by intracellularly reducing to Cr3+, releasing highly reactive oxygen free radicals in the process. In addition, the Cr3+ formed is able to bind and damage nuclear DNA, inducing genotoxicity. Chromium exposure and absorption through inhalation, ingestion, and dermal routes is primarily oc- cupational, with excretion occurring through the kidney. Clinical features Acute poisoning Soluble Cr6+ compounds include sodium and potassium chromate and dichromate and chromic acid (Cr6+ trioxide). Inhalation of these highly irritant compounds causes mucous membrane inflammation, cough, headache, chest pain, and dyspnoea; pulmonary oedema and respiratory failure may ensue. Ingestion of highly water-​soluble Cr6+ compounds causes a burning sensation in the mouth and throat, nausea, abdominal pain, diarrhoea, and a risk of gastrointestinal haem- orrhage. Hypovolaemic shock may follow. Methaemoglobinaemia, haemolysis, coagulopathy, and renal and hepatic failure have been re- ported. Chromic acid splashes produce severe burns. Percutaneous absorption may lead to systemic toxicity; fatalities have occurred. Chronic poisoning Inhalation of Cr6+ compounds has led to atrophy, ulceration, and perforation of the nasal septum. Pharyngeal and laryngeal ulcers may also occur. Asthma may be precipitated by exposure to fumes. Lung fibrosis, bronchitis, emphysema, and renal proximal tubular damage result from occupational exposure. Cr6+ is classified by the IARC as a group I carcinogen and chronic occupational exposure is strongly associated with an increased incidence of lung cancer. ‘Chrome ulcers’ may develop after repeated topical exposure to Cr6+ compounds. Cr6+ compounds are also skin sensitizers and con- tribute to the development of cement dermatitis and contact derma- titis from paint primer, tanned leather, tattoo pigments, and matches. Treatment The principal management of chromium poisoning is avoidance of exposure. Inhalational exposure should be treated conventionally. Despite claims that topically applied ascorbic acid and sodium cal- cium edetate protect against dermal toxicity, there is insufficient evidence to advocate their use. Immediate surgical assessment is recommended in cases of severe ingestion, as resection of necrotic gastrointestinal tissue may be life-​saving. There is no evidence that any chelating agent improves outcome in cases of systemic poi- soning. Haemodialysis removes chromium from the blood, but the high tissue uptake limits the value of this treatment when used alone. Cobalt Cobalt is an essential trace element and is a constituent of vitamin B12 (cyanocobalamin). Cobalt salts have been used as blue colour- ants for thousands of years. Cobalt composited with tungsten car- bide (‘hard metal’) is a very durable and temperature-​resistant metal used in the manufacture of drills and other tools. Historically, cobalt salts were used in brewing to enhance the ‘head’ on beer and in the treatment of anaemia. In recent years, poorly functioning cobalt-​ containing hip prostheses have become an important further source of exposure. Cobalt exerts its toxicity through generating reactive oxygen species, inflicting DNA damage and disrupting ionic, enzymatic, and haemato- poietic homeostasis. Cobalt can be absorbed orally and by inhalation and most undergoes renal excretion over 7 days, but a small proportion is retained with a biological half-​life of approximately 2 years. Clinical features Acute poisoning Acute poisoning is rare, though ingestion causes gastrointestinal irritation. Chronic poisoning Hard metal lung disease is a now rare form of interstitial lung dis- ease that occurs in susceptible patients exposed to hard metal and in some diamond workers who use cobalt-​containing polishes. Patients usually present with exertional dyspnoea and cough. There may be associated constitutional symptoms of fever, weight loss, or malaise. Inspiratory crackles are the earliest physical sign, but finger club- bing, cyanosis, and eventually cor pulmonale can ensue. Interstitial fibrosis is seen on chest X-​ray (primarily the lower zones), and a restrictive ventilatory defect is often present. Cobalt is also a recog- nized cause of occupational asthma. Historically, those consuming large amounts of cobalt-​ contaminated beer developed ‘beer-​drinkers’ cardiomyopathy’ with heart failure often accompanied by a pericardial effusion and poly- cythaemia. Systemic cobalt toxicity also developed in patients re- ceiving cobalt chloride as treatment for anaemia (cobalt stimulates erythropoietin release), with manifestations including hypothyroid goitre (cobalt inhibits the uptake of iodine by the thyroid gland), deafness, visual disturbances, and/​or peripheral neuropathy. More recently, systemic cobalt toxicity has been encountered occasionally in recipients of cobalt-​containing hip prostheses. This is far more likely in those with an ill-​fitting prosthesis where there is increased friction between metal surfaces, and in those who have a metal-​ containing hip as a revision of a damaged ceramic prosthesis (re- sidual ceramic shards abrading the metal surface). Cobalt is classified as a group 2B carcinogen by the IARC, with limited evidence to suggest a causal relationship between cobalt and cancer.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1754 Treatment Hard metal pneumoconiosis may respond to steroid therapy. Systemic features are usually at least partly reversible providing the cobalt source is removed. There is no antidote for systemic poisoning. Copper Copper is used for pipes and roofing materials, in alloys and as a pig- ment. It is a component of several endogenous enzymes, including tyrosinase and cytochrome oxidase, and is essential for the utiliza- tion of iron. Copper sulphate is used as a fungicide, an algicide, and in some fertilizers. Following ingestion, copper transport across the intestinal mucosa is facilitated by cytosolic metallothionein. In blood, copper is initially albumin-​bound and transported via the hepatic portal circulation to the liver where it is incorporated into caeruloplasmin. Ninety-​ eight (98%) per cent of copper in the systemic circulation is bound to caeruloplasmin and free copper is excreted via a lysosome-​to-​bile pathway. This process is essential to normal copper homeostasis and provides a protective mechanism in acute copper poisoning. An impaired or overloaded biliary copper excretion system results in hepatic copper accumulation, as occurs in Wilson’s disease (see Chapter 12.7.2) and copper poisoning. Free reduced Cu(I) can bind to sulfhydryl groups and inactivates enzymes such as glucose-​6-​phosphate dehydrogenase and gluta- thione reductase. In addition, copper may interact with oxygen spe- cies (e.g. superoxide anions and hydrogen peroxide) and catalyse the production of reactive toxic hydroxyl radicals. Copper(II) ions can oxidize haem iron to form methaemoglobin. Clinical features Acute poisoning Acute copper poisoning usually results from the ingestion of contam- inated foods or from accidental or deliberate ingestion of copper salts. Copper salt ingestion causes profuse vomiting with abdominal pain, diarrhoea, headache, dizziness, and a metallic taste. Gastrointestinal haemorrhage, haemolysis, and hepatorenal failure may ensue and fa- talities have occurred. Body secretions may be green or blue. Occupational exposure to copper fumes (during refining or welding) or to copper-​containing dust causes ‘metal fume fever’ with upper respiratory tract symptoms, headache, fever, and myalgia. Chronic poisoning Chronic copper poisoning has been reported predominantly as ‘vineyard sprayer’s lung’ in those spraying fungicides containing copper sulphate. Features include progressive dyspnoea, cough, wheeze, myalgia, malaise, anorexia, micronodular, and reticular opacities on chest X-​ray (which may coalesce), and a restrictive lung function defect. Lung biopsy may show pulmonary granulomata and fibrosis. Other features include hepatic copper-​containing granulomas, hypergammaglobulinaemia, and hepatomegaly. There is no convincing evidence that copper is carcinogenic in humans. Treatment Blood copper concentrations correlate well with the severity of in- toxication following acute ingestion. Serum caeruloplasmin con- centrations will also be increased in acute copper salt poisoning. Supportive measures are paramount following copper salt inges- tion. Early endoscopy or CT scan with contrast is recommended if corrosive damage is suspected. An early surgical opinion should be sought if there are clinical signs of an acute abdomen or deep ulcers and/​or areas of necrosis on endoscopy or CT. Methaemoglobinaemia should be treated with intravenous methylthioninium chloride 1–​2 mg/​kg. Oral D-​penicillamine, 1.5–​2 g daily, enhances urinary copper elimination in patients with Wilson’s disease but confirmed benefit in acute copper salt poisoning has not been demonstrated. Experimental studies suggest unithiol (DMPS) may be the most ef- fective antidote in copper poisoning but the presence of acute kidney injury in severely poisoned patients often limits the value of anti- dotes which enhance urine copper excretion. Extracorporeal elim- ination techniques do not enhance copper elimination significantly. Exchange transfusion has been undertaken successfully in patients with copper sulphate-​induced haemolysis. Lead Occupational lead exposure occurs mainly by inhalation, for ex- ample, in the reclamation of lead from scrap metal, in the demoli- tion and flame-​cutting of structures painted with lead-​containing paint, in the manufacture of storage batteries, ceramics, and pig- ments and in radiation shielding. Non​occupational lead exposure predominantly involves ingestion and important sources include ‘traditional’ remedies (particularly among ethnic minorities) and children with pica who pick at surfaces coated with lead-​containing paint or eat lead-​contaminated soil. Application of lead-​containing cosmetics such as ‘surma’ has also resulted in lead intoxication. Both ingested and inhaled lead are absorbed readily. Most (98.5%) lead is deposited in the bones and teeth, where it remains for 10–​ 15 years. Of the lead in the blood, 99% is associated with erythro- cytes. Lead can cross both the blood–​brain and placental barriers. Elimination is predominantly renal. As the body accumulates lead over many years, even small doses can accumulate over time and cause toxicity. There are two principal mechanisms of lead toxicity. First, lead complexes with important functional chemical groups including –​ COOH, –​NH2, and –​SH, and so disrupts the function of enzymes and other biologically important molecules. Lead inhibits several enzymes involved in haem synthesis (including δ-​aminolaevulinate dehydratase and ferrochelatase) and erythrocyte maturation (erythrocyte pyrimidine 5’ nucleotidase), causing a microcytic or normocytic hypochromic anaemia. Secondly, lead substitutes for di- valent ions, particularly calcium, which explains why lead accumu- lates in bone. The critical role of calcium in neuronal differentiation, myelination, and synapse development and functionality explains why lead poisoning can induce devastating neurotoxicity. The developing nervous system is particularly susceptible to irrevers- ible damage. Lead substitution for calcium also causes widespread chemical interactions, disrupts second messenger cellular signalling and triggers calcium-​activated apoptosis. Clinical features Lead poisoning frequently presents with non​specific features, including abdominal pain, anorexia, constipation, headache, and lethargy. Anaemia presents due to impaired haem synthesis and re- duced erythrocyte lifespan. Classically, lead poisoning presents with peripheral neuropathy in the form of foot or wrist drop, although this manifestation is now uncommon. In moderate intoxication, reversible renal tubular dysfunction occurs (causing glycosuria, aminoaciduria, and phosphaturia), which progresses to irreversible

10.4.1  Poisoning by drugs and chemicals 1755 interstitial fibrosis and progressive renal insufficiency in severe cases. Hypertension may result from renal toxicity. Lead encephal- opathy (delirium, seizures, and coma) only occurs in very severe poi- soning (blood lead concentrations >100 μg/​dl (4.8 micromol/​litre)) and is much more common in children than adults. Transplacental transfer of lead from mother to fetus results in reduced fetal viability, low birth weight, and premature birth. Despite unequivocal evidence that even low blood lead concen- trations are detrimental to health, the current practice in the United Kingdom is to only enforce stopping work with lead when a worker’s blood lead concentration exceeds 60 µg/​dl (2.9  micromol/​litre); 30 µg/​dl (1.4 micromol/​litre) for a woman of reproductive capacity; 50 µg/​dl (2.4 micromol/​litre) for an employee aged under 18 years. There is no safe blood lead concentration for children, particularly those below the age of 5 years. Treatment Primary prevention aimed at eliminating lead hazards for chil- dren and workers is crucial. The importance of primary preven- tion in children is emphasized particularly by the observation that chelation does not improve scores on tests of cognition, behav- iour, and neuropsychological function in children with moderate lead poisoning (blood lead concentrations of 22–​45 µg/​dl (1.0–​2.2 micromol/​litre)). The social dimension of the problem must also be recognized: simply giving children chelation therapy and then re- turning them to a contaminated home environment is of no value. Similarly, if an occupational source of lead exposure is implicated, a thorough evaluation of the workplace, other exposed workers and the systems for handling lead at work are appropriate. The decision to use chelation therapy is based on the symptoms present and the blood lead concentration. All symptomatic patients with blood lead concentrations of 50 µg/​dl (2.4 micromol/​litre) or higher should be considered for chelation therapy. Parenteral so- dium calcium edetate, 75 mg/​kg per day, has been the chelating agent of choice for more than 50 years but oral succimer (DMSA) 30 mg/​kg per day is of similar efficacy. Mercury Mercury is the only metal that is liquid at room temperature. It exists in three forms: metallic (Hg0), mercury(I) (mercurous), and mercury(II) (mercuric). Metallic mercury is very volatile and when spilt has a large surface area so that high atmospheric concentrations may be pro- duced in enclosed spaces, particularly when environmental temperat- ures are high. In addition to simple salts, such as chloride, nitrate, and sulphate, mercury(II) forms organometallic compounds where mer- cury is covalently bound to carbon, such as methyl-​, ethyl-​, phenyl-​, and methoxyethyl mercury. Inorganic mercury is used extensively in industrial and pharmaceutical settings; exposure is predominantly occupational though minor non​occupational exposure occurs via dental amalgam. By contrast, exposure to organomercury compounds most commonly occurs from dietary intake, as organomercury can accumulate up the food chain of aquatic species. The absorption of mercury depends on its chemical form. Inhaled mercury vapour is absorbed rapidly and oxidized to mercury (II) in erythrocytes and other tissues. Prior to oxidation, absorbed mercury vapour can cross the blood–​brain barrier, but the divalent ion oxida- tion product serves to trap mercury in the brain. Mercury vapour is also absorbed via the skin. Less than 1% of an ingested dose of metallic mercury reaches the systemic circulation. Organic mercuric salts are better absorbed following ingestion than are inorganic mercuric salts. Organic mercury compounds cross the blood–​brain barrier readily. In contrast, the kidney is the main storage organ for inorganic mercury compounds. In vivo mercury is bound to metallothionein, which serves a protective role, since renal damage is caused only by the unbound metal. Mercury is excreted mainly in urine and faeces although a small amount of absorbed inorganic mercury is exhaled as mercury vapour. The half-​life of most body mercury is 1–​2 months, but a small fraction has a half-​life of several years. The exact mechanism of toxicity of mercury remains unclear, but involves binding of the Hg2+ form to the sulfhydryl groups present on structural proteins, receptors, enzymes, intracellular organelles and DNA and to selenoproteins. The central nervous system is par- ticularly susceptible. Clinical features Acute poisoning Acute mercury vapour inhalation causes headache, nausea, cough, chest pain and bronchitis/​pneumonitis. Repeated exposure to low mercury vapour concentrations presents typically with characteristic neurological features including fine tremor, lethargy, memory loss, insomnia, personality changes, and ataxia. Other features include stomatitis, gingivitis, hypersalivation, and renal tubular damage. Mixed motor and sensory peripheral neuropathy may develop. Ingestion of metallic mercury is usually without systemic effects as it is poorly absorbed from the gastrointestinal tract. However, inges- tion of inorganic mercury (II) (mercuric) or aromatic mercuric salts causes an irritant gastroenteritis with corrosive ulceration, which may lead to circulatory collapse and shock. Inorganic mercury(I) (mer- curous) compounds are less soluble, less corrosive, and less toxic than mercuric salts. Ingestion of mercurous chloride in teething powder has led to ‘pink disease’ or acrodynia in infants. This is a hypersen- sitivity reaction characterized by a desquamating erythematous rash of the extremities, irritability, profuse sweating, tachycardia, and hypertension. Systemic toxicity in the form of renal and neurological damage can present following exposure to mercury salts. There are reports of deliberate intravenous or subcutaneous me- tallic mercury injection. Accidental injection also has occurred after injury from broken thermometers. Intravascular mercury may result in pulmonary venous or peripheral arterial embolism. Subcutaneous mercury initiates a soft-​tissue inflammatory reaction with granuloma formation. Signs of systemic mercury toxicity are rare following metallic mercury injection. Chronic poisoning Chronic poisoning from inorganic mercury compounds or mercury vapour causes anorexia, insomnia, abnormal sweating, headache, lassitude, increased excitability, tremor, peripheral neuropathy, gin- givitis, hypersalivation, personality changes, and memory or intel- lectual deterioration. Glomerular and tubular damage may occur, and renal tubular acidosis has been described in children. Most cases of human poisoning from alkyl mercury compounds result from ingestion of contaminated foods over a long period. There is often a latent period of several weeks between exposure and the development of symptoms which are predominantly neurological, with paraesthesiae of the lips, hands, and feet, ataxia, tremor, dysarthria, constriction of visual fields, and emotional and intellectual changes. Gastrointestinal disturbances may precede or

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1756 accompany these features. Seizures, coma, and death have occurred. Chronic exposure to methylmercury has been associated with an in- creased incidence of liver cancer, cirrhosis, renal disease, and cere- bral haemorrhage. Treatment Although there are no controlled clinical data to show that chela- tion therapy improves outcome in patients with neurological fea- tures of mercury poisoning, unithiol 30 mg/​kg/​day intravenously (or 77  mg/​kg/​day orally) increases urinary mercury elimination and reduces blood mercury concentrations. Where extracorporeal renal support is required for the treatment of renal failure, there is evidence that continuous veno–​venous haemodialfiltration is more effective than haemodialysis at removing unithiol–​mercury com- plexes. Substantial exposure to corrosive mercury salts may warrant immediate surgical assessment, as resection of necrotic gastrointes- tinal tissue may be life-​saving. Nickel Nickel is a ubiquitous trace metal mined in the form of sulphide ore. It is used primarily for producing stainless steel and other alloys. Nickel forms inorganic soluble (sulphate, chloride) and insoluble (oxide, sulphide) salts, used in electroplating and battery manufac- ture. Nickel carbonyl (Ni(CO)4) is a colourless, volatile liquid used as a catalyst in the petroleum, plastic, and rubber industries. It is an intermediate compound in nickel purification and is released as fumes when nickel is thermally decomposed. Nickel metal and inorganic salts can be absorbed orally and by inhalation, though absorption is generally poor. By contrast, nickel carbonyl is highly lipophilic and rapidly absorbed. Nickel is principally bound to al- bumin in the blood and is concentrated in the kidneys, liver, and lungs prior to renal excretion. The mechanism of nickel toxicity is thought to involve the induction of oxidative stress through reactive oxygen species production. Clinical features Acute poisoning Nickel carbonyl inhalation leads within a few minutes to dizziness, headache, vertigo, nausea, vomiting, cough, and dyspnoea. In many cases these symptoms disappear and there follows a symptom-​free period lasting 12–​36 h before tachypnoea, dyspnoea, haemoptysis, cyanosis, chest pain, vomiting, tachycardia, weakness, and muscle fatigue supervene. Paraesthesiae, diarrhoea, abdominal distension, delirium, and convulsions have also been reported. Death from car- diorespiratory failure may occur 4 to 11 days after exposure. At high concentrations, soluble nickel salts are primary skin, gut, and eye irritants. Workers at an electroplating plant who drank water accidentally contaminated with nickel sulphate experienced nausea, vomiting, diarrhoea, abdominal pain, headache, cough, and breathlessness, which persisted for up to 2 days. A 2-​year-​old child died 4 h after ingesting 15 g nickel sulphate crystals. Chronic poisoning Chronic exposure to aerosols of nickel salts may lead to chronic rhin- itis and sinusitis and, in rare cases, anosmia and perforation of the nasal septum. Inhaled nickel can produce a type I hypersensitivity re- action, manifest as bronchial asthma with circulating IgE antibodies to nickel. Pulmonary eosinophilia (Loeffler’s syndrome) due to a type III hypersensitivity reaction to nickel has also been described. A significant increase in deaths from non​malignant respiratory disease or pneumoconiosis has been observed in nickel refinery workers. Nickel compounds are classified by the IARC as class I car- cinogens, with evidence that occupational exposure increases the risk of cancer of the lung and nasal sinuses. Metallic nickel and nickel salts cause allergic contact dermatitis in up to 10% of females and 1% of males and is due to a type IV delayed hypersensitivity. Treatment Blood nickel concentrations immediately following exposure to nickel carbonyl provide a guide to severity of exposure and the need for chelation therapy. Unithiol (DMPS) enhances the urinary excre- tion of nickel in nickel-​intoxicated animals. Diethyldithiocarbamate and disulfiram (which is metabolized to diethyldithiocarbamate) are effective agents in the treatment of nickel dermatitis, but their role in the treatment of acute severe nickel carbonyl poisoning has not been confirmed in a controlled clinical study. Phosphorus Elemental phosphorus exists in several crystalline forms (allo- tropes), of which yellow phosphorus (sometimes referred to as white) is the most important toxicologically. Phosphorus oxidizes spontaneously in contact with air to form phosphorus pentoxide which, by an exothermic reaction, forms phosphoric acid on contact with water. Hence, dermal and gastrointestinal exposures to phos- phorus rapidly become exposures to phosphoric acid. Clinical features Typically, patients who have ingested phosphorus present with ei- ther gastrointestinal features (most commonly) or central nervous system features; 20% have a combination of both. Features gener- ally begin within minutes of ingestion and include nausea, vomiting, abdominal pain, burns of the pharynx, oesophagus, and stomach, which may lead to gastrointestinal haemorrhage. Shock in part due to fluid loss and GI haemorrhage follows. In other cases, central nervous system features (restlessness, irritability, delirium, coma, convulsions, and cerebral oedema) predominate. Metabolic compli- cations (metabolic acidosis, hypoglycaemia, hyperphosphataemia and hypocalcaemia) and hepatorenal failure ensue. Cardiovascular collapse and arrhythmias are the most common cause of death following ingestion, but in other cases cerebral oe- dema and haemorrhage complicating fulminant hepatic failure are responsible. Treatment Treatment is supportive. Hypotension/​shock should be corrected vigorously with intravenous fluid and inotropes. If metabolic acid- osis is not responsive to fluid resuscitation, give intravenous sodium bicarbonate. Early fibreoptic endoscopy and CT is indicated to grade the severity of the injury in any patient who is symptomatic or has evidence of oropharyngeal burns. Thallium Thallium sulphate was previously used as a rodenticide but is now banned for this use in many countries. Thallium salts have also been employed in the manufacture of optical and electrical equipment, as catalysts in organic synthesis, and in isotopic form for medical imaging of the myocardium.

10.4.1  Poisoning by drugs and chemicals 1757 Clinical features Initial symptoms (if ingested) include nausea, vomiting, abdominal pain, and, less commonly, gastrointestinal bleeding. Constipation follows in most patients. After a few days (usually between two and five), paraesthesiae develop, which start in the feet and progress to the hands and fingers; painful and tender extremities (‘burning feet syndrome’) and ascending sensory neuropathy then supervene. In severe cases confusion, delirium, convulsions, renal failure, respira- tory failure, heart failure, and coma occur; the mortality is high. If death does not occur within the first week, tremor, ataxia, and (usually lower limb) muscle weakness develops, due to the onset of motor neuropathy, which is usually distal. Ocular features include nystagmus, ptosis, and abnormalities of gaze due to involvement of the third, fourth, and sixth cranial nerves. Retrobulbar neuritis, fa- cial paralysis, decreased visual acuity, optic atrophy, and defective colour vision may develop. Characteristically, alopecia develops within 1–​3 weeks and it is often this sign which leads to the diagnosis being made. If the patient survives, the hair usually regrows, but is often abnormally fine and unpigmented. Nail growth is impaired with the develop- ment of ridges, Mees’ lines, and erosion of the proximal parts of the nails. Treatment As thallium ions are excreted into the gastrointestinal tract via the saliva, the bile, and through the intestinal mucosa, it is possible to sequester thallium ions in the gut and prevent reabsorption by the oral administration of colloidally soluble Prussian blue (potassium ferric hexacyanoferrate (II)) 250–​300  mg/​kg/​day (approximately 10 g twice daily for an adult). Thallium ions are exchanged for po- tassium ions in the lattice of the Prussian blue molecule and are sub- sequently excreted in faeces. During treatment with Prussian blue, plasma concentrations of thallium fall and urine excretion declines exponentially. In contrast, faecal excretion of thallium is detectable even when urine excretion of the metal has ceased and, therefore, administration of Prussian blue should be continued until thallium can no longer be detected in the faeces. Zinc Zinc oxide fumes are emitted in any process involving molten zinc and are the most common cause of metal fume fever. Exposure to zinc chloride occurs in soldering; in the manufacture of dyes, paper, and deodorants; and on military exercises when it is used as a smoke screen. Poisoning has followed the accidental or deliberate inges- tion of elemental zinc and zinc chloride and fatal intoxication has followed inadvertent intravenous administration. Inhalation of zinc chloride and oxide may lead to nasopharyngeal and respiratory tox- icity. Zinc may be absorbed through broken skin when zinc oxide paste is used to treat wounds and burns. Clinical features Zinc sulphate ingestion causes gastrointestinal irritation, sometimes in association with headache and dizziness. Zinc chloride is highly corrosive, and ingestion has led to erosive pharyngitis, oesophagitis, and haematemesis. Acute renal failure and pancreatitis have also been recorded after ingestion of zinc salts. Topical exposure to zinc chloride causes ulceration and dermatitis of the exposed skin. Zinc chloride is highly irritant to the eye. In contrast to the relatively mild clinical course after zinc oxide in- halation, exposure to zinc chloride ammunition bombs (hexite) pro- duces a chemical pneumonitis with marked dyspnoea, a productive cough, fever, chest pain, and cyanosis. The acute respiratory distress syndrome may ensue. Metal fume fever occurs most commonly in individuals who per- form welding involving zinc. It presents generally with influenza-​like symptoms, fever, shaking chills, arthralgias, myalgias, headache, and malaise, some 4–​10 h following exposure. In patients with ongoing metal fume exposure over the course of a workweek, tachyphylaxis occurs resulting in improvement in symptoms over the course of the workweek and maximal symptoms occurring after an exposure-​free period such as a weekend. Treatment Management is supportive. Endoscopy and CT should be performed following zinc chloride ingestion to assess the severity of oesopha- geal or gastric burns. Pesticides Aluminium and zinc phosphides Aluminium and zinc phosphides are highly effective insecticides and rodenticides, which are used to protect grain during transport and storage. The phosphide interacts with moisture in the surrounding air to liberate phosphine, which is the active pesticide. Acute poi- soning, therefore, results either from the ingestion of the salts them- selves or inhalation of the phosphine generated during their use; the latter is discussed later on in this chapter. In cases of poisoning by phosphide ingestion, toxic effects are due to phosphine release when the phosphide comes into contact with gut fluids. Phosphine is absorbed through the alimentary mucosa and widely distributed to tissues. Clinical features Early features include nausea, vomiting, retrosternal, and epigas- tric pain, gastric, or duodenal erosions causing haematemesis and dyspnoea. Diarrhoea is less common. Shock and circulatory failure occurring early in the course of poisoning are of ominous prognostic importance, as circulatory failure is a common and frequent cause of death. Impaired myocardial contractility and global dyskinesia are frequent in those severely poisoned. This group of patients is characterized by severe hypotension, reduced cardiac output, raised systemic venous pressure, normal pulmonary artery wedge pressure and inadequate systemic vasoconstriction. Severe metabolic acid- osis, renal failure, and disseminated intravascular coagulation are common accompaniments. Treatment Treatment is symptomatic and supportive. Gastric lavage should be avoided as it might increase the rate of disintegration of the product ingested and increase toxicity. Activated charcoal does not bind metal phosphides. Anticoagulant rodenticides Warfarin was widely used as a rodenticide until target species de- veloped resistance to it. The newer anticoagulant rodenticides

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1758 (sometimes termed ‘super warfarins’), such as brodifacoum, bromodialone, chlorophacinone, coumatetralyl, difenacoum, diphacinone and flocoumafen, are more potent and longer-​acting antagonists of vitamin K1 than warfarin. While accidental ingestion of small amounts rarely results in altered coagulation, deliberate in- gestion may result in prolongation of the INR for several weeks or months and fatal haemorrhage has occurred. These anticoagulants inhibit vitamin K1-​2,3-​epoxide reductase and thus the synthesis of vitamin K and subsequently clotting fac- tors II, VII, IX, and X. There is no anticoagulant effect until existing stores of vitamin K and clotting factors are depleted. The greater po- tency and duration of action of long-​acting anticoagulant rodenti- cides compared to warfarin is attributed to their greater affinity for vitamin K1-​2,3-​epoxide reductase, their ability to disrupt the vitamin K1-​epoxide cycle at more than one point, hepatic accumulation, and unusually long biological half-​lives due to high lipid solubility and enterohepatic circulation. Clinical features Gastrointestinal bleeding, haematuria, and bruising are the most common features, though the most common site of fatal haemor- rhage is intracranial. The onset of bleeding may be delayed for sev- eral days since the peak anticoagulant effect does not occur until some 72–​96 h after ingestion. Treatment Routine measurement of the INR is generally not indicated in chil- dren as the amounts they ingest are almost invariably small. In all other cases, the INR should be measured on presentation and 36 to 48 h after exposure. If the INR is normal at this time, no further action is required. If a patient presents within 1 h of a large ingestion, the adminis- tration of activated charcoal (50 g for adults; 10–​15 g for children) should be considered, as it is known that warfarin is adsorbed to charcoal. In patients with severe poisoning who have ingested a long-​acting formulation, oral cholestyramine 4 g three times daily for an adult should be considered in order to shorten the plasma half-​life of the rodenticide. If active bleeding occurs, dried prothrombin complex (which contains factors II, VII, IX, and X) 25–​50 units/​kg, or fresh frozen plasma 15 ml/​kg (if no concentrate is available) should be given, to- gether with phytomenadione 5 mg by slow intravenous injection (100 µg/​kg body weight for a child). If active bleeding occurs in a pa- tient who is being prescribed an anticoagulant, warfarin (or another anticoagulant) should be discontinued. If there is no active bleeding and the INR is less than 4.0, treat- ment with phytomenadione is not required. If the INR is more than 4, phytomenadione 5 mg by slow intravenous injection (100 µg/​kg body weight for a child) should be administered, unless the patient is anticoagulated for therapeutic reasons. If the patient is prescribed anticoagulants, the INR is more than 8, and there is no active bleeding or only minor bleeding, stop warfarin (restart when the INR <5), give phytomenadione 0.5 mg by slow intravenous injection and repeat the dose if the INR is more than 8 12–24 h later. If the INR is between 6.0 and 8.0, and there is no active bleeding or only minor bleeding, warfarin should be discontinued and restarted when the INR is less than 5. Patients who require reversal of coagulopathy after exposure to long-​acting formulations should have their INR measured for two weeks after cessation of treatment. Failure to do so may result in recurrence of anticoagulation and risk haemorrhage being missed. Bipyridyl herbicides The bipyridyl herbicides include diquat, morfamquat, and para- quat. Paraquat has been removed from the market in the EU; diquat remains widely available and morfamquat is not readily available. Clinical features The features of toxicity are largely dependent on the amount of para- quat swallowed. Ingestion of more than 6 g paraquat ion is likely to be fatal within 24–​48 h, while 3–​6 g is likely to lead to a more pro- tracted, but still fatal, outcome. After the ingestion of more than 6 g paraquat ion, nausea, vomiting, abdominal pain, and diarrhoea, are rapidly followed by peripheral circulatory failure, metabolic acidosis, impaired consciousness, convulsions, and increasing breathlessness secondary to acute pneumonitis. Breathlessness, tachypnoea, wide- spread crepitations, and central cyanosis progress relentlessly until the patient dies from hypoxia a few days later. Mild jaundice may be seen and renal failure is usually severe. After 3–​6 g paraquat ion, the cardiovascular and central nervous system complications are not seen, and the course of poisoning is dominated by alimentary features, particularly painful ulceration of the mouth, tongue, and throat, which makes it difficult to swallow, speak, and cough. Perforation of the oesophagus with subsequent mediastinitis has been reported. Ingestion of 1.5–​2.0 g of paraquat causes nausea, vomiting, and diarrhoea, mild renal tubular necrosis, and pain in the throat. Respiratory involvement may not be apparent until 10–​21 days after ingestion, but may progress till the patient dies of respiratory failure as late as 5 or 6 weeks after taking the paraquat. The features of diquat poisoning are similar. In severe and usu- ally fatal cases, gastrointestinal mucosal ulceration, paralytic ileus, hypovolaemic shock, acute renal failure, and coma predominate. Treatment The diagnosis of paraquat poisoning can be confirmed by a simple qualitative test on urine passed within 4 h of ingestion using alka- line sodium dithionite (a blue colour indicates paraquat is present); a negative test indicates that not enough paraquat has been taken to cause problems. In the case of diquat poisoning, the urine goes a green colour in the alkaline sodium dithionite test. The outcome of paraquat poisoning can be predicted with reasonable confidence within a few hours of ingestion by relating the plasma paraquat concentration to the time after ingestion, but this assay is not readily available in Europe following the withdrawal of paraquat from the market. There is no evidence that the outcome of paraquat or diquat poi- soning can be altered by any form of intervention. Symptomatic measures including antiemetics, mouth washes and analgesics are indicated, and intravenous fluids may be necessary to replace gastro- intestinal losses. Carbamate insecticides Carbamate insecticides inhibit acetylcholinesterase, causing ac- cumulation of acetylcholine at central and peripheral cholinergic nerve endings, including neuromuscular junctions. The duration of this effect is comparatively short-​lived (compared to organophos- phorus insecticides) as the carbamate–​enzyme complex tends to dissociate spontaneously.

10.4.1  Poisoning by drugs and chemicals 1759 Clinical features After substantial carbamate ingestion, patients usually develop cho- linergic symptoms within a few minutes, and in most severe cases, constriction of the pupils, muscle twitching, profound weakness, profuse sweating, excessive salivation, bronchorrhoea, chest tight- ness, coughing, incontinence, confusion, and progressive cardiac and respiratory failure ensue. Seizures are relatively uncommon as a primary complication, because carbamate penetration into the cen- tral nervous system is limited, though they may occur secondary to hypoxia. Death is usually due to respiratory failure. In less substan- tial ingestions, cholinergic symptoms are evident within 2 h in most cases and typically resolve within 24 h. Treatment Bronchorrhoea requires removal of secretions by suction and prompt relief with intravenous atropine 2 mg (0.02–​0.1 mg/​kg in a child) intravenously together with supplemental oxygen to maintain arterial PaO2 greater than 10 kPa (>75 mm Hg). The atropine dose should be titrated to control rhinorrhoea and bronchorrhoea. If the initial dose produces only a partial response, it should be doubled and doubled again if there is only a limited clinical response. If these measures fail, the patient should be intubated and mechanical ven- tilation instituted. At present there is insufficient evidence to either recommend or ad- vise against the use of pralidoxime in severe poisoning with carbamate insecticides. Pralidoxime seldom should need to be administered in less severe cases since carbamates have a relative short duration of action. However, if intoxication is life-​threatening and unresponsive to atro- pine and supportive measures, pralidoxime chloride 30 mg/​kg body weight by intravenous injection over 20 minutes should be given. Chloralose Chloralose is marketed for amateur use as cereal or paste baits con- taining 2 to 4% rodenticide. Technical α-​chloralose (c. 90% pure) is used by professionals against bird pests and rodents. Clinical features Toxic amounts of chloralose cause severe central nervous system ex- citation with hypersalivation, increased muscle tone, hyperreflexia, opisthotonus, myoclonic jerks, and convulsions. Rhabdomyolysis is a potential complication. Coma, generalized flaccidity, and respira- tory depression may follow. Treatment Children who ingest small amounts of baits (amateur formulations) containing chloralose are unlikely to develop symptoms. In contrast, patients who have deliberately ingested large amounts of bait or the technical compound are likely to require admission to intensive care for management of convulsions, myoclonus, and coma. Chlorates Sodium and potassium chlorates are non​selective herbicides. Potassium chlorate is also used in matchstick heads, explosives, and fireworks. Sodium chlorate and potassium chlorate are powerful oxi- dizing agents that induce methaemoglobinaemia and haemolysis. Clinical features Features of chlorate toxicity may develop within as little as two hours of ingestion and are usually due to gastrointestinal irritation. Early features include nausea, vomiting, diarrhoea, abdominal pain, and cyanosis secondary to methaemoglobinaemia. The combination of methaemoglobinaemia and haemolysis results in varying degrees of hypoxaemia and symptoms such as general weakness, fatigue, diz- ziness, agitation, anxiety, confusion, and headache. Chest pain and dyspnoea may also be experienced. Intravascular haemolysis causes hyperkalaemia and jaundice. Poisoning with chlorate is also com- monly complicated by acute renal failure, though the underlying mechanisms are not fully understood. Treatment Methaemoglobinaemia can be corrected by slow intravenous injec- tion of methylthioninium chloride (methylene blue) 1–2 mg/​kg body weight as a 1% solution, although this treatment is less effective in the presence of major intravascular haemolysis. Blood transfusion may be required. Plasma potassium concentrations should be moni- tored and reduced if necessary. Haemodialysis/​haemodialfiltration may remove chlorate and will also be required for the management of renal failure and hyperkalaemia. Plasmapheresis and plasma ex- change or exchange transfusion have also been employed to remove chlorate, circulating free haemoglobin, and red cell stroma, but data are too limited to make a firm recommendation. Chlorophenoxy herbicides Chlorophenoxy (phenoxyacetate) herbicides are weed killers that act as synthetic auxins (plant ‘hormones’) and cause plant death by disrupting nutrient transport and growth. They comprise an aliphatic carboxylic acid moiety attached to a chlorine-​ or methyl-​substituted aromatic ring. Important examples are listed in Table 10.4.1.10. These herbicides are usually formulated as salts or esters of the active compound and sometimes coformulated with the chemically related herbicides ioxynil, bromoxynil and/​or dicamba. Most in- stances of serious poisoning have been due to deliberate ingestion, mainly in the developing world. Mechanisms of toxicity include dose-​dependent cell membrane damage, chemical mimicry of acetyl coenzyme A and uncoupling of oxidative phosphorylation. Clinical features Ingestion causes nausea and vomiting which may be accompanied by burning in the mouth and throat and abdominal pain. Severe corro- sive injury to the gastrointestinal tract is rare. Hypotension, which is common, is due predominantly to intravascular volume loss, although vasodilation and direct myocardial toxicity may also contribute. Coma, hypertonia, hyperreflexia, ataxia, nystagmus, miosis, hallu- cinations, convulsions, fasciculation, and paralysis may then ensue. Hypoventilation is commonly secondary to central nervous system Table 10.4.1.10  Chlorophenoxy herbicides Chemical name Other names 2,4-​Dichlorophenoxy acetic acid 2,4-​D 4-​(2,4-​Dichlorophenoxy) butyric acid 2,4-​DB 2-​(2,4-​Dichlorophenoxy) propionic acid 2,4-​DP, dichlorprop 4-​Chloro-​2-​methylphenoxyacetic acid MCPA 4-​(4-​Chloro-​2-​methylphenoxy) butyric acid MCPB 2-​(4-​Chloro-​2-​methylphenoxy) propionic acid Mecoprop 2-​(2, 4-​Dichloro-​m-​tolyoxy) propionanilide Clomeprop

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1760 depression, but respiratory muscle weakness is a factor in the devel- opment of respiratory failure in some patients. Myopathic symptoms, including limb muscle weakness, loss of tendon reflexes, myotonia, and increased creatine kinase activity, have been observed. Metabolic acidosis, rhabdomyolysis, renal failure, increased aminotransferase activities, pyrexia, and hyperventilation have been reported. Treatment In addition to supportive care, urine alkalinization with high-​flow urine output will enhance herbicide elimination and should be con- sidered in all seriously poisoned patients. Haemodialysis produces similar herbicide clearances to urine alkalinization and should be considered if high-​flow urine alkalinization cannot be performed for clinical reasons. Glyphosate Glyphosate-​containing herbicides are very popular because their mode of action is plant-​specific (by inhibition of an enzyme pathway not present in mammals), they act only on contact with plant foliage and are inactivated on contact with soil. Formulations usually contain the isopropylamine salt of glyphosate, together with a surfactant. The latter is often an animal fat derivative (a tallow amine) polyoxyethylene amine (POEA), which contributes substantially to the toxicity of the formu- lation. Dilute, ready-​to-​use glyphosate/​POEA preparations are rarely associated with systemic toxicity, which usually requires the deliberate ingestion of a concentrate (typically 41% glyphosate/​15% POEA). Clinical features Ingestion of more than 85 ml of the concentrated glyphosate/​POEA formulation is likely to cause significant toxicity in adults. Fatal in- gestions have usually involved more than 200 ml. The most prom- inent effects are on the alimentary tract with burning in the mouth, throat, nausea, vomiting, dysphagia, and diarrhoea. Upper gastro- intestinal haemorrhage is a much less common complication. Renal and hepatic impairment are also frequent and usually reflect reduced organ perfusion. Respiratory distress, impaired consciousness, pul- monary oedema, infiltration on chest radiograph, shock, arrhyth- mias, renal failure requiring haemodialysis, metabolic acidosis, and hyperkalaemia may supervene in severe cases and, together with advancing age and late presentation, are poor prognostic indicators. Treatment Management is symptomatic and supportive. Intravenous fluids or blood transfusion may be required. Respiratory and renal failure should be managed conventionally. The fatality rate in case series of glypho- sate concentrate ingestion is approximately 6%. Death most frequently ensues within 72 h and is related to refractory cardiovascular collapse. Metaldehyde Metaldehyde in the form of pellets is used widely for killing slugs and in some countries as a solid fuel. Clinical features Nausea, vomiting, diarrhoea, abdominal pain, agitation, dizzi- ness, and tachycardia are common, but it is central nervous system and skeletal muscle complications that characterize the most se- vere toxicity. Convulsions may develop within 3 h of ingestion and recur frequently, and over days. Consciousness is impaired. Tremor, hypertonia, exaggerated limb reflexes, muscle twitching or spasms, including jaw clenching in the absence of overt seizures, superior lateral gaze fixation, and opisthotonus have all been described. This increase in motor activity, in turn, results in raised creatine kinase ac- tivity, rhabdomyolysis, hyperpyrexia, and metabolic acidosis in some cases. Less frequent complications include hyperventilation, respira- tory alkalosis, minor elevation of transaminase activities, upper gastrointestinal bleeding, and mildly deranged coagulation. Treatment Treatment is supportive. Intravenous diazepam 10 mg IV or lor- azepam 4 mg IV in an adult should be given to suppress convulsions and a clear airway and adequate ventilation ensured, using endo- tracheal intubation if necessary. Rhabdomyolysis and its complica- tions should be managed conventionally. Methyl bromide The uses of methyl bromide (bromomethane) have gradually be- come restricted as it is an ozone-​depleting chemical. It is currently used to fumigate mills, warehouses, shipping containers, stored products, and soil in greenhouses to eradicate pests such as wood- worm and rodents that damage crops and buildings. In most coun- tries application is now restricted to trained and licensed personnel. Methyl bromide is absorbed readily through the lungs and is ex- creted largely unchanged by the same route. Clinical features Inhalation causes respiratory tract irritation with shortness of breath, cough, and varying degrees of pulmonary oedema. Hypoxaemia and respiratory failure are inevitable consequences. Neurotoxicity manifests hours or days later with agitation, delirium, ataxia, in- tention tremor, nystagmus, dysdiadochokinesis, and hyperreflexia. Abnormal movements of the limbs are common and convulsions occur frequently. Consciousness may be impaired to the point of coma in severe cases. Proteinuria, oliguria (due to renal tubular and cortical necrosis), and jaundice have been described. Treatment The casualty should be removed promptly from the contamin- ated atmosphere and undressed, as methyl bromide can penetrate clothing and rubber gloves. Contaminated skin should be washed with water. Treatment is supportive. Systemic uptake can be quanti- fied by measuring serum and urine bromide concentrations. Neonicotinoids Neonicotinoids are now employed widely as systemic insecticides and some (imidacloprid, dinotefuran, thiamethoxam, nitenpyram) are used as flea control agents for dogs and cats. They block postsynaptic nicotinic receptors (nAChRs), particularly the α4ß2 subtype. The high specificity of neonicotinoids for insect nicotinic receptors, their low affinity for human nAChRs and relatively poor penetration of the human blood–​brain barrier should result in much lower toxicity to humans than nicotine-​containing pesticides. Human poisoning with neonicotinoids is well recognized and may be severe and fatal, though the solvents and surfactants present in many formulations may also contribute to toxicity. Clinical features Poisoning with neonicotinoids is characterized by the rapid onset of symptoms including nausea and vomiting, fever, sweating, increased

10.4.1  Poisoning by drugs and chemicals 1761 salivation, bronchorrhoea, agitation, lack of coordination, disorienta- tion, muscle weakness, seizures, and coma. Breathlessness, depressed respiration, cyanosis, and respiratory arrest have been reported. Bradycardia is sometimes present, but tachycardia is observed more frequently; ventricular tachycardia/​fibrillation has occurred occasion- ally, as has cardiac arrest. Severe hypotension and shock supervene in those severely poisoned. Miosis is present in some cases but not the majority. Metabolic acidosis and renal failure have been observed. Treatment In patients who are unconscious a clear airway should be established and, if ventilation is impaired, assisted ventilation should be com- menced. Hypotension and cardiac dysrhythmias should be managed conventionally and acid–​base and electrolyte balance corrected. Since the clinically important features, notably bronchorrhoea, are caused by cholinergic overactivity, atropine sulphate 2 mg intraven- ously (in an adult) should be given and the dose repeated until the signs of atropinization are present (dry skin and sinus tachycardia). There is some evidence that activated charcoal can bind neonicotinoids and if administered within 1 h of ingestion may reduce absorption; gastric lavage is contraindicated because of the presence of solvents unless it can be performed with a cuffed endotracheal tube in situ. Organophosphorus insecticides Organophosphorus insecticides are among the most widely used pesticides throughout the world. They inhibit acetylcholinesterase causing accumulation of acetylcholine at central and peripheral cholinergic nerve endings, including neuromuscular junctions. The clinical presentation and severity of OP poisoning depends not only on the pesticide and the magnitude of exposure but also on several other factors, including the route of exposure, the age of the patient, whether exposure was a suicidal attempt (when a substantial inges- tion is more likely), and the presence of a solvent in the formula- tion. Not only may skin absorption of the OP itself be enhanced by the presence of the solvent, but also ingestion of a solvent may in- duce vomiting with risk of aspiration; depressed consciousness may follow. In addition, there is increasing evidence that the solvents in formulations are responsible for the high morbidity and mortality. Clinical features The first symptom of mild poisoning, particularly in individuals oc- cupationally exposed, is often a feeling of exhaustion and weakness. Vomiting, cramping abdominal pain, sweating, and hypersalivation may follow. Constriction of one or both pupils and a sensation of tightness in the chest during inspiration also may occur at an early stage, but these signs are not reliable indices of the severity of systemic poisoning because they may be caused by local anticholinesterase ef- fects of spray mist on the eye or bronchi. In cases of more severe poisoning, the nicotinic features tend to appear first, but a combination of muscarinic, nicotinic, and cen- tral nervous system symptoms is apparent in many severe cases. Muscle twitching may affect the eyelids, tongue, face muscles, and calf muscles; respiratory muscles then become involved, and gen- eral muscle weakness ensues. Convulsions may occur, though OPs vary in their potency to induce seizures. Bronchial hypersecretion/​ bronchorrhoea, with bronchoconstriction, is followed in severe cases by cyanosis, respiratory depression, and coma. Death may follow from respiratory failure. Coma is usually due to direct central nervous system depression by the pesticide and solvents in the com- mercial formulation. Aspiration pneumonia is common. Though bradycardia would be expected from the mode of action of organophosphorus insecticides, it is present in only about 20% of cases; sinus tachycardia is more common. Rarely, complete heart block and arrhythmias occur. Relapse after apparent resolution of cholinergic symptoms has been reported, particularly in patients who have ingested highly lipophilic insecticides, and is termed the intermediate syndrome. This involves the onset syndrome of muscle paralysis affecting particularly upper limb muscles, neck flexors, and cranial nerves some 24–​96 h after exposure, though there are reports of paralysis occurring before 24 h and even after 96 h. It is often associated with the development of respiratory failure. Delayed neuropathy is a rare complication of acute exposure to some insecticides. It results from phosphorylation and subse- quent ageing of at least 70% of neuropathy target esterase (NTE). Only a small number of marketed insecticides, for example methamidophos, are capable of causing this syndrome. The fea- tures resulting from distal degeneration of some axons in both the peripheral and central nervous systems occur 1–​4 weeks after ex- posure. Cramping muscle pain in the lower limbs, distal numbness, and paraesthesiae are followed by progressive weakness, depres- sion of deep tendon reflexes in the lower limbs and, in severe cases, in the upper limbs. Signs include a high-​stepping gait from bilat- eral foot drop and, in severe cases, quadriplegia with foot and wrist drop, as well as pyramidal signs. In time, there might be significant recovery of peripheral nerve function but, depending on the de- gree of pyramidal involvement, spastic ataxia may be permanent The diagnosis is confirmed by measuring erythrocyte acetyl- cholinesterase activity; plasma butyrylcholinesterase activity is a less preferable alternative. Treatment Management involves supportive measures and judicious admin- istration of antidotes. Bronchorrhoea requires prompt relief with intravenous atropine 2 mg (0.02–​0.1 mg/​kg in a child). The atropine dose should be titrated to control rhinorrhoea and bronchorrhoea, to raise the pulse rate above 80 bpm, and restore systolic blood pressure to more than 80 mm Hg. If the initial dose produces only a partial response, it should be doubled and doubled again if there is only a limited clinical response. In addition, supplemental oxygen should be given to maintain PaO2 greater than 10 kPa (75 mm Hg). If neces- sary, the patient should be intubated and mechanical ventilation (with positive end-​expiratory pressure) should be instituted. There are consistent animal data supporting the effectiveness of oximes, when given early. While there are also some clinical studies which support the benefit of oxime therapy, others do not. Recent studies indicate that solvents in the formulations play a crucial role in toxicity, which could explain why oximes seem to be less effective clinically than in experimental studies where pure insecticide is often employed. Pralidoxime chloride 30 mg/​kg by intravenous injection over 20 min, repeated at 4–​6 h intervals, should be administered as soon as possible in any patient requiring atropine; alternatively, an intravenous infusion of 8–​10 mg/​kg/​h in an adult may be employed after the bolus injection. Administration of pralidoxime should con- tinue for as long as atropine is required, that is, until clear, irrevers- ible clinical improvement is achieved, which may take many days while residual insecticide is cleared from the body stores.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1762 The use of diazepam 5–​10 mg intravenously in an adult will reduce anxiety and restlessness, but larger doses may be required to control convulsions; diazepam also reduces morbidity and mortality. Phosphine Phosphine is used extensively as a fumigant to control rodents and a wide variety of insects in sealed containers or structures. It is also used in the semiconductor industry. Clinical features The initial symptoms are often alimentary rather than respiratory. Indeed, the nausea, vomiting, diarrhoea, and epigastric pain may be so striking that physicians are misled into making a diagnosis of acute gastroenteritis. However, respiratory complaints do occur. Chest tightness, breathlessness, chest pain or soreness, and palpitations are commonly reported. Inhaled phosphine is as cardiotoxic as in- gested metal phosphides (see earlier). Acute heart failure, pulmonary oedema (which may be both cardiogenic and non​cardiogenic) and cardiac arrhythmias have been observed, particularly in children. Convulsions, ataxia, and intention tremor have also been reported. Treatment The casualty should be removed from exposure as soon as possible. Thereafter, treatment is supportive and symptomatic. Pyrethroids Pyrethroids are used widely as insecticides both in the home and com- mercially, and in medicine for the topical treatment of scabies and head lice. In tropical countries, mosquito nets are commonly soaked in pyrethroid solutions as part of antimalarial strategies. Pyrethroid sprays are used to ‘disinsect’ the interiors of aircraft. Despite their ex- tensive worldwide use, severe poisoning with pyrethroids is extremely rare. The most important mechanism of toxicity is modification of the gating characteristics of voltage-​sensitive sodium channels, causing delayed closure. A protracted sodium influx ensues, which, if it is suf- ficiently large and/​or long, lowers the action potential threshold and causes repetitive firing which manifests as paraesthesiae. Clinical features Pyrethroids are best known for their ability to cause facial paraes- thesiae following occupational cutaneous exposure; these symptoms last only a few hours at most. Inhalation of pyrethroid-​containing dust or aerosol droplets may cause respiratory tract irritation, but systemic toxicity is unusual. Ingestion causes irritation of the gastro- intestinal tract with nausea and vomiting, increased salivation, and mouth ulceration. Coma, convulsions, and pulmonary oedema may ensue in the most severe cases and fatalities have occurred rarely. Treatment Symptomatic and supportive measures should be employed, and reassurance given that facial paraesthesiae will not be a long-​term problem. Other chemicals Acetone Acetone is a clear liquid with a characteristic pungent odour and sweet taste, used widely in industrial and household products including paints, nail polish, and nail polish removers. It is absorbed rapidly through the lungs and gut and metabolized in the liver to pyruvate. Metabolism is saturable with elimination of the parent compound in expired air (it can be smelt on the breath) and urine at high doses. Clinical features There is irritation of mucous membranes of eyes, nose, and throat. Systemic toxicity causes headache, excitement, restlessness, chest tightness, incoherent speech, nausea and vomiting and, occasion- ally, gastrointestinal bleeding, coma, convulsions, and hypergly- caemia (resulting from the metabolism of pyruvate to glucose). Treatment If toxicity has followed inhalation, remove from exposure, give sup- portive treatment, and correct hyperglycaemia. Since acetone is a small molecule, there may be a role for dialysis in the management of seriously poisoned patients, particularly if plasma acetone con- centrations are high. Acids Acids commonly involved in cases of poisoning include the inor- ganic acids hydrochloric, hydrofluoric (see ‘Hydrogen fluoride/​ hydrofluoric acid’, further on in this chapter), nitric, phosphoric, and sulphuric acids; and organic acids such as acetic, formic, lactic, and trichloroacetic acids. Car battery acid typically contains 28% sulphuric acid. Proprietary cleaning agents and antirust compounds often comprise a mixture of hydrochloric and phosphoric acids. Inorganic acids generally are of concentrations more likely to be cor- rosive at the normally available solution. Clinical features Acid burns of the skin cause erythema, blistering and, in severe cases, ulceration and necrosis. In the eyes, intense pain and bleph- arospasm are common, and corneal burns may occur. When ingested, acids tend to damage the stomach more than the oe- sophagus, but oropharyngeal and oesophageal injuries may also occur. Immediate pain is followed by vomiting and/​or haematemesis. Severe injury results in inability to swallow saliva with drooling. Gastric and oesophageal perforation may occur, resulting in chemical peritonitis and shock. Other effects include hoarseness, stridor, and respiratory distress secondary to laryngeal and epiglottic oedema, metabolic acid- osis leucocytosis, acute tubular necrosis, renal failure, hypoxaemia, re- spiratory failure, intravascular coagulation, and haemolysis. Treatment Acid burns to the skin should be irrigated liberally with water or sa- line and managed as a thermal burn. Skin grafting may be necessary and specialist advice should be sought. After ocular exposure with acid, the eye should be irrigated, pref- erably with saline for 15–​30 min. Topical local anaesthetic is usually required to relieve pain and to overcome blepharospasm. Specialist advice should be sought. After ingestion, a clear airway should be established. Opioids are often necessary for analgesia. Dilution and/​or neutralization is contraindicated. Features of severe tissue injury (severe abdom- inal pain, abdominal distension, circulatory collapse, or lactic acid- osis) may indicate the presence of bowel necrosis or perforation. Immediate surgical assessment is recommended because early re- section of necrotic tissue and intraluminal stenting has been shown

10.4.1  Poisoning by drugs and chemicals 1763 to improve survival and reduce the risk of oesophageal stricture for- mation. Both CT scan and fibreoptic endoscopy have been shown to be useful in assessing the severity of injury, risk of mortality, and risk of subsequent stricture formation. These two imaging modalities are complimentary and when combined provide the best understanding of the injury and risk. If there are severe clinical features, then en- doscopy is best performed by a surgeon capable of undertaking de- finitive treatment. Corticosteroids confer no benefit and may mask abdominal signs of perforation; antibiotics should be given for es- tablished infection only. Acid ingestion may result in antral, pyloric, or jejunal strictures, achlorhydria, protein-​losing enteropathy, and gastric carcinoma. Alkalis Alkalis are commonly found in the home and those encountered in cases of poisoning include sodium hydroxide (drain, lavatory, pipe cleaners), sodium carbonate, sodium silicate, sodium perborate, sodium phosphate, sodium carbonate (denture cleaning tablets), sodium dichloroisocyanurate (water sterilizing tablets), sodium hypochlorite (a bleaching agent), and alkaline batteries. Clinical features The features of eye, skin and laryngeal contamination with alkalis are similar to those produced by acids (see ‘Acids’) though when ingested, alkalis are more likely to damage the oesophagus. Oropharyngeal pain, together with epigastric pain are followed by vomiting and diar- rhoea. Oesophageal ulceration with or without perforation may be complicated by mediastinitis or pneumonitis. Oesophageal perfor- ation may result in catastrophic aorto-​enteric fistula formation. Treatment The treatment of corrosive injuries caused by alkalis is largely the same as for those produced by acids (see ‘Acids’). In severe cases, following resuscitation and stabilization, early assessment by endos- copy and/​or CT imaging is the priority. Alkali ingestion may result in stricture formation and there is a risk of malignancy. The mean latent period for development of carcinoma of the oesophagus fol- lowing alkali ingestion is more than 40 years. Arsine Arsine is a colourless, non​irritating gas. Arsine binds with oxidized haemoglobin causing massive intravascular haemolysis. Clinical features There is usually a delay of some 2–​24 h after exposure before the onset of headache, malaise, weakness, dizziness, breathlessness, migratory abdominal pain, fever, tachycardia, tachypnoea, nausea, and vomiting. A bronze skin colour is noted in some patients, but most have the typ- ical appearance of a jaundiced patient. Acute renal failure is observed by the third day after substantial exposure, and the urine is dark red, then brown (from haemoglobinuria), before anuria (due to acute renal tubular necrosis) ensues. Investigations will show leucocytosis, reticulocytosis, elevated plasma haemoglobin, and haemoglobinuria. Treatment If haemolysis is severe, the use of red cell exchange and plasma exchange may be more beneficial than red cell exchange alone, though plasma exchange alone is also effective in the treatment of intravascular haemolysis. Blood transfusion will be required in cases of severe haemolysis. If renal failure ensues, haemodialysis/​ haemodialfiltration should be undertaken. Antidotes to remove ar- senic are of no value. Benzene Benzene is a colourless, volatile liquid with a pleasant odour. It is an ingredient in many paints and varnish removers, and in some petrols. About 10% of inhaled benzene is excreted unchanged in the breath. The remainder is metabolized by mixed function oxidase en- zymes predominantly in the liver, but also in the bone marrow, the target organ of benzene toxicity. Benzene is a human carcinogen. Clinical features Acute exposure Following inhalation or ingestion, euphoria, dizziness, weakness, headache, blurred vision, mucous membrane irritation, tremor, ataxia, chest tightness, respiratory depression, cardiac arrhyth- mias, coma, and convulsions occur. Direct skin contact with liquid benzene may produce marked irritation. Chronic exposure The toxic effects of chronic poisoning may not become apparent for months or years after initial contact and may develop after all ex- posure has ceased. Anorexia, headache, drowsiness, nervousness, and irritability are well described. Anaemia (including aplastic anaemia), leucopenia, thrombocytopenia, pancytopenia, leukaemia, lymphomas, chromo- somal abnormalities, and cerebral atrophy have been reported. There is also an association between occupational benzene exposure and non-​Hodgkin’s lymphoma. Patients have recovered after as long as a year of almost complete absence of formation of new blood cells. A dry, scaly dermatitis may develop on prolonged or repeated skin exposure to liquid benzene. Treatment Following removal from the contaminated atmosphere, treatment should be directed towards symptomatic and supportive measures. Gastric lavage is hazardous as aspiration is likely to occur. Benzyl alcohol Benzyl alcohol has been used as a preservative in intravascular flush solutions and in drug formulations, which has led to severe toxicity in neonates. Benzyl alcohol is metabolized to benzoic acid, which is then con- jugated with glycine in the liver and excreted as hippuric acid. The immature liver’s capacity to metabolize benzoic acid is limited and, when exceeded, leads to accumulation of this metabolite and meta- bolic acidosis. Clinical features In 1982, a syndrome consisting of metabolic acidosis, convulsions, neurological deterioration (due to intraventricular haemorrhage), gasping respirations, hepatic and renal abnormalities, cardiovas- cular collapse, and death was described in small premature infants between 2 to 14 days of age. This was due to IV solutions containing benzyl alcohol. In contrast, healthy adult humans are able to tolerate as much as 30 ml of 0.9% benzyl alcohol by rapid intravenous in- fusion without signs of toxicity. However, a 5-year-old girl devel- oped hypernatraemia and metabolic acidosis due to the infusion of

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1764 diazepam for 36 hours to control status epilepticus. The presence of benzyl alcohol has been used as marker of toluene exposure. Treatment Metabolic acidosis, hepatic and renal failure should be treated con- ventionally and administration of fluids or drugs containing benzyl alcohol should be discontinued. Carbon dioxide Carbon dioxide is a colourless gas that is also available commer- cially as a solid for refrigeration purposes (‘dry ice’). High concen- trations may accumulate in wells, silos, manholes, mines, and in several volcanic lakes in Africa. In 1986, Lake Nyos in Cameroon emitted a cloud of carbon dioxide that killed 1700 villagers and 3500 of their livestock. Clinical features Dyspnoea, cough, headache, dizziness, sweating, restlessness, par- aesthesiae, and sinus tachycardia are features after modest exposure. Higher concentrations (>10%) produce psychomotor agitation, myoclonic twitches, eye flickering, coma, and convulsions. Death occurs from acute cardiorespiratory depression, typically at concen- trations exceeding 17%. Skin contact with solid carbon dioxide (dry ice) may result in frostbite and local blistering. Treatment The casualty should be removed from the contaminated envir- onment and oxygen administered. Thereafter, care is supportive. Dry ice burns are treated similarly to other cryogenic burns, with thawing of the affected tissue and suitable analgesia. Carbon disulphide Carbon disulphide is used as a fumigant for grain and as a solvent, particularly in the rayon industry. It is a clear, colourless, volatile li- quid with an odour like that of decaying cabbage. Clinical features Acute exposure Acute poisoning is rare. Absorption occurs through the skin as well as by inhalation. Because of its potent defatting activity, carbon di- sulphide causes reddening, cracking, and peeling of the skin, and a burn may occur if contact continues for several minutes. Splashes in the eye cause immediate and severe irritation. Acute inhalation may result in irritation of the mucous membranes, blurred vision, nausea and vomiting, headache, delirium, hallucinations, coma, tremor, convulsions, and cardiac and respiratory arrest. Chronic exposure There is an increased incidence of cardiovascular disease (hyperten- sion, arteriosclerosis, ischaemic heart disease, elevated cholesterol) among workers exposed to carbon disulphide. In addition, sleep disturbances, fatigue, anorexia, and weight loss are common com- plaints. Intellectual impairment, cerebellar signs, diffuse vascular encephalopathy, parkinsonism, peripheral polyneuropathy, hepatic damage, and permanent impairment of reproductive performance have been described. Treatment Treatment involves removal from exposure, washing contaminated skin, irrigation of the eyes with water and supportive measures. In most cases, however, preventive measures to keep carbon disul- phide concentrations in the workplace as low as possible are more important. Carbon monoxide Carbon monoxide is a tasteless, odourless, colourless, non​irritating gas produced by incomplete combustion of organic materials. Normal endogenous carbon monoxide production is sufficient to maintain a resting carboxyhaemoglobin concentration of 1 to 3% in urban non​smokers and 5–​6% in smokers. Common sources of carbon monoxide are car exhaust fumes (in the absence of a catalytic converter), improperly maintained and ventilated heating systems, and smoke from all types of fire, typically charcoal barbecues used indoors. Carbon monoxide derived from domestic heating systems is a major cause of accidental death in the developing world. Inhalation of methylene chloride (from paint strippers) may lead to carbon monoxide poisoning due to break- down of the parent compound. Symptoms and signs that follow inhalation of carbon monoxide are the result of tissue hypoxia. The affinity of haemoglobin for carbon monoxide is approximately 240 times greater than that for oxygen. Carbon monoxide combines with haemoglobin to form carboxyhaemoglobin, reducing the total oxygen-​carrying capacity of the blood. In addition, the oxygen dissociation curve shifts to the left due to modification of oxygen-​binding sites. As a result, the affinity of the remaining haem groups for oxygen is increased, the oxygen dissociation curve is distorted as well as being shifted and the resulting tissue hypoxia is thus far greater than that which would result from simple loss of oxygen-​carrying capacity. Carbon monoxide may also inhibit cellular respiration as a result of reversible binding to cytochrome oxidase a3. At higher concentra- tions, carbon monoxide causes brain lipid peroxidation and it has been suggested this may be relevant to the development of delayed neuropsychiatric sequelae. Clinical features The clinical features of carbon monoxide poisoning may be divided into those caused acutely, predominantly due to hypoxia, and those that result from tissue damage by the mechanisms detailed earlier. These later toxicities are therefore related to the initial amounts of carbon monoxide inhaled and length of time before rescue and treatment. In acute poisoning, organs with high oxygen demand are at special risk of damage and this includes, in particular, the heart and brain. The symptoms of mild to moderate exposure are non​specific and include headache, nausea, and confusion. These non​specific symp- toms require a high degree of suspicion in patients at potential risk of poisoning with the gas. As concentrations increase, metabolic acidosis ensues from interference with metabolic processes, and central nervous system features progress to cause loss of conscious- ness with hypertonia and hypereflexia, extensor plantar responses, papilloedema, and convulsions. Cardiovascular changes include ar- rhythmias and ischaemic myocardial damage. Other complications tend to be detected later and include per- sistent neurological damage in any part of the brain, which may

10.4.1  Poisoning by drugs and chemicals 1765 result in either paralysis or midbrain damage causing parkinsonism or akinetic mutism, deafness due to central ischaemia of the brain stem nuclei and cochlea, muscle necrosis causing rhabdomyolysis and renal failure and skin changes (bullae) due to prolonged unconsciousness. The degree of intoxication is correlated to some extent with carb- oxyhaemoglobin concentrations but, by the time patients arrive in hospital, they will often have received oxygen in an ambulance, which may lower carbon monoxide concentrations from those pre- sent at the scene of the injury. Very severe features are to be expected with carboxyhaemoglobin above 60%, but significant features would not generally be expected at concentrations below 30%. Neuropsychiatric problems may occur after recovery from carbon monoxide intoxication and are said to develop insidiously over several weeks. Defining limits of these changes may be difficult in patients with relatively mild exposure, and ascribing causation is particularly difficult in low-​level exposures where formal studies suggest no effect. Treatment Removal from exposure and administration of 100% oxygen using a tightly fitting facemask are essential. If patients are unconscious, endotracheal intubation and mechanical ventilation is required. Prolonged administration of oxygen is necessary to ensure carbon monoxide bound in tissues is released. Traditionally, hyperbaric oxygen has been used in carbon mon- oxide poisoning, although there remains significant controversy about its efficacy. A trial in the United States of America has shown some suggestion of benefit, but commentators are divided about the relevance of the relatively small changes noted, and the wider appli- cation of these results. In part, this is because patients were brought for treatment for quite some distance, and the debate remains as to whether any potential small therapeutic benefit warrants transfer to a distant treatment facility. Cyanide Hydrogen cyanide and its derivatives are used widely in industry and are released during the thermal decomposition of polyurethane foams. Cyanide poisoning may also result from the ingestion of the cyanogenic glycoside amygdalin, which is found in the kernels of almonds, apples, apricots, cherries, peaches, plums, and other fruits. Cyanide reversibly inhibits cellular enzymes which contain ferric iron notably cytochrome oxidase a3 so that electron transfer is blocked, the tricarboxylic acid cycle is paralysed, and cellular respiration ceases. Clinical features Acute poisoning The ingestion by an adult of 50 ml of (liquid) hydrogen cyanide or 200–​300 mg of one of its salts is likely to prove fatal. Inhalation of hydrogen cyanide gas may produce symptoms within seconds and death within minutes. Acute poisoning is characterized by dizziness, headache, palpitation, anxiety, a feeling of constriction in the chest, dyspnoea, pulmonary oe- dema, confusion, vertigo, ataxia, coma, and paralysis. Cardiovascular collapse, respiratory arrest, convulsions, and metabolic acidosis are seen in severe cases. Cyanosis may occur, and the classical ‘brick red’ colour of the skin is noted occasionally. There is sometimes an odour of bitter almonds on the breath, but the ability to detect this is genetically determined and some 40% of the population are unable to do so. Chronic exposure Chronic exposure results predominantly in neurological damage which can include ataxia, peripheral neuropathies, amblyopia, optic atrophy, and nerve deafness. Treatment The immediate administration of oxygen is of paramount import- ance, as there is evidence that it prevents inhibition of cytochrome oxidase a3 and accelerates its reactivation. Treatment thereafter de- pends on the severity and type of exposure. If hydrogen cyanide has been inhaled and the patient remains conscious 10 min after ex- posure has ceased, no antidotal treatment is required. In more severe cases an antidote is invariably necessary. Methaemoglobin binds cyanide, forming cyanmethaemoglobin. Methaemoglobinaemia may be induced efficiently by the adminis- tration of intravenous sodium nitrite 300 mg over 5 min. Because the effect of sodium nitrite is relatively rapid and methaemoglobin formation slower, the benefit of sodium nitrite may also be from its vasodilator action and the consequent improved tissue perfusion. Intravenous sodium nitrite is usually given with intravenous sodium thiosulfate; they have been shown to act synergistically in experi- mental cyanide poisoning by providing sulphane sulphur to en- hance endogenous metabolism. Sodium thiosulfate is administered intravenously in a dose of 12.5 g over 10 min. Dicobalt edetate solutions contain free cobalt, which complexes six times more cyanide than dicobalt edetate. Cobalt is toxic, how- ever, and use of this formulation in the absence of cyanide poisoning may cause cobalt toxicity. Dicobalt edetate should, therefore, be given only when the diagnosis is certain. Dicobalt edetate is admin- istered intravenously in a dose of 300 mg over 1 min, with a further 300 mg being given if recovery does not occur. One mole of hydroxocobalamin inactivates one mole of cyanide, but, on a weight-​for-​weight basis, 50 times more hydroxocobalamin is needed than cyanide because hydroxocobalamin is a far larger molecule. If available, hydroxocobalamin 5 g is given intravenously over 30 min; a second dose (5 g) may be required in severe cases. Diethylene glycol Diethylene glycol is used as a coolant, as a building block in organic synthesis and as a solvent. It can be also found in some hydraulic fluids and brake fluids. Occupational exposure is by the dermal route, but the most common route of exposure is ingestion, often unintention- ally as a result of contamination of medicines. Diethylene glycol has been responsible for several mass poisonings in Australia, Bangladesh, Haiti, India, Nigeria, South Africa, and the United States of America. Diethylene glycol is metabolized by alcohol dehydrogenase to 2-​hydroxyethoxyacetaldehyde and by aldehyde dehydrogenase to 2-​hydroxyethoxyacetate and a small amount of diglycolic acid. The metabolic acidosis observed in diethylene glycol poisoning is pri- marily due to 2-​hydroxyethoxyacetate, but lactate also plays a part in severe poisoning. Diglycolic acid is the metabolite responsible for the development of proximal tubular necrosis. Clinical features Nausea and vomiting, headache, abdominal pain, coma, seizures, metabolic acidosis, and acute renal failure have most commonly been reported. Renal dysfunction in those who do not develop dialysis de- pendence often improves over several months. Pancreatitis and hepa- titis have been observed, together with cranial neuropathies, such as

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1766 facial nerve motor deficits, and demyelinating peripheral neuropathy, demonstrated clinically as bilateral lower extremity numbness. Treatment Supportive measures to correct metabolic acidosis should be in- stituted promptly. If the patient presents early after ingestion, the priority is to inhibit metabolism using either intravenous fomepizole or ethanol. Fomepizole requires less monitoring, but is more expensive than ethanol. After a loading dose of fomepizole 15 mg/​kg over 30 min, four 12-​hourly doses of 10 mg/​kg should be given, followed by 15 mg/​ kg 12-​hourly until the glycol concentration is not detectable. If haemodialysis is used, the frequency of dosing should be increased to 4-​hourly as fomepizole is dialysable. Alternatively, a loading dose of intravenous ethanol 50 g for an adult (50 ml of absolute ethanol in 1 L 5% dextrose, i.e. a 5% ethanol solution) should be given, followed by an intravenous infusion of ethanol, 10–​12 g/​h (most easily given as 1 L 5% ethanol solution over 4–​5 h), to achieve a blood ethanol concentration of approxi- mately 1000 mg/​litre. Administration of ethanol should be con- tinued until the glycol is undetectable in the blood. If haemodialysis/​ haemodialfiltration is used, greater amounts of ethanol (17–​22 g/​h) must be given, because ethanol is readily dialysable. Haemodialysis/​haemodialfiltration will remove diethylene glycol, but it is not known whether the metabolites are also removed. Ethylene glycol Ethylene glycol has a variety of commercial applications and is com- monly used as an antifreeze fluid in car radiators. Its sweet taste and ready availability have contributed to its popularity as a suicide agent and as a poor man’s substitute for alcohol. It is thought that the minimum lethal dose of ethylene glycol is about 100 ml for an adult, although recovery after treatment has been reported following the ingestion of up to 1 L. The toxicity of ethylene glycol depends predominantly on its metabolites (Fig. 10.4.1.5) though the initial inebriation is due to ethylene glycol itself. Central nervous system symptoms coincide with the peak production of glycolaldehyde; aldehydes inhibit many aspects of cellular metabolism. Glycolate is largely responsible for the marked acidosis seen in severe cases; lactate concentrations are generally not very high. Lactate is produced as a result of the large amount of NADH formed by the oxidation of ethylene glycol and by inhibition of the tricarboxylic acid cycle by the condensation products of glyoxylate. There is increasing evidence that calcium ox- alate monohydrate crystals are the cause of renal failure and cerebral oedema. Clinical features The clinical features of ethylene glycol poisoning may be divided into three stages depending on the time after ingestion (Table 10.4.1.11). Although the three stages are useful theoretical descriptions of ethylene glycol poisoning, the onset and progression of the clinical course is frequently not as consistent or predictable. After a brief period of inebriation due to the intoxicating effect of ethylene glycol itself, metabolic acidosis develops, followed by tachyp- noea, coma, seizures, hypertension, and hypocalcaemia, together with calcium oxalate crystalluria, the appearance of pulmonary infiltrates, and oliguric renal failure. If untreated, death from multiorgan failure usually occurs 24–​36 h after ingestion. Severe acidosis, hyperkalaemia, seizures, and coma carry a poor prognosis. A serum ethylene glycol concentration more than 500 mg/​litre indicates severe poisoning. Treatment Supportive measures to combat cardiorespiratory depression should be employed and metabolic acidosis, hypocalcaemia, and renal failure should be treated conventionally. (For further information on treatment with fomepizole or ethanol, please see the section ‘Diethylene glycol’.) If admission plasma concentrations show that the ethylene glycol in- gested has already been metabolized, fomepizole or ethanol adminis- tration will not be of benefit and ethanol might exacerbate the acidosis. Haemodialysis/​haemodialfiltration removes ethylene glycol, glycolaldehyde, and glycolate (but not oxalate), and will also correct acid–​base disturbances. Dialysis should be employed particularly if presentation is late and marked metabolic acidosis is present. It should be continued until the glycol and glycolate are no longer detectable in the blood. Ethylene glycol GO Ethanol Fomepizole ALDH AO ADH Glycolate Glycoaldehyde Oxalate Glyoxylate LDH or GO Fig. 10.4.1.5  Metabolism of ethylene glycol. ADH, alcohol dehydrogenase; ALDH, aldehyde dehydrogenase; AO, aldehyde oxidase; GO, glycolate oxidase; LDH, lactate dehydrogenase. Table 10.4.1.11  Clinical features of ethylene glycol poisoning Stage 1 (30 min–​12 h): gastrointestinal and nervous system involvement • Apparent intoxication with alcohol (but no ethanol on breath) • Nausea, vomiting, haematemesis • Coma and convulsions (often focal) • Nystagmus, ataxia, ophthalmoplegias, papilloedema, depressed reflexes, myoclonic jerks, tetanic contractions • Cranial nerve II, V, VII, VIII, IX, X, XII palsies Stage 2 (12–​24 h): cardiorespiratory involvement • Tachypnoea, tachycardia • Mild hypertension • Metabolic acidosis • Myocarditis • Pulmonary oedema • Congestive cardiac failure Stage 3 (24–​72 h): renal involvement • Flank pain, renal angle tenderness • Hypocalcaemia • Acute tubular necrosis • Calcium oxalate crystalluria

10.4.1  Poisoning by drugs and chemicals 1767 Formaldehyde Formaldehyde is a flammable, colourless gas with a pungent odour. It is most commonly available commercially as a 30–​50% w/​w aqueous solution and is an important raw material in the synthesis of organic compounds such as plastics and resins. It is added to cos- metics and foodstuffs as a preservative and antimicrobial agent and is used in embalming. Formaldehyde also occurs naturally in the environment. It is released during the combustion of organic ma- terials (e.g. in forest fires, wood-​burning stoves, and waste inciner- ators), and is a product of incomplete petrol combustion in internal combustion engines. Absorption may follow ingestion, inhalation, or dermal contact. Once absorbed, formaldehyde is oxidized rapidly to formate then converted more slowly to carbon dioxide and water. Clinical features Severe irritation of the mucous membranes of the eyes, nose, and upper airways occurs after minimal exposure to low (<5 ppm) for- maldehyde concentrations, which tends to prevent higher exposure in even the most tolerant subjects. Substantial exposure may result in severe bronchospasm, pulmonary oedema, and death. Formaldehyde is a recognized cause of occupational asthma. Formaldehyde solutions splashed into the eye have caused corneal damage and skin contamination has resulted in dermatitis. Spillage of phenol-​formaldehyde resin on to the skin has produced extensive necrotic skin lesions, fever, hypertension, adult respiratory distress syndrome, proteinuria, and renal impairment. Ingestion of for- maldehyde solution has resulted in severe corrosive damage to the buccal cavity and tonsils, oesophagus, and stomach with ulceration, necrosis, and subsequent fibrosis and contracture. Shock, metabolic acidosis (due in part to high formate concentrations), respiratory in- sufficiency, and renal impairment usually ensue. Death may follow ingestion of less than 100 ml in an adult. Treatment Supportive measures, including the correction of acid–​base disturb- ance, should be employed. Folinic acid 50 mg (1mg/kg in children) IV 6 hourly accelerates formate metabolism. Haemodialysis is only moderately effective in increasing formate elimination. n-​Hexane n-​Hexane is an extremely volatile liquid that is used as a solvent. It is metabolized oxidatively to several compounds, including 2,5-​ hexanedione, which is eliminated through the urine and is impli- cated in the neurotoxic effect of this solvent. Clinical features When ingested n-​hexane causes nausea, dizziness, central nervous system excitation, and then depression, and presents an acute as- piration hazard. Following inhalation, either inadvertently or delib- erately, similar symptoms occur. The development of a progressive sensorimotor neuropathy is the principal hazard of chronic exposure. Treatment Treatment is supportive and symptomatic. Hydrogen fluoride/​hydrofluoric acid Hydrogen fluoride is a corrosive, fuming, nearly colourless liquid (hydrofluoric acid) at atmospheric pressures and temperatures below 19°C; above 19°C it is gaseous. Hydrogen fluoride is very soluble in cold water and for this reason it fumes strongly in moist air. Aqueous solutions dissolve glass. Hydrogen fluoride is particularly dangerous because of its unique ability among acids to penetrate tissue. The reason for this is the high electronegativity of fluorine, which forms a strong covalent bond with the hydrogen. The result is a weak acid that exists predomin- antly in the undissociated state. In this state, hydrogen fluoride can penetrate skin and soft tissue by non​ionic diffusion. Once in the tis- sues, hydrogen fluoride dissociates and causes liquefactive necrosis of soft tissue, bony erosion, and extensive electrolyte abnormalities by binding the cations calcium and magnesium. Clinical features Hydrogen fluoride can cause severe systemic toxicity from even rela- tively small dermal exposures. Inhalation or ingestion of hydrogen fluoride causes severe corrosive damage similar to other acids (see ‘Acids’ section). Following absorption by whatever route, fluoride chelates calcium and lowers the serum ionized calcium concentra- tion and causes weakness, paraesthesiae, tetany, and convulsions. Hypotension and cardiac arrhythmias, including ventricular fibrilla- tion, may be observed. Central effects of fluoride include confusion, clouding of consciousness, and coma. Hepatic and renal failure may develop. Skin contact with anhydrous hydrogen fluoride produces li- quefactive necrosis and severe burns that are felt immediately. Concentrated aqueous solutions also cause an early sensation of pain but more dilute solutions may give no warning of injury. If the solution is not removed promptly, penetration of the skin by fluoride ion may occur, leading to painful ulcers that heal only slowly. Treatment Inhalation Following inhalation of hydrogen fluoride, the casualty should be removed immediately from the contaminated atmosphere. Further treatment is symptomatic and supportive. Mechanical ventilation with positive end-​expiratory pressure may be needed to treat pul- monary oedema. Ingestion If hydrofluoric acid has been ingested, management is as for other acids (see ‘Acids’). An intravenous injection of 10 ml of 10% calcium gluconate solution should be given. Eye and skin exposure Skin contact requires immediate thorough washing of the af- fected area, for 1 min, even if there is no apparent burn or pain. Contaminated clothing should be removed, with rescuers protecting their hands with suitable gloves. Skin burns should be coated repeat- edly with 2.5% calcium gluconate gel; the gel should be massaged continuously into the skin until at least 15 min after pain is relieved. The area should then be covered with a dressing soaked in the gel and lightly bandaged. Eye contact requires immediate thorough washing with water. An urgent opthalmological opinion should be sought. Hydrogen sulphide Hydrogen sulphide is a colourless gas that smells of rotten eggs, al- though high concentrations cause olfactory nerve paralysis. The gas

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1768 is also found in mines and sewers and is liberated from decomposing fish (a hazard in fishing boats if the hold is filled with ‘trash’ fish used for making fish meal) and liquid manure systems. The serious sequelae following exposure to high concentrations of hydrogen sulphide are due principally to inhibition of cytochrome oxidase a3, in which respect it is more potent than cyanide. Clinical features Exposure to low concentrations leads to blepharospasm, pain and redness of the eyes, blurred vision, and coloured haloes round lights. Headache, nausea, dizziness, drowsiness, sore throat, and cough may also occur. With exposure to higher concentrations, cyanosis, con- fusion, pulmonary oedema, coma, and convulsions are common. Death ensues in some 6% of cases. Treatment The casualty should be moved to fresh air from the contaminated atmosphere by a rescuer who has donned breathing apparatus be- forehand. Thereafter the treatment is symptomatic and supportive. Irritant gases (ammonia, chlorine, sulphur dioxide) Ammonia, chlorine, and sulphur dioxide are water soluble (forming ammonia water, hypochlorite and hydrochloric acid, and sulphurous acid, respectively) and, therefore, dissolve easily in the mucus of the upper airways. These compounds react with elements of the cell walls, resulting in release of mediators causing an inflammatory cascade that alters vascular permeability and acts as chemotactic factors. The altered vascular permeability may lead to influx of plasma that can decrease airway calibre, and consequently increase airway resistance. Clinical features Inhalation Following exposure to these gases, the clinical symptoms appear immediately and consist of lacrimation, nasal discharge, broncho- spasm, increased mucus production, and cyanosis. In more severe cases bronchospasm, bronchial oedema, glottal oedema, and in- creased mucus production may be present. Although pulmonary oedema is observed, it is never the sole presenting feature. Patients with chronic bronchitis or asthma are usually more susceptible. Eye and skin exposure Exposure of the skin and eyes to concentrated ammonia water (li- quid ammonia) may cause corrosive damage; evaporation of am- monia may cause extreme cooling when spilled on the skin or eyes; cold burns may result. Ingestion Ingestion of ammonia water induces severe caustic lesions of the mucous membranes of the oropharynx, oesophagus, and stomach. Oesophageal or gastric perforation may occur, causing mediastinitis or peritonitis, respectively. Treatment Inhalation The priority is to remove the casualty from exposure. Early inspec- tion of the upper airways is important because the mucosal mem- branes may be very oedematous in severe cases, precluding oral endotracheal intubation; tracheostomy or coniotomy (emergency airway puncture) may be necessary. Pulmonary complications should be treated with humidified supplemental oxygen, broncho- dilators and, if necessary, assisted ventilation with positive end-​ expiratory pressure. Early administration of corticosteroids for a few hours may be beneficial, though later administration offers no benefit. Prophylactic antibiotics have not been shown to be of value. Eye and skin exposure If there is no mucosal irritation of the eyes or nose, it can be con- cluded that the exposure was not severe and that the individual is not at risk of delayed pulmonary oedema, and there is no need to admit such patients to hospital for observation. If the eyes are affected, they should be irrigated with water or saline 0.9% for 15–​30 min and an ophthalmic opinion sought. Eye irrigation is facilitated by the use of a topical anaesthetic. If exposed, the skin should be irrigated with water. Ingestion Neutralizing agents should not be administered after ingestion of ammonia water because the resultant exothermic reaction may worsen the injury. Induction of vomiting and gastric lavage are contraindicated. Urgent endoscopy and CT should be performed to identify the damage and to insert a nasogatric tube under direct observation. If endoscopy is delayed in severe cases, the mucous membranes become very swollen, increasing the risk of perforation. Healing of oesophageal lesions is often accompanied by strictures. Systemic corticosteroids do not improve the outcome in patients with severe oesophageal lesions. Isopropanol Isopropanol is used as a sterilizing agent and as ‘rubbing alcohol’. It is also found in aftershave lotions, disinfectants, and window-​cleaning solutions. Intoxication can result from both ingestion and skin ab- sorption. The accidental use of an isopropanol-​containing enema has resulted in death. Isopropanol is oxidized in the liver to acetone. Clinical features The major features of severe poisoning are due to central nervous system and respiratory depression, shock, and circulatory collapse. The most common metabolic effects are an increased osmolal gap, ketonaemia, and ketonuria. There may also be the odour of acetone on the breath, gastritis, haematemesis, hypothermia, renal tubular necrosis, acute myopathy, and haemolytic anaemia. Treatment In addition to supportive measures, haemodialysis/​haemodialfiltra­ tion should be employed in severely poisoned patients as it removes isopropanol and acetone. No advantage is gained by administering ethanol or fomepizole to block alcohol dehydrogenase, because the toxicity of isopropanol is caused principally by the parent com- pound and not by acetone. Moreover, such treatment will enhance the toxicity of isopropanol. Methanol Methanol is used widely as a solvent. It is also found in antifreeze solutions, paints, duplicating fluids, paint removers and varnishes, and shoe polishes. The ingestion of as little as 10 ml of pure methanol has caused permanent blindness and 30 ml is potentially fatal, al- though individual susceptibility varies widely. Toxicity may also occur as a result of inhalation or percutaneous absorption.

10.4.1  Poisoning by drugs and chemicals 1769 Methanol is metabolized by alcohol dehydrogenase and catalase enzyme systems to formaldehyde and formate (Fig. 10.4.1.6). The concentration of formate increases greatly and is accompanied by ac- cumulation of hydrogen ions, causing metabolic acidosis. Clinical features Ingested alone, methanol causes mild and transient inebriation and drowsiness. After a latent period of 8–​36 h, nausea, vomiting, ab- dominal pain, headaches, dizziness, and coma supervene. Blurred vision and diminished visual acuity may occur and the presence of dilated pupils, unreactive to light, suggests that permanent blindness is likely to ensue. A severe metabolic acidosis may develop, and this may be accompanied by hyperglycaemia and raised serum amylase activity. A blood methanol concentration of more than 500 mg/​litre confirms serious poisoning. Mortality increases with the severity and duration of the metabolic acidosis. Survivors may show permanent neurological sequelae including blindness, rigidity, hypokinesis, and other parkinsonian-​like signs; these features follow the development of optic neuropathy and necrosis of the putamen. Treatment The treatment of methanol poisoning is directed towards the inhib- ition of methanol metabolism by the administration of fomepizole or ethanol (see the section ‘Diethylene glycol’), the correction of metabolic acidosis and the removal of circulating methanol and its toxic metabol- ites by haemodialysis or haemodialfiltration. Substantial quantities of bicarbonate may be required and since this must be accompanied by sodium, hypernatraemia, and hypervolaemia may result. If admission plasma concentrations show that the methanol in- gested has already been metabolized, fomepizole, or ethanol ad- ministration will not be of benefit and ethanol might exacerbate the acidosis. Dialysis is indicated when a patient has ingested more than 30 g of methanol, or develops metabolic acidosis, mental, visual, or fundoscopic abnormalities attributable to methanol, or a blood methanol concentration in excess of 500 mg/​litre. Folinic acid 50 mg (1 mg/​kg in children) intravenously 6-​hourly may protect against ocular toxicity by accelerating formate metabolism. Methylene chloride (dichloromethane) Methylene chloride is a common ingredient in paint removers and is used as a solvent for plastic films and cements and also as a degreaser and aerosol propellant. Exposures usually follow inhalation, though deliberate ingestion is recognized. Methylene chloride is metabol- ized to CO2 and carbon monoxide. Carboxyhaemoglobin concen- trations of 3–​10% (exceptionally 40%) are attained. Clinical features Skin contact with liquid methylene chloride can cause a chemical burn. Following inhalation, dizziness, tingling and numbness of the extremities, throbbing headache, nausea, irritability, fatigue, and stupor have been reported. Severe and prolonged exposure may lead to irritant conjunctivitis, lacrimation, and respiratory depression. Hepatorenal dysfunction and pulmonary oedema have also been described. Fatalities have occurred. If high concentrations of carboxyhaemoglobin are present, the features of acute carbon monoxide poisoning may also ensue, al- though these tend to be mild even in the presence of such high concentrations. Methylene chloride ingestion causes corrosive injury to the gastrointestinal tract, agitation, diaphoresis, and drowsiness with rapid progression to coma in severe cases. Consciousness is typically regained after several hours unless hypoxic encephalopathy ensues. Pancreatitis, hepatic dysfunction, and renal and respiratory failure are potential complications. Carboxyhaemoglobin concentrations may remain raised for days. Treatment Prompt removal from exposure prior to death usually results in complete recovery. Thereafter, treatment is supportive and should include the use of supplemental oxygen. Nitrites Volatile alkyl nitrites, for example, amyl and butyl (predominantly isobutyl) nitrite, are recreational drugs marketed as aphrodisiacs or ‘room odourizers’. They are alleged to improve sexual performance by enhancing and prolonging orgasm and/​or as a smooth muscle re- laxant to relax the anal sphincter. They also are claimed to promote a sense of increased well-​being with temporary detachment from reality. The intended route of exposure is inhalation, but they are oc- casionally also ingested, either accidentally or deliberately. Alkyl nitrites cause vasodilatation via nitric oxide mediated vas- cular smooth muscle relaxation. Vasodilatation accounts for many of the effects observed or described by users following abuse. More important toxicologically is the ability of these agents to oxidize fer- rous haem from the Fe2+ to the ferric (Fe3+) state, resulting in meth- aemoglobinaemia after substantial inhalation or ingestion. Clinical features These reflect vasodilation with headache, flushing, blurred vision, postural hypotension, and syncope, followed by reflex vasoconstric- tion with sinus tachycardia. With continued exposure, methaemo- globinaemia results. Irritant effects including burning in the nose and eyes; cough and facial dermatitis are recognized; and transient ECG changes (T wave inversion and ST segment depression) have been reported. Methaemoglobin concentrations less than 20% are usually asymptomatic though they cause slate-​grey ‘cyanosis’ due predomin- antly to the presence of pigmented methaemoglobin. When 20–​40% total haemoglobin is replaced by methaemoglobin, there may be diz- ziness and headache, features not dissimilar to those caused by vaso- dilatation. Higher methaemoglobin concentrations reflect increasing tissue hypoxia and are unusual following volatile nitrite abuse, unless inhalation is substantial or ingestion has occurred. However, in these circumstances, life-​threatening methaemoglobinaemia may result. Treatment The vasodilatory effects of volatile nitrite abuse are not usually se- vere and can be managed supportively. In healthy adults, meth- aemoglobin concentrations less than 30% total haemoglobin are unlikely to warrant specific treatment. At higher methaemoglobin Methanol Formaldehyde Ethanol Fomepizole Formate CO2 + H2O 10-FTS FDH ADH Fig. 10.4.1.6  Metabolism of methanol. ADH, alcohol dehydrogenase; FDH, formaldehyde dehydrogenase; 10-​FTS, 10-​formyltetrahydrofolate synthetase.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1770 concentrations, or where clinical features suggest tissue hypoxia, antidotal therapy with intravenous methylthioninium chloride (methylene blue) 1–​2 mg/​kg body weight as a 1% solution should be given over 5–​10 min. Treatment is effective within 30 min and a second dose is required rarely. Nitrogen dioxide Nitrogen dioxide is the most toxic nitrogen oxide and causes hyp- oxic asphyxia by displacing oxygen. It dissolves poorly in water and therefore penetrates deeper into the lung (i.e. to the alveoli and ter- minal bronchioles). The ciliated cells of the bronchioles and the al- veolar type I cells are especially susceptible to injury. Following the alveolar damage, an influx of plasma and inflammatory cells occurs, causing acute lung injury. Clinical features The clinical features following acute exposure depend on the con- centration and duration of exposure to the gas. Modest acute ex- posure (<50 ppm) for a short time often produces no immediate symptoms, although throat irritation, cough, transient choking, tightness in the chest, and sweating have been observed. By contrast, exposure to a massive concentration of nitrogen dioxide, such as that found in a silo, can produce severe and immediate hypoxaemia, which may be fatal. As symptoms can be absent during the first hours after exposure, physical examination of the patient immediately after exposure may not provide information regarding the full extent of the clinical se- verity of the intoxication. However, if 6 h after exposure, the patient has normal arterial blood gases and chest X-​ray, there is little like- lihood that life-​threatening lung damage will develop. Patients can then be discharged with instructions that they must undergo med- ical observation again if increased dyspnoea occurs after discharge. Bronchiolitis obliterans may develop within 2–​6 weeks. Treatment Adequate supportive therapy such as supplemental oxygen and bronchodilators should be given. Early administration of cortico- steroids for a few hours may be beneficial, though later adminis- tration offers no benefit. Use of prophylactic systemic antibiotics is not recommended because of the increased risk of infection with resistant organisms. When intubation is required, the largest prac- ticable tube should be introduced to allow adequate bronchial toilet. Assisted ventilation with positive end-​expiratory pressure offers the best hope of reducing the mortality. Paraffin oil (kerosene) Paraffin oil has three physical properties accounting for its tox- icity. Its low viscosity and surface tension allow it to spread rapidly throughout the lungs when aspirated after ingestion, and its low va- pour pressure makes it unlikely to cause poisoning by inhalation. Clinical features Repeated local application to the skin results in dryness, dermatitis, and, rarely, epidermal necrolysis. Paraffin ingestion causes a burning sensation in the mouth and throat, vomiting, diarrhoea, and ab- dominal pain. Pulmonary features may occur within 1 h of ingestion with cough, tachypnoea, tachycardia, basal crackles, and cyan- osis. Non​segmental consolidation or collapse is seen radiologic- ally. Pneumatocoele formation, pneumothorax, pleural effusion, or pulmonary oedema may occur. Other complications include hepatic dysfunction and, in severe cases, atrial fibrillation and ventricular fibrillation. Treatment Gastric lavage and emesis should be avoided because of the increased risk of chemical pneumonitis. There is no evidence that corticoster- oids and antibiotics reduce morbidity or mortality; mechanical ven- tilation with positive end-​expiratory pressure may be necessary in severe cases of aspiration. Petrol (gasoline) Petrol is a complex mixture of volatile hydrocarbons containing a small proportion of non​hydrocarbon additives. Clinical features Following the inhalation of petrol, dizziness, and irritation of the eyes, nose and throat may occur within 5 min followed by euphoria, headache, and blurred vision. If inhalation continues, or if significant quantities of petrol are ingested, then excitement and depression of the nervous system occurs; incoordination, restlessness, excitement, confusion, disorientation, hallucinations, ataxia, nystagmus, tremor, delirium, coma, and convulsions may be seen. Inhalation of high concentrations of petrol may cause immediate death, probably from ventricular fibrillation or respiratory failure. Chemical pneumonitis may occur as in paraffin oil ingestion (see ‘Paraffin oil (kerosene)’) and the clinical features and management are then identical. Treatment Following removal from exposure, supportive measures provide the basis of treatment. Phenol Phenol (‘carbolic acid’) is nearly always recognizable by its odour and, distinctively, the pain to which it gives rise is much less than might be expected. This is due to its ability to damage afferent nerve endings. Clinical features If phenol is spilt on the skin, pain is followed promptly by numb- ness. The skin becomes blanched, and a dry opaque eschar forms over the burn. When the eschar sloughs off, a brown stain remains. Phenol penetrates intact skin rapidly and is well absorbed through the lungs. After ingestion, nausea, vomiting, and abdominal pain result. Depending on the concentration of the solution corrosive injury may result in bleeding, perforation, and subsequent stricture formation. Systemic toxicity may follow exposure by any route. Signs of sys- temic toxicity include sweating, headache, tinnitus, and dizziness. An initial rise in blood pressure is followed by hypotension precipi- tated by phenol-​induced loss of vasoconstrictor tone. Loss of con- sciousness, respiratory depression, coma, seizures, and shock follow, which may result in renal failure. An initial phase of central nervous system stimulation, and rarely convulsions, has sometimes been ob- served in children. Phenol poisoning is associated with grey or black urine and though this is due in part to metabolites of phenol, Heinz body haemolytic anaemia, as well as methaemoglobinaemia and hyperbilirubinaemia, contribute.

10.4.1  Poisoning by drugs and chemicals 1771 Treatment Fluid resuscitation and prompt assessment of the extent of corro- sive damage is crucial following ingestion. Management is otherwise supportive. Skin and eye contamination, renal failure and meth- aemoglobinaemia are managed conventionally. Phosgene Phosgene is used in the synthesis of isocyanates, polyurethane and polycarbonate resins, and dyes. It is also produced in fires and has been used as a chemical weapon. When combined with water, phosgene produces hydrogen chloride and carbon dioxide, although as the gas is poorly soluble in water, only small amounts of hydrochloric acid are produced under normal physiological conditions. It is thought that this is only relevant in causing mucus membrane and eye symptoms when phosgene is present at relatively high concentrations. Biologically, acylation and free radical-​mediated reactions occur between phos- gene and important cellular constituents. Acylation reactions involving phosgene occur with biological molecules containing sulfhydryl, amino and hydroxyl moieties. Clinical features Exposure to phosgene causes irritation of the eyes, dryness or burning sensation in the throat, cough, chest pain, and nausea and vomiting. There is usually a latent period lasting between 30 min and 24 h (rarely, 72 h) during which the casualty suffers little discomfort and has no abnormal chest signs. Subsequently, pulmonary oedema develops due to increased capillary permeability; circulatory col- lapse may follow. Treatment Treatment is supportive. There are experimental data to sug- gest that an intravenous bolus of high-​dose corticosteroid (e.g. methylprednisolone 1 g) < 6h after exposure may be of benefit. There is no benefit from nebulized steroid even when administered 1 h after exposure. Consideration should also be given to administra- tion of nebulized acetylcysteine 1–​2 g, though there is no substantive evidence of benefit outside a small animal, isolated lung model. If the oxygen saturation falls below 94%, patients should receive the lowest concentration of supplemental oxygen to maintain their SaO2 in the normal range. Once patients require oxygen, nebulized β-​agonists (e.g. salbutamol (albuterol) 5 mg by nebulizer every 4 h) may reduce lung inflammation if administered within 1 h of exposure. Elective intubation should be considered early using an ARDSNet protective ventilation strategy. Propylene glycol Propylene glycol is used widely as a preservative and solvent for oral, intravenous, and topical medications. It is oxidized to lactic acid and pyruvate via hepatic alcohol and aldehyde dehydrogenases in a similar way to the metabolism of other glycols such as ethylene glycol. Clinical features The ingestion of substantial quantities of propylene glycol or its ad- ministration to neonates, those in renal failure, or in exceptionally large doses (such as patients requiring massive parenteral doses of propylene glycol-​containing benzodiazepines in the management of acute alcohol withdrawal) may cause convulsions, coma, cardiac arrhythmias, hepatorenal damage, intravascular haemolysis, meta- bolic acidosis, and increased serum osmolality. Treatment Metabolic acidosis, renal failure, and respiratory depression should be treated conventionally. Haemodialysis removes propylene glycol efficiently. Ethanol or fomepizole may be used to inhibit propylene glycol metabolism in a similar way to their use in ethylene glycol poisoning, but in practice the diagnosis is often not made until a sig- nificant acidosis is present and thus it is too late for antidotal treat- ment to be useful. Tetrachloroethylene Tetrachloroethylene is used widely as an industrial solvent, par- ticularly for dry-​cleaning and degreasing. Poisoning may occur by inhalation or ingestion. A considerable proportion of an inspired dose is exhaled unchanged, and that retained is excreted only slowly (half-​life c.144 h), mainly by metabolism to trichloroacetic acid, the major urine metabolite. Clinical features Following inhalation or ingestion, there is depression of the central nervous system; nausea and vomiting may occur and persist for sev- eral hours. Irritation of the eyes, nose, and throat may occur from direct exposure. Hepatic and renal dysfunction may also develop and ventricular arrhythmias and non​cardiogenic pulmonary oe- dema have been reported. Treatment After removal from exposure, treatment is supportive and symptomatic. Toluene Toluene has much lower volatility and toxicity than benzene. It is used extensively as a solvent in the chemical, rubber, paint, glue, and pharmaceutical industries and as a thinner for inks, perfumes, and dyes. Following inhalation or ingestion, toluene is oxidized to ben- zoic acid, then to hippuric acid benzoylglucuronates, which are ex- creted in the urine. Clinical features Acute poisoning results in euphoria, excitement, dizziness, confu- sion, increased lacrimation, headache, nervousness, nausea, tin- nitus, ataxia, tremor, and coma. A review of adults who had abused toluene indicated three major patterns of presentation: (1) muscle weakness, (2) gastrointestinal complaints (abdominal pain, haematemesis), and (3) neuropsychi- atric disorders (altered mental status, cerebellar abnormalities, per- ipheral neuropathy). In addition, hypokalaemia, hypophosphataemia and hyperchloraemia were common. Rhabdomyolysis occurred in 40% of cases. Distal renal tubular acidosis and urinary calculi were also reported. Cardiac and haematological toxicity due to toluene appears to be uncommon. Treatment If poisoning results from inhalation, whether accidental or in- tentional, the patient should be removed from the contaminated

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1772 environment. Thereafter, treatment consists of symptomatic and supportive measures. 1,1,1-​Trichloroethane 1,1,1-​Trichloroethane is a liquid of high volatility used as a solvent in industry. Most of an inhaled dose is expired unchanged, though small amounts of trichloroacetic acid and trichloroethanol are formed. Concomitant ingestion of ethanol is known to enhance toxicity. Clinical features Following inhalation of a sufficiently large dose, central nervous system depression occurs; hepatic and renal dysfunction may also result. Deaths have followed exposure to very high concentrations in unventilated tanks. In such cases, death may either be due to cen- tral nervous system depression, culminating in respiratory arrest, or to fatal arrhythmias as a result of myocardial sensitization to circu- lating catecholamines in the presence of hypoxia. Treatment The casualty should be removed from the contaminated environ- ment. Thereafter treatment is symptomatic and supportive. Trichloroethylene Trichloroethylene is a volatile liquid used as an industrial solvent, particularly in metal degreasing and extraction processes. Following exposure, it is excreted unchanged in the breath and metabolized via chloral hydrate to trichloroethanol and trichloroacetic acid, which are excreted in the urine. Clinical features Following exposure by any route, central nervous system depres- sion occurs with nausea and vomiting, hepatic and renal dysfunc- tion, cranial nerve damage, cerebellar dysfunction, and convulsions have been described. ‘Degreaser’s flush’ (in which the skin of the face and arms becomes markedly reddened) may occur if ethanol is consumed shortly before or after exposure to trichloroethylene. Treatment The casualty should be removed from the contaminated environ- ment. Thereafter treatment is symptomatic and supportive. Household products Automatic dishwashing tablets The traditional tablets for automatic dishwashing machines are contained within an external wrapper that requires removal prior to loading the enclosed tablet into the machine. Soluble film auto- matic dishwashing tablets are enclosed by a water-​soluble poly- vinyl alcohol film and are loaded straight into the dishwashing machine, unlike their traditional counterparts. However, the in- tegrity of the soluble film can be compromised and the contents of the tablet can be released prematurely when in contact with moist hands or saliva. The tablets most commonly contain a source of hydrogen per- oxide (often as sodium percarbonate) and non​ionic surfactants. Other constituents in some formulations include sodium carbonate, sodium tripolyphosphate, and sodium silicate, which reduce water hardness. The pH once dissolved in water is alkaline. Clinical features Toxicity from hydrogen peroxide occurs as a result of its corro- sive effects and release of oxygen causing embolism. Ingestion may cause irritation of the gastrointestinal tract with nausea, vomiting, foaming at the mouth, paraesthesia around the mouth, blistering in the mouth, stomatitis, mouth bleeding, laryngitis, pharyngitis, and haematemesis. The foam may then obstruct the respiratory tract re- sulting in stridor or pulmonary aspiration. Sodium carbonate in- gestion has led to stridor, drooling, coughing, and oedematous lips. Treatment If several tablets have been ingested, upper gastrointestinal endos- copy and CT oesophagus should be considered. Batteries Batteries are usually swallowed accidentally. They contain metal salts, usually of nickel, lithium, cadmium, manganese, zinc, or silver. Code numbers on button batteries will help identify the content. While most batteries are passed without complications, problems may arise if the battery becomes lodged, particularly in the oe- sophagus, where local necrosis, bleeding, and perforation are po- tential complications. Fatalities have occurred in young children following button battery ingestion, although toxicity from metal content is a potential risk, in reality this is vanishingly rare. Clinical features These relate to whether or not the battery lodges or leaks. Symptoms are uncommon. Lodged batteries may perforate within 2–​4 h, and delayed bleeding is a further hazard after battery removal. Treatment A chest X-​ray is necessary in those who have swallowed a battery, and those lodged in the oesophagus must be removed as soon as possible. Further X-​rays 48 h later may be needed if batteries have not passed per anus. If symptoms develop, particularly of GI bleeding or obstruction, urgent surgical intervention is required, if possible by endoscopy. Bleaches Most household bleaches contain sodium hypochlorite, but some chlorine–​free bleaches contain 6% hydrogen peroxide. Chlorine is not released from bleach solutions in appreciable amounts under normal use conditions. However, the mixing of hypochlorite with acids (e.g. when cleaning the toilet bowl) can result in the substantial release of chlorine and lung injury. Mixing bleach with ammonia produces chloramine compounds (mainly monochloramine) which can produce severe chemical pneumonitis. Clinical features Following the ingestion of weak concentrations of sodium hypo- chlorite (<5%), symptoms are usually mild. With stronger bleaches, particularly of over 10% sodium hypochlorite, features are more se- vere. Small amounts cause a sensation of burning. Larger doses cause nausea, retching, vomiting, diarrhoea and, rarely, haematemesis. In

10.4.1  Poisoning by drugs and chemicals 1773 severe cases a hypernatraemic hyperchloraemic acidosis, hypoten- sion, coma, convulsions, and cardiorespiratory arrest can occur. The gastrointestinal mucosa may become haemorrhagic, ulcerated, and perforated. For the features of hydrogen peroxide, please see the earlier section ‘Automatic dishwashing tablets’. Treatment If concentrated bleach (>10% sodium hypochlorite) or hydrogen peroxide-​containing bleaches are ingested, upper gastrointes- tinal endoscopy and CT should be considered. Detergents Liquid laundry detergent capsules (also called single-​use detergent sacs; laundry pods) are a pouch of concentrated liquid laundry de- tergent in a water-​soluble polyvinyl alcohol membrane that can be placed directly in washing machines. In Europe, these liquid detergents most commonly contain anionic surfactants (20–​35% per capsule), non​ionic surfactants (10–​20%), propylene glycol (8–​20%) and ethanol (2–​5%), and have a pH of 7–​9. The capsules are designed to release their contents when they come into contact with water and this can happen prematurely if they come into con- tact with moisture (e.g. in the hands or mouth). Clinical features As a result, there have been a substantial number of exposures to laundry liquid detergent capsules, predominantly involving children less than 5 years of age. Although most patients remain asymptom- atic or suffer only minor features, a small proportion develop fea- tures such as central nervous system depression, stridor, pulmonary aspiration and/​or airway burns following ingestion and conjunctiv- itis leading to corneal ulceration from eye exposure. Treatment Treatment is symptomatic and supportive. Disinfectants Disinfectants are antimicrobial agents that contain chlorophenol or chloroxylenols (dichlorometaxylenol and parachlorometaxylenol), quarternary ammonium compounds (such as benzalkonium chloride, cetyl trimethylammonium bromide, cetylpyridinium chloride, and benzethonium chloride). Sodium hypochlorite and hydrogen peroxide (see bleaches) are also effective disinfectants because they release chlorine and oxygen, respectively, which oxi- dizes the cell membrane of microorganisms. Clinical features Chloroxylenols and chlorophenol cause a burning sensation in the mouth and throat, vomiting, coma, hypothermia, hypotension, and respiratory depression. Metabolic acidosis and bradycardia can occur and aspiration pneumonia and pulmonary oedema have been reported; ingestion of a large quantity can cause renal impairment. Concentrated solutions of quarternary ammonium compounds can cause immediate burning pain in the mouth, throat, and ab- domen, hypersalivation, and ulceration of mucous membranes, followed by vomiting, haematemesis, diarrhoea, and confusion. In severe cases there can be hypotension, shock, convulsions, respira- tory paralysis, and coma. Metabolic acidosis and increased liver enzyme activities can occur. Dermal burns have been reported and eye exposures can cause corneal damage. Treatment Treatment is symptomatic and supportive. FURTHER READING Epidemiology Bateman DN, et al. (2006). Legislation restricting paracetamol sales and patterns of self-​harm and death from paracetamol-​containing preparations in Scotland. Br J Clin Pharmacol, 62, 573–​81. Dart RC, et al. (2015). Poisoning in the United States: 2012 emergency medicine report of the national poisons data system. Ann Emerg Med, 65, 416e22. Martins SS, et al. (2015). Worldwide prevalence and trends in unin- tentional drug overdose: a systematic review of the literature. Am J Public Health, 105, e29e49. McCarthy M (2015). Drug overdose has become leading cause of death from injury in US. Br Med J, 350, h3328. Morrison EE, Dear JW, Sandilands EA (2015). Self-​poisoning in the elderly: a 10-​year observational study. Clin Toxicol, 53, 284e5. Immediate treatment Barceloux D, et al. (2004). Position paper: cathartics. Clin Toxicol, 42, 243–​53. Benson B E, et al. (2013). Position paper update: gastric lavage for gastrointestinal decontamination. Clin Toxicol, 51, 140–​6. Chyka PA, et al. (2005). Position paper: single-​dose activated charcoal. Clin Toxicol, 43, 61–​87. Höjer J, et al. (2013). Position paper update: ipecac syrup for gastro- intestinal decontamination. Clin Toxicol, 51, 134–​39. Thanacoody R, et al. (2015). Position paper update: whole bowel ir- rigation for gastrointestinal decontamination of overdose patients. Clin Toxicol, 53, 5–​12. Methods to increase poison elimination Eddleston M, et  al. (2008). Multiple-​dose activated charcoal in acute self-​poisoning: a randomised controlled trial. Lancet, 371, 579–​87. Kay TD, Playford HR, Johnson DW (2003). Hemodialysis versus con- tinuous veno–​venous hemodiafiltration in the management of se- vere valproate overdose. Clin Nephrol, 59, 56–​8. Proudfoot AT, Krenzelok EP, Vale JA (2004). Position paper on urine alkalinization. Clin Toxicol, 42, 1–​26. Vale JA, et al. (1999). Position statement and practice guidelines on the use of multi-​dose activated charcoal in the treatment of acute poi- soning. Clin Toxicol, 37, 731–​51. Drugs Angiotensin-​converting enzyme (ACE) inhibitors Prasa D, et al. (2013). Angiotensin II antagonists—​an assessment of their acute toxicity. Clin Toxicol, 51, 429–​34. Antibacterial agents Dharnidharka VR, et al. (1998). Ciprofloxacin overdose: acute renal failure with prominent apoptotic changes. Am J Kidney Dis, 31, 710–​12. Holdiness MR (1989). A review of the red man syndrome and rifam- picin overdose. Med Toxicol Adverse Drug Exp, 4, 444–​51. Jones DP, et al. (1993). Acute renal failure following amoxycillin over- dose. Clin Pediatr, 32, 735–​9.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1774 Rezkalla MA, Pochop C (1994). Erythromycin induced torsades de pointes: case report and review of the literature. S D J Med, 47,161–​4. Vannaprasaht S, et  al. (2006). Ceftazidime overdose-​related non­ convulsive status epilepticus after intraperitoneal instillation. Clin Toxicol, 44, 383–​6. Anticoagulants Bateman DN, Page CB (2016). Antidotes to coumarins, isoniazid, methotrexate and thyroxine; toxins that work via metabolic pro- cesses. B J Clin Pharm, 81, 437–​45. Pollack CV, et al. (2015). Idarucizumab for dabigatran reversal. N Eng J Med, 373, 511–​20. Anticonvulsants Vale A (2016). Anticonvulsants. Medicine GBR, 44, 133–​4. Antidepressants Bateman DN (2005). Tricyclic antidepressant poisoning: central ner- vous system effects and management. Toxicol Rev, 24, 181–​6. Bradberry SM, et al. (2005). Management of the cardiovascular com- plications of tricyclic antidepressant poisoning:  role of sodium
bicarbonate. Toxicol Rev, 24, 195–​204. Isbister GK, et al. (2004). Relative toxicity of selective serotonin re- uptake inhibitors (SSRIs) in overdose. Clin Toxicol, 42, 277–​85. Thanacoody HK, Thomas SH (2005). Tricyclic antidepressant poi- soning: cardiovascular toxicity. Toxicol Rev, 24, 205–​14. van Gorp F, et al. (2012). Population pharmacokinetics and pharma- codynamics of escitalopram in over-​dose and the effect of activated charcoal. Br J Clin Pharmacol, 73, 402–​10. Whyte IM, Dawson IM, Buckley NA (2003). Relative toxicity of venlafaxine and selective serotonin reuptake inhibitors in overdose compared to tricyclic antidepressants. Q J Med, 96, 369–​74. Antiabetic agents Darracq MA, et al. (2014). A retrospective review of isolated gliptin-​ exposure cases reported to a state poison control system. Clin Toxicol, 52, 226–​30. Glatstein M, Scolnik D, Bentur Y (2012). Octreotide for the treatment of sulfonylurea poisoning. Clin Toxicol, 50, 795–​804. von Mach MA, et al. (2006). Antidiabetic medications in overdose: a comparison of the inquiries made to a regional poisons unit re- garding original sulfonylureas, biguanides and insulin. Int J Clin Pharmacol Ther, 44, 51–​6. Antihistamines Pragst F, Sieglinde H, Bakdash A (2006). Poisonings with diphenhydramine—​a survey of 68 clinical and 55 death cases. Forens Sci Int, 161, 189–​97. Antimalarials Langford NJ, et al. (2003). Quinine intoxication reported to the Scottish Poisons Information Bureau 1997–​2002: a continuing problem. Br J Clin Pharmacol, 56, 576–​8. Mégarbane B, et al. (2010). Blood concentrations are better predictors of chloroquine poisoning severity than plasma concentrations:  a prospective study with modelling of the concentration/​effect rela- tionships. Clin Toxicol, 48, 904–​15. Antipsychotics Burns MJ (2001). The pharmacology and toxicology of atypical anti- psychotic agents. Clin Toxicol, 39, 1–​14. Glassman AH, Bigger Jr JT (2001). Antipsychotic drugs: prolonged QTc interval, torsade de pointes, and sudden death. Am J Psychiatry, 158, 1774–​82. Isbister GK, et  al. (2010). Amisulpiride overdose is frequently as- sociated with QT prolongation and torsade de pointes. J Clin Psychopharmacol, 30, 391–​5. Strachan EM, Kelly CA, Bateman DN (2004). Electrocardiogram and cardiovascular changes in thioridazine and chlorpromazine poi- soning. Eur J Clin Pharmacol, 60, 541–​5. Benzodiazepines Höjer J, Baechrendtz S, Gustafsson L (1989). Benzodiazepine poi- soning:  experience of 702 admissions to an intensive care unit during a 14-​year period. J Int Med, 226, 117–​22. Penninga E, et al. (2016). Adverse events associated with flumazenil treatment for the management of suspected benzodiazepine intoxication—​a systematic review with meta-​analyses of random- ised trials. Basic & Clinical Pharmacology & Toxicology, 118, 37–​44. β-​Blockers Bailey B (2003). Glucagon in β-​blocker and calcium channel blocker overdoses: a systematic review. Clin Toxicol, 41, 595–​602. β2-​Adrenoceptor agonists Lewis LD, et al. (1993). A study of self-​poisoning with oral salbutamol—​ laboratory and clinical features. Hum Exp Toxicol, 12, 397–​401. Vale A (2016). β2-​Agonists. Medicine GBR, 44, 146. Bismuth chelate (tripotassium dicitratobismuthate) Slikkerveer A, et al. (1998). Comparison of enhanced elimination of bis- muth in humans after treatment with meso-​2,3-​dimercaptosuccinic acid and d,l-​2,3-​dimercaptopropane-​1-​sulfonic acid. Analyst, 123, 91–​2. Stevens PE, et al. (1995). Significant elimination of bismuth by haemo- dialysis with a new heavy-​metal chelating agent. Nephrol Dial Transplant, 10, 696–​8. Calcium channel blockers Graudins A, Lee HM, Druda D (2016). Calcium channel antagonist and beta-​blocker overdose: antidotes and adjunct therapies. Br J Clin Pharmacol, 81, 453–​61. Dapsone Kim Y-​J, et al. (2016). Difference of the clinical course and outcome between dapsone-​induced methemoglobinemia and other toxic-​ agent-​induced methemoglobinemia. Clin Toxicol, 54, 581–​4. Digoxin and digitoxin Chan BS, et  al. (2016). Efficacy and effectiveness of anti-​digoxin antibodies in chronic digoxin poisonings from the DORA study (ATOM-​1). Clin Toxicol, 54, 488–​94. Roberts DM, et al. (2016). Pharmacological treatment of cardiac glyco- side poisoning. Br J Clin Pharmacol, 81, 488–​95. Diuretics Lip GY, Ferner RE (1995). Poisoning and anti-​hypertensive drugs: di- uretics and potassium supplements. J Hum Hypertens, 9, 295–​301. Iron Chang TP, Rangan C (2011). Iron poisoning: a literature-​based review of epidemiology, diagnosis and management. Pediatr Emerg Care, 27, 978–​85. Tenenbein M (1996). Benefits of parenteral deferoxamine for acute iron poisoning. Clin Toxicol, 34, 485–​9. Isoniazid Bateman DN, Page CB (2016). Antidotes to coumarins, isoniazid, methotrexate and thyroxine; toxins that work via metabolic pro- cesses. B J Clin Pharm, 81, 437–​45.

10.4.2 Injuries, envenoming, poisoning, and allerg

10.4.2 Injuries, envenoming, poisoning, and allergic reactions caused by animals 1778

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1778 10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals David A. Warrell ESSENTIALS Mechanical injuries Attacks by wild and domesticated animals are increasing worldwide. They are best prevented by taking local advice about minimizing ex- posure. Injuries usually occur in places remote from medical care. They may involve extensive trauma, haemorrhagic shock, and a high risk of bacterial contamination. First aid consists of resuscitation, control of bleeding and perforating injuries, intravenous fluid re- placement, and rapid evacuation to hospital for emergency surgery and treatment of infection. Venomous snakes Bites by venomous snakes can cause death or permanent physical and mental morbidity. Snakebite is largely an occupational/​environ- mental hazard of agricultural workers and their children in rural areas of the tropics. Bites are commonly inflicted on the lower limbs and could be prevented by wearing protective footwear, by using a light and prodding with a stick while walking at night, and by sleeping off the ground or under a mosquito net. Snake venoms are complex mixtures of toxic proteins causing ne- crosis, shock, haemostatic disturbances, paralysis, rhabdomyolysis, and acute kidney injury. Envenoming by Elapidae (cobras, kraits, mambas, coral snakes, Australian snakes, and sea snakes) can cause descending flaccid paralysis, starting with ptosis and progressing to respiratory paralysis. Some elapid venoms cause local necrosis, rhabdomyolysis, and haemostatic disturbances. Bites by Viperidae (vipers, adders, and pit vipers—​rattlesnakes, moccasins, lanceheads) can cause severe local swelling, bruising, blistering, and necrosis together with shock, consumption coagulopathy, spontaneous systemic bleeding, acute kidney injury, and, with some species, neuromyotoxicity. First aid involves reassurance, immobilization of the whole pa- tient, especially the bitten limb, rapid evacuation to the nearest hos- pital, and avoidance of dangerous traditional methods. When the necessary skills and equipment are available, pressure-​pad immo- bilization should be applied immediately unless the possibility of a neurotoxic elapid bite can be excluded. In hospital, specific antivenom (hyperimmune equine or ovine immunoglobulins) is given if there is evidence of systemic or se- vere local envenoming. Polyspecific antivenoms cover envenoming by medically important snakes in the geographical area for which they are intended. Early anaphylactic or pyrogenic reactions and late serum sickness antivenom reactions are common but not predict- able by hypersensitivity tests. Incidence of severe early reactions is reduced by prophylactic low-​dose subcutaneous adrenaline. After the initial dose of antivenom, the indication for repeated dosage is failure of restoration of blood coagulability after 6 h, or progression of other signs of envenoming. Assisted ventilation, renal dialysis, and cardiovascular support may be required. Necrotic tissue requires surgical debridement. Signs of compartment syndrome may be misleading and fasciotomy is almost never justified. Venomous fish Many fish of temperate and tropical seas can inflict dangerous stings—​stingrays, catfish, weevers, scorpionfish, stonefish, and lionfish. Prevention is by wearing foot protection when wading and avoiding contact with tropical reef fish. Immediate agonizing pain is alleviated by immersing the stung limb in uncomfortably hot but not scalding water (less than 45°C). Erythematous swelling and necrosis may develop with the risk of infection by marine bacteria. Stingray spines can cause fatal penetrating injuries. Systemic envenoming is uncommon. Stonefish antivenom is available. Poisonous aquatic animals Ciguatera poisoning from eating tropical reef fish is prevalent in Pacific and Caribbean regions. Fish acquire polyether toxins from dinoflagellates. Acute gastroenteritis develops 1–​6 h after ingestion, followed by neurotoxic and cardiovascular disturbances, notably persistent paraesthesiae and myalgias. Tetrodotoxin poisoning is attributable to the Japanese delicacy ‘fugu’ (puffer fish). Neurotoxic symptoms caused by this sodium channel blocker develop 10–​45 min after ingestion. Fatal respiratory paralysis may ensue 2–​6 h later. Scombroid poisoning results when bacterial decompos- ition of tuna and other dark-​fleshed fish generates histamine. Anaphylactic-​type symptoms develop within minutes to a few hours after ingestion. Paralytic shellfish poisoning is caused by eating bivalve molluscs contaminated with tetrahydropurine neurotoxins from dinofla- gellates whose mass blooming manifests as ‘red tide’. Neurotoxic symptoms appear within 30 min of ingestion, progressing to fatal respiratory paralysis within 12 h. Prevention is by not eating scaleless (tetrodotoxic) fish at any time, very large (ciguatera-​toxic) reef fish, and shellfish when there is a red tide. Correct processing prevents scrombroid poisoning. Cooking does not destroy any of these toxins. Venomous marine invertebrates Cnidarians (jellyfish, stinging corals, sea anemones, and so on) have tentacles studded with stinging nematocysts. Lethal species are Indo-​Australian box jellyfish, Irukandji, Portuguese man-​o’-​ war (Physalia), and Chinese Stomolophus nomurai. Prevention is by observing warning notices on affected beaches, bathing in ‘stinger-​ resistant’ enclosures, or wearing protective clothing. Stings produce immediately painful irritant weals. Box jellyfish cause the most se- vere systemic symptoms: respiratory and cardiac arrest, generalized convulsions, and pulmonary oedema within minutes of the acci- dent. ‘Irukandji’ syndrome is distinctive: severe persisting musculo- skeletal pain, anxiety, trembling, headache, piloerection, sweating, tachycardia, hypertension, and pulmonary oedema starting about 30 min after stings by tiny cubomedusoids. Vinegar inactivates box jellyfish and Irukandji nematocysts. Hot water relieves the pain of Physalia and box jellyfish stings. Box jellyfish antivenom is available in Australia.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1779 Echinoderm (starfish and sea urchin) spines become embedded in waders’ feet, sometimes penetrating bones and joints. Pain is re- lieved by hot water. Systemic envenoming is rare but there is a risk of marine bacterial infection. Molluscs—​cone shells and small Australasian blue-​ringed octo- puses can cause fatal envenoming. Venomous arthropods Hymenoptera—​stings by bees (Apidae); wasps, yellow jackets, and hornet (Vespidae), and ants commonly cause allergic reactions, while rare mass attacks (e.g. by Africanized ‘killer’ bees) can result in fatal direct envenoming. People in whom systemic anaphylaxis has been provoked by a hymenopteran sting should always carry—​and be competent to use—​self-​injectable adrenaline (epinephrine). Desensitization with purified venom should be considered if type I hypersensitivity is con- firmed by detecting venom-​specific IgE. Massive envenoming by Apidae or Vespidae causes histamine toxicity, generalized rhabdo- myolysis, intravascular haemolysis, hypertension, pulmonary oe- dema, myocardial damage, bleeding, hepatic dysfunction, and acute kidney injury. Lepidoptera—​stinging hairs of many species of moths and their caterpillars can excite cutaneous irritation and allergy, sometimes causing epidemics. In South America, caterpillars of atlas moths (Lonomia) cause many stings. Their venom contains antihaemostatic toxins causing spontaneous bleeding, polyarthralgia, and acute kidney injury. An antivenom is available in Brazil. Coleoptera—​contact with ‘Spanish fly’ and ‘Nairobi eye’ beetles causes blistering. Scorpions—​stings still cause numerous fatalities in North and South Africa, the Middle East, Mexico, Latin America, and India. Prevention is by excluding scorpions from homes. Severe local pain is the commonest symptom. Systemic symptoms vary ac- cording to the species of scorpion involved. ‘Autonomic storm’ is caused by massive release of acetylcholine and catecholamines by ion channel toxins. Cardiorespiratory effects include hyperten- sion, shock, tachy-​, and bradyarrhythmias, electrocardiographic changes, and pulmonary oedema. Neurotoxic effects include er- ratic eye movements, fasciculation, and muscle spasms (pseudo-​ convulsions) causing respiratory distress. Pain is best controlled by digital block with local anaesthetic. Antivenom is available in some countries, but pharmacological treatment with prazosin and other vasodilators is preferred elsewhere. Spiders—​bites are common in the Americas, Mediterranean, South Africa, and Australia but there are few fatalities. Only recluse spiders (Loxosceles) are reliably associated with necrotic araneism (arachnidism), but many innocent peridomestic species have been vilified. Local pain and swelling develop slowly, followed by the classic ‘red-​white-​and-​blue sign’ and eventually an eschar, which sloughs leaving a necrotic ulcer. Systemic symptoms, including fever, rash, haemolysis, and acute kidney injury, are unusual. Bites by cosmopol- itan black and brown widow spiders, Latin American banana spiders, and Sydney funnel web spiders and their relatives, cause neurotoxic araneism. Immediate pain is followed by sweating with gooseflesh at the site of the bite. Systemic symptoms quickly evolve: headache, nausea, vomiting, profuse generalized sweating, fever, priapism, and painful muscle spasms, tremors, and rigidity that may cause respiratory distress or simulate an acute abdomen. Antivenoms widely used for Loxosceles and neurotoxic bites are of uncertain effectiveness. Ticks—​mainly in North America and Australia, both ixodid (hard) and argasid (soft) ticks can inject a salivary neurotoxin during their blood meal, causing an ascending flaccid paralysis. The tick must be detached as soon as possible. Centipedes-cause painful stings in tropical countries, while toxic secretions of millipedes may be applied to skin, lips, and eyes by chil- dren who are handling or trying to eat them. Leeches Leeches have anticoagulant saliva. Land leeches infest rainforests and can invade clothing while aquatic leeches are swallowed in fresh water or they may penetrate body orifices of swimmers. Prevention is by applying repellents to skin, clothes, and footwear, by boiling or filtering drinking water and by avoiding affected waters. Clinical effects are local pain, itching, blood loss, secondary infection, and phobia. Ingested aquatic leeches may obstruct pharynx, bronchi, or oesophagus. Use of medicinal leeches may be complicated by Aeromonas hydrophila infection. Mechanical injuries caused by animals Epidemiology Many species of wild animals have mauled and killed humans. Attacks by wild mammals are increasingly reported. Tigers, lions, leopards, and other big cats, hyenas, domestic dogs, jackals, wolves, bears, elephants, rhinos, hippopotamuses, buffaloes, bison, moose, elk, other large deer and antelopes, wild pigs, tapirs, chimpan- zees, baboons, ostriches, and cassowaries have killed people. The big cats, wolves, bears, elephants, hippopotamuses, and buffaloes are the most dangerous. Since 2000, about 60–​80 confirmed un- provoked attacks by sharks with an average of 4.3 fatalities (case fatality c.8%) have been reported each year. Other fish, such as barracudas, moray, and conger eels, garfish, groupers, stingrays, and piranhas can inflict lethal injuries. Electric ‘eels’ Electrophorus electricus (Gymnotidae) (Fig. 10.4.2.1) of rivers and coastal waters in Florida and South American and marine torpedo rays (e.g. Torpedo spp., Torpediniformes) can impart stunning electric shocks but are unlikely to be lethal. Even the 5-​cm Amazonian catfish (genus Vandellia, Trichomycteridae; Spanish ‘canero’; Portuguese ‘candirú’), the only vertebrate human ectoparasite, can traumatize humans by burrowing into their urethra, vagina, or anus, causing pain, bleeding, and obstruction. Crocodilians (alliga- tors, caimans, and crocodiles) kill, eat, and scavenge dead humans in Africa, Asia, and Oceania. In the United States of America, es- pecially Florida, alligators Alligator mississippiensis are responsible for a few deaths but in Africa, Nile crocodiles Crocodilus niloti- cus kill about 1000 people each year, and in South Asia, northern Australia, and New Guinea the saltwater crocodile C. porosus kills hundreds each year. Giant pythons very rarely kill humans in Africa (Python sebae), India (Python molurus), Indonesia (Python reticula- tus), Australia (Morelia amethistina), and South America (Eunectes murinus). Occasional human deaths have been attribute to attacks by Komodo dragons (Varanus komodoensis). To put these incidents into perspective, in the United States, collisions between vehicles

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1780 and deer and injuries to horseback riders are much more common than attacks by wild animals. Bites by domestic dogs are common worldwide. In England and Wales, where the estimated dog population is 6 million, more than 200 000 bite victims attend hospital each year. In the United States, dogs are responsible for 80–​90% of all animal bites. They bite about 4.7 million people each year (1.8% of the population), 800 000 of whom (0.3% of the population) require medical attention, and 12 are killed. Children are especially vulnerable. Other domestic animals that have caused severe injuries or deaths include camels, cattle, water buffalo, sheep, pigs, cats, and even ferrets. Prevention It is essential to obtain local advice about these environmental haz- ards. Where dangerous wild animals abound, wandering alone and unprotected between dusk and dawn incurs the highest risk of attacks. Staying in a vehicle and travelling in groups reduces risk. Pet dogs may attract large predators. In bear country, hikers should travel in groups, making plenty of noise. Bears should never be ap- proached (e.g. for photography), especially if there are cubs. Faced by a charging bear, avoid eye contact and do not attempt to hide, run away, or climb a tree. At a distance of 30 feet (c.10 m), a bear may be repelled by discharging a commercial pepper spray (10% capsicum oleoresin) towards its eyes. If attacked by a dog, avoid eye contact, shout, and fight back with sticks and stones. Young children should not be left alone with dogs, even family pets, and notoriously dangerous breeds should be banned. Elephants are dan- gerous whether wild or tamed. They should be treated with extreme respect or avoided, especially if they are in ‘musth’. Swimming or canoeing in hippo-​infested waters or blocking their retreat to water is highly dangerous. To prevent crocodilian attacks, keep well away from the water’s edge, do not bathe between dusk and dawn, and avoid canoeing in croc-​infested waters. If attacked by a crocodilian on land, run; in the water, fight back, hitting the animal on its nose and eyes with any available weapon. To avoid shark attacks, never bathe in shark-​infested waters, between sand bars and the deep ocean, where dead fish have been dumped, flocks of sea birds are feeding, or sewage is discharged. If attacked by a shark, fight back, hitting it on the nose and clawing at its eyes and gills. Chemical and electrical-​field repellents and chain mail suits have been developed to protect divers. Clinical features Teeth, tusks, horns, claws, and spines gouge, tear, crush, avulse, and puncture soft tissues and break bones. Big cats, bears, pigs, pythons, crocodilians, and sharks will eat their victims. Bovines and elephants trample and kneel on the prostrate body. Body cavities may be punctured, resulting in pneumothorax, haemothorax, her- niation and strangulation of bowel, and rupture of liver and spleen. Horse and camel bites and kicks can fracture, dislocate, crush, and concuss. Wild and feral pigs, armed with lethal tusks, can inflict abdominal evisceration, pneumothorax and fractures and lacer- ations of tendons, arteries, and nerves. Giant pythons asphyxiate by constriction. Sharks amputate whole limbs, causing rapidly fatal haemorrhage. Garfish (needle fish) and sting rays can fatally im- pale. Infection is likely with all these traumas: rabies, tetanus, gas gangrene, cat scratch disease (Bartonella henselae), Pasteurella mul- tocida, Capnocytophaga canimorsus, leptospires, Spirillum minus, Streptobacillus moniliformis, and aquatic organisms such as Vibrio vulnificus and Aeromonas hydrophila (see Section 8, Infectious diseases). Treatment Since wild animal attacks are most likely to happen in areas remote from medical care, delayed hospital treatment makes first aid espe- cially crucial for the survival of the victim. Fig. 10.4.2.1  Electric ‘eel’ Electrophorus electricus (Gymnotidae).

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1781 First aid of severe injuries First, the patient and rescuers must be made safe from further danger and drowning. Bleeding is controlled by local pressure or tourniquets, perforating injuries are closed with pressure dress- ings, circulating volume repletion is started as soon as possible with intravenous fluids, and the casualty is evacuated promptly to hos- pital. Some regions have flying doctor services (e.g. AMREF in East Africa). All injuries inflicted by animals must be assumed to be in- fected by a range of organisms (see earlier paragraphs) and so it may be appropriate to start antibiotic treatment immediately. Medical treatment in the hospital Emergency surgery may be required. Blood loss should be replaced and attention given to local mechanical complications such as frac- tures, tension pneumothorax, damage to large blood vessels, per- foration of the bowel, and lacerations of other abdominal viscera. Thorough debridement or amputation of dead tissue may be re- quired with removal of foreign material, teeth, and so on, and irriga- tion and drainage. Except for wounds on the head and neck, which can be sutured immediately, primary suturing should be delayed for 48–​72 h, after which further debridement, suturing, or covering with split-​skin grafts should be considered. Wounds should be thoroughly cleaned with soap and water as soon as possible; suitable antiseptics include iodine and alcohol so- lutions. Prophylactic antimicrobials such as amoxicillin/​clavulanic acid, doxycycline, or erythromycin have proved effective in dog-​ and cat-​bite wounds and are indicated for multiple or severe wounds and bites on the face and hands. For other bites, use penicillin, an aminoglycoside and metronidazole and for marine or aquatic wounds, to cover unusual bacteria such as Vibrio and Aeromonas spp., doxycycline or co-​trimoxazole or, in severe cases, a combin- ation of tetracycline with an aminoglycoside (e.g. gentamicin) and cefotaxime, or tetracycline with aminoglycoside and a fluoroquino- lone. Specific infections, such as tetanus, rabies, and Herpes simiae virus (from monkey bites) must be considered and treated or pre- vented appropriately. Venomous animals For predation or defence, some animals inject venoms through fangs, chelicerae (venom jaws), stings, spines, hairs, nematocysts, and other specialized venom organs. ‘Spitting’ snakes, scorpions, and millipedes squirt venom on to absorbent mucous mem- branes. The flesh or skin of some animals contains poisons acquired through the food chain. Allergic reactions to injected venoms (e.g. of Hymenoptera and cnidarians) may cause more frequent and serious medical problems than their direct toxic effects. Venomous mammals Male duck-​billed platypuses (Ornithorhynchus anatinus) have erectile venomous spurs on their hind limbs. These aquatic, egg-​ laying mammals of eastern Australia sting at least one person each year in Victoria, but only 17 cases have been reported since 1817. There is immediate, agonizing, persistent local pain, as well as pro- longed local swelling, chronic pain on movement, hyperaesthesia, wasting, inflammation, and regional lymphadenopathy. These ef- fects are not life-​threatening in humans, but dogs have died of envenoming. In the absence of specific treatment, non​steroidal anti-​inflammatory agents (NSAIDs) or corticosteroids have proved effective. The venom contains a C-​type natriuretic peptide (which causes mast-​cell degranulation), nerve growth factor, several α-​ and β-​defensin-​like peptides, enzymes, and other peptides and proteins, including a sildenafil-​like phosphodiesterase-​5 inhibitor. Male ech- idnas, the other egg-​laying mammal, possess a similar but smaller venom apparatus. Several species of Insectivora produce venomous saliva con- ducted into bite wounds by curved and sometimes grooved lower incisors. Venomous species include the Hispaniolan and Cuban so- lenodons (Solenodon paradoxus, S. (Apotogale) cubanus), northern water shrew Neomys fodiens, southern water shrew N. anomalus, and North American short-​tailed shrew Blarina brevicauda. Their bites can kill rodents and lagomorphs, but in humans the effect is local pain, swelling, and inflammation. The saliva of vampire bats (Desmodontinae) contains permeability- ​increasing factors, a platelet inhibitor, draculin (an inhibitor of acti- vated factors X and IX), and a plasminogen activator which is being developed as a thrombolytic drug. The slow loris Nycticebus coucang (Primates; Lorisidae) possesses brachial glands whose secretion contains toxic protein very similar in structure to Fel d 1 cat allergen, which the lorises lick up and can inject when they bite. In humans, slow loris bites may be damaging, infective, or toxic, causing pain, swelling, and even anaphylaxis. Venomous snakes Fewer than 200 species of venomous snake (families Colubridae, Atractaspidinae, Elapidae, and Viperidae) have been responsible for severely envenoming humans, resulting in death or permanent disability. Since it may be difficult to distinguish venomous from nonvenomous species, unnecessary contact with all snakes should be avoided, and patients bitten by any species should be assessed carefully. Distribution The Antarctic; most islands of the western Mediterranean, Atlantic, Caribbean, and eastern Pacific (including Hawaii); Chile, Iceland, Ireland, Madagascar, New Caledonia, and New Zealand are free from venomous snakes. Elsewhere, venomous snakes are widely distributed up to altitudes of more than 4900 m in the Himalayas (Gloydius himalayanus), within the Arctic Circle (Vipera berus), in the Indian and Pacific oceans as far north as Siberia (Pelamis pla- tura/​Hydrophis platurus), and in some freshwater lakes (Hydrophis semperi). Classification Medically important species have one or more pairs of enlarged teeth (fangs) in their upper jaws, containing a groove or closed channel through which they inject venom into their prey.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1782 Colubridae The short, immobile fangs are at the back of the maxilla (Fig. 10.4.2.2). Most of the familiar snakes regarded as non-​ venomous (e.g. the British grass snake Natrix natrix helvetica and the smooth snake Coronella austriaca), belong to this large family. However, many colubrid species have proved capable of causing at least local envenoming and some have caused severe envenoming or death, such as three African species—​the boomslang Dispholidus typus and the vine, twig, bird, or tree snake or Voëlslang (Thelotornis kirtlandii and T. capensis); the Japanese yamakagashi Rhabdophis tigrinus; and the Southeast Asian red-​necked keel- back R. subminiatus. (Fig. 10.4.2.2). Atractaspidinae (family Lamprophiidae) The African and Middle Eastern burrowing asps, stiletto snakes, or burrowing or mole vipers or adders strike sideways, impaling their victims on one of their two long front fangs, protruding through the partially closed mouth (Fig. 10.4.2.3). Four species, Atractaspis microlepidota, A. engaddensis, A. corpulenta and A. irregularis, have proved capable of killing humans. Elapidae This family includes cobras (Fig. 10.4.2.4), kraits, mambas, shield-​ nose snakes, coral snakes (Fig. 10.4.2.5), garter snakes, all the venomous Australasian snakes (Fig. 10.4.2.6), and sea snakes (Fig. 10.4.2.7). The short front fangs are immobile (10.4.2.4a and Fig. 10.4.2.7). Several African and Asian species (rinkhals and spitting cobras) can eject venom from the tips of their fangs (Fig. 10.4.2.3) as a fine spray for a distance of a few metres into the eyes of a perceived aggressor. Viperidae The front fangs are long, curved, and capable of a wide range of movement (Fig. 10.4.2.8). The subfamily Crotalinae comprises the American rattlesnakes (Fig. 10.4.2.9), moccasins, lance-​headed vi- pers, and Asian pit vipers, which possess a heat-​sensitive pit organ behind the nostril (Fig. 10.4.2.10). The Old World vipers and adders (subfamily Viperinae) lack this pit organ. Incidence of snakebite and medically
important species Snakebite is a frequent medical emergency in rural areas of many tropical countries; its incidence is seriously underestimated by hospital returns, because many victims seek the help of trad- itional healers rather than western-​style doctors. In Kilifi District, Kenya, where snakebites cause 6.7 deaths per 100 000 per year (0.7% of all deaths), 68% of the victims had sought treatment from (a) (b) Fig. 10.4.2.2  Back-​fanged Colubroid snakes. (a) Back fangs of red-​ necked keelback (Rhabdophis subminiatus), a Southeast Asian colubrid snake (family Natricinae), capable of causing severe envenoming. (b) Baron’s green racer (Philodryas baroni), a South American colubrid snake (family Dispsadidae), capable of causing mild local envenoming. Copyright D. A. Warrell. Fig. 10.4.2.3  Slender burrowing asp Atractaspis aterrima, Nigeria, showing fang. Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1783 (a) (b) (c) Fig. 10.4.2.4  Common Indian cobra Naja naja: (a) short front fang, (b) and (c) showing defensive posture with open hood with ‘spectacle’ marking (specimen in Sri Lanka). Copyright D. A. Warrell. Fig. 10.4.2.5  Painted coral snake Micrurus corallinus, Brazil. Copyright D. A. Warrell. Fig. 10.4.2.6  Papua New Guinean taipan Oxyuranus scutellatus showing the distinctive dorsal red stripe. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1784 traditional healers. In Africa, the saw-​scaled or carpet viper Echis spp., puff adder Bitis arietans, and spitting cobras (Naja nigricollis, N. mossambica, and so on) are the species of greatest medical im- portance. In the Benue Valley of north-​east Nigeria, E. ocellatus
(Fig. 10.4.2.11) causes 500 bites per 100 000 population per year, with a 12% mortality. Vipers of the genus Echis, whose geo- graphical range extends through Africa north of the equator, the Middle East, and eastern Asia to India, are responsible for many bites and deaths. In India, the most important species are cobras Naja naja, N. kaouthia (Fig. 10.4.2.4), common krait Bungarus caeruleus, Russell’s viper Daboia russelii (Fig. 10.4.2.8), and the saw-​scaled viper E. carinatus. A well-​designed, nationwide study in India established that 46 000 people were killed by snakes in 2005. In children aged 5–​14 years, snakebites caused 3% of all deaths. Among fatalities, 97% died in rural areas, only 23% of them in hospitals. In Bangladesh, a community-​based study es- timated 600 000 bites and 6000 deaths each year. In Southeast Asia, the Malayan pit viper Calloselasma rhodostoma, D. siamen- sis, green pit vipers (e.g. Trimeresurus T. albolabris), and cobras N.  kaouthia and N.  siamensis cause most bites and deaths. In Myanmar, Russell’s viper bite is a common cause of acute kidney injury and is responsible for most of the estimated 1000 snakebite deaths each year. In the United States of America, there are about 7000–​8000 bites each year with about five deaths. Rattlesnakes, especially Crotalus adamanteus, C. atrox, C. horridus, C. oreganus, C. scutulatus, C. viridis and Sistrurus miliaris, are the most dan- gerous species. In Mexico there are about 27 000 bites each year. In Central and South America, medically important species include rattlesnakes (e.g. Crotalus simus, C. durissus) (Fig. 10.4.2.9) and the lance-​headed vipers Bothrops atrox (‘barba amarilla’), B. asper (‘terciopelo’), B. bilineatus (‘papagaio’) and B. jararaca (‘jararaca’). There are an estimated 4000 bites each year in Central America, fewer than 100 in the Caribbean, and 45 000 in South America. In the Amami and Ryukyu islands of Japan, the habu, Protobothrops flavoviridis, inflicted an average of 610 bites with 5.6 deaths per year during the 1960s. In the United Kingdom, the adder or viper Vipera berus is the only venomous species (Fig. 10.4.2.12). More than 100 people are bitten each year, but only 14 deaths have been reported since 1876, the last in 1975. Scandinavia has hundreds of adder bites each year, but very few deaths. V. aspis
causes most bites in France, while V. ammodytes is important in Eastern Europe. In Australia, there are about 1000 bites (4.76/​100 000 popula- tion) and 2–​5 deaths (0.1–​0.2/​100 000) per year. Recently, almost all fatalities have been attributed to brown snakes Pseudonaja spp. Other important species are tiger snakes (Notechis scuta- tus, and so on), taipan Oxyuranus scutellatus (Fig. 10.4.2.6), and death adders Acanthophis spp. There are several 100 deaths each year in New Guinea, mostly caused by taipans. The highest snakebite mortalities, up to 24% of all adult deaths, are recorded among hunter–​gatherer tribes of Brazil (Kashinawa), Venezuela (Yanomamo), Ecuador (Waorani), Tanzania (Hadza), and Papua New Guinea. Epidemiology Most snakebites are inflicted on the lower limbs of farmers, plan- tation workers, herdsmen, and hunters in rural areas of tropical developing countries. The snake is usually trodden on at night or in undergrowth. Some species such as the Asian kraits Bungarus spp. and African spitting cobras N. nigricollis enter human dwell- ings at night in search of their natural prey and may bite people who roll over on to them while sleeping on the floor or in response to human odour or warmth. Snakes strike if inadvertently trodden upon or touched. In Europe, North America, and Australia, exotic venomous snakes are increasingly popular pets: their owners are sometimes bitten on their hands, especially when inebriated. In 120 9 8 7 6 5 4 3 2 1 0 110 100 90 80 70 60 50 40 30 20 10 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Average rainfall Study deaths (n = 562) Month Snakebite deaths in the study Average rainfall (in mm) Fig. 10.4.2.7  Monthly snakebite deaths in randomly selected study areas of India in 2005 (solid line) and average rainfall (dashed line), showing peak mortality during the monsoon. From Mohapatra B, et al. (2011). Snakebite mortality in India: a nationally representative mortality survey. PLoS Negl Trop Dis, 5(4), e1018.

1785 (a) (b) (c) Fig. 10.4.2.8  Eastern Russell’s viper Daboia siamensis, Ban Mi, Thailand: (a) showing ‘chain’ pattern (scale in cm); (b) showing long, hinged front fangs (reserve fang on the left side) in dental sheath; (c) dissection of venom apparatus. Copyright D. A. Warrell. Fig. 10.4.2.9  South American tropical rattlesnake or cascabel Crotalus durissus cascavella. Specimen from Brazil. Copyright D. A. Warrell. Fig. 10.4.2.10  Southeast Asian white-​lipped green pit viper Trimeresurus (Trimeresurus) albolabris showing heat-​sensitive pit organ between eye and nostril. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1786 the United States, 25% of bites result from snakes being attacked or handled. Serious bites by back-​fanged (colubrid) snakes usu- ally occur under these circumstances. Seasonal peaks in the inci- dence of snakebite are associated with agricultural activities, such as ploughing before the annual rains in the West African Sahel and the rice harvest in Southeast Asia. In India, most snakebite deaths occur during the May to October monsoon season (Fig. 10.4.2.7). Other factors are fluctuations in the activity or population density of venomous snakes, severe flooding that concentrates the human and snake populations in decreasing areas of dry land (e.g. Bangladesh, Pakistan, India, Nepal, Myanmar, Vietnam, and Colombia) and clearing of primary forests during construction of new highways and irrigation and hydroelectric schemes (e.g. Brazil, Sri Lanka). Snakebite or injection of snake venom has been employed for sui- cide and homicide. Venom apparatus Venom glands of Elapidae and Viperidae are situated behind the eye, surrounded by compressor muscles (Fig. 10.4.2.8c). A venom duct leads to the base of the fang in which venom is conducted along a groove or through a closed canal. In Colubridae, or more broadly, Colubroid snakes, venom secreted by Duvernoy’s gland tracks down grooves in the anterior surfaces of fangs at the posterior end of the maxilla (Fig. 10.4.2.2). The average dry weight of venom injected at a strike is approximately for N. naja 60 mg, E. carinatus 13 mg, D. russelii 63 mg, and Daboia palaestinae 32 mg. The amount in- jected when a snakebites a human is highly variable. In a proportion of bites there is negligible envenoming (‘dry bites’): more than 50% of those bitten by Malayan pit vipers C. rhodostoma or Russell’s vi- pers; less than 10% bitten by Echis spp.; but more than 75% bitten by common brown snakes (Pseudonaja spp.) in Australia. The Palestine viper Daboia palaestinae expends only about one-​tenth of the cap- acity of its venom gland at each consecutive strike, whereas Daboia siamensis exhausts more than three-​quarters of its reservoir at the first strike. The popular belief that snakes are less dangerous after they have eaten is incorrect. Prevention of snakebite To reduce the risk of bites, snakes should never be disturbed, attacked, cornered, or handled, even if they are thought to be a harmless species or appear to be dead. Venomous species should never be kept as pets or as performing animals. In snake-​infested areas, boots, socks, and long trousers should be worn for walks in undergrowth or deep sand, and gloves for exploring foliage. A light should always be carried at night together with a stick for prodding (a) (b) (c) Fig. 10.4.2.11  Saw-​scaled or carpet vipers-genus Echis. (a) Echis ocellatus from West Africa. (Specimen from Nigeria). (b) Echis pyramidum from East Africa (specimen from Kenya). (c) Echis carinatus sochureki from the Middle East and South Asia (specimen from Oman). Copyright D. A. Warrell. Fig. 10.4.2.12  European adder or viper Vipera berus, the only venomous British snake (specimen from Wales). Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1787 the ground ahead. Lightweight boots that were resistant to Russell’s viper strikes were developed in Burma and proved acceptable to rice farmers for use during the high risk harvesting season. Collecting firewood; dislodging logs and boulders with bare hands; pushing sticks or fingers into burrows, holes, and crevices; climbing rocks and trees covered with dense foliage; and swimming in overgrown lakes and rivers are particularly hazardous activities. Unlit paths and gutters are especially dangerous after heavy rains. Sleeping on the ground carries a risk of nocturnal krait bites in South Asia and of spitting cobra bites in Africa, but well tucked-​in mosquito nets are protective. To prevent sea-​snakebites, fishermen should not touch these animals when they are caught in nets or on lines. Swimmers and divers should not aggravate them and should avoid wading in the sea, especially in muddy estuaries, in sand, or near coral reefs. It is futile and ecologically undesirable to attempt to exterminate venomous snakes. Various substances toxic to snakes, such as in- secticides and methylbromide, have been used to keep human dwellings free of these animals. However, no effective yet harmless snake repellent has been discovered. Immunization against envenoming To be effective, high titres of a neutralizing antibody would have to be circulating at the time of the bite. This has been achieved in animals used for antivenom production but only by frequent im- munization, which would not be practicable even in the highest-​risk human populations. An accelerated (anamnestic) secondary rise in antibody levels, stimulated by envenoming, would be too slow to pre- vent envenoming. An antirattlesnake vaccine for domestic dogs, of dubious efficacy, is marketed in the United States and pre-​exposure immunization of farmers in Japan, against habu (Protobothrops flavoviridis) venom was ineffective. Properties of snake venoms More than 90% of the dry weight of venom consists of more than 100 different proteins: enzymes, non​enzymatic polypeptide toxins, and non​toxic proteins such as nerve growth factor. Enzymes consti- tute 80–​90% of viperid and 25–​70% of elapid venoms. They include digestive hydrolases, hyaluronidase, and activators or inactivators of physiological processes. Most venoms contain l-​amino acid oxidase, phosphomono-​ and diesterases, 5′-​nucleotidase, DNAase, NAD-​nucleosidase, phospholipase A2, and peptidases. Elapid venoms also contain acetylcholine esterase, phospho- lipase B, and glycerophosphatase, while viperid venoms have metalloproteinases, endopeptidase, arginine ester hydrolase, kininogenase, as well as thrombin-​like, factor X, and prothrombin-​ activating enzymes. Phospholipase A2 (lecithinase) is the most widespread and extensively studied of all venom enzymes. It damages mitochondria, red blood cells, leucocytes, platelets, per- ipheral nerve endings, skeletal muscle, vascular endothelium, and other membranes, produces presynaptic neurotoxic activity, opiate-​like sedative effects, the autopharmacological release of histamine, and may be anticoagulant. The acetylcholinesterase found in most elapid venoms does not contribute to their neuro- toxicity. Hyaluronidase promotes the spread of venom through tissues. Proteolytic enzymes (metalloproteinases, endopeptidases, or hydrolases) are responsible for local changes in vascular permeability leading to oedema, blistering, and bruising, and to necrosis. Venom l-​amino acid oxidases are homodimeric flavoenzymes that catalyse the oxidative deamination of an l-​ amino acid substrate to an α-​keto acid, ammonia, and hydrogen peroxide. They are widely distributed in venoms of Viperidae and Elapidae. Their reported biological activities include induction of apoptosis, oedema, and haemolysis, antibacterial function, and platelet activation or inhibition. Polypeptide toxins (neurotoxins) Postsynaptic (α) neurotoxins such as α-​bungarotoxin and cobrotoxin contain about 60–​62 or 66–​74 amino acids. They bind to acetyl- choline receptors at the motor endplate. Presynaptic (β) neuro- toxins, such as β-​bungarotoxin, crotoxin, and taipoxin, contain about 120–​140 amino acids and a phospholipase A subunit. These release acetylcholine at the nerve endings at neuromuscular junc- tions and then damage the endings, preventing further release of transmitter. Venom pharmacology The smaller neurotoxins of the Elapidae are rapidly absorbed into the bloodstream, whereas the larger phospholipase A2 presynaptic toxins and Viperidae toxins are taken up more slowly through the lymphatics. Venoms of the spitting cobras and rinkhals can be ab- sorbed through the intact cornea, causing systemic envenoming and even death in animals. Envenoming after ingestion of snake venom has not been reported in humans. Most venoms are concentrated and bound in the kidney, and some components are eliminated in the urine. Crotaline venoms are selectively bound in the lungs, con- centrated in the liver, and excreted in bile, while polypeptide neuro- toxins, such as α-​bungarotoxin, are tightly bound at neuromuscular junctions. Most venom components do not cross the intact blood–​ brain barrier and so central nervous system effects of venom toxins are controversial. Pathophysiology Swelling and bruising of the bitten limb result from increased vascular permeability induced by proteases, phospholipases, membrane-​damaging metalloproteinases (haemorrhagins), and endogenous autacoids released by the venom, such as hista- mine, 5-​hydroxytryptamine, and kinins. Venoms of some of the North American rattlesnakes and viperine species cause a gen- eralized increase in vascular permeability resulting in hypovol- aemia, haemoconcentration, hypoalbuminaemia, albuminuria, serous effusions, pulmonary oedema, and, in the case of Burmese D.  siamensis, conjunctival and facial oedema (Fig. 10.4.2.13). Tissue necrosis near the site of the bite is caused by myotoxic and cytolytic factors:  in some cases, ischaemia resulting from thrombosis, intracompartmental syndrome, or a tight tourni- quet may contribute. Causes of hypotension and shock include hypovolaemia, vasodilatation, and myocardial dysfunction. Some venoms release vasodilating autacoids such as histamine and ki- nins. Venom of the Brazilian jararaca B. jararaca was found to activate bradykinin and, through a bradykinin-​potentiating pep- tide, to prolong its hypotensive effect by inactivating the peptidyl dipeptidase responsible both for destroying bradykinin and for converting angiotensin I to angiotensin II. This observation led

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1788 to the synthesis of angiotensin-​converting enzyme (ACE) in- hibitors. Bradykinin-​potentiating and ACE-​inhibiting peptides have also been found in several other crotaline venoms (genera Bothrops and Agkistrodon). Four sarafotoxins have been isolated from the venom of the Israeli burrowing asp Atractaspis engad- densis (Fig. 10.4.2.3). They show 60% sequence homology with the endothelins, which are also 21-​amino acid polypeptides. Sarafotoxins and endothelins are potent vasoconstrictors of cor- onary and other arteries, delay atrioventricular conduction, and are positively inotropic. Snake venoms can cause haemostatic defects in several different ways. Venom procoagulant enzymes, many of them serine pro- teases, activate the blood clotting cascade at various sites. Some Viperidae venoms contain thrombin-​like fibrinogenases, which remove fibrinopeptides from fibrinogen directly. Others acti- vate endogenous plasminogen. Venoms may induce or inhibit platelet aggregation. Spontaneous systemic bleeding is caused by haemorrhagins, metalloendopeptidases, some with disintegrin-​ like and other domains, which damage vascular endothelium (Fig. 10.4.2.14). The combination of consumptive coagulopathy, thrombocytopenia, and vessel wall damage can result in mas- sively incontinent bleeding, a common cause of death after bites by Viperidae, Australasian Elapidae, and the few medically important Colubridae. Many venoms are haemolytic in vitro, but clinically significant intravascular haemolysis, apart from the microangiopathic haemolysis associated with disseminated intravascular coagulation described in victims of viperine and Australian brown snake Pseudonaja bites, is seen only after bites by D. russelii (Sri Lanka and India), and some Bothrops and colu- brid species. Acute renal tubular necrosis may be caused by severe hypotension, disseminated intravascular coagulation (D. russelii, D. siamensis), a direct nephrotoxic effect of the venom (D. siamen- sis), and myoglobinuria secondary to generalized rhabdomyolysis (sea snakes, D. russelii in Sri Lanka and India, and tropical rattle- snakes). Neurotoxic polypeptides and phospholipases block neuromuscular transmission causing death through bulbar or re- spiratory paralysis. Clinical features Fear, effects of treatment, and the venom contribute to the symp- toms and signs of snakebite. Even patients who are not envenomed may feel flushed, dizzy, and breathless, and may notice constriction of the chest, palpitations, sweating, and acroparaesthesiae. Tight tourniquets may produce swollen and ischaemic limbs; local inci- sions at the site of the bite may cause bleeding and sensory loss and herbal medicines often induce vomiting. The earliest symptoms dir- ectly attributable to the bite are local pain and bleeding from the fang punctures, followed by pain, tenderness, swelling, and bruising extending up the limb, lymphangitis, and tender enlargement of re- gional lymph nodes. An anaphylaxis-​like syndrome of early syncope, vomiting, colic, diarrhoea, angio-​oedema, and wheezing may follow bites by European Vipera, Russell’s vipers, Bothrops spp., Australian elapids, and Atractaspis engaddensis. Nausea and vomiting are common early symptom of systemic envenoming. Bites by Colubridae (back-​fanged snakes) Severe envenoming causes repeated vomiting, colicky abdom- inal pain, headache, systemic bleeding with widespread ecchym- oses and the risk of intracerebral haemorrhage, incoagulable blood, intravascular haemolysis, and acute kidney injury. Local swelling and bruising may be the only results of envenoming. The first symptoms of envenoming may be delayed for 24–​72 h after the bite. Bites by Atractaspidinae (burrowing asps or stiletto snakes) Local effects include pain, swelling, blistering, necrosis, and tender enlargement of local lymph nodes. Violent gastrointestinal symp- toms (nausea, vomiting, and diarrhoea), anaphylaxis (dyspnoea, respiratory failure), and electrocardiographic (ECG) changes (atrioventricular block, ST, T-​wave changes) have been described in patients envenomed by A.  engaddensis and A.  microlepidota andersoni. Bites by Elapidae (cobras, kraits, mambas, African garter snakes, coral snakes, Australasian snakes, and sea snakes) Bites by kraits, mambas, coral snakes, and some cobras (e.g. N. haje, N. nivea, and N. philippinensis) produce minimal local effects, but the venoms of African spitting cobras (N.  nigri- collis, N.  mossambica, and so on) and Asian cobras (N.  naja, N. kaouthia, N. sumatrana, and so on) cause tender local swelling, Fig. 10.4.2.14  Haemorrhagin activity revealed clinically as gingival haemorrhage in a patient bitten by a saw-​scaled or carpet viper Echis ocellatus in Nigeria. Copyright D. A. Warrell. Fig. 10.4.2.13  Gross bilateral conjunctival oedema (chemosis) in a Burmese rice farmer 48 h after being bitten by a Russell’s viper (Daboia siamensis). Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1789 blistering, and superficial necrosis, which may be extensive (Fig. 10.4.2.15). ‘Skip’ lesions, separated by apparently normal areas of skin, may occur (Fig. 10.4.2.16). However, elapid venoms are best known for their neurotoxic effects. Early symptoms, be- fore there are objective neurological signs, include vomiting, ‘heaviness’ of the eyelids, blurred vision, paraesthesiae around the mouth, hyperacusis, headache, dizziness, vertigo, hypersalivation, congested conjunctivae, and ‘gooseflesh’. Paralysis is first detect- able as ptosis and external ophthalmoplegia appearing as early as 15 min after the bite, but sometimes it is delayed for 10 h or even more than 24 h. Later the face, palate, jaws, tongue, vocal cords, neck muscles, and muscles of deglutition may become paralysed (Fig. 10.4.2.17). The pupils are dilated. Respiratory failure may be precipitated by airway obstruction at this stage, or later after paralysis of intercostal muscles and the diaphragm. Neurotoxic effects are completely reversible, either acutely in response to antivenom or anticholinesterases—​for example, following bites by Asian cobras, some Latin American coral snakes Micrurus spp., and Australasian death adders Acanthophis spp.—​or they may wear off spontaneously in 1 to 7 days. Excruciating pain and paraesthesiae radiating up the bitten limb have been described with bites by coral snakes (Micrurus tener and M. lemniscatus), explained by specific activation of acid-​sensing ion channels by a venom toxin (MitTx). Severe, non​colicky, crescendo abdominal pain, attributable to the smooth muscle stimulating effects of an AVIT toxin, is often the most striking initial symptom in victims of krait bites (Bungarus caeruleus). Envenoming by terrestrial Australasian elapids produces four main groups of symptoms: neurotoxicity (Fig. 10.4.2.18), haemo- static disturbances and, rarely, generalized rhabdomyolysis, and acute kidney injury. Painful regional lymph nodes are a useful sign of impending systemic envenoming, but local signs are usually mild, except after bites by the king brown or Mulga snake Pseudechis aus- tralis. Early symptoms include vomiting, headache, and syncopal attacks. Fig. 10.4.2.15  Extensive necrosis of skin and subcutaneous tissues in a Nigerian girl bitten nine days previously on the elbow by a black-​necked or spitting cobra Naja nigricollis. Copyright D. A. Warrell. Fig. 10.4.2.16  Zimbabwean girl showing ‘skip lesions’ separated by areas of intact skin after envenoming by a Mozambique spitting cobra Naja mossambica. By courtesy of the late Revd Dr Robbie McCabe. Fig. 10.4.2.17  Neurotoxic envenoming. Ptosis, external ophthalmoplegia, and facial paralysis in a Sri Lankan patient envenomed by a common krait Bungarus caeruleus. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1790 Patients ‘spat’ at by spitting elapids may develop venom ophthalmia. There is intense pain in the eye, blepharospasm, pal- pebral oedema, and leucorrhoea (Fig. 10.4.2.19a). Corneal ero- sions can be seen by slit-​lamp or fluorescein examination in more than half of patients spat at by N. nigricollis. Rarely, venom is ab- sorbed into the anterior chamber causing hypopyon and anterior uveitis. Secondary infection of corneal abrasions may lead to per- manent blinding opacities or panophthalmitis (Fig. 10.4.2.19b). Bites by sea snakes and sea kraits Patients envenomed by sea snakes notice headache, a thick feeling of the tongue, thirst, sweating, and vomiting. Between 30 min and 3.5 h after the bite, there is generalized aching, stiffness, and ten- derness of the muscles. Trismus is common. Later there is gener- alized flaccid paralysis. Myoglobinuria appears 3–​8 h after the bite. Myoglobin and potassium released from damaged skeletal muscles can cause acute kidney injury, while hyperkalaemia may precipi- tate cardiac arrest. Bites by Viperidae (vipers, adders, rattlesnakes, lance-​headed vipers, moccasins, and pit vipers) Viper venoms usually produce more severe local effects than do those of other snakes. Swelling may become detectable within 15 min and usually by 2 hours, but is sometimes delayed for sev- eral hours. It spreads rapidly with bruising, sometimes involving the whole limb, adjacent trunk and, in children, the whole body (Fig. 10.4.2.25). There is associated pain and tenderness in regional lymph nodes, with bruising of overlying tissues and lymphangitic lines. Blistering, and necrosis may appear during the next few days (Fig. 10.4.2.20). Necrosis can be severe fol- lowing bites by some rattlesnakes, lance-​headed vipers Bothrops spp., Asian pit vipers, and the large African Bitis species. When the envenomed tissue is contained in a tight fascial compartment such as the pulp space of digits or the anterior tibial compart- ment, ischaemia may result (Fig. 10.4.2.21). Absence of swelling 2 h after a viper bite suggests that there has been no envenoming. However, fatal envenoming by a few species can occur in the ab- sence of local signs (e.g. C. d. terrificus, C. scutulatus, and Burmese Russell’s viper). Haemostatic abnormalities are characteristic of envenoming by Viperidae. Persistent bleeding from fang punc- ture wounds, venepuncture, or injection sites, other new and par- tially healed wounds, and postpartum, indicates that the blood is incoagulable. Spontaneous systemic haemorrhage is most often detected in the gingival sulci. Epistaxis, haematemesis, cutaneous ecchymoses, haemoptysis, and subconjunctival, retroperitoneal, and intracranial haemorrhages (Fig. 10.4.2.22) are also seen. Patients envenomed by Burmese and Indian/​SriLankan Russell’s vipers may suffer haemorrhagic infarction of the anterior pitu- itary, resulting in acute or chronic pituitary/​adrenal insufficiency (Sheehan’s-​like syndrome) (Fig. 10.4.2.23). Hypotension and shock are common in patients bitten by North American rattle- snakes (e.g. C. adamanteus, C. atrox, and C. scutulatus), Bothrops, Daboia, and Vipera species (e.g. D. palaestinae and V. berus). The Fig. 10.4.2.18  Mozambique spitting cobra (Naja mossambica) in the act of ejecting venom from the tips of its fangs. Courtesy Dr David J. Williams. (a) (b) Fig. 10.4.2.19  Venom ophthalmia caused by the black-​necked spitting cobra Naja nigricollis: (a) Acute venom ophthalmia showing intense painful inflammation and discharge. (b) Neglected cobra spit ophthalmia complicated by corneal erosion and pan ophthalmitis requiring enucleation to prevent sympathetic ophthalmia of the other eye. Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1791 central venous pressure is usually low and the pulse rate rapid, suggesting hypovolaemia resulting from extravasation of fluid into the bitten limb. Patients envenomed by Russell’s vipers show evidence of generally increased vascular permeability. Direct myocardial involvement is suggested by an abnormal ECG or car- diac arrhythmia and reduced ejection fraction detected by echo- cardiography. Patients envenomed by some species of the genera Daboia, Vipera, Crotalus, Bothrops, and Australasian elapids may experience early transient and recurrent syncopal attacks, associ- ated with features of an autopharmacological or anaphylactic re- action, such as vomiting, sweating, colic, diarrhoea, shock, and angio-​oedema. These symptoms may appear as early as 5 min or as late as many hours after the bite. Early collapse after bites by Australian brown snakes (Pseudonaja spp.) and tiger snakes (Notechis spp.) has been attributed to coronary and pulmonary thromboembolism but this seems unlikely. Acute kidney injury is a common mode of death in patients envenomed by Viperidae. Victims of Russell’s viper may become oliguric within a few hours of the bite and complain of loin pain, suggesting renal ischaemia at a time when their plasma renin activity is high. Neurotoxicity, resembling that seen in patients bitten by Elapidae, is a feature of envenoming by a few species of Viperidae (e.g. C. d. terrificus, Gloydius spp., berg adder Bitis atropos and other small Bitis spe- cies, and Sri Lankan D. russelii) (Fig. 10.4.2.24). There is evidence of generalized rhabdomyolysis (Fig. 10.4.2.24), but progression to respiratory or generalized paralysis is unusual. Fig. 10.4.2.20  Swelling, blistering, and necrosis that required amputation in a Thai woman bitten on the hand four days earlier by a Malayan pit viper Calloselasma rhodostoma. There are widespread ecchymoses. Copyright D. A. Warrell. Fig. 10.4.2.21  Extensive necrosis of skin and muscle including the contents of the anterior tibial compartment in a patient bitten by a lancehead Bothrops marajoensis in Brazil 27 days earlier. Copyright D. A. Warrell. Fig. 10.4.2.22  CT scan showing intracranial haemorrhage in a child bitten by a common lancehead Bothrops atrox in Ecuador. The fluid level in the larger collection of blood indicates that the blood was incoagulable. Copyright D. A. Warrell. Fig. 10.4.2.23  Haemorrhagic infarction of the anterior pituitary in a Burmese patient who died after being bitten by a Russell’s viper Daboia siamensis. By courtesy of Dr U Hla Mon, Yangon, Myanmar.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1792 Envenoming by European vipers The common viper or adder V. berus (Fig. 10.4.2.12), the only venomous snake found in the United Kingdom, occurs in England, Wales, Scotland, and northern Europe, extending into the Arctic Circle and through Asia as far east as Sakhalin island and south to northern Korea. There are four other vipers that are widely distributed in mainland Europe:  the nose-​horned or sand viper V. ammodytes in the Balkans, Italy, Austria, and Romania; the asp viper V. aspis in France (south of Paris), Spain, Germany, Switzerland, and Italy; Lataste’s viper V.  latastei in Spain and Portugal, and Orsini’s viper V. ursinii in southeastern France, central Italy, and Eastern Europe. The Montpellier snake Malpolon monspessulanus is a large back-​fanged colubrid snake whose bite can cause transient mild local symptoms and rarely neurotoxicity. Clinical features of European viper bite • Local envenoming: immediate sharp pain is followed by spreading pain, tenderness, and tender enlargement of regional lymph nodes within hours. Reddish lymphangitic lines and bruising appear, and the whole limb may become swollen and bruised within 24 h with involvement of the trunk and, in children, the whole body (Fig. 10.4.2.25). Intracompartmental syndromes and necrosis are rare. • Systemic envenoming:  dramatic anaphylactic symptoms may appear between 5 min and many hours after the bite: nausea, retching, vomiting, abdominal colic, diarrhoea, incontinence of urine and faeces, sweating, fever, vasoconstriction, tachy- cardia, lightheadedness, shock with loss of consciousness, angio-​oedema of the face, lips, gums, tongue, throat, and epi- glottis, urticaria, and bronchospasm. These symptoms may persist or fluctuate for as long as 48 h in the absence of treat- ment. Hypotension is a dangerous sign that usually develops within 2 h and may resolve spontaneously, persist, recur, or progress fatally. Clinical features of a bleeding diathesis are unusual, but bleeding from the gums and nose and into the lungs, gastrointestinal and genitourinary tracts, and serosal cavities and retroperitoneally can occur. The risk of bleeding is greatly increased by misguided treatment with heparin. Fatal haemothorax, massive haematemesis and melaena, haematuria, and intrauterine fetal death are rare tragedies. Acute kidney injury is not uncommon in children. Increased capillary per- meability is reflected by the local and sometimes generalized oedema, as well as the more focal angio-​oedema that can lead to fatal occlusion of the upper airway, and pulmonary, and cere- bral oedema. Coma and seizures are attributable to hypoten- sion, cerebral oedema, hyponatraemia, hypoalbuminaemia, or hypoxaemia secondary to respiratory distress. Cardiac arrest, acute gastric dilatation, paralytic ileus, and acute pancreatitis are other reported complications. Classic mild neurotoxicity (ptosis, external ophthalmoplegia) has been reported after bites by several species of European Vipera, including V. aspis, V. berus, and V. ammodytes in certain geographical areas. • Laboratory findings: neutrophil leucocytosis is common. Serum creatine kinase (CK), transaminases, urea, and creatinine con- centrations may be raised, and bicarbonate may be reduced. Thrombocytopenia and mild coagulopathy; reflected by pro- longed prothrombin time, activated partial thromboplastin time, hypofibrinogenaemia, and raised fibrin degradation prod- ucts or D-​dimer; is sometimes detected. Severe coagulopathy is uncommon. Electrocardiographic changes include tachy-​ and bradyarrhythmias, atrial fibrillation, flattening or inversion of T-​ waves, ST elevation or depression, second-​degree heart block, and frank myocardial infarction. • Prognosis: most adder bites cause only trivial symptoms, but pa- tients must be assessed individually and deaths have occurred between 6 and 60 (average 34) h after the bite. Children may be severely envenomed: in a French series, there were three deaths in a group of seven children aged between 2.5 and 10 years. The dangers of adder bite should not be underestimated. The anti- venom treatment of adder bite is discussed in the following paragraphs. Laboratory investigations The peripheral neutrophil count may be raised to 20 000 cells/​µl or more in severely envenomed patients. The blood film may show evidence of microangiopathic haemolysis. Initial Fig. 10.4.2.24  Brazilian girl bitten 24 h earlier by a tropical rattlesnake Crotalus durissus terrificus. She has bilateral ptosis, paralysis of the facial muscles, and gross myoglobinuria resulting from generalized rhabdomyolysis. Copyright D. A. Warrell. Fig. 10.4.2.25  Generalized swelling and bruising in a 4-​year-​old child bitten by a European adder Vipera berus in Sweden. Courtesy of Dr H Persson.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1793 haemocon­centration, resulting from extravasation of plasma (Crotalus species and Burmese D. siamensis), is followed by anaemia caused by bleeding or, more rarely, haemolysis. Thrombocytopenia is common following bites by pit vipers (e.g. Calloselasma rho- dostoma, Crotalus oreganus helleri) and some Viperidae (e.g. Bitis arietans and Russell’s vipers), but is unusual after bites by Echis species. 20-​minute whole blood clotting test (20WBCT): this is a simple bedside test for venom-​induced defibrinogenation or anticoagulation. A  few millilitres of venous blood are placed in a new, clean, dry, glass vessel, left undisturbed for 20 min, and then tipped once to see if it has clotted or not. Positive 20WBCT (incoagulable blood) indicates systemic envenoming, either con- sumptive coagulopathy (plasma fibrinogen concentration below 0.5 g/​litre) or effects of an anticogulant toxin (e.g. envenoming by Australian black snake, Pseudechis spp.). It may be diagnostic of a particular species (e.g. Echis spp. in Africa north of the equator). The only equipment required for the test is a new glass tube, but this may be difficult to find in modern hospitals where glass has been replaced by plastics. Glass is essential to contact-​activate Hageman factor (factor XII) which initiates the ‘intrinsic’ coagulation pathway. Glass washed with soap or detergent may lose this property. A posi- tive 20WBCT had a positive predictive value of 89.7%, negative pre- dictive value of 93.5%, sensitivity of 92.9%, and specificity of 90.6% for plasma fibrinogen concentrations of less than 0.5 g/​litre. Point of care INR coagulometers and D-​dimer tests are not reliable in snake-​ envenomed patients. Laboratory tests of blood coagulation (prothrombin time, ac- tivated partial thromboplastin time, fibrinogen concentration) and fibrinolysis (fibrin/​fibrinogen degradation products, D-​ dimer) are more sensitive but take much longer and are more demanding in equipment than the simple 20WBCT. Patients with generalized rhabdomyolysis show a steep rise in serum creatine kinase, myoglobin, and potassium levels. Black or brown urine suggests generalized rhabdomyolysis and/​or intravascular haem- olysis; in both cases, positive urine sticks tests will not distinguish between blood, haemoglobin, and myoglobin. Concentrations of serum enzymes, such as CK and aspartate aminotransferase, are moderately raised in patients with severe local envenoming, due to muscle damage at the site of the bite. High concentra- tions (CK >2000  U/​litre) suggest generalized rhabdomyolysis. Urine should be examined for blood/​haemoglobin, myoglobin, and protein, and for microscopic haematuria and red cell casts. Electrocardiographic abnormalities such as sinus bradycardia, ST–​T changes, various degrees of atrioventricular block, and hyperkalaemic changes may be seen. Immunodiagnosis Specific snake venom antigens have been detected in wound swabs, aspirates or biopsies, serum, urine, cerebrospinal fluid, and other body fluids. Venom antigenaemia can be quantitated using en- zyme immunoassay (EIA), providing the most accurate prognosis. Under ideal conditions, relatively high venom antigen concentra- tions (wound swabs or aspirates) may be detected quickly enough (15–​30 min) to allow the selection of the appropriate monospe- cific antivenom. A commercial venom detection kit for Australian elapids is produced by Seqirus (formerly CSL), Melbourne. For retrospective diagnosis, including forensic cases, tissue around the fang punctures, wound and blister aspirate, serum, and urine should be stored for EIA immunodiagnosis. Polymerase chain re- action (PCR) to detect venom gland RNA or snake DNA in the bite wound are under development. Management of snakebite First aid The patient should be reassured and moved to the nearest hospital or dispensary as quickly, comfortably, and passively as possible. The whole patient should be immobilized, especially the bitten limb, using a splint or sling. Most traditional first aid methods are potentially harmful and should not be used. Local incisions and suction do not remove venom effectively and may introduce infection, damage tissues, and cause persistent bleeding. Vacuum extractors, potassium perman- ganate, and ice packs may potentiate local necrosis. Electric shocks are dangerous and have not been proved beneficial. Tourniquets and compression bands are potentially dangerous as they can cause gangrene, increased fibrinolysis, and bleeding in the occluded limb, peripheral nerve palsies, compartmental ischaemia, and intensifica- tion of local signs of envenoming. Pressure-​immobilization (P-​I) methods In animal studies, compressing superficial veins and lymph- atics in the bitten limb delayed the spread of larger molecular weight toxins such as the presynaptic phospholipase A2 toxins of Australian elapid venoms. This delay might prevent development of life-​threatening respiratory paralysis before the victim has had time to reach medical care. P-​I is, therefore, indicated after bites by neurotoxic elapids but also, in cases of bites by unknown species, P-​I should be applied immediately unless a bite by a neurotoxic elapid can, with confidence, be excluded. 1 Anker’s (Monash) pressure-​pad immobilization method (Fig. 10.4.2.26a). A pad of any available material, approximately 5 × 5 × 3 cm, is applied directly over the bite wound, at a pres- sure of about 70 mm Hg, and the limb is splinted. This delayed systemic envenoming, as assessed by measurements of venom antigenemia, in a preliminary field trial in Burmese Russell’s viper bite victims. 2 Sutherland’s original pressure-​bandage immobilization method (Fig. 10.4.2.26b) involves bandaging the bitten limb with a series of long, 10-​cm-​wide elastic bandages, ‘as firmly as for a sprained ankle’ (about 55 mm Hg), starting distal to the bite site, continuing up to the groin or axilla and incorporating a splint. Although never subjected to formal clinical trials, the method was con- sidered effective, based on anecdotal reports of delayed systemic envenoming and rapid deterioration after release of the bandage, in some cases supported by measurements of venom antigenemia. However, in practice, the technique has proved difficult to apply, even in Australia, and it is demanding on equipment and training. External compression increases intracompartmental pressure and might accentuate the local effects of some necrotic snake venoms, but animal studies found little evidence that this was deleterious and confirmed the life-​saving effects of lymphatic and venous compression.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1794 (a) (b) Fig. 10.4.2.26  Pressure-​immobilization methods. (a) Pressure-​pad immobilization. A pad of whatever material is immediately available is placed directly over the bite wound and bound on very firmly with an inelastic bandage. The whole limb is then splinted to prevent movement at any of its joints. (b) Pressure-​bandage immobilization. The bitten limb is firmly bound with long, wide (4 cm) elastic bandages, starting distal to the bite site and ending at the armpit or groin. A splint is incorporated. Courtesy of Dr David J Williams.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1795 Inhibition of the intrinsic lymphatic pump A study of lymphatic flow in human volunteers and in rats showed that nitric oxide (NO)–​donating drugs, such as glyceryl trinitrate (GTN), applied topically to the bitten limb, substantially slowed lymphatic flow, despite movement of the limb. Topical application of NO-​donating drugs might prove a useful adjunct to pressure-​pad or pressure-​bandage first-​aid methods. Pursuing and killing the snake is not recommended, but if the snake has been killed, it should be taken with the patient to hospital. It must not be handled as even a severed head can inject venom. Patients being transported to hospital should lie on their left side in the recovery position to prevent aspiration of vomit. Persistent vomiting can be treated with chlorpromazine by intramuscular in- jection (25–​50 mg in adults, 1 mg/​kg in children; intravenous injec- tion risks hypotension), or chlorpromazine or prochlorperazine by intrarectal suppository. Syncope, shock, angio-​oedema, and other autonomic symptoms can be treated with 0.1% adrenaline by intra- muscular injection (0.5 ml for adults, 0.01 ml/​kg for children) and an antihistamine such as chlorphenamine maleate by intravenous injection (10 mg for adults, 0.2 mg/​kg for children). Patients with incoagulable blood will develop haematomas after intramuscular and subcutaneous injections, and so the intravenous route should be used whenever possible except in the case of adrenaline. Respiratory distress and cyanosis should be treated by clearing the airway, giving oxygen, and, if necessary, assisted ventilation. If the patient is uncon- scious and no femoral or carotid pulses can be detected, cardiopul- monary resuscitation must be started immediately. Hospital treatment Clinical assessment In most cases of snakebite, uncertainties about the species and the quantity and composition of venom injected can be resolved only by admitting the patient to hospital for at least 24 h of observation. Local swelling is usually detectable within 15 min of pit viper envenoming and within 2 h of envenoming by most other vipers, but may not de- velop in patients bitten by some vipers, colubrids, and elapids such as kraits, coral snakes, and sea snakes. Fang marks are sometimes invisible. Tender enlargement of regional lymph nodes draining the bitten area is an early sign of envenoming by Viperidae and some Elapidae, notably Australasian elapids. All the tooth sockets should be examined meticulously as this is usually the first site of spontan- eous bleeding: other common sites are the nose, conjunctiva, skin, and gastrointestinal tract. Persistent bleeding from venepuncture sites and other wounds implies incoagulable blood. Hypotension and shock are important signs of hypovolaemia, vasodilatation, or cardiotoxicity, seen particularly in patients bitten by North American rattlesnakes and some Viperinae (e.g. V. berus, Russell’s vipers, D. palaestinae). Ptosis is the earliest sign of neurotoxic envenoming (Fig. 10.4.2.17). Respiratory muscle power should be assessed objectively and repeat- edly, for example, by measuring vital capacity. Trismus and general- ized myalgia with muscle tenderness suggest rhabdomyolysis (sea snakes). If a procoagulant venom is suspected, the coagulability of whole blood should be checked at the bedside using the 20WBCT. Antivenom treatment In managing cases of snakebite, the most important decision is whether or not to give antivenom, the only specific antidote for envenoming. There is abundant evidence that in patients with se- vere envenoming, the benefits of this treatment outweigh the risks of antivenom reactions (see following paragraphs). Antivenom has reduced the mortality of systemic envenoming by Echis ocellatus in Nigeria from 20% to 3% and by C. d. terrificus in Brazil from 74% to 12%. Antivenoms are effective in reversing hypotension caused by V.  berus envenoming and coagulopathies caused by Bothrops species, Russell’s vipers, C. rhodostoma, Trimeresurus T. albolabris, and Oxyuranus scutellatus. Antivenom, also known as antivenin, antivenene, antisnakebite serum, and anti-snakevenom (ASV) is the partially purified immunoglobulin (whole IgG, F(ab′)2, or Fab fragments) of horses or sheep that have been hyperimmunized with venom. Antivenoms are in short supply in sub-​Saharan Africa and New Guinea; elsewhere, they are of variable efficacy and safety and are often used inappropriately. Indications for antivenom Antivenom is indicated if there are signs of systemic envenoming such as: • haemostatic abnormalities:  spontaneous systemic bleeding, incoagulable blood, or thrombocytopenia • neurotoxicity: descending paralysis starting with ptosis and ex- ternal ophthalmoplegia • hypotension and shock, abnormal ECG, or other evidence of severe cardiovascular dysfunction • generalized rhabdomyolysis or massive intravascular haemolysis: black urine Supporting evidence of severe envenoming is a neutrophil leuco- cytosis, elevated serum enzymes such as CK and aminotransferases, haemoconcentration, severe anaemia, myoglobinuria, haemo- globinuria, methaemoglobinuria, hypoxaemia, and acidosis. In the absence of systemic envenoming, local swelling involving more than half the bitten limb, extensive blistering or bruising, bites on digits, and rapid progression of swelling are indications for anti- venom, especially in patients bitten by species whose venoms are known to cause local necrosis (e.g. Viperidae, Asian cobras, and African spitting cobras). Patients bitten by European Vipera spp. who show any evidence of envenoming should be admitted to hospital for observation for at least 24 h. Antivenom should be given whenever there is evidence of systemic envenoming (see earlier), even if its appearance is delayed for several days after the bite. Prediction of antivenom reactions Hypersensitivity testing by intradermal or subcutaneous injection or intraconjunctival instillation of diluted antivenom was widely practised in the past. However, these tests delay the start of anti- venom treatment, are not without risk, and have no predictive value for early (anaphylactic) or late (serum sickness-​type) antivenom reactions, because they are not usually the result of acquired IgE-​mediated type I hypersensitivity Prevention of early antivenom reactions Prophylactic antihistamines (anti-​H1 and anti-​H2), corticosteroids, and adrenaline have been widely used, singly or in combination, without convincing evidence of effectiveness. However, premedica- tion of 1007 Sri Lankan snakebite victims with promethazine,

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1796 hydrocortisone, and adrenaline in a subcutaneous adult dose of 0.25 ml of 1:1000 was compared, each alone and in various com- binations. Compared with placebo, adrenaline significantly reduced severe reactions to antivenom by 43% (95% CI, 25–67) at 1 hour and by 38% (95% CI, 26–49) up to and at 48 hours after antivenom administration. Hydrocortisone and promethazine were ineffective, and addition of hydrocortisone negated the benefit of adrenaline. Routine prophylaxis with low-​dose subcutaneous epinephrine (adult dose, 0.25 mg of 1: 1000 solution), given before starting anti- venom infusion, should now be generally recommended based on this convincing evidence. Contraindications to antivenom Atopic patients and those who have reacted previously to equine antiserum are at increased risk of developing severe antivenom re- actions. In such cases, antivenom should be given only if there is definite systemic envenoming. Reactions may be prevented or ameliorated by pretreatment with subcutaneous adrenaline (see earlier). There is no time for rapid desensitization. Selection and administration of antivenom Antivenom should be given only if its stated range of specificity in- cludes the species thought to be responsible for the bite. Whatever the stated expiry date on the ampoule, opaque solutions should be discarded, as precipitation of protein indicates loss of activity and an increased risk of reactions. However, expiry dates quoted on ampoules are often unnecessarily short, for commercial reasons; provided that the antivenom has been kept refrigerated and the so- lution is clear, a high proportion of its original activity is retained for 5 years or more. Monospecific (monovalent) antivenom is ideal if the biting species is known. Polyspecific (polyvalent) antivenoms are used in many countries because of the difficulty in identifying the species responsible for bites. Polyspecific and monospecific anti- venoms can be equally effective. Antivenoms may exhibit a range of paraspecific neutralizing activity. For example, the South African Vaccine Producer’s (formerly SAIMR) ‘polyvalent antivenom’, which is raised against the venoms of 10 species, has paraspecific activity against a further five species. It is almost never too late to give antivenom while signs of sys- temic envenoming persist, but, ideally, it should be given as soon as it is indicated. Antivenom has proved effective up to 2 days after sea snake bites and, in patients still defibrinogenated, weeks after bites by Viperidae. In contrast, local envenoming is probably not amenable unless antivenom is given within a few hours of the bite. The intravenous route is far more effective than intramuscular (Fig. 10.4.2.27). An infusion of antivenom diluted in approxi- mately 5 ml of isotonic fluid/​kg body weight may be easier to con- trol than an intravenous ‘push’ injection of undiluted antivenom given at the rate of about 4 ml/​min. However, there is no evidence that dilution, or slower administration of antivenom within the range 10–​120 minutes, affects the incidence or severity of early antivenom reactions. Dose of antivenom Manufacturers’ recommendations are based on mouse protection tests and may be very misleading. Few clinical trials have been per- formed to establish appropriate initial doses, and in most coun- tries this is judged empirically. Many clinical severity grading and scoring systems are in use to guide choice of the initial dose of anti- venom but none has been tested for its prognostic significance. The patient’s condition may deteriorate suddenly, making these rigid and unproven prescriptions unreliable. Many hospitals in the rural tropics give a standard dose of 1 to 2 ampoules to every patient who claims to have been bitten, irrespective of clinical severity. This practice squan- ders scarce, expensive antivenom, and exposes non-​envenomed pa- tients to the risk of reactions. Some suggested initial doses are given in Table 10.4.2.1. Children must be given the same dose as adults. Response to antivenom Often, there is marked symptomatic improvement soon after anti- venom has been injected. In shocked patients, the blood pressure may rise and consciousness returns (C. rhodostoma, V. berus, Bitis arietans). Neurotoxic signs may improve within 30 min (Acanthophis spp., N.  kaouthia), but the response usually take several hours. Spontaneous systemic bleeding usually stops within 15 to 30 min and blood coagulability is restored within a median time of 6 h after antivenom treatment, provided a neutralizing dose has been given. More antivenom should be given if severe signs of envenoming persist after 1 to 2 h, or if blood coagulability is not restored within about 6 h. Systemic envenoming may recur hours or days after an initially good response to antivenom. This is explained by the continuing absorption of venom from the injection site after clearance of antivenom from the bloodstream or redistribution of venom from the tissues into the vascular compartment. The apparent serum half-​lives of antivenoms in envenomed patients range from 26 to 95 h. Envenomed patients should therefore be assessed daily for at least 3 or 4 days. Antivenom reactions Early (anaphylactic) reactions  These reactions develop within 10 to 180 min of starting antivenom in between 3% and 84% of patients, depending on which antivenom is used. The inci- dence increases with dose and is lowest in antivenoms lacking 0.5 0.4 0.3 Fab antivenom (ul/ml) 0.2 0.1 0 0 1 3 6 12 24 48 Time after antivenom (hr) 1 VIAL Fab ANTIVENOM I.VI. COMPARED WITH 1 VIAL I.M.I. (9 PATIENTS VIA EACH ROUTE) Fab AV I.V. Fab AV I.M. Fig. 10.4.2.27  Serum therapeutic antivenom concentrations in two groups of patients with mild envenoming given the same dose of a Fab fragment antivenom by intramuscular or slow intravenous injection. Intramuscular administration resulted in delayed peak concentrations (at 24 h) sixfold less than by intravenous injection. Theakston RDG, Warrell DA, unpublished data.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1797 Table 10.4.2.1  Guide to initial dosage of selected important antivenoms (key to abbreviations at foot of table) Species Latin name Common name Manufacturer, antivenom Initial dose (approximate) Acanthophis spp. Death adders CSL,a monospecific 1–​3 vials Agkistrodon piscivorus, A. contortrix Copperhead and cottonmouth moccasins BTGg ‘CroFab’, Laboratorio Silanes (Mexico) Antivipmyn 4–​6 vials Bitis arietans Puff adder SAVPb polyspecific, ICPc EchiTAb-​plus-​ICP 80 ml Bothrops asper Terciopelo ICPc polyvalent, Laboratorio Silanes (Mexico) Antivipmyn TRI 50–​100 ml Bothrops atrox Common lancehead Butantan, FEDd Antibotropico 20 ml Bothrops (Bothriopsis) bilineatus Papagaio Butantan Antibotropico 20 ml Bothrops jararaca Jararaca Instituto Butantan, FEDd Antibotropico 20 ml Bothrops lanceolatus and B. caribbaeus Lesser Antillean fer de lance Sanofi-​Pasteur BothroFav 2–​6 vials Bungarus caeruleus Common krait Indian manufacturerse, polyvalent 100 ml Calloselasma (Agkistrodon) rhodostoma Malayan pit viper Thai Red Cross monovalent or haemato-​polyvalent 100 ml Thai Government Pharmaceutical Organization monovalent 50 ml Cerastes spp. Desert (horned) vipers NAVPCf polyvalent 30–​50 ml Vacsera AntiViper or polyvalent 30–​50 ml Crotalus adamanteus Eastern diamondback rattlesnake BCGg ‘CroFab’, Laboratorio Silanes (Mexico) Antivipmyn 7–​15 vials C. atrox Western diamondback rattlesnake BTGg ‘CroFab’, Laboratorio Silanes (Mexico) Antivipmyn 7–​15 vials C. oreganus and C. viridis sspp. Western rattlesnakes BTGg ‘CroFab’ or Laboratorio Silanes (Mexico) Antivipmyn 7–​15 vials C. simus and C. durissus sspp. Central and Southern American rattlesnakes ICPc or Laboratorio Silanes polyvalent 5–​15 vial Butantan, FEDd Anticrotalico, or Antibotropico-​crotalico 5–​20 vials Daboia (Vipera). Palaestinae Palestine viper Rogoff Medical Research Institute, Tel Aviv, Palestine, viper-​monospecific 50–​80 ml Daboia (Vipera) russelii Western Russell’s viper Indian manufacturerse polyspecific 100 ml D. siamensis Eastern Russell’s viper Thai Red Cross monovalent or haemato-​polyvalent 50 ml Myanmar Pharmaceutical Factory monovalent 80 ml Dendroaspis spp. Mambas SAVPb Dendroaspis or polyvalent 50–​100 ml Echis ocellatus, E. leucogaster, E. pyramidum (Africa) African saw-​scaled or carpet vipers SAVPb Echis monovalent ICP, EchiTAb-​plus-​ICP, MicroPharm EchiTAb-​G 20 ml, 3 vials, 1 vial Echis carinatus sspp. Asian saw-​scaled viper Indian manufacturerse, polyvalent 50 ml Echis spp. Middle East Middle Eastern saw-​scaled vipers NAVPCf Polyvalent Snake Antivenom Vacsera Polyvalent and Anti-​Viper Venom Antiserum 50 ml Hydrophiinae Sea snakes CSLa sea snake antivenom 1–​10 vials Lachesis spp. Bushmasters ICPc polyspecific, FEDd Antibotropico laquetico, Butantan Antiophidico 10–​20 vials Micrurus spp. Central American and Brazilian coral snakes ICPc monovalent 1–​5 vials Butantan Antielapidico 1–​5 vials Naja kaouthia Monocellate Thai cobra Thai Red Cross cobra monovalent or neuro-​polyvalent 100 ml N. naja Indian cobra Indian manufacturers,e polyvalent 100 ml N. nigricollis, N. mossambica, and so on African spitting cobras SAVPb polyvalent, ICPc EchiTAb-​plus-​ICP 100 ml Notechis scutatus Tiger snake CSL,a monospecific 1–​3 vials (continued)

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1798 complement-​activating aggregates. Fewer reactions occur when administration is by intramuscular rather than intravenous injec- tion. The symptoms are itching, urticaria, cough, nausea, vomiting, other autonomic manifestations, fever, and tachycardia. Up to 40% of patients with early reactions develop systemic anaphyl- axis: hypotension, bronchospasm, and angio-​oedema. Deaths are rare, but individual cases, such as the asthmatic boy who died from anaphylactic shock after receiving Pasteur antivenom in England in 1957, have been widely publicized and have led to an unreason- able rejection of antivenom treatment. Early antivenom reactions are unlikely to be type I, IgE-​mediated hypersensitivity reactions to equine serum protein. They result from complement activation by immune complexes or aggregates of IgG. Pyrogenic reactions  Pyrogenic reactions result from contam- ination of the antivenom with endotoxin-​like compounds. Fever, rigors, vasodilatation, and a fall in blood pressure develop 1 to 2 h after treatment. In children, febrile convulsions may be precipitated. Late serum sickness-​type reactions  Late reactions of serum sickness type may develop between 5 and 24 (mean 7) days after antivenom therapy. The incidence of these reactions and the speed of their de- velopment increases with the dose of antivenom. Clinical features include fever, itching, urticaria, arthralgia (sometimes involving the temporomandibular joint), lymphadenopathy, periarticular swell- ings, mononeuritis multiplex, albuminuria, and rarely, encephalop- athy. This is a classic immune complex disease. Treatment of antivenom reactions Adrenaline is the effective treatment for early reactions; 0.5 to 1.0 ml of 0.1% (1 in 1000, 1 mg/​ml) is given by intramuscular injection into the lateral thigh to adults (children 0.01 ml/​kg) at the first signs of a reaction. The dose may be repeated if the reaction is not controlled. Patients with profound hypotension, severe bronchospasm, or la- ryngeal oedema may be given adrenaline by slow intravenous in- jection (0.5 mg diluted in 20 ml of isotonic saline over 10–​15 min). For bronchospasm, a β2 agonist such as salbutamol should be given by inhaler or nebulizer, together with oxygen. A histamine anti-​H1 blocker, such as chlorphenamine maleate (10 mg for adults; 0.2 mg/​ kg for children) can be given by intravenous injection to combat the effects of histamine release during the reaction, but this is less urgent. Pyrogenic reactions are treated by physically cooling the patient and giving antipyretics. Late reactions respond to an oral antihistamine such as chlorphenamine (2 mg every 6 h for adults; 0.25 mg/​kg per day in divided doses for children) or to oral prednis- olone (5 mg every 6 h for 5 to 7 days for adults; 0.7 mg/​kg per day in divided doses for children). Supportive treatment Neurotoxic envenoming Bulbar and respiratory paralysis may lead to death from aspiration, airway obstruction, or respiratory failure. A clear airway must be maintained and, if bulbar muscle weakness results in pooling of secretions, or respiratory distress develops, a cuffed endotracheal tube, laryngeal mask airwayor i-​gel supraglottal airway should be inserted or a tracheostomy performed. Provided they are ad- equately ventilated, patients with neurotoxic envenoming remain fully conscious with intact sensation and can respond to spoken questions by flexing a finger or toe. Lifting their paralysed eyelids so that they can see is very reassuring. Patients have been effectively ventilated manually (by Ambu bag or anaesthetic bag), as in the 1952 poliomyelitis epidemic in Copenhagen, for 30 days and have recovered after 10 weeks of mechanical ventilation. Although artifi- cial ventilation was first suggested for neurotoxic envenoming more than 100 years ago, patients continue to die because they are denied this simple procedure. Anticholinesterases have a variable but potentially useful ef- fect in patients with neurotoxic envenoming, especially when postsynaptic neurotoxins are involved. However, recent media claims that intranasal neostigmine might provide a universal first-​ aid method for snakebite victims are unsubstantiated, misleading, and fanciful. The ‘Tensilon test’ should be performed in all cases of severe neurotoxic envenoming, as with suspected myasthenia gravis. Atropine sulphate (0.6 mg for adults; 50 µg/​kg for children) or glycopyrronium is given by intravenous injection followed by edrophonium chloride (Tensilon) by slow intravenous injection in an adult dose of 10 mg, or 0.25 mg/​kg for children or neostigmine bromide or methylsulphate (Prostigmin) by intramuscular injec- tion (0.02 mg/​kg for adults, 0.04 mg/​kg for children). The ‘ice test’ is a possible alternative to the Tensilon test. In myasthenia gravis, application of an ice-​filled plastic glove to one eye for 2 minutes results in improvement in ptosis on that side, due to inhibition of anticholinesterase. Patients who respond convincingly can be maintained on neostigmine methylsulphate, 0.5 to 2.5 mg every Species Latin name Common name Manufacturer, antivenom Initial dose (approximate) Oxyuranus scutellatus Australian and New Papuan Taipan CSL,a monospecific 1–​6 vials Pseudonaja textilis Eastern brown snake CSL,a monospecific 1–​3 vials Trimeresurus (Trimeresurus) albolabris, T. (T.) macrops White-​lipped and large-​eyed green pit viper Thai Red Cross monovalent or haemato-​polyvalent 100 ml Vipera berus and other European Vipera European adder and other vipers MicroPharm ‘ViperaTAb’, Sanofi-​Pasteur ViperFav 100–​200 mg, 4 ml a CSL, Commonwealth Serum Laboratories, Australia; b SAVP, South African Vaccine Producers (formerly SAIMR: South African Institute for Medical Research); c ICP, Instituto Clodomiro Picado, Costa Rica; dFED, Fundação Ezequiel Dias, Brazil; eIndian manufacturers (Vins, Bharat, Premium Serums and Vaccines, and so on); f NAVCP, National Antivenom and Vaccine Production Center, KSA; g BTG (formerly Protherics). Table 10.4.2.1  Continued

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1799 1 to 3 h up to 10 mg/​24 h maximum for adults or 0.01 to 0.04 mg/​kg every 2 to 4 h for children by intramuscular, intravenous, or sub- cutaneous injection. Hypotension and shock If the central venous pressure is low or there is other clinical evidence of hypovolaemia, isotonic saline should be infused. If there is evidence of increased capillary permeability (e.g. facial and conjunctival oedema, serous effusions, haemoconcentration, hypoalbuminaemia, and so on) it may be safer in the long term to rely on a selective vasoconstrictor such as dopamine (starting dose 2.5–​5 µg/​kg per min by intravenous infusion). Delayed hypotension that develops about 1 week after bites by Burmese D. siamensis is a consequence of acute pituitary-​adrenal insufficiency (Sheehan’s-​like syndrome) responds to intravenous hydrocortisone. Oliguria and acute kidney injury Urine output, serum creatinine, urea, and electrolytes should be measured each day in patients with severe envenoming, and in those bitten by species known to cause acute kidney injury (e.g. Russell’s vipers, hump-​nosed pit viper (Hypnale hypnale), C.d. terrificus, Bothrops spp., sea snakes). If urine output drops below 400 ml in 24 h, urethral and central venous catheters should be inserted. If urine flow fails to increase after cautious rehydration, patient should be placed on strict fluid balance. Dopamine (2.5 µg/​kg per min by intravenous infusion) has proved effective in some patients bitten by Russell’s vipers, although its use has been largely abandoned by nephrologists. Renal replacement therapy (peritoneal or haemo- dialysis or haemofiltration) will usually be required. In Rangoon, Burma, the mortality of acute kidney injury following D. siamensis envenoming has been reduced to less than 30% by using peritoneal dialysis, usually for only 72 h. Local infection at the site of the bite After bites by some species (e.g. Bothrops spp., C. rhodostoma), local infections caused by unusual bacteria derived from the snake’s venom or fangs develop in 10% or more cases. A booster dose of tetanus toxoid should be given, but prophylactic antibiotics are not indicated unless the wound has been incised or tampered with in any way or if there is necrosis with the associated risk of Clostridium tetani and other anaerobes. If a local abscess develops, it should be drained and the pus cultured. Penicillin, chloram- phenicol, or erythromycin are usually effective. An aminoglycoside such as gentamicin should be given for 48 h if there is evidence of local necrosis. Management of local envenoming Bullae are best left intact. The bitten limb should be nursed in the most comfortable position but not elevated excessively as this in- creases the risk of intracompartmental ischaemia. Once definite signs of necrosis have appeared (blackened anaesthetic area with putrid odour or signs of sloughing), surgical debridement, imme- diate split-​skin grafting, and broad-​spectrum antibiotic cover are indicated. Intracompartmental syndrome and fasciotomy Swelling of envenomed tissues within tight fascial compartments such as the digital pulp spaces and anterior tibial compartment may cause ischaemia that adds to the risk of venom-​induced ne- crosis. This may explain why digital bites are so often necrotic. The classic signs of ‘compartment syndrome’ are excessive pain, weak- ness and tenderness of the compartmental muscles, and pain when they are passively stretched, hypoaesthesia of skin supplied by nerves running through the compartment, and obvious tenseness of the compartment. Misleadingly, these signs are frequently pre- sent in snakebitten limbs in which intracompartmental pressures are normal. Recent studies in the United States failed to demon- strate any benefit of fasciotomy in snakebite victims. In any case, fasciotomy is absolutely contraindicated until blood coagulability has been fully restored (by adequate doses of antivenom followed by clotting factors). Surgery must be justified by demonstrating that intracompartmental pressure is consistently raised to less than 30 mm Hg below mean arterial pressure, or it exceeds 45 mm Hg in adults or 30 mm Hg in children, when measured directly with a Stryker transducer. Haemostatic disturbances Once specific antivenom has been given to neutralize venom pro- coagulants, restoration of coagulability and platelet function may be accelerated by giving (reliably screened) fresh whole blood, fresh frozen plasma, cryoprecipitates (containing fibrinogen, factor VIII, fibronectin, and some factors V and XIII), or platelet concentrates. Heparin has been used to treat a variety of snakebites, usually with disastrous results. Heparin did not prove beneficial in patients envenomed by Echis ocellatus. Other drugs Corticosteroids, antifibrinolytic agents (aprotinin and ε-​aminocaproic acid), antihistamines, trypsin, and a variety of traditional herbal rem- edies have all been used, but none has proved effective and most are potentially harmful. Treatment of snake venom ophthalmia caused
by spitting cobras and rinkhals First-​aid treatment involves urgent decontamination of the af- fected eye(s) using large volumes of water or any other available bland fluid (even urine!). A single topical administration of local anaesthetic drops such as 0.4% oxybuprocaine hydrochloride, 4% lidocaine hydrochloride, or tetracaine hydrochloride drops cures the agonizing pain. Adrenaline (0.1%) drops are also effective. Corneal abrasions must be excluded by fluorescein staining and/​ or slit-​lamp examination. A  prophylactic topical antibiotic such as tetracycline, chloramphenicol, soframycin, ciprofloxacin, or gatifloxacin should be instilled. Posterior synechiae, ciliary spasm, and discomfort are prevented with 2% atropine, scopol- amine, or homatropine. In case of allergic keratoconjunctivitis in someone previously spat at, topical antihistamines are used. Topical or intravenous antivenom and topical corticosteroids are contraindicated. Interval between bite and death Exceptionally, patients may die ‘within a few minutes’ (reputedly after a bite by the king cobra Ophiophagus hannah) or as long as 41  days (Echis carinatus) after snakebite. However, most deaths occur about 8 h after cobra bites (N.  naja), 18 h after krait bites (Bungarus caeruleus), 16 h after North American rattlesnake bites

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1800 (Crotalus spp.), 3 days after Russell’s viper bites, and 5 days after Echis bites. Venomous lizards Two species of venomous lizard (genus Heloderma) have proved capable of envenoming humans. Venom from submandibular glands pools in labial gutters in the lower jaw and is conducted along grooves in the lower teeth. The Gila monster H.  suspec- tum (Fig. 10.4.2.28), which is striped with a short thick tail and grows to 55  cm in length, occurs in the south-​western United States and adjacent areas of Mexico. The Mexican beaded lizard or escorpión H. horridum, which is spotted with a relatively long thin tail and reaches 1 m in length, is found in western Mexico south to Guatemala. Heloderma venoms contain lethal glycopro- tein toxins, Gila, and horridum toxins, phospholipase A2, and 5 bioactive peptides of great interest, including helospectin (a vaso- active intestinal peptide analogue) and exendins-​3 and -​4, which are glucagon-​like peptide-​1 (GLP-​1) homologues that stimulate insulin secretion and inhibit glucagon secretion. A  synthetic homologue of exendin-​4, exenatide, is a high affinity GLP-​1 re- ceptor agonist which has been developed for treatment of type 2 diabetes mellitus. Bites are rare and are usually inflicted on the fingers, hands, and forearms of inebriated young men who are handling or trying to catch the lizards. The lizard hangs on with its powerful jaws and is difficulty to disengage. Expert opinion currently favours levering the jaws apart with a screw driver, run- ning the cold tap over the attached lizard, placing its four feet on the ground or introducing some alcohol into its mouth. There is immediate severe throbbing or burning local pain that radiates up the limb with tender swelling and regional lymphadenopathy. Systemic symptoms include weakness, dizziness, tachycardia, hypotension, syncope, angio-​oedema, sweating, rigors, tinnitus, nausea, and vomiting. There may be leucocytosis, coagulopathy, electrocardiographic changes, myocardial infarction, and acute kidney injury. No fatal cases have reliably been reported. Specific antivenom is not generally available. A powerful analgesic may be required. Hypotension should be treated with fluids, adrenaline, or a pressor agent such as dopamine. Recently, venomous salivary secretion have been demonstrated in other groups of lizards such as iguanas (Iguanidae), glass/​alli- gator lizards (Anguidae), and monitors (Varanidae), notably the Komodo dragon Varanus komodoensis that has been responsible for human fatalities that were attributed to trauma of infection of the bite wounds. Poisonous amphibians and birds Poisonous amphibians The moist skin of amphibians such as frogs, toads, newts, and sala- manders is an accessory respiratory organ, which is protected from microorganisms by highly toxic secretions containing amines, peptides, proteins, steroids, and alkaloids. Some compounds are synthesized de novo, while others are sequestered from prey such as ants, beetles, and millipedes. The bitter flavour and lethal effects of these secretions and the vivid warning colouration of many spe- cies defend them against predators. The skin of ‘poison-​dart’ frogs (Dendrobatidae) of Central and South America secrete lipophilic alkaloids such as batrachotoxins (Phyllobates spp.), which activate sodium channels; histrionico toxins (Dendrobates histrionicus) (Fig. 10.4.2.29), which block nicotinic receptors; pumiliotoxins (D.  pumilio), which affect sodium channels; and epibatidine (Epipedobates tricolor), a powerful analgesic and nicotinic receptor Fig. 10.4.2.28  Gila monster Heloderma suspectum. Copyright D. A. Warrell. Fig. 10.4.2.29  Poison frog Dendrobates histrionicus (Dendrobatidae) Bahia Solauo, Colombia. Its skin secretion contains potent nicotinic receptor antagonists, histrionicotoxins. Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1801 agonist. Two Colombian tribes, the Embará and Noanamá Chocó, use the skin poisons of three species of Phyllobates to coat the tips of their blow-​gun darts. Some toads can squirt from their parotid glands venom containing bufadienolides which affect membrane Na+, K+-​ATPase. When licked or put in the mouth by dogs or chil- dren, or when ingested as Chinese traditional medicines such as Kyushin, Yixin Wan, or the topical aphrodisiac Ch’an-​Su, the poi- sons can cause fatal digoxin-​like poisoning. Symptoms include hypersalivation, cyanosis, cardiac arrhythmias, and generalized convulsions. Antidigoxin antibodies (‘Digibind’, ‘DigiTAb’) have some therapeutic effect. The skin of three species of newts, genus Taricha, from the western United States, contains tarichatoxins identical to tetro- dotoxin, which also occurs in some toads, frogs, fish, crustaceans, and octopuses (see following paragraphs). Tetrodotoxin can be absorbed through the gastric mucosa, explaining the death of a man who swallowed a 20-​cm long Oregon rough-​skinned newt Taricha granulosa. He developed paraesthesia of the lips, pro- gressing to more generalized numbness and weakness, and had a cardiopulmonary arrest about 2 h after swallowing the newt. Poisonous birds The feathers, skin, and breast muscles of five species of pitohui or thickhead, passerine birds from New Guinea (genus Pitohui; Pachycephalidae) and the blue-​capped ifrita or ifrit (Ifrita kow- aldi; Cinclosomatidae) contain homobatrachotoxin, a potent steroidal alkaloid that activates sodium channels and was ori- ginally isolated from the skin of South American poison-​dart frogs (Phyllobates—​see earlier paragraphs). The birds may ac- quire the poison by eating melyrid beetles (Choresine spp.). Poisonous pitohuis have an unpleasant peppery odour, and their skin has a bitter flavour. Contact with their feathers causes numbness and burning of the tongue, lip or skin wounds, and sneezing. This may be a protective mechanism, and the striking ‘warning’ colouration of the hooded pitohui (P.  dichrous) (Fig. 10.4.2.30) may be the subject of Müllerian mimicry by less poisonous species. Venomous fish About 200 species of fish inhabiting temperate and tropical seas possess a defensive venom-​injecting apparatus that can inflict dangerous stings, but more than 1200 species are now thought to be venomous. Fatal stings have been reported from cartilagenous fish (class  Chondrichthyes), such as sharks and dogfish (order Squaliformes) and stingrays and mantas (order Rajiformes), and from bony fish (superclass Osteichthyes), such as ray-​finned fish (class Actinopterygii) of the orders Siluriformes (catfish), Perciformes (families Trachinidae (weever fish), Uranoscopidae (stargazers or stone-​lifters), and others) and Scorpaeniformes (scorpion fish, stonefish, lion fish Synanceja/​Synanceia spp.) (Fig. 10.4.2.31). Two species of lion fish have been introduced into the Atlantic ocean and now occupy the SE coast of the United States and the Caribbean, posing an ecological threat to the region. The Indo-​Pacific region and other tropical waters have the richest venomous fish fauna, but dangerous species such as sharks, chimaeras, and weevers also occur in temperate northern waters, and several large rivers in South American, West Africa, and Southeast Asia are inhabited by freshwater stingrays Potamotrygon spp. (Fig. 10.4.2.32). Venom glands are embedded in grooves in the spines or, in the case of stingrays, lie beneath a membrane covering the long barbed precaudal spine. Incidence and epidemiology Weever fish are common around the coast of the British Isles, especially off Cornwall. Hundreds of stings occur in some years, with a peak incidence in August and September. It has been esti- mated that there are 1500 stings by rays and 300 stings by scorpion fish in the United States each year. Stings by venomous freshwater rays (Potamotrygon hystrix, P. motoro) are common in the Amazon region of Brazil. Ornate, but aggressive and venomous members of the genera Pterois and Dendrochirus (lion, zebra, tiger, turkey, or red fire fish) (Fig. 10.4.2.31), which are popular aquarium pets, often sting their owners on the fingers. Most fish stings are inflicted on the soles of the feet of people wading near the shore Fig. 10.4.2.30  Hooded pitohui Pitohui dichrous. Varararta National Park near Port Moresby, Papua New Guinea. By courtesy of Dr Ian Burrows. Fig. 10.4.2.31  Lion fish Pterois volitans (Scorpionidae). Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1802 or in the vicinity of coral reefs. Venomous fish are effectively cam- ouflaged (Synanceja spp.) or lie partly covered by sand. Stingrays lash their tails at the intruding limb and usually impale the ankle (Fig. 10.4.2.33). Fatal fish stings are very rarely reported. Prevention Fish stings can be prevented by employing a shuffling gait when wading, by avoiding handling living or dead fish, and by keeping clear of fish in the water, especially in the vicinity of tropical reefs. Footwear protects against most species except stingrays. Venom composition The instability of most fish venoms at normal ambient temperat- ures has made them difficult to study. Stingray and weeverfish venoms contain peptides, enzymes, and a variety of vasoactive compounds such as kinins, 5-​hydroxytryptamine, histamine, and catecholamines. Pharmacological effects include local necrosis, direct actions on cardiac, skeletal, and smooth muscle, resulting in ECG changes, hypotension, paralysis, and central nervous system depression. Clinical features Immediate sharp, agonizing pain is the dominant symptom. Hot, erythematous swelling extends up the stung limb and may per- sist with pain for several days and be complicated by necrosis (Fig. 10.4.2.33) and secondary infection by marine Vibrio spp. (such as V.  vulnificus), freshwater species (such as Aeromonas hydrophila), and other unusual bacteria, particularly if the spine remains embedded in the wound. Stingray spines, which are up to 30 cm long, can cause severe lacerating injuries, especially to the lower legs, but if the victim inadvertently lies on the ray or falls on to it, the spine may penetrate the thoracic or abdominal cavities with fatal results. Systemic effects are uncommon after weever stings (Trachinidae), but people stung by rays or Scorpaenidae (scorpion-​ and stonefish) may develop nausea, vomiting, signs of autonomic nervous system stimulation; such as diarrhoea, sweating, and hypersalivation; car- diac arrhythmias, hypotension, respiratory distress, neurological signs, and generalized convulsions. Patients have died within 1 h of being stung by Synanceja verrucosa. Treatment Pain is alleviated by immersing the stung limb in water, which is uncomfortably hot yet not scalding (<45°C; the 50°C recommended by some authorities will cause a full thickness scald!). Temperature can be assessed with the unstung limb. Addition of magnesium sul- phate is unnecessary. Injection of a local anaesthetic is less effective even when applied as a ring block in the case of stung digits, but a local nerve block with 0.5% of plain bupivacaine is effective. The venomous spine (which may be barbed), fragments of membrane, and other foreign material should be removed as soon as possible. Systemic effects must be treated symptomatically. An adequate airway should be established, and cardiopulmonary resuscitation may be needed. Severe hypotension may respond to adrenaline, bradycardia to atropine. Seqirus (formerly CSL) in Australia manu- facture an antivenom specific for Synanceja trachynis, S. verrucosa, and S. horridus. This has paraspecific activity against the venoms of the North American scorpion fish (Scorpaena guttata) and some other members of the Scorpaenidae. One ampoule (2 ml or 2000 units) is given intravenously for each two puncture marks found at the site of the sting. The dose is increased for patients with se- vere symptoms. Antibiotic treatment for secondary infections should take into account the range of possible marine pathogens. Doxycycline or co-​trimoxazole covers Vibrio and Aeromonas spp. Poisoning by ingestion of aquatic animals Acute gastrointestinal symptoms (‘food poisoning’) after eating sea- food are usually caused by bacterial or viral infections such as Vibrio parahaemolyticus (crustaceans, especially shrimps), V.  cholerae (crabs and molluscs), non-​O group 1 V. cholerae (oysters), V. vulnifi- cus (oysters), Aeromonas hydrophila (frozen oysters), Plesiomonas shigelloides (oysters, mussels, mackerel, cuttlefish), Shigella spp. Fig. 10.4.2.32  Fresh water stingray Potamotrygon sp. Copyright D. A. Warrell. Fig. 10.4.2.33  Necrotic and secondarily infected wound at the site of a sting by a freshwater ray Potamotrygon hystrix in a Brazilian patient. By courtesy of Dr João Luiz Costa Cardoso.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1803 (molluscs), Campylobacter jejuni (clams), Salmonella typhi (mol- luscs), hepatitis A virus (molluscs, especially clams, and oysters), Norwalk virus (clams and oysters), and astro-​ and calici-​ viruses (cockles and other molluscs). Botulism has been caused by eating smoked fish and canned salmon; and in Japan and elsewhere, fish and molluscs became contaminated with methyl mercury from in- dustrial waste, causing severe neurological damage and fetal abnor- malities (‘Minamata disease’). Natural toxins acquired in the food chain, originally from bac- teria can contaminate the tissues of a variety of fish, shellfish (bivalve molluscs) and other marine animals, giving rise to the several dis- tinctive syndromes of seafood poisoning. Ciguatera fish poisoning Symptoms develop between 1 and 6 h (extreme range, min to 30 h) after eating fish such as groupers, snappers, parrot fish, mackerel, moray eels, barracudas, and jacks. These are warm-​water shore or reef fish. The global incidence is thought to exceed 50 000 cases per year. Up to 2% of the population may be affected each year (e.g. in Kiribati, Tokelau, and Tuvalu in the Pacific region) with a case fatality of 0.1%. The toxins responsible are polyethers such as ciguatoxin (activates Na+ channels), maitotoxin (activates Ca2+ channels), and scaritoxin, ultimately derived along the food chain from benthic dinoflagellates such as Gambierdiscus toxicus. They are concentrated in the liver, viscera, and gonads, especially of large carnivorous fish. The increasing market for exotic fish from the Caribbean and else- where has led to cases of ciguatera in the United Kingdom. Acute gastrointestinal symptoms—​nausea, vomiting, diarrhoea, abdominal pain and cramps, and a metallic taste in the mouth—​ are followed by neurological symptoms—​paraesthesiae around the mouth and extremities, reversed hot-​cold sensation (dysesthesia), increased salivation, dilatation of the pupils, strabismus, ptosis, weakness, and ataxia, usually resolve within a few hours, but paraes- thesiae and myalgia may persist for a week, or even months. Pruritus of the soles and palms and rashes may occur. Cardiovascular fea- tures include braycardia, hypotension, and hypovolemia. Similar symptoms (chelonitoxication) may follow ingestion of marine tur- tles in the Indo-​Pacific area, with a much higher case fatality. Tetrodotoxin poisoning Scaleless fish, such as porcupine, sun, puffer, and toad fish (order Tetraodonitiformes) may become highly poisonous at certain seasons, such as May to June, the spawning season in Japan. Tetrodotoxin, an aminoperhydroquinazoline, is one of the most potent non​protein toxins known. It produces neurotoxic and cardiotoxic effects by blocking voltage-​gated sodium ion chan- nels. It is found concentrated in the ovaries, viscera, and skin of tetraodontiform fish; in the skin of newts (genus Taricha), frogs, and toads (genera Colostethus, Atelopus, Bracycephalus), and salamanders; in the saliva of octopuses; in the digestive glands of several species of gastropod molluscs; in a starfish, flatworm Planorbis spp., and nemertine worms in Japan; and is produced by some bacteria (Pseudomonas, Pseudoalteromonas, Vibrio spp., and so on). Puffer fish (‘fugu’) is particularly popular in Japan where, des- pite stringent regulations, there are still cases of tetrodotoxin poi- soning, with about four deaths each year. Nausea and abdominal pain occur but usually no vomiting or diarrhoea, or there may be no gastrointestinal symptoms. Neurotoxic symptoms character- ized by rapid onset, within 10–​45 minutes, of weakness, dizziness, paraesthesiae of the lips, tongue, throat and, later, the limbs. Pallor, sweating, and increased salivation may be present. Tachycardia, hypotension, difficulty breathing, and flaccid ascending paralysis may lead to respiratory paralysis; death usually occurs 2–​6 hours after eating the fish. Usually, consciousness is retained throughout, although victims may appear comatose. Development of fixed di- lated pupils and brain stem areflexia suggests brain death, but com- plete recovery is possible with mechanical ventilation. Freshwater puffer fish poisoning in northern Thailand has been attributed to saxitoxin. Histamine-​like syndrome (scombrotoxic poisoning) The dark red flesh of scrombroid fish (tuna, mackerel, bonito, skip- jack) and of canned non​scrombroid fish (sardines, pilchards) may be decomposed by the action of bacteria, such as Proteus morgani and Klebsiella pneumoniae, which decarboxylate muscle histidine into saurine, histamine, cadaverine, and other unidentified toxins: 100 g of spoiled fish may contain almost 1 g of histamine. Histamine ab- sorbed from the gut is normally broken down by N-​methyl trans- ferase and diamine oxidase (histaminase), but if the histamine concentration is very high, or the patient is taking a diamine oxi- dase inhibitor such as isoniazid (as antituberculosis chemotherapy), scombrotoxic poisoning may result. Toxic fish may produce a tin- gling or smarting sensation in the mouth when eaten. Within min- utes or up to a few hours after ingestion, flushing, burning, sweating, urticaria, and pruritis may develop with headache, abdominal colic, nausea, vomiting, diarrhoea, bronchial asthma, giddiness, and hypotension. Poisoning by ingesting carp gallbladder In parts of East Asia, the raw bile and gallbladder of various spe- cies of freshwater carp (e.g. the grass carp Ctenopharyngodon idel- lus, ‘plaa yeesok’ Probarbus jullienii) are believed to have medicinal properties. Patients in China, Taiwan, Hong Kong, Japan, Thailand, and elsewhere have developed acute abdominal pain, vomiting, and watery diarrhoea 2 to 18 h after drinking the raw bile or eating the raw gallbladder of these fish. One patient developed flushing and dizziness. Hepatic and renal damage may develop, progressing to oliguric or non​oliguric acute kidney injury (acute tubular necrosis). The hepatonephrotoxin has not been identified, but is heat stable and may be derived from the carp’s diet. Paralytic shellfish poisoning Five main clinical syndromes of shellfish poisoning are recognized. Many of the causal toxins, derived from dinoflagellates and diatom algae, have been identified but no specific antidotes have been discovered. Diarrhoeal shellfish poisoning:  Symptoms indistinguish- able from acute infective gastroenteritis evolve usually within 30 minutes to 12 hours of eating the contaminated shellfish—​ diarrhoea, nausea, vomiting, and abdominal colic with recovery over a few days. The toxins responsible include okadaic acid and other dinophysis toxins, pectenotoxins, and yessotoxins, many of which are protein phosphatase inhibitors. They occur in mus- sels, cockles, scallops, oysters, whelks, and green crabs in Japan, Europe, and Africa.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1804 Neurotoxic shellfish poisoning:  Milder gastrointestinal and neurotoxic symptoms without paralysis develop 1–​3 hours after ingestion of molluscs contaminated by brevetoxins from Karenia brevis (formerly known as Gymnodinium breve or G. brevis) dino- flagellates which can also cause ‘red tides’. Neurotoxic symptoms, reminiscent of ciguatera fish poisoning, include paraesthesiae, cold allodynia (pain or hyperaesthesia on touching cold objects), my- algia, vertigo, and ataxia. In the United Kingdom there have been several outbreaks of neurotoxic red-​whelk (Neptunea antiqua) poi- soning attributable to tetramine. Paralytic shellfish poisoning: Bivalve molluscs, such as mussels, clams, oysters, cockles, and scallops (and also xanthid, coconut, and horseshoe crabs) may acquire tetrahydropurine neurotoxins such as saxitoxin and gonyautoxins from dinoflagellates (Alexandrium spp., Pyrodinium bahamense var. compressum and Gymnodinium catena- tum). These may be sufficiently abundant between latitudes 30 ºN and 30 ºS during the warmer months of May to October to produce a ‘red tide’. The dangerous season is signalled by the deaths of large numbers of fish and sea birds. Symptoms develop within 30 min of ingestion. Descending paralysis may progress to fatal respiratory paralysis within 12 h in 8% of cases. Amnesic shellfish poisoning: Develops after ingestion of mussels and other molluscs contaminated with domoic acid from diatoms (Pseudonitzschia spp.). Gastroenteritis starts within 24 h of exposure and, in in about half the cases, neurotoxic symptoms develop within 48 h. Severe headache and short-​term memory loss are common. Amnesia is sometimes permanent. Severe symptoms include agi- tation, seizures, coma, profuse respiratory secretions, circulatory instability, and death. Elderly patients and those with underlying illnesses were most vulnerable. Azaspricid poisoning: Acute gastrointestinal symptoms develop within 6–​18 hours of Ingestion and last up to 5 days. Diagnosis and treatment of seafood poisoning The differential diagnosis includes bacterial and viral food poi- soning and allergic reactions. No specific treatments or antidotes are available, but gastrointes- tinal contents should be eliminated by emetics and purges if this can be achieved safely and within 1 to 2 h of ingestion. Activated char- coal adsorbs saxitoxin and other shellfish toxins. Atropine is said to improve gastrointestinal symptoms and sinus bradycardia in patients with gastrointestinal and neurotoxic poisoning. Calcium gluconate may relieve mild neuromuscular symptoms. Oximes and anticholinesterases appear ineffective in ciguatera and tetrodo- toxin poisoning, respectively. Mannitol has been advocated for early treatment of ciguatera poisoning. In cases of paralytic poisoning, endotracheal intubation and mechanical ventilation and cardiac re- suscitation have proved life-​saving. The symptoms of scrombrotoxic poisoning can be alleviated with antihistamines and bronchodilators. Prevention of poisoning by ingestion of aquatic animals Ciguatera toxin, tetrodotoxin, scombrotoxins, and most other marine toxins are heat stable, so cooking does not prevent poi- soning. Some toxins are fairly water soluble and may be leached out by soaking. Therefore, water in which fish are cooked should not be drunk. In tropical areas, the flesh of fish should be separ- ated as soon as possible from the head, skin, intestines, gonads, and other viscera, in which toxins are concentrated. All scaleless fish should be regarded as potentially tetrodotoxic, and very large fish carry an increased risk of being ciguatera toxic. Moray eels and parrot fish (Scaridae) should never be eaten because of the high risk of unusually rapid and severe ciguatera and scaritoxic fish poi- soning. Scrombroid poisoning can be prevented by eating fish fresh or by freezing fish as soon as possible after they are caught. Shellfish should not be eaten during the dangerous seasons and when there are red tides. Venomous marine invertebrates Cnidarians (Coelenterata) These include jellyfish, cubomedusoids, sea wasps, Portuguese-​men-​ o’-​war, or bluebottles, hydroids, stinging corals, and sea anemones. Their tentacles are armed with millions of nematocysts (stinging capsules). When triggered by contact or chemicals, stinging hairs are everted at enormous acceleration and force, penetrating the skin as far as the epidermo-​dermal junction and producing lines of painful irritant weals. Cnidarian venoms contain peptides and other vasoactive substances such as 5-​hydroxyhistamine, histamine, pros- taglandins, and kinins, which cause immediate excruciating pain, inflammation, and urticaria. Epidemiology The most dangerous species, the box jellyfish, cubomedusoid, sea wasp, or indringa Chironex fleckeri of northern Australia, has caused more than 70 deaths since 1883. Most stings occur in December and January. Chiropsalmus quadrumanus and C. quadrigatus have caused fatal jellyfish stings in the Indo-​Pacific region. Portuguese men-​o’-​war (Physalia spp.), Chinese jellyfish (Stomolophus nomu- rai) and tiny cubomedusoids (Carukia barnesi) responsible for many ‘Irukandji stings’ in northern Queensland, Florida, and Guadeloupe in the Caribbean, have also caused fatalities. Pelagia noctiluca may swarm in vast numbers off the northern Adriatic coast, stinging many swimmers. The North American sea nettle (Chrysaora quin- quecirrha) is widely distributed throughout the Atlantic and Indo-​ Pacific oceans and is especially abundant in Chesapeake Bay on the Maryland coast. There are millions of stings each year but no fatalities Prevention Bathers, especially children, should keep out of the sea at times of the year when dangerous cnidarians are prevalent, especially when warning notices are displayed; or they should bathe in ‘stinger-​ resistant’ enclosures, although these do not exclude Irukandji. Wetsuits or Lycra garments, nylon stockings, and other clothing will protect against nematocyst stings. Clinical features Immediate severe pain is the commonest symptom. Nematocyst stings may leave a diagnostic pattern on the skin: C. fleckeri produces wide, striated brownish-​purple weals (Fig. 10.4.2.34), whereas Carukia barnesi causes a transient erythematous macule, and the Portuguese man-​o’-​war (genus Physalia) produces chains of oval weals surrounded by erythema. Chirodropids (genera Chironex and Chiropsalmus) cause the most severe systemic symptoms—​ respiratory arrest, generalized convulsions, pulmonary oedema,

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1805 and cardiac arrest within minutes of the accident. Other systemic effects include cough, nausea, vomiting, abdominal colic, diar- rhoea, rigors, severe musculoskeletal pains, and profuse sweating. ‘Irukandji syndrome’ consists of severe musculoskeletal pain, anx- iety, trembling, headache, piloerection, sweating, tachycardia, hypertension, and pulmonary oedema starting about 30 min after a sting by C. barnesi (and by other species of tiny cubomedusoids) and persisting for hours. Physalia species can also cause severe sys- temic envenoming, including intravascular haemolysis, peripheral gangrene, and acute kidney injury. Treatment Victims of box jellyfish stings may die within minutes and so first aid is urgent. The victim is taken out of the water to prevent drowning, adherent tentacles are washed off with sea water or removed by shaving the skin and hot water is applied to relieve pain as for fish stings (see earlier). Undischarged nematocysts in adherent tentacles should be inhibited. For Chironex and other cubozoans, including Irukandji, commercial vinegar or 3–​10% aqueous acetic acid may inhibit further nematocysts discharge, although this has become controversial, but is not recommended for stings by Physalia or Stomolophus. For Chrysaora stings, baking soda and water (50% w/​ v) is used. In vitro, several popular remedies, such as alcohol (in sun lotion), ammonia, acetic acid, and meat tenderizer, caused massive discharge of Chrysaora quinquecirrha and Physalia physalis tent- acles. However, 5–​15% lignocaine hydrochloride prevented dis- charge and relieved the pain of Chiropsalmus quadrumanus and Chrysaora quinquecirrha stings, in proportion to the concentra- tion applied. Pressure-​immobilization with a crepe bandage may increase the amount of venom injected and is not recommended. Cardiopulmonary resuscitation has proved life-​saving in several Australian patients stung by C. fleckeri who became cyanosed, co- matose, and pulseless. A specific ‘sea wasp’ antivenom for C. fleckeri is manufactured in Australia but its efficacy is being questioned. Treatment with verapamil is not recommended. Starfish and sea urchins (Echinodermata) These animals are protected by hard exoskeletons with numerous long, sharp projecting spines (Fig. 10.4.2.35) and grapples (globiferous pedicellariae), which can release venom and a violet-​ coloured liquid when embedded in the skin. Severe pain and local swelling may result, and sometimes systemic effects such as syn- cope, numbness, generalized paralysis, aphonia, respiratory dis- tress, cardiac arrhythmias, and even death. Embedded fragments of spines may penetrate bones and joints and lead to secondary infection and chronic granulomas. Treatment Hot water (see earlier paragraphs) may relieve the pain. Skin penetrated by the spines, usually the soles of the feet, should be softened with 2% salicylic acid ointment or acetone. An attempt should then be made to squeeze out the spines or removed them surgically but this may prove difficult. No antivenoms are avail- able. There is a risk of marine bacterial infections (see earlier paragraphs). Cone shells and octopuses (Mollusca) The 500 species of cone shells (genus Conus) are carnivorous marine snails that harpoon their prey (fish, polychaete worms, and other molluscs), implanting a radular tooth charged with venom containing a mixture of small (10–​30 amino acid) peptide toxins (Fig. 10.4.2.36), conotoxins, that block acetylcholine recep- tors and voltage-​sensitive calcium and sodium ion channels. Cone shells are attractive and valuable collectors’ items, but col- lectors may be stung. Symptoms of envenoming include nausea, vomiting, paraesthesia, and numbness of the lips and site of sting, numbness, dizziness, ptosis, diplopia, dysarthria, dyspnoea, and loss of consciousness. In a series of 35 cases mostly stung by Conus geographus reported in Japan (1896–​1996), 10 died within 2 to 5 h of the sting. Several species of small octopus found in the Australian and West Pacific region (blue-​ringed octopuses, Hapalochlaena spp.) Fig. 10.4.2.35  Black long-​spined sea urchin Diadema setosum (Diadematidae) with spines 35 cm long, Madang, Papua New Guinea. Copyright D. A. Warrell. Fig. 10.4.2.34  Extensive weals from contact with the stinging tentacles of the box jellyfish Chironex fleckeri in an Australian stung in Darwin. By courtesy of Drs B. Currie and P. Nitschke.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1806 (Fig. 10.4.2.37) inject salivary tetrodotoxin when they bite swimmers with their powerful beaks. Bites are painful and cause local bleeding, swelling, and inflammation. Severe neurotoxic symptoms, and even fatal generalized paralysis, may develop within 15 min of the bite. Treatment No antivenoms are available. Cardiopulmonary resuscitation and mechanical ventilation may be required. Venomous arthropods Bees, wasps, yellowjackets, hornets,
and ants (Hymenoptera) The most frequent and severe hymenoptera stings are inflicted by members of the families Apidae (honey bees Apis mellifera, A. cerania, A. dorsata, and so on, and bumble bees Bombus spp.), Vespidae (wasps, yellow jackets, white-​faced ‘hornets’, paper wasps genera Vespula, Dolichovespula, Polistes, and true hornets genus Vespa) (Fig. 10.4.2.38), and Formicidae (American fire ants genus Solenopsis, Australian bull or bulldog ants genus Myrmecia). Allergic reactions to single stings from hymenoptera are common, whereas envenoming resulting from many stings is rare. Venom allergens in- clude phospholipases A, hyaluronidase, acid phosphomonoesterases, and polypeptide neurotoxins such as apamin and melittin (A. mel- lifera). Non​allergenic compounds include vasoactive amines, such as histamine, 5-​hydroxytryptamine, catecholamines and kinins, cholinesterase (in the venom of Vespula germanica), pheromones, 2-​methylpiperidine alkaloids (in venoms Solenopsis), and anti-​ inflammatory peptides from honey bee venom. Epidemiology Each year, fewer than five people die from identified hymenopteran sting anaphylaxis in England and Wales, 2–​3 per year in Australia, and 40–​50 in the United States. The incidence of systemic reactions to stings by hymenoptera has been reported as 0.4 to 0.8% in chil- dren. In one adult population in the United States, the prevalence of systemic allergic sting reactions was found to be 4%; 20% of this population showed evidence of venom hypersensitivity (skin tests or radioallergosorbent test, RAST). In the United Kingdom, most patients allergic to bee venom are beekeepers or their relatives. Since the escape of swarms of African honey bees A. m. scutellata in Brazil, in 1957, this aggressive strain has spread throughout Latin America and north as far as Las Vegas in the United States. About 30 deaths from mass attacks by these bees have been reported each year. Two species of fire ants, Solenopsis richteri and S. invicta, were imported into the United States from South America in 1918 and have now spread to 13 southern states where an estimated 2.5 million people are stung each month. The incidence of systemic allergic re- actions is about 4 per 100 000 population per year, and there have Fig. 10.4.2.37  Southern/​lesser blue-​ringed octopus Hapalochlaena maculosa, Point York, Southern Australia. Copyright D. A. Warrell. Fig. 10.4.2.38  Stinger of European wasp Vespula vulgaris. Copyright D. A. Warrell. (a) (b) Fig. 10.4.2.36  Cone shell Conus bullatus harpooning and then ingesting a small fish. Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1807 been fatalities. In Tasmania and southern Australia, the jack jumper ant Myrmecia pilosula causes 90% of all ant stings. About 2–​3% of the population are hypersensitive, and there have been deaths from anaphylaxis. Prevention Patients who have a history of systemic anaphylaxis following a sting and who have evidence of hypersensitivity to the venom of the same family of hymenoptera (venom-​specific IgE detect- able in the serum or a positive skin test) should be considered for desensitization with purified venoms. This treatment proved significantly more effective than placebo or the previously used whole-​body extracts of hymenoptera in preventing anaphylactic reactions to sting challenge. Desensitization usually involves weekly visits to hospital for at least 8 weeks for the administra- tion of gradually increasing doses of venom. When protection has been demonstrated by the patient’s ability to tolerate 100 µg of venom (equivalent to two stings) they are ready for maintenance therapy, usually 100 µg of venom every 4–​8 weeks. A period of 2–​5 years of maintenance desensitization is recommended, after which more than 90% of subjects will remain protected against systemic reactions after stopping treatment. Desensitization is complicated by systemic reactions in 5–​15% of patients and by local reactions in 50%. Wasps are attracted by sweet things and meat in kitchens, green- grocers’ shops, orchards, vineyards, brightly coloured floral pat- terns, and perfumes. Hornets are attracted by light. Some hornets (e.g. Asian Vespa mandarina) are so aggressive that their nests must be eradicated before the area can be farmed. The risk of mass attacks by apids and vespids can be reduced by vigilance. Observing increasing numbers of vespids can lead to discovery and destruction of their nests. Attacks on farm animals and a ten- dency for bees to pursue apiarists walking away from the hives are signs of an increasingly aggressive colony, prompting replacement of the queens. Clinical features Toxic effects In non​sensitized people, a sting, which, in the case of Vespidae and Apidae, introduces about 50 µg of venom, will rapidly produce a hot, red, painful swelling and weal a few centimetres in diameter, which persists for no more than a few hours. These effects are dangerous only if the airway is obstructed, following stings on the tongue. As few as 30 stings can cause fatal systemic envenoming in children, but children and adults have survived more than 1 000 stings by A. mellifera. Symptoms suggest histamine toxicity—​vasodilatation, hypotension, vomiting, diarrhoea, throbbing headache, coma, and bronchoconstriction. In Latin America, victims of mass attacks by A. m. scutellata develop generalized rhabdomyolysis (grossly ele- vated serum CK, aminotransferases, and myoglobin), intravascular haemolysis, hypercatecholaminaemia (hypertension, pulmonary oedema, myocardial damage), bleeding, hepatic dysfunction, and acute kidney injury. In non​sensitized individuals, stings from Solenopsis and Myrmecia spp. produce pain, itching, swelling, and erythema around a central weal, which last a few hours, and later vesicles or pustules. In an unsensitized patient, an estimated 10 000 S. invicta stings caused no systemic envenoming. Allergic effects: Systemic anaphylaxis Clinical suspicion of dangerous venom hypersensitivity arises when systemic symptoms follow a sting. Systemic symptoms in- clude tingling scalp; itching, initially of the palms, soles, axillae, and perineum, becoming generalized; flushing; dizziness; syncope; wheezing; abdominal colic (uterine colic in women), violent diar- rhoea, incontinence of urine and faeces; tachycardia and visual dis- turbances; all developing within a few minutes of the sting. Over the next 15–​20 min, urticaria, angio-​oedema of the lips, gums, and tongue, a generalized redness of the skin with swelling, oedema of the glottis, profound hypotension, and coma may develop. The median time to first cardiac arrest is 10–​20 min after the sting but deaths have occurred after only 2 min. A few patients develop serum sickness a week or more after the sting. Some patients with sting allergy have other evidence of an atopic disposition. Reactions are enhanced by β-​blockers. Severe local reactions to stings: a separate subset of patients who become allergic to hymenoptera venoms develop delayed and some- times massive and persistent local swelling hours or days after the sting without showing any systemic features of anaphylaxis. Such reactions, which may be progressively more severe after successive stings, do not predict increased risk of anaphylaxis. Diagnosis of anaphylaxis and venom hypersensitivity Detection of a raised mast-​cell tryptase concentration is useful in confirming the diagnosis of anaphylaxis and excluding panic attacks and other causes of collapse. Serum/​plasma concentrations peak 0.5–​1.5 h after the attack, but persist for 6–​8 h. Type I hyper- sensitivity is confirmed by detecting venom-​specific (Vespidae, Apidae, Formicidae) IgE in the serum using RAST or by prick-​skin tests. Among hymenoptera venoms, there is strong cross reactivity between bumble bee and honey bee venoms and between wasp, yellow-​jacket, and true hornet venoms, but not between venoms of Apidae and Vespidae. Patients who have suffered a systemic re- action have a 50–​60% risk of a reaction to their next sting. Local reactions, even massive ones involving persistent swelling of the whole stung limb, in the absence of systemic symptoms, do not pre- dict a systemic reaction following subsequent stings. Children who have generalized urticaria after a sting have only a 10% chance of a systemic reaction when restung. Hypersensitivity to venom may be lost spontaneously in some children and young adults but this is unpredictable and unreliable. In some countries, live insect-​sting challenge is used to assess hypersensitivity and response to im- munotherapy. The RAST test can be used for a post-​mortem diag- nosis of hymenoptera sting anaphylaxis. Treatment The barbed stings of Apidae remain embedded at the site of the sting and continue to inject venom, so they should be removed immedi- ately by any means possible. Vespids can withdraw their stings and sting repeatedly. Wasp stings may become infected because some species feed on rotting meat. Domestic meat tenderizer (papain) di- luted roughly 1:5 with tap water is said to produce immediate relief of pain. Ice packs and aspirin are also effective. Systemic but not topical antihistamines can be used for more severe local reactions. Massive local reactions may require aspirin, non​steroidal anti-​inflammatory agents, or even corticosteroids.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1808 Systemic anaphylaxis First, the patient must be laid down and kept flat, ideally in the re- covery position. Immediate cardiopulmonary resuscitation may be needed. Adrenaline should be given as soon as possible: 0.1% (1:1000) (0.5–​1 ml for adults; 0.01 mg/​kg for children) given by intramuscular injection into the anterolateral thigh, or, if the patient is unconscious or pulseless, diluted 1:100 000, by slow intravenous injection. In rare cases, blood pressure fails to respond to even re- peated doses of adrenaline and fluid resuscitation. These patients should be given cardiopulmonary resuscitation, selective bron- chodilators such as salbutamol, pressor agents such as dopamine, and intravenous histamine H1 blockers such as chlorphenamine maleate (10 mg for adults; 0.2 mg/​kg for children). Corticosteroids probably have no effect in acute anaphylaxis, but may prevent re- lapses a few hours later. Patients who know that they are hypersen- sitive should wear an identifying tag or bracelet (such as provided by Medic-​Alert or Medi-​Tag in Britain) as they may be discovered unconscious after being stung. They should be trained to give them- selves adrenaline using an ‘EpiPen’ or similar apparatus, but a high proportion of those who carry these kits are unable to use them effectively through lack of training. Shock and upper-​ or lower-​ airway obstruction are the main modes of death following insect-​ sting anaphylaxis. Severe envenoming from multiple stings by hymenoptera should be treated with adrenaline, intravenous antihistamines (doses as mentioned earlier), and corticosteroids. Intensive care is essential. Intravenous fluids may protect the kidneys from the damaging ef- fects of myoglobinuria and haemoglobinaemia (‘pigment neph- ropathy’), as in patients with the crush syndrome. Experimental antivenoms have been produced but are not yet commercially avail- able. Exchange transfusion or plasmapheresis might be considered to remove venom in severe cases. Renal dialysis is often needed. Butterflies and moths (Lepidoptera) The stinging hairs of some species of adult moths can cause con- tact dermatitis and urticaria (‘lepidopterism’), while caterpillars can produce local or even systemic effects (‘erucism’). Venomous lepidoptera are found in all parts of the world, but most cases of lepidopterism are reported from Middle and Southern America. Severe cutaneous urticating eruptions can be caused by cater- pillars of oak processionary moths Thaumetopoea processionea (Thaumetopoeidae) in central/​southern Europe and of the genus Megalopyge (Megalopygidae—​called ‘puss caterpillars’ in the southern United States) and by adult female moths of the genus Hylesia (Saturniidae), which have barbed setae (‘flechettes’) on their abdomens (Fig. 10.4.2.39). Epidemics of stings by these moths have been described, especially from coastal areas of Brazil, Mexico, Peru, and Venezuela. In Brazil, Colombia, Guyana, Paraguay, Peru, and Venezuela, caterpillars of atlas or emperor moths (Lonomia obliqua, L. achelous; Saturniidae) cause thousands of stings each year. A tourist died of Lonomia envenoming died a few days after returning to Canada from Peru where she had trodden on some of these caterpillars. Venom injected through their bristles con- tains fibrinolytic (factor XIII activator); anticoagulant; procoagu- lant (activators of prothrombin, factor X, factor V), kallikrein-​like, metalloproteinase, and phospholipase A2 activities resulting in defibrinogenation and spontaneous bleeding. The case fatality of about 2% is usually attributable to cerebral haemorrhage. Symptoms include local burning, erythema, swelling, inflammation, head- ache, nausea, vomiting, malaise, bleeding from nose, gums, gut, genitourinary tract, and partly healed scars, polyarthralgia, and acute renal failure. Laboratory findings in envenomed patients are decreased plasma fibrinogen, factor V, factor XIII, and plasminogen concentrations, as well as increased fibrin/​fibrinogen degradation products and fibrinolytic activity but a normal platelet count. An effective antivenom (‘Soro antilonômico’) is produced by Instituto Butantan, São Paulo, Brazil. Beetles (Coleoptera) The most notorious vesicating beetle is ‘Spanish fly’ Lytta vesicato- ria (Meloidae—​blister beetles) whose venom contains cantharidin, which causes blistering 2–​3 h after application to the skin. ‘Nairobi eye’ and similar blistering conditions in Australia and Southeast Asia are caused by Paederus (Staphylinidae) species 5–​10 mm in length (Fig. 10.4.2.40). The typical skin lesions (dermatitis linearis), whose appearance may be delayed 12–​96 h after contact, consist of erythema, itching, and blistering caused by inadvertently crushing and smearing the beetle. Systemic symptoms such as fever, arth- ralgia, and vomiting may arise in severe cases. The active principle Fig. 10.4.2.39  Lesions caused by urticating abdominal hairs of female moths Hylesia spp. in Brazil. Copyright D. A. Warrell. Fig. 10.4.2.40  Vesicating beetle Paederus crebripunctatus (Staphylinidae) responsible for causing ‘Nairobi eye’. Courtesy of Dr John Paul.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1809 pederin is the most complex non​proteinaceous insect toxin known. Treatment is palliative. The toxin is easily spread to other sites such as the eye by fingers. Scorpions (Scorpiones, Buthidae, Hemiscorpiidae) Species capable of inflicting fatal stings occur in North Africa and the Middle East (Androctonus, Buthus, Hemiscorpius (Fig. 10.4.2.41), Leiurus spp.) South Africa (Parabuthus spp.); India, Sri Lanka, and Nepal (Hottentotta (formerly Mesobuthus) tamulus); North, Central and South America, Trinidad and Tobago (Centruroides (Fig. 10.4.2.42), Tityus (Fig. 10.4.2.43) spp.). Scorpion toxins target Na+, K+, Ca2+, and Cl-​ ion channels causing direct effects and the re- lease of neurotransmitters such as acetylcholine and catecholamines. Epidemiology In Mexico, 250 000 stings with 70 deaths are reported each year, attributed to Centruroides limpidus, C. noxius, C. suffusus, and so on. Brazil recorded 91 000 scorpion stings and 121 deaths in 2016, more for snake-bites. In Khuzestan Province, Iran, 25 000 stings (Hemiscorpius lepturus, Androctonus spp., and Buthus spp.) are treated each year and are the fourth major cause of death. Algeria and Tunisia report tens of thousands of stings and hundreds of deaths. In the United States 15 000 stings, mainly by Centruroides exilicauda (Fig. 10.4.2.42) are reported in Arizona each year but deaths are rare. In Brazil, the important species are Tityus serru- latus (Fig. 10.4.2.43) and other Tityus spp. In 2005, among 36 558 reported stings, there were only 50 deaths (case fatality 0.14%). In India, many people are stung by the red scorpion Hottentotta (for- merly Mesobuthus) tamulus with fatalities in adults and children. Prevention Scorpions can be excluded from houses by incorporating a row of ceramic tiles into the base of the outside wall, making the doorsteps at least 20 cm high, and using residual insecticides, such as carba- mate or organophosphate sprays or dusts indoors. Clinical features Intense local pain is the commonest symptom. There may be slight local oedema and tender enlargement of regional lymph nodes, but stings by Hemiscorpius lepturus (Iran, Iraq, Pakistan, and Yemen) are relatively painless. Systemic symptoms usually develop within minutes but may be delayed for as much as 24 h. They vary, ac- cording to the species of scorpion involved. Most scorpion venoms stimulate the release of acetylcholine and catecholamines, often re- sulting in initial cholinergic and later adrenergic symptoms. Early symptoms include vomiting, profuse sweating, piloerection, alter- nating brady-​ and tachycardia, abdominal colic, diarrhoea, loss of sphincter control, and priapism. Later, severe life-​threatening car- diorespiratory effects may appear: hypertension, shock, tachy-​ and bradyarrhythmias, ECG changes, and pulmonary oedema with or without evidence of myocardial dysfunction. Severe cardiovascular complications are particularly associated with stings by Androctonus Fig. 10.4.2.41  Hemiscorpius lepturus (Hemiscorpiidae), Iran. Courtesy of Dr M. Radmanesh. Fig. 10.4.2.42  Arizona bark scorpion Centruroides (Sculpturatus) exilicauda. Copyright D. A. Warrell. Fig. 10.4.2.43  Brazilian yellow scorpion Tityus serrulatus, São Paulo, Brazil. Scale in cm. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1810 spp., Leiurus quinquestriatus, Hottentotta tamulus, and Tityus spp. (Fig. 10.4.2.44). Neurotoxic effects such as erratic eye movements, fasciculation, and muscle spasms, which can be misinterpreted as tonic-​clonic convulsive movements, and respiratory distress are a particular feature of stings by Centruroides (sculpturatus) exilicauda in Arizona. Parabuthus transvaalicus envenoming in southern Africa is more likely to cause ptosis and dysphagia with death from respiratory paralysis. Hemiplegia and other neurological lesions have been attributed to fibrin deposition resulting from dissem- inated intravascular coagulation, for example, after stings by Nebo hierichonticus in the Middle East. Hypercatecholaminaemia could explain hyperglycaemia and glycosuria but in the case of stings by the black scorpion of Trinidad (Tityus trinitatis) there is severe abdominal pain with nausea, vomiting, and haematemesis, hyperglycaemia, and biochemical evi- dence of acute pancreatitis attributable to simultaneous spasm of the sphincter of Oddi and pancreatic exocrine hypersecretion. In Iran and Iraq, stings by Hemiscorpius lepturus (Hemiscorpiidae) produce a unique clinical syndrome. The sting is painless but macular erythema, pupura, and bullae develop at the site with induration in 39% of cases, swelling and necrosis that requires surgery in 20% of cases (Fig. 10.4.2.45). Systemic symptoms include dry mouth, thirst, dizziness, nausea, vomiting, fever, cardiac arrhythmias, ST depres- sion on ECG, hypoglycaemia, confusion and convulsions, leucocyt- osis, thrombocytopenia, coagulopathy, haemolytic anaemia with haemoglobinuria, proteinuria, and acute kidney injury. Twenty per cent (20%) of paediatric cases required dialysis. Early treatment with Rhazi Institute antivenom proved effective. Treatment Pain responds to local infiltration or ring block with local anaes- thetic. Parenteral opiate analgesics, such as pethidine or morphine, may be required, but are said to be dangerous in victims of C. exil- icauda (sculpturatus). Antivenom is recommended. In a recent trial in children stung by C. sculpturatus in Arizona, antivenom treatment was associated with more rapid resolution of symptoms and less requirement for midazolam sedation than placebo. In India, two studies found that addition of a new antivenom raised against H. tamulus concanensis venom produced more rapid recovery than prazosin alone. Antivenom should be administered intravenously as soon as possible in patients with systemic envenoming and in young children stung by dangerous species, even before the development of these symptoms. For patients with cardiovascular symptoms (hypertension, bradycardia, and early pulmonary oedema), vasodilators such as the α1-​blocker prazosin are recommended. Patients who develop left ventricular failure despite early prazosin therapy benefit from dobutamine. The use of atropine (except in cases of life-​threatening sinus bradycardia), cardiac glyco- sides and β-​blockers is not recommended. Spiders (Araneae) All but one family of this enormous order are venomous, but only about 20 species have proved dangerous to humans. Many others have been wrongly accused of inflicting harmful bites. Spiders bite with a pair of small fangs, the chelicerae, to which the venom glands are connected. Medically important genera include Loxosceles, causing necrotic araneism, and Latrodectus, Phoneutria, Atrax, Hadronyche, and Missulena spp., causing neurotoxic araneism. Epidemiology Spider bites are common in some parts of the world but there are now few fatalities. In Brazil in 2016 there were 29 000 spider bites with 22 fatalities. In Central and South America, Loxosceles spp. such as L. laeta and L. gaucho (Fig. 10.4.2.46) are widely distributed and cause many bites. In Chile, the case fatality of loxoscelism ranges from 1 to 17%. In the southern and south-​central United States, the brown recluse spider L. reclusa caused at least 200 bites and six deaths during the last century. Bites by L. rufescens have been re- ported in the Mediterranean region, North Africa, and Israel. Most bites from Loxosceles spp. occur in bedrooms while people are asleep or dressing. Fig. 10.4.2.44  Twenty-​six-​day-​old child stung on right axilla by a Brazilian yellow scorpion (Tityus serrulatus) in urban São Paulo: showing agitation and pulmonary oedema. Copyright D. A. Warrell. (a) (b) Fig. 10.4.2.45  (a) Local swelling, blistering, and ‘purpuric plaque’ caused by the sting of Hemiscorpius lepturus in Iran. (b) Progressing to necrosis with granulation tissue. Courtesy of Dr M Radmanesh.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1811 Black and brown widow spiders are cosmopolitan in distribu- tion. Latrodectus tredecemguttatus (sometimes referred to, loosely, as ‘tarantula’) lives in fields in Mediterranean countries and has been responsible for epidemics of bites. The Australian redback spider (L. hasselti) (Fig. 10.4.2.47) causes up to 340 bites each year in Australia; 20 deaths have been reported. It has colonized parts of Japan, New Zealand, UAE, and New Caledonia. In the United States, L.  mactans was responsible for 63 deaths between 1950 and 1959. Several species of Latrodectus occur in Latin America (Fig. 10.4.2.48). Wandering, armed, or banana spiders Phoneutria spp. (Fig. 10.4.2.49), cause bites and a few deaths in Latin American coun- tries. They have been imported into temperate countries in bunches of bananas, causing a few bites and deaths. Highly dangerous funnel web spiders Atrax spp. are restricted to south-​eastern Australia and Tasmania. The Sydney funnel web spider (A. robustus) is found only within a 160-​mile (256-​km) ra- dius of Sydney. The aggressive males caused at least 13 deaths be- tween 1927 and 1980. Members of the related genera Hadronyche and Missulena may be equally dangerous. In England, mild neurotoxic araneism has been described after bites by Steatoda nobilis and S. grossa (Theridiidae) and the wood- louse spider Dysdera crocata. Necrotic araneism Skin lesions, varying in severity from mild localized erythema and blis- tering to extensive granulomas and tissue necrosis, have been falsely attributed to a large variety of familiar peridomestic species, such as the Australian white-​tailed spider Lampona cylindrata, North American hobo spider Tegenaria agrestis, European and South American wolf spiders Lycosa spp. (including the Italian ‘tarantula’ L. terentula), and cosmopolitan sac spiders Cheiracanthium spp. However, only mem- bers of the genus Loxosceles have proved capable of causing ‘necrotic Fig. 10.4.2.46  South American recluse spider Loxosceles laeta, Brazil. Copyright D. A. Warrell. Fig. 10.4.2.47  Australian redback spider Latrodectus hasselti, Adelaide, showing the dorsal red hourglass marking. Copyright D. A. Warrell. Fig. 10.4.2.48  Curaçao black widow spider Latrodectus curacaviensis, Brazil. Copyright D. A. Warrell. Fig. 10.4.2.49  Brazilian armed, wandering, or banana spider Phoneutria nigriventer. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1812 arachnidism/​araneism’. Venom sphingomyelinase D is implicated in the pathogenesis of dermonecrosis. Neutrophils adhere to the endo- thelium of cutaneous capillaries and degranulate. The bite itself is usually painless and unnoticed. Burning develops over several hours at the site of the bite, with swelling and development of a character- istic macular lesion, the red-​white-​and-​blue sign (Fig. 10.4.2.50) showing areas of red vasodilatation, white vasoconstriction, and blue prenecrotic cyanosis. A blackened eschar develops, which sloughs in a few weeks, leaving a full thickness necrotic ulcer. Sometimes an entire limb or area of the face is involved. Facial bites cause much swelling. Some 13% of cases have systemic symptoms such as fever, headaches, scarlatiniform rash (Fig. 10.4.2.51), jaundice, methaemoglobinaemia, and haemoglobinuria resulting from intravascular haemolysis. Renal failure may ensue. The average case fatality is about 5%. Neurotoxic araneism The bite is very painful immediately, but local signs are minimal (L. mactans) or moderate (L. hasselti). After about 30 min, there is painful regional lymphadenopathy, then headache, nausea, vomiting, and local sweating with piloerection (‘gooseflesh’, ‘goose bumps’), a sign highly suggestive of neurotoxic araneism (Fig. 10.4.2.52). Envenoming by L. mactans and L. tredecemguttatus, causes profuse generalized sweating and fever with painful muscle spasms, tremors, and rigidity. This may be sufficiently severe to embarrass respiration. The classic ‘facies latrodectismica’ is an agonized grimace, caused by facial spasm and trismus, associated with swollen eyelids, congested conjunctivae, flushing, and sweating (Fig. 10.4.2.53). Abdominal rigidity may simulate an acute abdomen and prompt laparotomy. Fig. 10.4.2.50  ‘Red-​white-​and-​blue’ sign developing 18 h after a bite by the Brazilian recluse spider Loxosceles gaucho. Copyright D. A. Warrell. Fig. 10.4.2.51  Blanching generalized scarlatiniform rash appearing 3 days after a bite above the left hip by a Brazilian recluse spider Loxosceles gaucho. By courtesy of Dr João Luiz Costa Cardoso. Fig. 10.4.2.52  Intense local sweating and piloerection at the site of a Brazilian banana spider bite Phoneutria nigriventer 30 min earlier. Copyright D. A. Warrell. Fig. 10.4.2.53  ‘Facies latrodectismica’ with profuse sweating and painful muscle spasms, persisting 24 h after a bite by Latrodectus mactans (‘viuda negra’, black widow) near Cusco, Peru. Copyright D. A. Warrell.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1813 Other features include tachycardia, hypertension, restlessness, irrit- ability, psychosis, priapism, and rhabdomyolysis. A localized or dif- fuse rash may appear several days later. Envenoming by Phoneutria and Atrax spp. produces similar features. Treatment First aid In Australia, pressure-​immobilization (see Fig. 10.4.2.26) is recom- mended for bites by A. robustus and Hadronyche species. Specific treatment Antivenoms for envenoming by Latrodectus spp. are made in Australia, Mexico, South Africa, Brazil, and some other South American coun- tries; for Atrax spp. in Australia; for Loxosceles spp. in Argentina, Brazil, and Peru; and for Phoneutria spp. in Brazil. Despite decades of use, there is no decisive evidence for the efficacy of Loxosceles antivenoms, but neurotoxic araneism is more obviously responsive to antivenom. Supportive treatment Oral dapsone (100 mg twice daily) is said to reduce the extent of necrotic lesions by inhibiting neutrophil degranulation and calcium gluconate (10 ml of a 10% solution, given by slow intravenous in- jection) is said to relieve the pain of muscle spasms caused by the venom of Latrodectus spp. rapidly and more effectively than muscle relaxants such as diazepam or methocarbamol. Evidence for the efficacy of these drugs is lacking. Antihistamines, corticosteroids, α-​blockers, and atropine have also been advocated. For necrotic araneism caused by Loxosceles spp., early surgical debridement, corticosteroids, antihistamines, and hyperbaric oxygen all have their advocates, but there is no basis for recommending their use. Ticks (Acari) Taxonomy and epidemiology Ticks, with mites, form the order Acari of the class Arachnida. Adult females of about 34 species of hard tick (family Ixodidae) and imma- ture specimens of nine species of soft ticks (family Argasidae) have been implicated in human tick paralysis. The tick’s saliva contains a neurotoxin, which causes presynaptic neuromuscular block and decreased nerve-​conduction velocity. The tick embeds itself in the skin with its barbed hypostome introducing the salivary toxin while it engorges with blood. Although tick paralysis has been reported from all continents, including Europe, most cases occur in western North America (Dermacentor andersoni), eastern United States (D.  variabilis), and eastern Australia from north Queensland to Victoria (Ixodes holocyclus—​known as the bush-​, scrub-​, paralysis-​, or dog-​tick). In British Columbia there were 305 cases with a 10% case fatality be- tween 1900 and 1968. About 120 cases have been reported in the United States, and in New South Wales there were at least 20 deaths between 1900 and 1945. Clinical features Ticks are picked up in the countryside or from domestic animals, particularly dogs, in the home. Aalmost all fatal cases are children. After the tick has been attached for about 5 or 6 days a progressive ascending lower motor neurone paralysis develops with paraesthe- siae. Often a child, who may have been irritable for the previous 24 h, falls on getting out of bed first thing in the morning and is found to be weak or ataxic. Paralysis increases over the next few days: death results from bulbar and respiratory paralysis and aspiration of stomach contents. Vomiting is a feature of the more acute course of Ixodes holocyclus envenoming. This clinical picture has often been misinterpreted as poliomyel- itis. Other neurological conditions, including Guillain–​Barré syn- drome, paralytic rabies, Eaton–​Lambert syndrome, myasthenia gravis, or botulism, may also be suspected. Diagnosis and cure depends on finding the tick, which is likely to be concealed in a crevice, orifice, or hairy area of the body. The scalp is the commonest place. Fatal tick paralysis has been caused by a tick attached to the tympanic membrane. Treatment The tick must be discovered and detached without being squeezed. It can be painted with ether, chloroform, paraffin, petrol, or turpen- tine, or prised out between the partially separated tips of a pair of small, curved forceps. Following removal of the tick there is usu- ally a rapid and complete recovery; but in Australia, patients have died even after the tick had been detached. The antivenoms, raised in dogs and rabbits in Australia, are no longer produced. Centipedes and millipedes (subphylum Myriapoda) Centipedes (class Chilopoda) Epimorph centipedes have 15–​191 pairs of legs and move rapidly and distractedly. They occur in most parts of the world including the Arctic Circle. The largest, Scolopendra gigantea of South America, can grow to more than 30 cm in length. Many species can inflict painful stings through a pair of modified claws (forcipules) on the postcephalic segment (Fig. 10.4.2.54). More than 3000 stings are (a) (b) Fig. 10.4.2.54  (a) Thai centipede (Scolopendra dehaani) (b) showing venom ‘claws’ or forcipules which are modified limbs. Copyright D. A. Warrell.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1814 reported each year in Brazil. Venoms contain serotonin, histamine, lipids, polysaccharides, proteases, and peptides that are neurotoxic to insects. Stings cause intense radiating pain, swelling, inflamma- tion, erythema, and lymphangitis, and sometimes local necrosis. Systemic effects such as vomiting, sweating, headache, cardiac ar- rhythmias, myocardial ischaemia, rhabdomyolysis, proteinuria, acute renal failure, and convulsions are extremely rare. The risk of mortality was probably greatly exaggerated in the older literature. Hypersensitivity may have played a role in these reactions. Reports of documented fatalities remain elusive but are said to occur on some Indian Ocean islands. The most important genus is Scolopendra which is distributed throughout tropical countries. Local treatment is the same as for scorpion stings. No antivenom is available. Millipedes (class Diplopida) Millipedes are widely distributed. They may exceed 35 cm in length, have hundreds of legs (not a thousand, despite their name), move sluggishly, and tend to coil into a ball. Most species possess glands in each of their body segments which secrete, and in some cases squirt out, irritant liquids for defence. These contain hydrogen cyanide and a variety of aldehydes, esters, phenols, and quinonoids. Members of at least eight genera of millipedes have proved injurious to humans, including Rhinocricus (Caribbean), Spirobolus (Tanzania and Papua New Guinea), Spirostreptus and Iulus (Indonesia), and Polyceroconas (Salpidobolus) (Papua New Guinea). Children are at risk when they handle or try to eat these large arthropods. When venom is squirted into the eye, intense conjunctivitis results, and there may be corneal ulceration and, allegedly, blindness. Skin lesions initially stain brown (‘mahogany stains’) or purple, blister after a few days, and then peel (Fig. 10.4.2.55). They have been mistaken for signs of child abuse. First aid is generous irrigation with water. Eye injuries should be treated as for snake venom ophthalmia (see earlier in this chapter). Leeches (phylum Annelida, class Hirudinea) Leeches are bloodsucking, hermaphroditic, egg-​laying annelids, which have elongated annulated bodies. They attach themselves to leaves, rocks, or the host by a posterior sucker. To feed, the leech applies its anterior sucker containing the mouth armed with three radially arranged jaws which make a Y-​shaped incision. Blood is sucked out by the action of the muscular pharynx. To prevent blood clotting, the saliva contains a histamine-​like vasodilator and anticoagulants, such as: hirudin from the medicinal leech Hirudo medicinalis, which inhibits thrombin and factor IXa; hementin from Haementeria ghilianii, which is directly fibrinolytic; and hementerin from H. depressa (= H. lutzi), a plasminogen activator. Other en- zymes include esterases, antitrypsin, antiplasmin, and antielastase. Recombinant hirudin is now produced as a therapeutic anticoagu- lant. The medicinal leech is still used by plastic surgeons to reduce haematomas under skin grafts; the wound may become infected with Aeromonas hydrophila, which lives symbiotically in the leech’s gut. Two groups of leeches cause human morbidity and even mortality in tropical countries. Land leeches Species of the genera Haemadipsa and Phyrobdella are 1–​8 cm long. They infest, often in enormous numbers, the damp leaf litter and low vegetation of rainforests, choosing game trails and watering places. By standing on the posterior sucker and waving the anterior sucker, they can sense their prey with amazing efficiency. They drop on to the prey or pursue it with a looping or lashing motion. Leeches usually attach themselves to the lower legs or ankles and are adept at penetrating clothing, even long trousers tucked into socks and lace-​up boots. The bite is usually painless and infested individuals may not realize what has happened until they hear a squelching sound, notice that their feet are warm and wet, and see blood welling over the tops of their boots. Land leeches ingest about 1 ml of blood in 1 h and then drop off, but the wound continues to bleed for some time and forms a fragile clot. Aquatic leeches These species may be swallowed by individuals who drink stagnant water or even mountain stream water, or they may attack bathers, entering the mouth, nostrils, eyes, vulva, vagina, urethra, or anus. Hirudo medicinalis can ingest 5–​15 ml of blood, increasing its initial weight up to 10 times. The enormous brightly coloured buffalo leech Hirudinaria manillensis of Southeast Asia, is up to 16 cm long and can ingest 1 ml of blood in 10 min. Limnatis nilotica occurs around (a) (b) Fig. 10.4.2.55  Skin and mucosal lesions (a) caused by application of giant Papua New Guinea millipede (b) (Spirobolus vogesi). Copyright Dr Bernie Hudson.

10.4.2  Injuries, envenoming, poisoning, and allergic reactions caused by animals 1815 the Mediterranean, Middle East, and North Africa. Myxobdella af- ricana occurs in East Africa. Dinobdella ferox (5 cm long) is found in Asia. Some aquatic leeches are very slow feeders and may remain attached for days or even weeks. L. nilotica and D. ferox have been implicated in ‘halzoun’. However, in Lebanon, leech infestation con- tracted from spring water (‘alack’) is distinguished from ‘halzoun’ following ingestion of raw offal (see Section 7 Infectious Diseases). Prevention Leech intrusion can be reduced by impregnating clothing, especially the bottoms of trousers and socks, with repellents such as dibutyl phthalate and diethyl toluamide and applying them to the skin and the inside and outside of footwear. If these compounds are not avail- able, invasion of footwear during jungle walks can be prevented, ra- ther messily, by rolling a rope of tobacco in the tops of the socks and keeping the feet well soaked with water or using an aqueous ex- tract of tobacco leaves. Women’s pantyhose are said to prevent leech attachment, but may be damaged by DEET-​containing repellents. Effective leech-​proof light cotton socks are available commercially. Children should be discouraged from bathing in leech-​infested waters and all drinking water should be boiled or filtered. Clinical features The main effect is blood loss, but other symptoms include pain caused by the bite, secondary infection, a residual itching, and phobia. Ingested aquatic leeches usually attach to the pharynx but may penetrate the bronchi or oesophagus. H. manillensis entering via the anus can reach the rectosigmoid junction of the bowel causing perforation and periton- itis. Patients with a leech in the pharynx often have a feeling of movement at the back of the throat with cough, hoarseness, stridor, breathlessness, epistaxis, haemoptysis, and haematemesis. Fatal upper airway obstruc- tion may result. The leech Limnatis nilotica is no longer thought to be a cause of ‘halzoun’ (Lebanon) or ‘marrara’ (Sudan) (see Chapter 8.13 on pentastomiasis) but aquatic leech infestation could be a differential diag- nosis of those dramatic syndromes of pharyngeal obstruction. Bleeding may persist for up to a week after the leech has dropped off. In rural Thailand, vaginal bleeding in girls who have swum in ponds or canals is often attributable to infestation by aquatic leeches. Sexual abuse may be wrongly inferred if this diagnosis is not considered. Transmission of rinderpest and other viruses, leptospirosis, and Trypanosoma cruzi has been suggested but not proved. Secondary infection of medicinal leech bites by Aeromonas hydrophila has been described. Treatment Leeches are best scraped off with a fingernail. Traditional methods such as applying a grain of salt, a lighted match or a cigarette, al- cohol, turpentine, or vinegar make the leech regurgitate into the wound, creating a risk of infection. Local bleeding can be stopped by applying a styptic, such as silver nitrate or a firm dressing. Aquatic leeches that have penetrated the respiratory, upper gastro- intestinal, genitourinary tracts, or rectum must be removed by endoscope. Spraying with 30% cocaine, 10% tartaric acid, or di- lute (1:10 000) adrenaline makes the leech detach from the naso- pharynx, larynx, trachea, or oesophagus, while irrigation with a concentrated salt solution may be effective in the genitourinary tract and rectum. Leeches should not be pulled off so roughly that the mouth parts are left in the wound as this will lead to a chronic infection. Antimicrobial treatment of secondary bacterial infec- tions (e.g. of Aeromonas hydrophila with cefuroxime or a quin- olone) may be required. FURTHER READING Mechanical injuries caused by animals Auerbach PS (ed) (2016). Wilderness medicine, 7th edition. Mosby Elsevier, Philadelphia, PA. Packer C, et al. (2011). Fear of darkness, the full moon and the noc- turnal ecology of African lions. Plos One, 6, e22285. Spotte S (2002). Candirú: life and legend of the bloodsucking catfish. Creative Arts Books Company, Berkeley, CA. Woodroffe R, Thirgood S, Rabinowitz A (eds) (2005). People and wild- life: conflict or coexistence? Cambridge University Press, Cambridge. Websites Polar bears: http://​kho.unis.no/​doc/​Polar_​bears_​Svalbard.pdf Shark attacks: http://​www.flmnh.ufl.edu/​fish/​Sharks/​ISAF/​ISAF.htm Venomous mammals Madani G, Nekaris KA (2014). Anaphylactic shock following the bite of a wild Kayan slow loris (Nycticebus kayan): implications for slow loris conservation. J Venom Anim Toxins Incl Trop Dis, 20, 43. Whittington CM, et al. (2008). Defensins and the convergent evolution of platypus and reptile venom genes. Genome Res, 18, 986–​94. Venomous snakes de Silva HA, et al. (2011). Low-​dose adrenaline, promethazine, and hydrocortisone in the prevention of acute adverse reactions to anti- venom following snakebite: a randomised, double-​blind, placebo-​ controlled trial. PLoS Med, 8, e1000435. Gutiérrez JM, et al. (2017). Snakebite envenoming. Nat Rev Dis Primers, 3, 17063. Meier J, White J (eds) (1995). Handbook of clinical toxicology of animal venoms and poisons. CRC Press, Boca Raton, FL. Mohapatra B, et  al. (2011). Snakebite mortality in India:  a
nationally representative mortality survey. PLoS Negl Trop Dis, 5, e1018. Sutherland SK, Tibballs J (2001). Australian animal toxins: the crea- tures, their toxins and care of the poisoned patient, 2nd edition. Oxford University Press, Melbourne. Warrell DA (2004). Epidemiology, clinical features and
management of snakebites in Central and South America. In: Campbell J, Lamar WW, Greene H (eds) Venomous reptiles of the Americas, 2nd edition, pp. 709–​61. Cornell University Press, Ithaca, NY. Warrell DA (2010). Snake bite. Lancet, 375, 77–​88. Weinstein SA, et al. (eds) (2011). ‘Venomous’ bites from non-​venomous snakes: a critical analysis of risk and management of ‘colubrid’ snake bites. Elsevier, London. White J (2013). A clinician’s guide to Australian venomous bites and stings. CSL, Parkville. Websites Antivenoms World Health Organization (WHO) 2018 Guidelines: http:// ​www.who. int/​bloodproducts/​snake_​antivenoms/​snakeantivenomguide/​en/​ Envenoming worldwide http://​www.toxinology.com/​ http://​www.vapaguide.info/​ Snakebite in South and South East Asia (WHO SEARO 2016 Guidelines), 2nd edition: http://apps.who.int/iris/handle/10665/249547 Snakebite in Africa (WHO AFRO 2010 Guidelines): http://apps.who. int/medicinedocs/en/d/Js17810en/ Venomous lizards Beck DD (2005). Biology of Gila monster and beaded lizards. University of California Press, Berkeley.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1816 Poisonous amphibians Daly JW, Spande TF, Garraffo HM (2005). Alkaloids from amphibian skin: a tabulation of over eight hundred compounds. J Nat Prod, 68, 1556–​75. Myers CW, Daly JW, Malkin B (1978). A dangerously toxic new frog (Phyllobates) used by the Emberá Indians of Western Colombia with discussion of blowgun fabrication and dart poisons. Bull Am Mus Nat History, 161(Art 2), 307–​66. Poisonous birds Dumbacher JP, et  al. (1992). Homobatracho-​toxin in the genus Pitohui: chemical defense in birds? Science, 258, 799–​800. Dumbacher JP, et al. (2004). Melyrid beetles (Choresine): a putative source for the batrachotoxin alkaloids found in poison-​dart frogs and toxic passerine birds. Proc Natl Acad Sci U S A, 101, 15857–​60. Venomous fish Halstead BW (1988). Poisonous and venomous marine animals of the world, 2nd revised edition. Darwin Press, Princeton, NJ. Jouiaei M, et al. (2015). Ancient venom systems: a review on cnidaria toxins. Toxins (Basel), 7, 2251–71. Williamson JA, et al. (eds) (1996). Venomous and poisonous marine an- imals: a medical and biological handbook. University of New South Wales Press, Sydney. Poisoning by ingestion of aquatic animals Bagnis RA, et al. (1979). Clinical observations on 3009 cases of ci- guatera (fish poisoning) in the Southern Pacific. Am J Trop Med Hygiene, 28, 1067–​73. Friedman MA, Fernandez M, Backer LC, et al. (2017). An updated review of ciguatera fish poisoning: clinical, epidemiological, envir- onmental, and public health management. Mar Drugs, 5, pii: E72. Halstead BW (1988). Poisonous and venomous marine animals of the world, 2nd revised edition. Darwin Press, Princeton, NJ. Williamson JA, et al. (eds) (1996). Venomous and poisonous marine an- imals: a medical and biological handbook. University of New South Wales Press, Sydney. Venomous marine invertebrates Halstead BW (1988). Poisonous and venomous marine animals of the world, 2nd revised edition. Darwin Press, Princeton, NJ. Little M, Fitzpatrick R, Seymour J (2016). Successful use of heat as first aid for tropical Australian jellyfish stings. Toxicon, 122, 142–4. Olivera BM, Teichert RW (2007). Diversity of the neurotoxic Conus peptides: a model for concerted pharmacological discovery. Mol Interv, 7, 251–​60. Williamson JA, et al. (eds) (1996). Venomous and poisonous marine an- imals: a medical and biological handbook. University of New South Wales Press, Sydney. Venomous arthropods Hymenoptera França FOS, et  al. (1994). Severe and fatal mass attacks by ‘killer’ bees (Africanised honey bees—​Apis melliferascutellata in Brazil): clinicopathological studies with measurement of serum venom con- centrations. Q J Med, 87, 269–​82. Hunt J Jr., et al. (1978). A controlled trial of immunotherapy in insect hypersensitivity. N Engl J Med, 2991, 157–​61. Mueller UR (1990). Insect sting allergy: clinical picture, diagnosis and treatment. Gustav Fischer, Stuttgart. Piek T (1986). Venoms of the hymenoptera: biochemical, pharmacolog- ical and behavioural aspects. Academic Press, London. Winston ML (1992). Killer bees:  the Africanized honey bee in the Americas. Harvard University Press, Cambridge, MA. Lepidoptera da Silva WD, et al. (1996). Development of an antivenom against toxins of Lonomia obliqua caterpillars. Toxicon, 34, 1045–​9. Kelen EMA, Picarelli ZP, Duarte AC (1995). Hemorrhagic syn- drome induced by contact with caterpillars of the genus Lonomia (Saturniidae, Hamileucinae). J Toxicol Toxin Rev, 14, 283–​308. Coleoptera Aoun O, et al. (2018). Morning blisters: cantharidin-related Meloidae burns. J Travel Med, 25(1). Aoun O, et al. (2018). Mind the eye-squirter! An Anthia sexmaculata sexmaculata-related necrotic burn. J Travel Med, 25(1). Eisner T, et al. (1990). Systemic retention of ingested cantharidin by frogs. Chemoecol, 1, 57–​62. Frank JH, Kanamitsu K (1987). Paederus sensu lato (Coleoptera: Staphylinidae):  natural history and medical importance. J Med Entomol, 24, 1555–​91. Southcott RV (1989). Injuries from Coleoptera. Med J Aust, 151, 654–​9. Scorpions Bettini S (1978). Athropod venoms: handbook of experimental pharma- cology, Vol. 48, p, 977. Springer-​Verlag, Berlin. Boyer LV, et  al. (2009). Antivenoms for critically ill children with neurotoxicity from scorpion slings. N Engl J Med, 360, 2090–​8. Brownell P, Polis G (eds) (2001). Scorpion biology and research. Oxford University Press, New York, NY. Freire-​Maia L, Campos JA, Amaral CFS (1996). Treatment of scor- pion envenoming in Brazil. In: Bon C, Goyffon M (eds) Envenomings
and their treatments, pp. 301–​10. Edition Fondation Marcel Mérieux, Lyon. Ismail M (1995). The scorpion envenoming syndrome. Toxicon, 33, 825–​58. Natu VS, et al. (2010). Efficacy of anti-​scorpion venom serum over prazosin in the management of severe scorpion envenomation.
J Postgrad Med, 56, 275–​80. Polis GA (ed) (1990). The biology of scorpions. Stanford University Press, Stanford, CA. Spiders Maretic Z, Lebez D (1979). Araneism with special reference to Europe. Novit, Pula-​Ljubjan, Yugoslavia. Pauli I, et al. (2011). The efficacy of antivenom in loxoscelism treat- ment. Toxicon, 48, 123–​37. Southcott RV (1976). Arachnidism and allied syndromes in the Australian region. Rec Adelaide Child Hosp, 1, 97–​186. Sutherland SK, Tibballs J (2001). Australian animal toxins: the crea- tures, their toxins and care of the poisoned patient, 2nd edition. Oxford University Press, Melbourne. Warrell DA, et al. (1991). Neurotoxic envenoming by an immigrant spider (Steatoda nobilis) in southern England. Toxicon, 29, 1263–​5. Ticks Gothe R, Kunze K, Hoogstraal H (1979). The mechanism of pathogen- icity in the tick paralyses. J Med Entomol, 16, 357–​69. Murnaghan MF, O’Rourke FJ (1978). Tick paralysis. In: Bettini S (ed) Arthropod venoms: handbook of experimental pharmacology, Vol. 48, p. 419. Springer-​Verlag, Berlin. Pearn J (1977). The clinical features of tick bite. Med J Aust, 2, 313. Centipedes and millipedes Bettini S (ed) (1978). Arthropod venoms: handbook of experimental pharmacology, Vol. 48, p. 977. Springer-​Verlag, Berlin. Radford AJ (1975). Millipede burns in man. Trop Geogr Med, 27, 279–​87. Radford AJ (1976). Giant millipede burns in Papua New Guinea. P N G Med J, 18, 138–​41. Leeches Adams SL (1988). The medicinal leech. A page from the annelids of internal medicine. Ann Int Med, 109, 399–​405. Cundall DB (1986). Severe anaemia and death due to the pharyngeal leech Myxobdella africana. Trans R Soc Trop Med Hyg, 80, 940–​4.

10.4.3 Poisonous fungi 1817

10.4.3 Poisonous fungi 1817

10.4.3  Poisonous fungi 1817 Karunaratne AH, et al. (2015). A Rare Case of Vaginal Bleeding in a Child Due to a Leech Bite and Review of the Literature. Wilderness Environ Med, 26, 579–84. Keegan HL (1963). Leeches as pests of man in the Pacific region. In: Keegan HL, McFarlane WR (eds) Venomous and poisonous ani- mals and noxious plants of the Pacific region, pp. 99–​104. Pergamon Press, Oxford. Sawyer RT (1986). Leech biology and behaviour. Oxford University Press, Oxford. 10.4.3  Poisonous fungi Hans Persson and David A. Warrell ESSENTIALS Phylum Ascomycota includes ergot and aflatoxin-​producing Aspergillus; phylum Basidiomycota includes the larger umbrella-​ shaped gilled mushroom/​toadstools (order Agaricales). Epidemiology—​most fungi are non​toxic and most fungal pois- onings are not severe, but morbidity and mortality remain high in Eastern and Central European countries and the Far East and may be increasing worldwide with globalization of exotic species. Fungal poisoning usually results from mistaking poisonous mushrooms for edible ones, it may be accidental in children, or intentional for hallu- cinogenic effects, suicide, or even homicide. Prevention—​educational campaigns should emphasize risks of careless harvesting and eating. Diagnosis: morphology and habitat of the ingested fungi (and residue in vomitus) and nature and timing of symptoms are informative. Poison information centres may be consulted. Laboratory methods can detect amatoxins, aflatoxins, and ergot alkaloids. Ergot (Claviceps purpurea poisoning)—​results from ingestion of contaminated grains, cereals, and foods (bread). Toxic sclerotia de- velop in ovaries of parasitized grass flowers. Ergot alkaloids cause uterine contraction and vasoconstriction, employed therapeutically in migraine and obstetrics. Lysergic acid derivatives are psychedelic. Chronic ingestion causes gastrointestinal and neurological symp- toms, peripheral vasocontriction, burning pain (‘St Anthony’s fire’) and peripheral gangrene. Larger doses cause acute gastrointestinal symptoms, paraesthesiae, hallucinations, convulsions, and death. Treatment involves vasodilator drugs and anticoagulants. Aspergillus aflatoxin poisoning—​aflotoxins from saprophytic Aspergillus flavus contaminate peanuts, maize, and other grains, seeds, and spices, especially in tropical countries. Outbreaks of aflatoxicosis-​induced hepatitis leading to fatal hepatic necrosis and portal hypertension occur in undernourished rural popula- tions and, in areas of hepatitis B endemicity, cause hepatocellular carcinoma. Mushroom poisonings—​classification is based on toxic effects and related symptoms. Group 1—​Cytotoxic mushroom poisoning 1A—​primary hepatotoxicity (amatoxin poisoning) is caused by Amanita, Lepiota, and Galerina species, notably death cap Amanita phalloides, responsible for 90% of mushroom fatalities (case fatality 10–​20%). Amatoxins block RNA polymerases, inhibiting protein synthesis and damaging intestinal mucosa, kidneys and especially liver. After an ominously long latent period of 8–​24 (mean 12) h, intense GI symptoms followed by transient improvement and inexorably fulminant hepatic failure may de- velop. Attempted decontamination employs multidose activated charcoal. Silibinin, high-​dose benzyl penicillin, and N-​acetyl cyst- eine have been tried and polymyxin B suggested. Liver transplant- ation is life-​saving. 1B—​early primary nephrotoxicity (aminohexadienoic acid poi- soning): Amanita smithiana and some other Amanita species cause acute gastrointestinal symptoms followed a few days later by acute kidney injury. 1C—​late primary nephrotoxicity (orellanine poisoning):  two Cortinarius species can cause insidious poisoning with acute kidney injury, 4–​15 (mean 8) days after ingestion. Group 2—​Neurotoxic mushroom poisoning 2A—​hallucinogenic mushrooms (‘magic mushrooms’) (psilocybin poisoning): Psilocybe, Panaeolus, Conocybe, and other species con- taining potent hallucinogens cause LSD-​like euphoria, hallucinations, depersonalization, and anxiety within 10–​60 minutes. Treatment is with diazepam. 2B—​autonomic-​toxicity mushrooms (muscarinic poisoning): some Inocybe, Clitocybe, Mycena, and Rubinoboletus species contain muscarine causing parasympathetic stimulation after 15 minutes to 2 h. Treatment is with atropine. 2C—​central nervous system-​neuroexcitatory mushrooms (ibotenic acid/​muscimol poisoning):  fly agaric (A.  muscaria) and some other Amanita species contain GABA agonists causing exhil- aration and euphoria alternating with anxiety, agitation, and hallu- cinations within 0.5–​1.5 h, sometimes with cholinergic symptoms. Treatment is with diazepam. 2D—​Morel neurological syndrome: inadequately cooked ‘morels’ (Morchella species), popular ‘edible’ mushrooms, cause GI symp- toms, while M. esculenta and M. conica can cause severe neurological symptoms (tremor, ataxia, visual disturbances, and so on), even if cooked. Group 3—​Myotoxic mushroom poisoning 3A—​early myotoxicity (cycloprop-​2-​ene carboxylic acid poisoning): Russula subnigricans causes gastrointestinal symptoms followed a few h later by generalized myalgia and rhabdomyolysis. 3B—​late myotoxicity (saponaceolide B and M poisoning):
large/​repeated meals of Tricholoma equestre/​ flavovirens, a popular ‘edible’ mushroom, cause generalized rhabdomyolysis 1–​3 days after ingestion. Group 4—​Metabolic toxicity mushroom poisoning 4A—​GABA-​blocking mushroom poisoning (gyromitrin poisoning): inadequately cooked ‘false morel’ (Gyromitra esculenta) and other Gyromitra species contain gyromitrin that impairs central ner- vous system GABA synthesis. After 5—​8 h gastrointestinal symp- toms, cerebellar disturbances, seizures, coma, hepatic damage, haemolysis, and hypoglycaemia may ensue. Treatment is with pyridoxine. 4B—​disulfiram-​like mushroom poisoning (coprine poisoning): antabuse syndrome occurs when alcohol is drunk up to a week after eating Coprinus atramentarius, and some other Coprinus, Clitocybe Boletus species.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1818 4C—​polyporic mushroom poisoning (polyporic acid poisoning): ingestion of the bracket mushroom Hapalopilus rutilans causes gastrointestinal symptoms, neurological effects, acute kidney injury, and passage of purple urine 12 h later. 4D—​trichothecene mushroom poisoning (trichothecene poi- soning): Podostroma cornudamae ingestion has caused fatal multi­ system poisoning with peeling skin in Japan and Korea. 4E—​hypoglycaemic mushroom poisoning:  ‘Yunnan Sudden Unexplained Death’ has been explained by ‘little white mushroom’ (Trogia venenata) poisoning. 4F—​hyperprocalcitoninaemic mushroom poisoning is caused by ‘Satan’s bolete’ (Rubroboletus satanas) associated increased cir- culating procalcitonin and C-​reactive protein causing GI symptoms and fever. 4G—​pancytopenic mushroom poisoning:  has been caused by Ganoderma neojaponicum, a popular herbal medicine in Asia. Group 5—​Gastrointestinal irritant mushroom poisoning This common type of poisoning can follow ingestion of many genera of mushrooms (Agaricus, Boletus, Entoloma, and so on) including many popular, ‘edible’ mushrooms. They cause gastrointestinal symptoms alone, starting within a few hours and usually self-​limiting. Fluid and electrolyte replacement may be necessary, especially in children and older people. In North America and Hawai’i, ‘false parasol’ (Chlorophyllum/​Lepiota molybdites) and ‘Morgan’s mush- room’ cause gastrointestinal symptoms and cholinergic effects within 30 min to 4 h. Group 6—​Miscellaneous adverse reactions to mushrooms 6A—​shiitake mushroom dermatitis (lentinan poisoning):  inad- equately cooked shiitake mushrooms (Lentinola edodes) can cause a generalized, linear prurigo 24 h after ingestion. 6B—​erythromelalgic mushroom poisoning (acromelic acid poi- soning): Clitocybe acromelalgia/​amoenolens can cause acromelalgia. 6C—​paxillus syndrome: severe gastrointestinal symptoms, haem- olysis, and acute kidney injury can follow repeated ingestion of Paxillus involutus. 6D—​encephalopathic syndrome Pleurocybella gyrometrinporrigens, Pleurotus eryngii and some other popular ‘edible’ mushrooms have caused encephalopathy in elderly people suffering from chronic renal or hepatic disease. Cyanide is implicated. Introduction This chapter covers poisoning by members of the fungal sub- kingdom Dikarya that includes most of the ‘higher fungi’. The phylum Ascomycota contains the medically important toxic fungi Claviceps purpurea, the cause of ergotism, Aspergillus flavus, a source of hepatotoxic and carcinogenic aflatoxins, the edible but potentially toxic morel mushrooms (Morchella species), Gyromitra species, and Podostroma cornudamae that has caused multisystem symptoms and signs. The phylum Basidiomycota includes the order Agaricales (gilled mushrooms/​toadstools or agarics) to which most of the med- ically important larger fungi belong. ‘Mushroom’ and ‘toadstool’ may suggest ‘edible’ and ‘poisonous’ respectively, but these terms are not strictly applied. They refer to the visible, above-​ground, fleshy, fruiting bodies of these fungi which are typically umbrella-​shaped with a cap (pileus) with spore-​bearing gills (lamellae) on its under- side, on top of a stalk (stipe), with or without a ring and a volva at its base. Poisonous species must be distinguished from field (Agaricus campestris) and cultivated (A. bisporus) mushrooms and from the many other mushrooms that are considered to be delicious and are passionately sought after by mycophiles worldwide. History Assyrian and pharaonic references to dangerous outgrowths in ears of grain date from 600 to 400 BC. In Europe, epidemics of gangrenous (‘St Anthony’s fire’) and convulsive (‘St Vitus’s dance’) ergotism, ini- tially killing tens of thousands, have occurred between the 9th and early 20th centuries AD. These followed the introduction of rye, a host of ergot, as a major food crop for impoverished rural popula- tions. The poisonous properties of mushrooms/​toadstools have been recognized since ancient Greek and Roman times. Among famous fatalities, homicidal, accidental, or mythical, attributed to poisonous mushrooms, were Siddhartha Gautama (the Buddha), the Roman Emperor Claudius, Pope Clement VII, the Holy Roman Emperor Charles VI, the Russian Tsar Alexis and his wife, and the 18th-​ century German composer Johann Schobert. Aetiology The most common cause of fungal poisoning is confusing poisonous mushrooms with edible ones. Safe mushroom hunting requires skill and experience, as there are many possible sources of dangerous confusion. Too many people harvest, cook, and eat mushrooms whose identity is uncertain. Toddlers may accidentally try mush- rooms and serious poisoning may occur, but this is fortunately un- common. Intentional ingestion of toxic fungi is mostly related to abuse of hallucinogenic fungi. Suicidal ingestion is rare. Illness after eating mushrooms is not necessarily related to poi- soning. Large mushroom fragments or raw mushrooms may prove hard to digest, there may be anxiety that toxic mushrooms might have been ingested, bacterial toxins may be present in mushrooms that have been stored for too long and fungi may be contaminated with toxic heavy metals. There are individual differences in sensibility to some fungal toxins. One example is false morels Gyromitra spp. Toxin contents may also vary between mushrooms of the same species. Epidemiology The incidence of fungal poisoning varies greatly worldwide. Availability of fungi, depending on climate and geographical con- ditions, economics, and lifestyle determine different local traditions for harvesting and eating mushrooms. In some parts of the world mushrooms are part of the normal diet, but more commonly they are eaten as a delicacy. In most places, mushrooms cannot be con- sidered indispensable for nutrition and so it remains an unaccept- able paradox that self-​harvested mushrooms, enjoyed as a delicacy,

10.4.3  Poisonous fungi 1819 still kill people in the 21st century. Most fungi are non​toxic and most fungal poisonings are not severe. However, in certain regions such as Russia and other Eastern and Central European countries, morbidity and mortality is high. Recent publications suggest an increasing incidence of poisoning worldwide, or at least increased recognition and reporting. As in other branches of medicine, there is globalization, such that exotic fungi are being eaten across ex- panding geographical areas and being introduced into places where their effects are unfamiliar. Prevention With a few exceptions, fungal poisoning is accidental. Many people develop severe and life-​threatening illness after mushroom meals. Some die; others suffer chronic, irreversible organ damage requiring transplantation. Since mushrooms are eaten mostly as a delicacy and not as a basic nutritional requirement, it is a paradox that poisoning is so common. The solution is prevention. Educational campaigns should be launched to improve knowledge about mushrooms and raise awareness of the risks involved in careless harvesting and eating. To many people mushroom hunting is an adventure, al- most a game, but this easy-​going attitude must be changed. Because the geographical distribution and appearance of certain fungi are variable, people who are not familiar with the local area tend to be overrepresented as victims of fungal poisoning. Educational mater- ials should therefore be multilingual and specifically addressed to immigrants. Diagnosis Diagnosis may be difficult, as the circumstances are often con- fusing. It is important to consider any disease that may mimic fungal poisoning. The history is crucial. Attention should be paid to the appearance of mushrooms ingested and the habitat where they were harvested. The speed of onset of symptoms and their character, intensity, and duration are often informative. Some fungal poisonings may present with characteristic symp- toms, for example, those caused by muscarine, psychotropic toxins, and amatoxins (see following paragraphs). Careful observation and evaluation of evolving clinical features may, in combination with the history, allow a diagnosis. In difficult cases where a dangerous poisoning cannot be excluded, macro-​ and microscopic examination of the fungi (using online keys and apps), including fragments recovered from vomitus, may prove diagnostic. In many countries, poison information centres may as- sist, either by identifying the mushrooms themselves or by obtaining advice from external experts. Laboratory methods: DNA-​based macroarrays have been used to identify mushroom fragments. α-​amanitin is detectable in serum by high-​performance liquid chromatography (HPLC), and amatoxins by RIA, EIA, EPLC, LC/​MS/​MS and mass LC/​MS in urine. Aflatoxins are detected by thin-​layer chromatography and HPLC (EIA is too in- sensitive) and ergot alkaloids by HPLC and HPLC-​MS/​MS. Specific types of fungal poisoning Ergotism (ergot Claviceps poisoning) Ergotism results from ingestion of uncleaned grains and cereals, or their products such as bread, contaminated with Claviceps species, an Ascomycota fungus. Ideal climatic conditions exist when a warm wet spring and summer follow a cold winter. Claviceps purpurea, the most important species, parasitises grasses and cereals, notably rye but also wheat; C. fusiformis pearl millet in Africa and East Asia; and C. africana sorghum in Africa. Claviceps spores infect the ovaries of flowering grasses from which the poisonous ‘ergot’ or sclerotium develops. It protrudes from the seed head, an odoriforous, hard, curved, blackish-​purple body about 4 cm long shaped like a rooster’s spur (Fig. 10.4.3.1). There are three groups of ergot alkaloids: amines (e.g. ergonovine/​ ergometrine); aminoacids (ergopeptides) (e.g. ergotamine); and semi­ synthetic dehydrogenated compounds (e.g. dihydroergametrine). Ergonovine/​ergometrine causes powerful uterine contraction; er- gotamine causes vasoconstriction, uterine contraction, α-​adrenergic blockade, and central emesis; while dihydroergotamine causes vaso- constriction, α-​adrenergic blockade, and emesis. Beneficial effects of vascular and uterine smooth muscle contraction have been im- plemented for treatment of migraine and to promote delivery of the placenta and prevent post-​partum haemorrhage. Psychedelic lysergic acid diethylamide (LSD) was first synthesized from ergotamine. The ergot alkaloid ergine (d-​lysergic acid amide—​LSA), also found in seeds of the flowers of ‘morning glory’ (Ipomoea tricolor, Rivea/​ Turbina corymbosa), has similar effects. Clinical features Ingestion of relatively small amounts of C. purpurea ergot over long periods causes chronic drowsiness, vomiting, diarrhoea, muscle twitching, ataxia, and hallucinations. Eventually, the vasocon- strictive effects of ergot alkaloids cause ischaemia and gangrene of the feet. The associated burning (ischaemic) pain led to the name ‘St Anthony’s fire’ (not to be confused with erysipelas known as ‘the rose’). Larger doses cause acute vertigo, tinnitus, headache, high fever, nausea, vomiting, other gastrointestinal disturbances, uterine con- tractions, paraesthesiae, hallucinations, behavioural abnormalities, weakness, convulsions, and death. Claviceps fusiformis poisoning causes self-​limiting acute gastro- enteritis and drowsiness. A similar range of symptoms is described in patients who took excessive doses of therapeutic ergot alkaloids. Treatment For peripheral ischaemia, parenteral nitroprusside or nitrogly- cerin, oral prazosin, or ACE-​inhibitors and anticoagulants are recommended. Prevention Selection of less susceptible strains, development of ergot-​resistant crops, improved crop husbandry through minimizing ergot-​ susceptible grass weeds (e.g. black grass), azole-​treatment of grains, surveillance of milling, and use of food processing and baking reduce the risk of ergotism.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1820 Aspergillus aflatoxin poisoning Apergillus species are conidial Ascomycota fungi that are ubiqui- tous saprophytes of soil and vegetation, including agricultural crops. Aspergillus flavus is the most widespread species. About 30% of its strains produce aflatoxins. These were first recognized, isolated, and characterized in 1960, after the death of more than 100 000 turkeys from ‘turkey X disease’, attributed to consumption of Aspergillus-​ contaminated peanut meal. Among at least 14 of these naturally occurring thermostable mycotoxins, the most important are the difurocoumarocyclopentenones: aflatoxins B1 (AFB1, the most dan- gerous) and AFB2, produced by Aspergillus flavus, A. parasiticus and A. nomius; and AFG1 and AFG2, produced by A. parasiticus and A. nomius. AFM1 and AFM2 in milk or urine of poisoned humans or animals results from 4-​hydroxylation of AFB1 and AFB2 in the liver. Foodstuffs that are commonly contaminated with aflatoxins include ground (peanuts) and tree nuts, maize, corn, and other grains, seeds, cassava, dried fish, chilli peppers and other spices, especially in India, the Far East, and South America. Clinical features Outbreaks of aflatoxicosis-​induced hepatitis have occurred in undernourished rural populations in tropical countries such as Fig. 10.4.3.1  Ergot (Claviceps purpurea) showing development of the sclerotium in the ovary of the host grass flower (3–​5), and fruiting bodies (6–​8). Reprinted from P. Esser (1910). Die Giftpflanzen Deutschlands. Braunschweig, Friedrich Vieweg und Sohn. Copyright © 1910, Springer Fachmedien Wiesbaden, with permission from Springer.

10.4.3  Poisonous fungi 1821 India (maize diet) and Kenya (corn diet). Jaundice, ascites, portal hypertension, and fatal portal variceal haemorrhage were features of these epidemics. High doses of aflatoxins B1 and G1 cause prolif- eration of bile duct epithelium, fatty infiltration, and centrilobular necrosis. In areas of hepatitis B endemicity, aflatoxicosis is a co-​ carcinogen for hepatocellular carcinoma. AFB1 is the most potent naturally occurring carcinogen. It is metabolized in the liver to a reactive epoxide, which forms DNA adducts at some guanine res- idues. The epoxide adduct catalyses a G → T mutation in the p53 tumour suppressor protein gene, causing a missense mutation that inactivates its product. Treatment is supportive. Prevention Risk of aflatoxicosis can be reduced by improving agricultural prac- tices, including storage, and surveillance of AF levels in key crops and foodstuffs. Classification of mushroom poisonings (mycetismus, mycetism, mycotoxicosis) Fungal toxins are a heterogeneous group, chemically and toxi- cologically. In clinical practice, the most relevant approach is a classification based on toxic effects and related symptoms (see Table 10.4.3.1). The scheme adopted here is the new classification recently proposed by White, Weinstein, De Haro, Bédry, Schaper, Rumack and Zilker (2016). These authors have also developed a useful diagnostic algorithm that is colour-​coded for key clinical features of poisoning. Group 1—​Cytotoxic mushroom poisoning 1A—​Primary hepatotoxicity (amatoxin poisoning) The highly poisonous amatoxins occur in species of the families Amanitaceae (genus Amanita), Agaricaceae (genus Lepiota), and Cortinariaceae (genus Galerina). The death cap Amanita phalloides (Fig. 10.4.3.2), destroying angel A. virosa (Fig. 10.4.3.3), eastern North American destroying angel A. bisporigera, Western North American destroying angel A. ocre- ata, Guangzhou destroying angel A. exitialis, and fool’s mushroom A.  verna are the most commonly involved in human poisoning. Galerina species such as L. marginata and Lepiota species such as L. helveola may also be implicated. Epidemiology Amatoxin poisonings are reported from all continents, but are most frequent in Europe, accounting for more than 90% of mushroom fatalities (case-​fatalities 18–​22% in adults, 33–​51% in children in the 1970s and 1980s). Case fatality has declined in Western countries but remains alarmingly high in other parts of the world. Pathogenesis The eight amatoxins, of which α-​amanitin is the most toxic, are cyclic octapeptides that inhibit transcription of DNA to mRNA by blocking nuclear RNA polymerases II and III ac- tivity. This results in defective protein synthesis and cell death. Amatoxins also act with endogenous cytokines to induce apop- tosis, and there is glutathione depletion. The main target organs are intestinal mucosa, liver, and kidneys. Hepatotoxicity deter- mines prognosis. Clinical features After a latent period of 8–​24 (mean 12) h after ingestion, gastro- intestinal symptoms start violently with intense, watery diarrhoea, and vomiting and persist for 36 h. This latency has great diagnostic significance. Patients become rapidly dehydrated and develop oli- guria, hypoglycaemia, hypokalaemia, and metabolic acidosis. After apparent improvement, biochemical signs of liver damage appear after 36–​48 h and progress over the next few days. Fulminant hepatic failure may develop. Initial disturbances of renal function will re- solve after rehydration, but within another 3–​4 days, renal function may again deteriorate because of toxic kidney damage, a sign of poor prognosis. Treatment Decontamination  Forced emesis or gastric lavage is performed if the patient is admitted within 4–​6 h and this can be accomplished safely. Multidose activated charcoal is always given. Toxin removal • Multiple-​dose activated charcoal is administered for 3 days after ingestion (20–​40 g every 3–​4 h or 50 g every 6 h). • Diuresis of about 200 ml/​h (adults) is maintained for the first 24–​ 48 h after ingestion. • Haemoperfusion or haemodialysis is not indicated unless the pa- tient has pre-​existing renal disease or is admitted very early and in the asymptomatic period (very rare). Reduction of  hepatic toxin  uptake  Silibinin (a component of silymarin from milk thistle—​Silybum marianum) (Legalon® SIL) is a free radical scavenger, an anti-​inflammatory agent that stimulates protein synthesis and inhibits amatoxin uptake by hepatocytes. 5 mg/​ kg is given by intravenous infusion over 1 h, followed by 20 mg/​kg/​ 24 h as a continuous infusion for 3 days after ingestion. The effective- ness of this treatment is not entirely established. Parenteral silibinin is not always available, even in Western countries. High-​dose benzyl penicillin (600 mg or 1 megaU/​kg/​day is an alternative. Recently, evidence has been produced that polymyxin B, which, prevents α-​ amanitin from binding to RNA polymerase II, might be a promising new treatment. Symptomatic and supportive care • Symptomatic care is crucial and includes cautious monitoring, fluid replacement, and correction of metabolic disturbances. Hepatic and renal support may be required. • There is some experimental, theoretical, and clinical support for the use of N-​acetylcysteine (300 mg/​kg as a continuous IV infu- sion over 21 h) as a liver-​protective agent. • If fulminant hepatic failure is pending, a liver unit should be consulted for advice on treatment and with a view to possible transplantation. Prognosis and comments  The prognosis is related to toxic dose and start of treatment. Case fatality is high after heavy exposure. Vigorous symptomatic and supportive care, maintenance of an ad- equate diuresis, and multiple-​dose activated charcoal are accepted treatments. Silibinin may modify toxicity to some extent through

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1822 Table 10.4.3.1  Classification of mushroom poisoning according to principal clinical features of poisoning and causative toxin(s) where known Group Toxin Commonly implicated species Clinical effect Time of onset 1—​Cytotoxic poisoning 1A—​1° hepatotoxic Amatoxins Amanita phalloides, A. virosa, A. bisporigera, A. ocreata, A. exitialis, A. verna, Galerina marginata, Lepiota helveola Intense gastroenteritis after 8–​24 h, liver and kidney damage 8–​12 h 1B—​1° early nephrotoxic Aminohexadienoic acid A. smithiana, A. proxima Gastroenteritis, headache, fatigue, myalgias, rash, AKI, cardiotoxicity 1–​12 h 1C—​1° late nephrotoxic Orellanine Cortinarius orellanus, C. rubellus (speciosissimus) Gastroenteritis, headache, drowsiness, sweating, rash, AKI 4–​15 days 2—​Neurotoxic poisoning 2A—​Hallucinogenic ‘magic mushrooms’ Psilocybins, psilocins, gymnopilins Psilocybe semilanceata, Panaeolus, Conocybe, Gymnopilus, Copelandia, Pluteus, Stropharia species Euphoria, anxiety, agitation, aggression, hallucinations, depersonalization, tachycardia, flushing 10–​60 minutes 2B—​Autonomic Muscarine Inocybe patouillardii, Clitocybe dealbata, C. rivulosa), Mycena, Rubinoboletus species Nausea, diarrhoea, sweating, salivation, miosis, bradycardia, hypotension, lacrimation, chills,
tremor, bronchospasm, and so on 15–​120 minutes 2C—​CNS toxic Isoxazoles (ibotenic acid, muscimol, muscazone) A. muscaria, A. pantherina, A. strobiliformis, A. regalis Inebriation, euphoria, confusion, agitation, anxiety, delusions, hallucinations, violent behaviour, seizures, tachycardia, mydriasis, urinary retention, Minutes—​3 h 2D—​Morel neurological syndrome ? Morchella Gastroenteritis, dizziness, ataxia, visual disturbances, headache, paraesthesiae, trismus, muscle spasms, drowsiness/​ confusion, dysarthria 12 h 3—​Myotoxic poisoning 3A—​Early myotoxic Cycloprop-​2-​ene
carboxylic acid Russula subnigrans Gastroenteritis, rhabdomyplysis,
2° AKI, collapse 2+ h 3B—​Late myotoxic Saponaceolides B & M Tricholoma equestre/​flavovirens Fatigue, myalgia, weakness, rhabdomyolysis, sweating, AKI, collapse 1–​3 days 4—​Metabolic toxicity poisoning 4A—​GABA-​blocking Gyromitrins/​ monomethylhydrazine Gyromitra esculenta Neurological, gastrointestinal, and hepatic symptoms; haemolysis 6–​12 h 4B—​Disulfiram-​like Coprines Coprinus atramentarius, Clitocybe clavipes, Boletus luridus Antabuse-​like reaction after alcohol 4C—​Polyporic Polyporic acid Hapalopilus rutilans Gastroenteritis then neurotoxic effects, purple urine, AKI, increased aminopeptidases 12 h 4D—​Trichothecene Trichothecenes Podostroma cornudamae Gastroenteritis, hypertension,
oliguria, pancytopenia, skin peeling, alopecia Hours–​days 4E—​Hypoglycaemic γ-​guanidinobutyric acid, 2R-​amino-​4S-​hydroxy-​5-​ hexynoic acid, 2R-​amino-​5-​ hexynoic acid Troiga venenata Profound hypoglycaemia (severe), dizziness, SOB, syncope 2 h 4F—​Hyperprocalcitoninaemic ? Boletus satanas Gastroenteritis, low grade fever, increased plasma procalcitonin,
C-​reactive protein 2 h 4G—​Pancytopenic ? Ganoderma neojaponicum Fever, pancytopenia ? 5—​Gastrointestinal irritant poisoning Many different and unidentified Agaricus, Entoloma, Boletus, Hebeloma, Tricholoma, Russula, Lactarius, Ramaria, Chlorophyllum/​Lepiota molybdites, Macrolepiota/​Lepiota morganii Gastrointestinal symptoms 1–​3 h (continued)

10.4.3  Poisonous fungi 1823 reduction of the hepatic uptake of amatoxin. In some cases, liver transplantation may be the ultimate way of saving the patient. 1B—​Early primary nephrotoxicity
(aminohexadienoic acid poisoning) Allenic norleucine (aminohexadienoic acid) and chlorocrotylglycine in Amanita smithiana, A. pseudoporphyria, A. proxima, and some other Amanita species can cause acute kidney injury (AKI). Initial gastrointestinal and other symptoms develop between 20 minutes and 12 h after ingestion, followed by AKI a few days later. This was reported from North America after ingestion of A. smithiana, mis- taken for Tricholoma magnivelare, and in southern France, in pa- tients who had eaten A. proxima, mistaken for A. ovoidea as part of ‘proximien syndrome’. ‘Proximien syndrome’ consists of gastro- intestinal symptoms starting 8–​14 h after ingestion, followed after 4 days by AKI with evidence of transient liver damage and severe cardiotoxicity in some cases. 1C—​Late primary nephrotoxicity (orellanine poisoning) Orellanine is a potent nephrotoxin present in certain species of the family Cortinariaceae, genus Cortinarius. C.  orellanus and C.  rubellus (speciosissimus) (‘cortinar’ or webcap”) (Figs. 10.4.3.4 and 10.4.3.5) are responsible for most poisonings. Orellanine is a bipyridine N-​oxide that may interfere with protein synthesis in the kidneys causing interstitial nephritis, tubular cell damage, basal cell membrane rupture and, eventually, irreversible fibrosis. Clinical features Orellanine poisoning is the most insidious of all mushroom poison- ings. Occasionally, there may be some mild gastrointestinal symp- toms within a couple of days after the meal, but as these symptoms Group Toxin Commonly implicated species Clinical effect Time of onset 6—​Miscellaneous reactions 6A—​Shiitake dermatitis Lentinans Lentinola edodes Linear prurigo 1–​2 days 6B—​Erythromelalgic Acromelic acid Clitocybe acromelalgia/​ amoenolens Acromelalgia Hours–​days 6C—​Paxillus syndrome ? Paxillus involutus Collapse, gastroenteritis, autoimmune haemolysis, AKI, DIC 1–​2 h 6D—​Encephalopathic syndrome Cyanide Pleurocybella porrigens Cramps, coma (in elderly patients with chronic renal/​hepatic disease) Days–​weeks Based on the new classification by White J, Weinstein SA, De Haro L, Bédry R, Schaper A, Rumack BH, Zilker T. (2016). Fig. 10.4.3.2  Death cap Amanita phalloides. Courtesy of Hans Marklund. Fig. 10.4.3.3  Destroying angel Amanita virosa. Courtesy of Hans Marklund. Table 10.4.3.1  Continued Fig. 10.4.3.4  Cortinarius rubellus (speciosissimus). Courtesy of Astrid Holmgren.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1824 are both discrete and inconsistent they are easily overlooked. However, usually symptoms do not appear until 4–​15 (mean 8) days after the mushroom meal, and, by then, reflect established kidney damage. Symptoms evolve insidiously and are difficult for the patient to interpret—​headache, fatigue, intense thirst, chills, muscular discomfort, abdominal, lumbar, and flank pain. After a polyuric phase, oliguria and anuria may follow. Laboratory tests on admission reveal elevated serum creatinine and urea, proteinuria, haematuria, and—​characteristically—​pyuria. The AKI may resolve or become chronic. Treatment Since patients are normally admitted late, therapeutic interven- tions can neither prevent nor reduce toxic damage. Renal function is monitored. Therapy is symptomatic with support of renal func- tion and treatment of uraemia, including dialysis while waiting for the kidneys to recover. In case of persistent renal insufficiency, the options are chronic dialysis or transplantation. However, transplant- ation should not be performed too early, as renal recovery may be considerably delayed. Very early suspicion of orellanine poisoning should prompt meas- ures to prevent absorption and promote elimination. Prognosis and comments End-​stage renal failure was observed in 11% of Polish, 17% of French, and 40% of Swedish patients. It should be emphasized that treatment measures discussed above are theoretically based and there is so far no clinical support for a rational treatment strategy of this ghostly poisoning, apart from supportive and symptomatic care. Group 2—​Neurotoxic mushroom poisoning 2A—​Hallucinogenic mushrooms (‘magic mushrooms’) (psilocybin poisoning) Psilocybin and related toxins occur particularly in Psilocybe spe- cies, for example, liberty cap Psilocybe semilanceata (Fig. 10.4.3.6), and Panaeolus, Conocybe, Gymnopilus, Copelandia, Pluteus, and Stropharia species. The toxins—​psilocybins, psilocins, and gymnopilins—​are tryptamine derivatives that increase sero- tonin levels in the central nervous system and act as potent hal- lucinogens. The effects mimic those of LSD. Ingestion is almost invariably related to abuse as these fungi are sought out for their hallucinogenic properties and can be purchased legally in some countries. Clinical features Within 10–​60 min, the patient will experience altered sense of time and space, euphoria, hallucinations, and depersonalization. Less pleasurable symptoms are anxiety, agitation, aggression, bizarre and terrifying hallucinations, tachycardia, mydriasis, headache, and flushing. Symptoms peak at around 2 h after ingestion and start vanishing after 4–​6 h. However, symptoms may persist and there may be flashbacks after weeks or months. Organic psychosis is a differential diagnosis. A reliable history may be available only after recovery. Treatment The patient should rest in a quiet environment and be sedated (e.g. with diazepam). If this is inadequate, haloperidol or chlorpro- mazine can be added. 2B—​Autonomic-​toxicity mushrooms (muscarinic poisoning) Toxic amounts of muscarine occur particularly in certain Inocybe species (e.g. I.  patouillardii), Clitocybe species (e.g. C.  dealbata, C. rivulosa), Mycena species, and Rubinoboletus species. Muscarine has also been detected in small, mostly insignificant amounts in other genera. Muscarine stimulates cholinergic receptors in the autonomic nervous system. Clinical features Symptoms of parasympathetic stimulation start within 15 minutes to 2 h. Nausea, diarrhoea, diaphoresis, hypersalivation, miosis, brady- cardia, and hypotension are common and rhinorrhoea, lacrimation, chills, tremor, central nervous system depression, bronchorrhoea, bronchospasm, and painful micturition have been reported. The pa- tient is often pale and feels sick and miserable. The clinical features are fairly diagnostic. Fig. 10.4.3.5  A common cause of poisoning by nephrotoxic Cortinarius spp. is confusion of Cortinarius rubellus (the lower three fungi in this picture) and funnel chanterelle Cantharellus tubaeformis. Courtesy of Astrid Holmgren. Fig. 10.4.3.6  Liberty cap/​‘magic mushroom’ Psilocybe semilanceata. Courtesy of Hans Marklund.

10.4.3  Poisonous fungi 1825 Treatment Intravenous atropine (adults 1–​2 mg, children 0.02–​0.05 mg/​kg) effectively counteracts the cholinergic symptoms. Repeated doses may be required. Symptomatic treatment is given as required. 2C—​CNS-​neuroexcitatory mushrooms
(ibotenic acid/​muscimol poisoning) Isoxazoles (ibotenic acid, muscimol, and muscazone) occur in certain Amanita species, for example, fly agaric A. muscaria (Figs. 10.4.3.7 and 10.4.3.8), panther cap A. pantherina (Fig. 10.4.3.9), A. strobili- formis, and A. regalis. The toxins act as GABA agonists. Clinical features Symptoms start within 0.5–​1.5 h, peak at around 3 h, and vanish gradually over the next 24 h. The symptoms are unpredictable: exhilaration, euphoria, drowsiness, and confusion alternate with anxiety, agitation, delusions, illusions, and hallucinations. Extreme agitation and violent behaviour may ensue. Occasionally myoclonic jerks, muscle fasciculations, and seizures are ob- served. Tachycardia, mydriasis, and urinary retention may occur. Cholinergic symptoms are attributable to trace amounts of mus- carine in some specimens. Panther cap more often causes central nervous system depression, whereas fly agaric is more likely to trigger excitation and bizarre behaviour. History and symptoms are often diagnostic. However, the his- tory is often obscure until patients are fit enough to tell their story. Differential diagnoses include organic psychosis and central ner- vous system infections. Treatment Treatment is symptomatic and supportive. Intravenous diazepam (adults 5–​10 mg, children 0.1–​0.2 mg/​kg) is given and repeated for sedation. Haloperidol or chlorpromazine may be useful as a complement in delirious and agitated patients. 2D—​Morel neurological syndrome (toxin unknown) ‘Morel’ (Morchella species) are highly relished edible mushrooms, but they contain toxic thermolabile hydrazine haemolysins. Ingestion of un/​inadequately cooked morels causes acute gastroenteritis (nausea, vomiting, abdominal pain, diarrhoea) after about 5 h. In France and Bavaria, people who ingested morels such as Morchella esculenta and M. conica, even cooked ones, developed neurological symptoms about 12 h later: tremor, dizziness/​inebriation, unsteadi- ness/​ataxia, visual disturbances, headache, paraesthesiae, trismus or muscle spasms, drowsiness/​confusion, and dysarthria. Group 3—​Myotoxic mushroom poisoning 3A—​Early myotoxicity (cycloprop-​2-​ene carboxylic
acid poisoning) Generalized rhabdomyolysis following ingestion of Russula sub- nigricans (order Russalales), a mushroom that stains red when Fig. 10.4.3.7  Fly agaric Amanita muscaria. Courtesy of Ole Högberg. Fig. 10.4.3.8  Variations in appearance of fly agaric Amanita muscaria—​may result in confusion with edible mushrooms. Courtesy of Ole Högberg. Fig. 10.4.3.9  Panther cap Amanita pantherina. Courtesy of Hans Marklund.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1826 damaged. It was first described in Taiwan in 2001 and subsequently in China, Japan, and Korea. Toxicity is attributable to unstable cycloprop-​2-​ene carboxylic acid. Poisoning causes self-​limiting nausea, vomiting, diarrhoea, and agitation about 2 h after ingestion in most victims, followed a few h later in a minority by generalized myalgia and rhabdomyolysis (serum creatine kinase concentration

20 000 U/​litre), with or without hyperkalaemia, hypocalcaemia, respiratory failure, AKI, pulmonary oedema, hypertension, ven- tricular tachycardia, and circulatory shock. 3B—​Late myotoxicity (saponaceolide B and M poisoning) Rhabdomyolysis has been reported in France and Poland, after ingestion of large or repeated meals of ‘man on horseback’ or ‘yellow knight’ (Tricholoma equestre/​ flavovirens) that grow under conifers. These puzzling observations, implicating a popular mushroom, have caused considerable attention and concern. Toxic saponaceolides B and M have been found in the related and widely-​consumed ‘grey knight’ or ‘dirty tricholoma’ (T. terreum) but not in T. equestre. Clinical features Progressive fatigue, muscle weakness, myalgia especially in the thighs, nausea, sweating, facial erythema, and evidence of generalize rhabdomyolysis develop 1–​3 days after the last meal of T. equestre. In severe cases there is progression to fatal acute respiratory failure, cardiac arrhythmias, and cardiovascular collapse. The case fatality in Poland was 20%. Severity correlated with the amount of ingested mushroom. Group 4—​Metabolic toxicity mushroom poisoning 4A—​GABA-​blocking mushroom poisoning
(gyromitrin poisoning) Gyromitrin occurs in fungi of the family Helvellaceae, genus Gyromitra. Most poisonings are caused by the ‘false morel’ or ‘Lorchel’ (Gyromitra esculenta), but the toxin is found also in other Gyromitra species. Gyromitrin decomposes in the stomach to form monomethylhydrazine. This reduces central nervous system pyridoxine and, hence, GABA synthesis. Glutathione depletion in erythrocytes and damage to hepatic macromolecules has also been postulated. Gyromitrin is water-​soluble and volatile, and so can be partly removed from the fungus by drying or boiling. Clinical features Systemic symptoms may follow inhalation of vapour, which also irritates eyes and airways. Most common, however, is poisoning by ingestion of false morels that have not been properly pre- pared. Symptoms are delayed 5–​8 h. Vomiting and diarrhoea may occur, but more typical features are vertigo, ataxia, nystagmus, diplopia, balance disturbances, diaphoresis, slurred speech, and drowsiness. Rare symptoms are delirium, seizures, and coma. Moderate hepatic damage, haemolysis, and hypoglycaemia have been observed. Treatment If neurological symptoms prevail, it is relevant to give pyridoxine 25 mg/​kg as an intravenous bolus infusion over 30 min. Repeated doses may be required. If the patient is convulsing and pyridoxine is not immediately available, diazepam is given initially. It is wise to have a glucose infusion running and maintain an adequate diuresis. 4B—​Disulfiram-​like mushroom poisoning (coprine poisoning) Antabuse syndrome An ‘antabuse syndrome’ may be induced by ‘common ink cap’ or ‘inky cap’ (Coprinus atramentarius), a few other Coprinus species, Clitocybe clavipes, and Boletus luridus. The toxin (coprine) acts like disulfiram, blocking acetaldehyde dehydrogenase. Consequently, drinking ethanol after eating these mushrooms will cause an antabuse syndrome—​flushing, sweating, nausea, anxiety, tachycardia, hypotension, and dyspnoea. The risk persists for about 1 week after the mushroom ingestion. If the mistake is discovered early, decontamination and adminis- tration of activated charcoal may be useful. Otherwise symptomatic and supportive care is given. 4C—​Polyporic mushroom poisoning (polyporic acid poisoning) An epiphytic bracket mushroom, ‘tender nesting polypore’, ‘purple dye polypore’, or ‘cinnamon bracket’ (Hapalopilus rutilans) con- tains polyporic acid that can cause gastrointestinal and neuro- logical effects about 12 h after ingestion, followed in severe cases by AKI, mild hepatotoxicity, proteinuria, pyuria, and diagnostic purple/​violet coloured urine. 4D—​Trichothecene mushroom poisoning
(trichothecene poisoning) Podostroma cornudamae ingestion has caused fatal multisystem poisoning, including AKI, pancytopenia, peeling skin, and alopecia in Japan and Korea attributable to trichothecenes. 4E—​Hypoglycaemic mushroom poisoning Deaths of some 400 villagers in Yunnan Province, China over 35 years (‘Yunnan Sudden Unexplained Death’) have been attrib- uted to a previously undescribed species of ‘little white mushroom’, Trogia venenata. It contains γ-​guanidinobutyric acid, 2R-​amino-​ 4S-​hydroxy-​5-​hexynoic acid, 2R-​amino-​5-​hexynoic acid which cause profound hypoglycaemia (like hypogycin from ackee fruit below). Inhibition of β-​oxidation prevents generation of glucose by gluconeogenesis from lipid substrate. 4F—​Hyperprocalcitoninaemic mushroom poisoning
(toxin unknown) Poisoning by Satan’s bolete (Rubroboletus/​Boletus satanas) in France, associated with high plasma concentrations of procalcitonin and C-​reactive protein, caused self-​limiting gastrointestinal symptoms and fever. 4G—​Pancytopenic mushroom poisoning (toxin unknown) Self-​limiting pancytopenia has been described in people taking Ganoderma neojaponicum, a common constituent of herbal medi- cines in China, NE, and SE Asia. Group 5—​Gastrointestinal irritant mushroom poisoning. Fungi solely causing gastroenteritis form the largest subgroup, comprising species from many genera (e.g. Agaricus, Boletus,

10.4.3  Poisonous fungi 1827 Entoloma, Hebeloma, Lactarius, Ramaria, Russula, and Tricholoma). Many of these genera include delicious, popular, and edible mush- rooms, increasing the risk of confusion. Few toxins in this group are chemically identified. They cause non​specific irritation of the gastrointestinal mucosae. Clinical features Vomiting and diarrhoea start within a few hours and generally resolve quickly. Intensity and duration may, however, vary. For ex- ample, ‘leaden entoloma’ (Entoloma sinuatum) may cause intense and long-​lasting symptoms. Depending on length and duration of symptoms, fluid and electrolyte imbalance may ensue. Gastrointestinal infection must be considered as a differential diagnosis, together with effects of gorging on large amounts of indi- gestible mushrooms (the death of the Roman Emperor Jovian was attributed to ‘a surfeit of mushrooms’). Delayed onset of intense symptoms until 8 to 24 h after the mushroom meal suggests poten- tially dangerous poisoning by amatoxin-​containing fungi (see fol- lowing paragraphs). Treatment Admission to hospital is seldom necessary, but if symptoms are more intense fluid and electrolyte replacement may be necessary, espe- cially in children and older people. The ‘false parasol’ or ‘green-​spored parasol’ (Chlorophyllum molyb- dites or Lepiota molybdites; also Morgan’s mushroom, Macrolepiota or Lepiota morganii) is the most commonly ingested poisonous mushroom in North America and Hawai’i. The toxin is irritant and also exerts α-​adrenergic blockade and cholinergic effects. Within 30 min to 2 (–​4) h, intense, watery, and sometimes bloody diarrhoea begins (also a feature of ‘tsukiyotake’ —​Omphalotus japon- icas/​guepiniformis—​ingestion mistakenly for shiitake mushrooms in Japan). This may result in dehydration, electrolyte imbalance, shock, and renal impairment. Occasionally miosis, pallor, diaphoresis, and hypotension are observed. Fluid replacement and other symptomatic and supportive care are given as required. Atropine is given in case of cholinergic symptoms. Group 6—​Miscellaneous adverse reactions to mushrooms 6A—​Shiitake mushroom dermatitis (lentinan poisoning) Inadequately cooked shiitake mushrooms (Lentinola edodes) that are cultivated in the Far East, can cause ‘shiitake dermatitis’, a gener- alized, light-​sensitive, erythematous, micro-​papular, linear, pruritic, urticarial rash that appears 24 hours after ingestion and that persists for 3–​21 days. It is attributed to the thermolabile polysaccharide, lentina. 6B—​Erythromelalgic mushroom poisoning (acromelic acid poisoning) Consumption of ‘paralysis funnel’ or ‘poison dwarf bamboo mush- room’ (Clitocybe acromelalgia/​amoenolens) can cause persisting acromelalgia, attributed to acromelic acid. 6C—​Paxillus syndrome (toxin unknown) After repeated ingestion of the ‘roll-​rim cap’ or ‘poison pax’ (Paxillus involutus), its antigens may induce a Paxillus syndrome:  severe gastroenteritis, haemolysis, and subsequent renal impairment. High plasma concentrations of bilirubin and aminotransferases due to haemolysis may suggest amatoxin poisoning. Symptomatic and supportive care includes fluid replacement, maintenance of adequate diuresis, and blood transfusions. 6D—​Encephalopathic syndrome (possibly cyanide poisoning) ‘Angel wing’ (Pleurocybella gyrometrinporrigens), ‘king trumpet mushroom’ (Pleurotus eryngii) and several other species of mush- rooms long considered safe for eating have been responsible for delayed encephalopathic poisoning in Japan in elderly people often with chronic renal or hepatic disease. Cyanide has been implicated. FURTHER READING Bédry R, et  al. (2001). Wild-​mushroom intoxication as a cause of rhabdomyolysis. N Engl J Med, 345, 798–​802. Bresinsky A, Besl H (1990). A colour atlas of poisonous fungi. A hand- book for pharmacists, doctors, and biologists. Wolfe Publishing, London. Cooper MR, Johnson AW (1998). Poisonous plants and fungi in Britain. Animal and human poisoning, 2nd edition. The Stationery Office, London. Dinis-​Oliveira RJ, et al. (2016). Human and experimental toxicology of orellanine. Hum Exp Toxicol, 35, 1016–​29. Enjalbert F, et al. (2002). Treatment of amatoxin poisoning: 20-​year retrospective analysis. J Toxicol Clin Toxicol, 40, 715–​57. Garcia J, et  al. (2015). A breakthrough on Amanita phalloides poi- soning: an effective antidotal effect by polymyxin B. Arch Toxicol, 89, 2305–​23. Garcia J, et al. (2015). Amanita phalloides poisoning: mechanisms of toxicity and treatment. Food Chem Toxicol, 86, 41–​55. Graeme KA (2014). Mycetism: a review of the recent literature. J Med Toxicol, 10, 173–​89. Holmdahl J, Blohmé I (1995). Renal transplantation after Cortinarius speciosissimus poisoning. Nephrol Dial Transplant, 10, 1920–​2. Karlson-​Stiber C, Persson H (2003). Cytotoxic fungi—​an overview. Toxicon, 42, 339–​49. Lee PT, et al. (2001). Rhabdomyolysis: an unusual feature with mush- room poisoning. Am J Kidney Dis, 38, E17. Saviuc P, et al. (2010). Can morels (Morchella sp.) induce a toxic neuro- logical syndrome? Clin Toxicol (Phila), 48, 365–​72. Wasson RG (1972). The death of Claudius or mushrooms for
murderers. Botanical Museum Leaflets Harvard University, 23, 101–​28. White J, et  al. (2016). Mushroom poisoning:  a proposed new
clinical classification. (in press). Presented at The 18th World Congress of the International Society on Toxinology, Oxford, UK, 25–​30 September 2015:  abstract 143. Toxicon 103 Supplement (2015), 85–​6. Yin X, et al. (2014). Chemical and toxicological investigations of a pre- viously unknown poisonous European mushroom Tricholoma ter- reum. Chemistry, 20, 7001–​9.1 Zhou ZY, et  al. (2012). Evidence for the natural toxins from the mushroom Trogia venenata as a cause of sudden unexpected death in Yunnan Province, China. Angew Chem Int Ed Engl, 51, 2368–​70. Mushroom identification app: http://​rogersmushroomsapp.com/

10.4.4 Poisonous plants 1828

10.4.4 Poisonous plants 1828

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1828 10.4.4  Poisonous plants Michael Eddleston and Hans Persson ESSENTIALS Many plants contain toxic substances—​heterogeneous in chemical composition and diverse in toxic effects. When classifying plant pois- onings, a pragmatic approach is to look at the main clinical effects, but it should be emphasized that few plant toxins produce just one type of symptom and that symptomatology is often multiple, with some features predominating. Ingestion of, or contact with, poisonous plants is common but ser- ious plant poisoning is rare worldwide because most plant exposures are accidental: the majority occur in small children, the ingested dose is usually small, and no treatment is required. Globally, severe plant poisoning usually results from intentional exposure. Toxic plants are ingested for self-​harm in certain regions (e.g. in Sri Lanka and India), where cardiac glycosides in yellow ole- ander Cascabela thevetia and odollam Cerbera manghas and diphyllin glycosides in oduvan Cleisthanthus collinis are responsible for much morbidity and mortality. Other intentional and serious poisonings occur with, for example, Aconitum and Colchicum spp.. Plants with psychotropic and hallucinogenic effects, for example, Datura and Cannabis spp., are abused as recreational drugs or used in food or drinks to render travellers unconscious for robbery. Severe unintentional poisoning has resulted from the use of herbal medicines or foods containing, or contaminated by, plant toxins
(e.g. aconitine in China, cyanide (cassava) in Africa, and aristolochic acid (Aristolochia spp.) in Europe). Food insecurity in lower in- come countries results in outbreaks of poisoning, often in children
(e.g. with pyrrolizidine alkaloids in Heliotrope species, mitochondrial toxins in Xanthium strumarium, a hepatotoxin in Blighia sapida, and cyanogenic glycosides in toxic Detarium senegalense). Treatment of severe plant poisoning includes careful de- contamination and symptomatic and supportive care. Specific antidotes are only available for poisoning by plants containing belladonna alkaloids (physostigmine), cardiac glycosides (digoxin-​ specific Fab fragments), and cyanogenic agents (dicobalt edetate, hydroxocobalamin). Aetiology and epidemiology The most common exposure to toxic plants, particularly in the West, is in young children as they explore their environment. Few of these exposures result in serious harm. In other parts of the world, par- ticularly Asia, the most common plant exposures are from self-​harm in adolescents and adults, sometimes with fatal outcome. Deaths also occur after children (although rarely adults) eat poisonous plants as food, especially where there is food insecurity. Herbal medicines containing toxic principles cause accidental plant poisoning, while recreational ingestion or smoking of psychoactive plants is popular in industrialized countries. A clinically oriented overview of the main plant toxins is given in Table 10.4.4.1. Neurotoxic plants Anticholinergic toxins Belladonna alkaloids (atropine, hyoscyamine, scopolamine) occur in a variety of plants, including deadly nightshade Atropa bella- donna, henbane Hyoscyamus niger, thorn apple/​jimson weed Datura stramonium, and angels’ trumpets Brugmansia suaveolens. These al- kaloids are muscarinic acetylcholine receptor antagonists, causing central and peripheral anticholinergic effects. Poisoning can occur unintentionally in children or, in some localities, as part of drugging a person for robbery; most commonly, toxicity occurs after recre- ational use in young people ingesting Datura or Brugmansia spp. The anticholinergic toxidrome includes tachycardia, fever, agita- tion, flushing, mydriasis, delirium, hallucinations, and urinary re- tention. The latter can make confused patients even more distressed and should be actively sought. Rarely, seizures and coma ensue. Differential diagnoses include sympathomimetic or serotinergic toxidromes, common after poisoning with amphetamines and other central nervous system stimulants, and central nervous system infection. Nurse in a quiet calm environment; diazepam may be required for sedation. Physostigmine given by slow intravenous push (adults Table 10.4.4.1  Classification of the main plant toxins Neurotoxins Anticholinergic alkaloids Hallucinogenic toxins Convulsants Nicotinic agonists Cardiotoxins Aconitine Grayanotoxins Cardiac glycosides Taxanes Cytotoxic agents Colchicine Toxalbumins Cyanogenic glycosides Diterpenoid glucosides Epidemic dropsy alkaloids Hepatotoxins Pyrrolizidine alkaloids Senna occidentalis toxins Hypoglycin Nephrotoxins Aristolochic acid Diphyllin glycosides Terpenes Antraquinone glycosides Oxalic acid Gastrointestinal irritants Calcium oxalate Oxalic acid Diterpene esters Dermatotoxins Calcium oxalate Oxalic acid Phototoxic psoralens

10.4.4  Poisonous plants 1829 1–​2 mg, children 0.02–​0.04 mg/​kg) can reverse marked central anticholinergic toxicity that is not settling with time and diazepam. The dose may be repeated as required. Physostigmine should be withheld if cardiotoxic agents have been co-​ingested, if the pa- tient has a bradycardia, or if there are signs of cardiac conduction abnormalities. Unilateral mydriasis may occur in people (often gardeners, hence ‘gardeners’ mydriasis’) who handle plants of this type and happen to rub their eye. This has caused confusion on presentation to hospital and unnecessary, expensive investigations. Hallucinogenic toxins Some plant toxins are particularly popular among abusers because of their hallucinogenic properties. Examples are tetrahydrocannabinols in cannabis Cannabis sativa, alkaloids in khat Catha edulis, mes- caline in peyote Lophophora williamsii, and myristicin in nutmeg Myristica fragrans. Ayahuasca is a hallucinogenic brew made from Banisteriopsis caapi vine and Psychotria viridis leaves in South America. After absorption, the usual first pass breakdown of dimethyltryptamine from P. viridis is inhibited by monoamine oxi- dase inhibitors in B caapi, markedly potentiating the effect of the former. Treatment is symptomatic, with a calm environment and benzodiazepines as necessary. Convulsants The γ-​aminobutyric acid (GABA) antagonists, cicutoxin and oenanthotoxin, are some of the most potent convulsants known. Cicutoxin occurs in cowbane Cicuta virosa, water hemlock C mac- ulata, and western water hemlock C douglasii, while oenanthotoxin occurs in hemlock water dropwort Oenanthe crocata. Severe poisoning has occurred in adults eating one of these plants after mistaking it for an edible plant. Typical symptoms include gastrointestinal upset, increased salivation, diaphoresis, and violent, recurrent, and long-​lasting tonic–​clonic convulsions. These may result in hypoxia, severe metabolic acidosis, coma, cir- culatory instability, rhabdomyolysis, joint dislocations, and rectal prolapse. Diagnosis is typically based on the presence of recurrent seiz- ures together with the history of plant ingestion. Treatment re- quires careful symptomatic and intensive care, with emphasis on combating convulsions with benzodiazepines, barbiturates, and general anaesthesia, correction of acidosis, and maintenance of urinary output. Other toxins reported to cause coma and/​or seizures include coriamyrtin in Coriaria myrtifolia in the Western Mediterranean, terpenes in chinaberry Melia azedarach in South East Asia, the alkaloid dauricine in moonseed Menispermum canadense and podophylloresin in may apple Podophyllum peltatum, both in North America, strychnine in the nux vomica or strychnine tree (Strychnos nux-​vomica) in South and South East Asia, and un- known toxins in Urobotrya siamensis Hiepko in South East Asia and star fruit Averrhoa carambola in patients with chronic kidney disease in East Asia. Treatment is symptomatic and supportive. Nicotinic agonists The tobacco plant Nicotiana tabacum and hemlock Conium macula- tum contain multiple alkaloids with nicotinic receptor agonist effects, particularly nicotine in the former and coniine and γ-​coniceine in the latter. Their unpleasant taste should reduce the risk of poisoning by ingestion; nevertheless, they are sporadically confused with herbs and eaten in salads. Early symptoms are vertigo, agitation, thirst, tachy- cardia, hypertension, salivation, diaphoresis, vomiting, and diarrhoea. Muscle fasciculation, convulsions, hypotension, bradydysrhythmias, ascending weakness, paralysis, and coma may follow. Careful symp- tomatic and supportive care, including assisted ventilation, may be required. Cytisine in Laburnum spp. (e.g. golden chain) and lobeline in Lobelia spp. cause mild nicotine-​like effects. However, the most common symptoms after ingestion are vomiting and diarrhoea. Childhood exposures are frequent and symptoms mostly mild or moderate. Treatment is symptomatic. Other neurotoxins Toxins in fruit of the buckthorn or tullidora bush, Karwinskia hum- boldtiana, in Central America produce a flaccid, symmetric, ascending paralysis of the lower limbs. The dimeric hydroxyanthracenone peroxisomicine A1 appears to be cytotoxic but the precise mechanism of the peripheral neuropathy is not yet known. Buckthorn poisoning can easily be mistaken for Guillain-​Barré syndrome unless there is a history of fruit ingestion. Treatment is supportive until the peripheral neuropathy resolves, with careful monitoring for impending ventila- tory failure. Gelsemine in lemuan Gelsemium elegans in China and yellow jessamine G sempervirens in North America is a glycine agonist. Patients present with dizziness and eye manifestations (blurred vi- sion, diplopia, nystagmus, ptosis), progressing to coma, seizures, and respiratory failure requiring mechanical ventilation. Treatment requires supportive and intensive care. Cardiotoxic plants Aconitine Aconitine is one of the most potent plant toxins known, occurring in multiple Aconitum spp. (e.g. monkshood, A napellus) native to mountainous parts of the northern hemisphere and now grown widely in gardens. The toxin binds to voltage-​gated sodium chan- nels causing persistent sodium influx and depolarization of cardiac and neurological tissue. Serious poisoning results from intentional ingestion of the plant, homicidal administration of aconitine in food, and from unintentional overdose of Asian herbal medica- tions. It is also used in arrow poisons (e.g. Bikh in Nepal). Ingestion results in rapid onset of burning and tingling in the lips, mouth, and pharynx, followed by numbness and paraesthesia of the limbs, hypersalivation, and gastrointestinal symptoms in particular severe and protracted vomiting. Many kinds of dys- rhythmias occur, but particularly ventricular ectopy leading to ventricular tachycardia and fibrillation that may be refractory to treatment. Cardiac failure and shock often develop; coma, mus- cular weakness, neuromuscular failure, and seizures also occur. Considering the extreme toxicity of this plant, gastrointestinal decontamination should be performed. Treatment includes optimal symptomatic and supportive care, directed at dysrhythmias and cardiac failure, including magnesium, flecainide, or lidocaine, and extracorporeal membrane oxygenation.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1830 Grayanotoxins Grayanotoxins occur in multiple Rhodedendron spp. Pollen from these plants are incorporated by bees into ‘mad’ honey particularly around the Black Sea and Himalayas. The toxic dose is reported to be 20–​200 g of honey, with a clear dose response. Similar to acon- itine, the toxin binds to and activates voltage-​gated sodium channels on nerve, muscle, and heart cells. Strong vagal effects cause brady- cardia. Toxicity often lasts for 2–​3  days, with occurrence of car- diac dysrhythmias, dizziness, diplopia, reduced consciousness, and rarely respiratory failure. Treatment is supportive. Grayanotoxins also occur in mountain laurel Kalmia latifolia, Menziesia spp., and Pieris spp. Cardiac glycosides Cardiac glycosides occur in multiple foxglove spp., including Digitalis purpurea and D lanata, common oleander Nerium ole- ander, yellow oleander Cascabela thevetia, sea mango or odollam tree Cerbera manghas, lily of the valley Convallaria majalis, and red squill Urginea maritima. Cardiac glycoside poisoning is particularly common in South Asia where C. thevetia and C. manghas are com- monly used for self-​harm, resulting in hundreds of deaths each year; poisoning with other cardiac glycoside containing plants is more usually unintentional. Ingestion results in severe gastrointestinal features, with ab- dominal pain, profuse vomiting, and diarrhoea common. The toxins’ inhibition of the Na+/​K+ ATPase on cardiomyocytes results in hyperkalaemia, as well as raised intracellular sodium and then raised intracellular calcium concentration. Bradycardia with sinus and AV blocks are common; death occurs in 5–​10% of patients, likely due to catecholaminergic polymorphic ventricular tachy- cardia. Where available, digitalis-​specific antibodies (ovine Fab fragments) are highly effective (Fig. 10.4.4.1). Where not avail- able, management focuses on symptomatic care and treatment of hyperkalaemia. Taxanes Taxin alkaloids in yew Taxus spp. block Na+ and Ca+ channels, inducing QRS prolongation, atrioventricular block, ventricular fib- rillation, and cardiac arrest. They also block Na+ channels and dis- rupt microtubule function, inhibiting cell division, causing central nervous system and gastrointestinal effects. Treatment involves supportive care, with extracorporeal membrane oxygenation if available. Other cardiotoxins Other cardiotoxins include veratrine in Veratrum and Zigadenus spp. and phoratoxin in American mistletoes Phoradendron spp. Cytotoxic plants Colchicine Colchicine occurs in autumn crocus/​meadow saffron Colchicum autumnale and glory lily Gloriosa superba. It binds to β-​tubulin, producing antimitotic effects on cells with high metabolism (e.g. gut and bone marrow), but also direct toxicity on heart, liver, and kidneys. Intentional exposures may result in severe poisoning. The clinical course has different phases. After an initial delay—​ sometimes of many hours—​there is onset of intense gastrointestinal symptoms, followed by dysrhythmias, circulatory failure, seizures, central nervous system depression, and muscular weakness. There may be signs of renal and hepatic damage and, after a few days, bone marrow depression. Patients who survive the acute phase may lose their hair and develop a peripheral neuropathy. Multiple-​dose activated charcoal may enhance elimination, but intensive care is crucial together with measures to encourage bone marrow recovery. Anticolchicine Fab fragments have been studied but are not yet available in clinical practice. Toxalbumins Ricin in the castor plant Ricinus communis and abrin in je- quirity bean Abrus precatorius are water-​soluble proteins known as toxalbumins. Ricin is so toxic that it has been placed on the Chemical Weapons Convention List. These compounds block protein synthesis by inhibiting ribosome function, causing cell death, with the gut as the primary target organ. Intact beans are not toxic as they pass through the gut without releasing toxin. However, beans that have been chewed are highly toxic, with just a few sufficient to cause severe poisoning. A few hours after inges- tion, severe gastroenteritis may occur with heavy fluid and elec- trolyte loss, resulting in renal failure, circulatory instability, and hepatic damage. Treatment is symptomatic with vigorous fluid replacement. Baseline t = 30 min t = 60 min t = 2 h t = 8 h t = 48 h Fig. 10.4.4.1  Resolution of atrioventricular conduction block after treatment with 1200 mg of digoxin-​specific Fab fragments. Reprinted from The Lancet, Vol. 355, No. 9208, Eddleston M et al., Anti-​digoxin Fab fragments in cardiotoxicity induced by ingestion of yellow oleander: a randomised controlled trial, pages 967–​72, Copyright © 2000, with permission from Elsevier.

10.4.4  Poisonous plants 1831 Cyanogenic glycosides Over 25 cyanogenic glycosides are known, occurring in more than 2500 species of plants distributed across the world. They occur in kernels or fruits of Prunus spp. such as bitter almonds, apricots, cherries, and peaches as well as non-​Prunus spp. such as loquat Eriobotrya japonica, cassava (Manihot spp.), elderberry Sambucus nigra, and toxic fruits of the tallow tree Detarium senegalense. Even apple pips Malus spp. contain small amounts of cyanogenic glyco- sides, but large amounts are required to cause poisoning. After the kernels are chewed and swallowed, barriers between the cyanogenic glycoside and enzyme break down, resulting in enzymatic release of cyanide in the stomach. This is a slow process, and symptoms of poisoning may be delayed for many hours. Cyanide poisoning from plants is unusual but, should it occur, treatment is as outlined else- where for cyanide (see Chapter 10.4.1). Inappropriately prepared cassava M. esculenta represents a spe- cial, large-​scale problem of chronic cyanide exposure, producing neurological disorders such as tropical ataxic neuropathy and konzo. It was observed in Nigeria in the 1930s, and subsequently in other African countries. Food insecurity results in atypical ways of preparing cassava, producing small outbreaks from time to time. Diterpenoid glucosides The diterpenoid glucosides, atractyloside and carboxyatractyloside, have been identified in bird-​lime or blue thistle Atractylis gummif- era in north Africa and in the cocklebur thistle Xanthium strumar- ium in Bangladesh (Fig. 10.4.4.2). These plants are usually eaten by children when food is short. These toxins block the adenine nu- cleotide translocator in mitochondria, inhibiting mitochondrial oxidative phosphorylation. Patients develop vomiting, abdom- inal pain, and diarrhoea, then headache, convulsions, coma, car- diovascular collapse, and liver failure. The published case fatality is high, with most deaths occurring within 24 hrs. Treatment is supportive. Epidemic dropsy alkaloids The alkaloids sanguinarine and dihydrosanguinarine in seeds of the Mexican prickly poppy Argemone mexicana cause epidemic dropsy in populations, particularly South Asian, that cook with mustard oil Brassica nigra. Dropsy occurs when B nigra seeds become contaminated with A mexicana seeds, either acciden- tally due to A mexicana growing in or near B nigra cultivation or due to intentional adulteration. The alkaloids damage capillaries, leading to vascular protein loss and oedema. Patients present with diarrhoea, cough, and marked bilateral pitting oedema of the legs; right sided congestive cardiac failure may occur. Treatment is symptomatic. Hepatotoxic plants Pyrrolizidine alkaloids Hepatotoxicity most commonly occurs when people ingest plants containing pyrrolizidine alkaloids. These toxins cause veno-​occlusive disease and occur in many Senecio, Crotalaria, Heliotropium, and Symphytum spp. Cases are reported from Afghanistan, countries of the old USSR, India, and Jamaica; large epidemics have occurred in west Asia when Heliotropium plants have contaminated grain fields producing contamination of flour with pyrrolizidine-​containing seeds. Outbreaks have also occurred when misidentification of plants has resulted in them being incorporated into herbal medi- cines. Treatment is supportive. Senna occidentalis toxins Epidemics of acute hepatomyoencephalopathy have been noted in young children in India. Previously thought to be due to virus in- fection, they are now recognized to be due to ingestion of beans from Senna (prev Cassia) occidentalis. The plant contains a var- iety of toxins (anthraquinones, emodin, glycosides, toxalbumins, alkaloids). Previously healthy children present with vomiting, agi- tation, and abnormal movements that rapidly progress to coma. Blood tests show markedly raised alanine transaminase and cre- atine kinase. In a study of 55 children, 42 (76%) died. Treatment is supportive. Hypoglycin Hypoglycin found in unripe fruit of the ackee Blighia sapida tree causes a reduction in liver fatty acid β oxidation, production of toxic metabolites, and liver steatosis. Patients present with vomiting, sometimes severe hypoglycaemia, coma, and convulsions. Many cases, including localized outbreaks, have been reported from West Africa, Haiti, and Jamaica (where it is termed the Jamaican vomiting disease). Treatment involves correction of hypoglycaemia and intensive care. OH O O O O O O H COOH OH CH2 H O O CH3 CH3 H3C O O S S HO HO (a) (b) Fig. 10.4.4.2  (a) Atractylis gummifera L. (chardon a glu, bird-​lime, or blue thistle). (b) The plant’s toxin, atractyloside, which is a mitochondrial adenine nucleotide translocator inhibitor. (a) Courtesy of Luis Nunes Alberto, licensed under the Creative Commons Attribution-​Share Alike 3.0 Unported license.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1832 Nephrotoxic plants Aristolochic acid Balkan endemic nephropathy is due to dietary exposure to seeds of the European birthwort Aristolochia clematitis, which grows intermingled with the wheat used for bread. During the early 1990s, hundreds of patients in Belgium and the United Kingdom developed renal failure after Aristolochia fangchi was mistakenly incorporated into a herbal weight loss treatment instead of Stephania tetrandra. The mechanism of aristolochic acid induced nephropathy is not known. Treatment is supportive. Diphyllin glycosides Self-​poisoning with leaves of the oduvan Cleistanthus colli- nus tree in south India is responsible for perhaps hundreds of deaths each year. The plant contains the diphyllin glycosides cleistanthins A  and B which cause severe hypokalaemia (via kaliuresis) and metabolic acidosis (due to renal tubular acid- osis), acute respiratory distress syndrome, cardiovascular shock, rhabdomyolysis, and neuromuscular weakness that may require ventilation. Deaths occur from cardiac dysrhythmias; the prog- nosis is worse when water from boiled leaves is drunk. There is no antidote although K+ replacement & acetylcysteine have been tried. Other nephrotoxins Spurge laurel Daphne laureda, mezereon Daphne mezereum, and savin Juniperus sabina contain terpenes that cause intense irritation and blistering in the mouth and gastrointestinal tract, but also renal inflammation with haematuria and proteinuria. Anthraquinone glycosides and oxalic acid in rhubarb Rheum rhabarbarum cause irritation in the mouth and gastrointestinal tract; transient renal impairment and metabolic acidosis may follow ingestion of large amounts of raw leaves or stems. Rumex spp. (docks and sorrels) also contain oxalates; food based on these plants may result in similar toxic effects. Gastrointestinal irritants Popular plants with sap containing calcium oxalate and oxalic acid are elephant’s ear Philodendron spp., cuckoo pint Arum maculatum, and dumb cane Dieffenbachia spp. The needle-​shaped calcium ox- alate crystals damage mucous membranes mechanically. Euphorbia and Daphne spp. contain irritating diterpene esters that cause pain, burning sensations in the mouth and pharynx, salivation, reddening, blistering, and uncommonly, in very large expos- ures, nephritis. Dysphagia, vomiting, and diarrhoea may follow. Treatment is rinsing of the mouth and oral fluids for dilution. Dermatotoxic plants Euphorbia and Dieffenbachia spp. damage skin as described earlier for mucous membranes. Hypersensitivity to plant allergens may also cause impressive skin reactions. Examples are poison ivy Rhus radicans, western poison oak Toxicodendron diversilobum, Primula obconica, and citrus plants and fruit. Treatment involves rinsing with water and symptomatic care. The giant hogweed and other Heracleum spp., rue Ruta graveo- lens, and gas plant Dictamnus albus contain phototoxic psoralens. Contact with sap and subsequent solar radiation can provoke in- tense phototoxic reactions with eczematous skin lesions and large, painful bullae. The best treatment is to rinse the skin directly after exposure and avoid sunlight. When skin damage is already estab- lished, treatment is the same as for chemical burns. FURTHER READING Chan TY (2009). Aconite poisoning. Clinical Toxicology, 47, 279–​85. Chauvin P, Dillon JC, Moren A (1994). Sante, 4, 263–​68. Chrispal A (2012). Cleistanthus collinus poisoning. J Emerg Trauma Shock, 5, 160–​6. Eddleston M, et al. (2000). Anti-​digoxin fab fragments in cardiotoxicity induced by ingestion of yellow oleander: a randomised controlled trial. Lancet, 355, 967–​72. Eddleston M, Persson H (2003). Acute plant poisoning and antitoxin antibodies. J Toxicol Clin Toxicol, 41, 309–​15. Gokmen MR, et al. (2013). The epidemiology, diagnosis, and manage- ment of aristolochic acid nephropathy: a narrative review. Ann Int Med, 158, 469–​77. Gurley ES, et al. (2010). Fatal outbreak from consuming Xanthium strumarium seedlings during time of food scarcity in northeastern Bangladesh. PLoS One, 5, e9756. Islam MN, et al. (2014). Toxic compounds in honey. J Appl Toxicol, 34, 733–​42. Krenzelok EP (2010). Aspects of datura poisoning and treatment. Clinical Toxicology, 48, 104–​10. Meda HA, et al. (1999). Epidemic of fatal encephalopathy in preschool children in Burkino Faso and consumption of unripe ackee (Blighia sapida) fruit. Lancet, 353, 536–​40. Tourdjman M, et al. (2009). Plant poisoning outbreak in the western area of Cambodia, 2005. Transactions of the Royal Society of Tropical Medicine and Hygiene, 103, 949–​51.

10.5 Podoconiosis (nonfilarial elephantiasis) 1833

10.5 Podoconiosis (nonfilarial elephantiasis) 1833

10.5 Podoconiosis (non​filarial elephantiasis) Gail Davey ESSENTIALS Podoconiosis is the lesser-​known of the two major forms of trop- ical lymphoedema. The disease affects genetically susceptible indi- viduals who go barefoot and whose feet and legs are exposed long term to red clay soils. It is found focally across tropical Africa, Central America, and North India where such soils coexist with high altitude, high rainfall, and low-​income populations. Prodromal symptoms include itching and a burning sensation in the foot; early changes include spreading or ‘splaying’ of the forefoot and leakage of col- ourless ‘lymph’ fluid’. Later stages are manifested by lymphoedema, which is usually bilateral and limited to below the knee. Early stage disease is reversible through good foot hygiene and use of socks and shoes. Late-​stage disease, which results in considerable economic and social difficulties, is treated with foot hygiene, bandaging, socks and shoes, elevation, and (if necessary) nodulectomy. The disease’s later stages result in considerable economic and social hardship and, despite treatment, may never fully resolve. Aetiology and pathogenesis Spatial analysis has been used to link environmental factors with occurrence of podoconiosis. These include altitude (between 1000 and 2500 metres above sea level), annual precipitation (>1250  mm), higher vegetation index, and population density. Similar approaches have been used on a smaller scale to investi- gate associations between specific mineral constituents of soil and disease, with phyllosilicate materials (smectite clays and mica), quartz, iron oxide, and zirconium associated with podoconiosis prevalence. Not everyone exposed to these soils develops podoconiosis, but family clustering has long been observed. Recent genetic studies demonstrated high heritability of the trait and segregation analysis suggests the presence of an autosomal codominant major gene con- ferring susceptibility to podoconiosis. A genome-​wide association study demonstrated association between variants in HLA class II loci and podoconiosis, suggesting that the disease may be a T-​cell-​ mediated inflammatory disease. The best evidence to date suggests that podoconiosis is the re- sult of an abnormal inflammatory reaction to mineral particles that penetrate the skin and accumulate in the lower leg. Colloid-​sized particles of elements common in irritant clays (aluminium, silicon, magnesium, and iron) have been demonstrated in the lower limb lymph node macrophages of both patients and non​patients living barefoot on the clays. Electron microscopy of lymph nodes shows evidence of follicular hyperplasia, with an increased number of pro- liferative germinal centres. An exaggerated lymphocyte population (predominantly CD4+ T lymphocytes) is seen in the paracortex sinuses of some of the nodes. Epidemiology Podoconiosis has been reported from highland areas of tropical Africa, Central America, and north-​west India (Fig. 10.5.1). Africa appears to be the most endemic World Health Organization (WHO) region:  high prevalence areas have been confirmed in Ethiopia, Cameroon, Rwanda, Uganda and Kenya, and were previously re- ported in Tanzania, Equatorial Guinea, the islands of Bioko, São Tomé, and Principe, and Cabo Verde. Recent nationwide mapping of podoconiosis in Ethiopia, the most heavily endemic country in the world, suggests that 1.54 million individuals, or 4% of the adult population, live with podoconiosis, and almost half the population (47.8%) live in areas environmentally suitable for the development of podoconiosis. The condition has been reported in the Central American highlands from Mexico, south to Ecuador, but recent work suggests very few new cases in these areas. Podoconiosis has also been reported in north-​west India, Sri Lanka, and Indonesia. Large-​scale population studies in Ethiopia indicate that being fe- male, older, unmarried, washing the feet less frequently than daily, and being semi-​skilled or unemployed are all associated with in- creased risk of podoconiosis. All major community-​based studies have shown onset in the first or second decade and a progressive increase in prevalence up to the sixth decade.

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1834 Clinical, economic, and social burdens Studies have documented that 77.4% to 97% of patients have experi- enced ‘acute attacks’ (acute dermatolymphangioadenitis or ADLA) at least once per year. These are episodes of inflammatory swelling of lymphoedematous legs that may be triggered by bacterial, viral, or fungal superinfection. They are characterized by hot, painful, and reddened swelling and lead to loss of productivity. According to a study in southern Ethiopia in an area with 1.7 mil- lion residents, the annual economic cost of podoconiosis was more than 16 million USD per year, which when extrapolated to the country as a whole, suggests cost of more than 200 million USD per annum for Ethiopia. People with podoconiosis were found to be half as pro- ductive as those without podoconiosis but with the same occupations. The social impact of podoconiosis is also significant. Qualitative studies in southern Ethiopia have shown that widespread miscon- ceptions about the causes, prevention, and treatment of podoconiosis have contributed to it being considered the most stigmatizing health problem in endemic areas. This is manifested in patients being ex- cluded from school, denied participation in local meetings, churches and mosques, and excluded from marrying unaffected individuals. Affected individuals report lower quality of life, higher levels of mental distress, and higher levels of depression than their healthy neighbours. Clinical features The disease process is usually described in three phases: prodromal, early, and advanced. Prodromal symptoms include itching of the skin of the forefoot, a burning sensation in the foot and lower leg, ‘chills’, or generalized joint pains. The affected person may describe exacerbation of symptoms when they try to walk long distances or do hard physical work. Early changes include spreading or ‘splaying’ of the forefoot, swelling of the sole of the foot, leakage of colourless ‘lymph’ fluid from the foot, and changes in the skin so that it looks like moss or velvet (Fig. 10.5.2). The affected person may report limb aches, heaviness of the lower leg and foot, and odour from the lymph leak, which may attract flies. As the lymphoedema progresses, both legs and feet will be af- fected, though one is usually more severely affected than the other. Fig. 10.5.2  Early changes—​oedema, block toes, changes in skin colour (here related to the pattern of the open sandal worn), lymph ooze, and early mossy changes around interdigital clefts. 0.21–0.30 0.31–0.99 1.00–2.51 2.52–4.62 Average Prevalence (%) based on available data Case report Not reporting Reported Presence Fig. 10.5.1  Geographical distribution of prevalence of podoconiosis—dots represent island nations. Source data from Deribe K, et al. (2015). The feasibility of elimination of podoconiosis. Bull WHO, 93, 712–18.

10.5  Podoconiosis (nonfilarial elephantiasis) 1835 In advanced disease, the swelling may be soft and fluid swelling (‘water-​bag’ type, Fig. 10.5.3), hard and fibrotic swelling (‘leathery’ type, Fig. 10.5.4), or a mix of these two. Differential diagnosis Podoconiosis must be distinguished from lymphoedema caused by filariasis (see Table 10.5.1), leprosy, onchocerciasis, rheumatic heart disease, pelvic surgery, and Milroy’s disease. Clinical investigation In the community, diagnosis is usually based on the features given in Table 10.5.1. A  more comprehensive algorithm that excludes lymphatic filariasis through antigen and antibody testing has been developed for mapping studies. Each foot should be assessed for mossy changes, interdigital maceration, wounds, and stage of dis- ease according to the five-​level Tekola staging system. Prevention Evidence suggests that primary prevention should consist of avoiding prolonged contact between the skin and irritant soils through the regular wearing of closed shoes. Daily foot hygiene is also important for prevention, as is covering mud floors of traditional houses. ‘Community conversations’ have been suc- cessfully used in endemic areas to provide information about the causes of podoconiosis and to encourage preventive behavioural changes. Treatment Treatment of podoconiosis lymphoedema is aimed at reducing the frequency of acute attacks, restoration of the barrier function of the skin, and reduction of exposure to irritant soil. A simple package of self-​care measures including foot hygiene, ointment, Fig. 10.5.3  ‘Water-​bag’ type swelling: patient in her early 20s. Fig. 10.5.4  ‘Leathery’ type advanced podoconiosis in patient aged 14. Table 10.5.1  Distinguishing podoconiosis from lymphatic filariasis Characteristic LF Podoconiosis Area of residence <1000 m above sea level

1500 m above sea level Chief cause Mosquito-​borne parasite Genetic susceptibility plus long-​term exposure to irritant highland soil Diagnosis Field ‘ICT’ blood test Exclusion of other cause of leg swelling Site of first symptom Any part of limb except foot Toes and foot Lymph node ‘attacks’ Precede swelling of limb Follow swelling of limb Chief site of swelling Above and below knee Below knee Prevention Mosquito nets Mass Drug Administration (MDA) Protective footwear Floor coverings Foot washing Treatment 2 drugs (albendazole + DEC) Foot hygiene Foot hygiene, bandages, exercises, socks and shoes

SECTION 10  Environmental medicine, occupational medicine, and poisoning 1836 bandages, elevation, exercises, and use of socks and shoes has re- cently been shown to reduce the incidence and duration of acute attacks. Charles’s operation (removal of skin, subcutaneous tissue, and deep fascia to lay the muscles and tendons bare, followed by grafting of healthy skin), is no longer recommended, as long-​term results are disappointing. Nodulectomy may be required if one or two nodules prevent use of footwear, and follow-​up has shown skin healing postnodulectomy to be good. Prognosis Recent follow-​up during a community-​based trial suggests excess mortality associated with podoconiosis. Untreated patients have severely reduced mobility and work capacity by their mid-​40s, chiefly due to episodes of acute dermatolymphangioadenitis. Global awareness Podoconiosis was included in the WHO list of ‘Other Neglected Tropical Conditions’ in February 2011, and is currently addressed under the lymphatic filariasis programme of the Department for Control of Neglected Tropical Diseases (NTDs). Elimination of podoconiosis as a public health problem is achievable be- cause no biological agent or vector involved in podoconiosis has been identified, the global scale of the problem is relatively small, and the strategies for podoconiosis prevention and control are safe. A national network, the National Podoconiosis Action Network, and a global initiative, Footwork, the International Podoconiosis Initiative (https://​podo.org), were launched in 2012. These aim to advocate for and coordinate integration of podoconiosis inter- ventions into NTD control strategies at national and international levels. Together with non​government partners, Footwork has coord- inated training of health workers in podoconiosis management in Cameroon and Uganda and will provide similar training in Rwanda and Burundi. Likely research developments The Global Atlas of Podoconiosis project was launched in 2017 to further understanding of the geographical distribution and spatial epidemiology of the disease. Studies examining host im- munological and inflammatory responses are currently underway. A range of devices to enable better diagnosis and volume meas- urement are being developed and tested, while a trial examining the effectiveness of doxycycline has recently been launched in Cameroon. Implementation research exploring optimal integration of podoconiosis care into existing health systems is also underway. FURTHER READING Deribe K, Cano J, Trueba ML, Newport MJ, Davey G. (2018). Global epidemiology of podoconiosis: a systematic review. PLoS Negl Trop Dis, 12(3), e0006324. Deribe K, et al. (2015). The feasibility of elimination of podoconiosis. Bull WHO, 93, 712–​18. Negussie H, et al. (2018). Lymphoedema management to prevent acute dermatolymphangioadenitis in podoconiosis in northern Ethiopia (GoLBet): a pragmatic randomised controlled trial. The Lancet Global Health, 6(7), e795–​e803. Tekola F, et  al. (2006). Economic costs of podoconiosis (endemic non-​filarial elephantiasis) in Wolaita Zone, Ethiopia. Trop Med Int Health, 11, 1136–​44. Tekola F, et al. (2012). The HLA class II locus confers susceptibility to podoconiosis. N Engl J Med, 388, 1200–​8.

SECTION 11 Nutrition Section editor: Katherine Younger 11.1 Nutrition: Macronutrient metabolism  1839 Keith N. Frayn and Rhys D. Evans 11.2 Vitamins  1855 Tom R. Hill and David A. Bender 11.3 Minerals and trace elements  1871 Katherine Younger 11.4 Severe malnutrition  1880 Alan A. Jackson 11.5 Diseases of affluent societies and the need
for dietary change  1891 J.I. Mann and A.S. Truswell 11.6 Obesity  1903 I. Sadaf Farooqi 11.7 Artificial nutrition support  1914 Jeremy Woodward