8.6.13 Haemophilus influenzae 1066

8.6.13 Haemophilus influenzae 1066

section 8  Infectious diseases 1066 other diarrhoeas) by having a reduced concentration of sodium (75 mmol/​litre). This hypo-​osmolar solution is also acceptable for cholera. ORS are easily prepared by adding the following simple ingredients to 1 litre of water: 2.6 g sodium chloride, 2.9 g trisodium citrate, 1.5 g potassium chloride, and 13.5 g glucose (or 50 g boiled and cooled rice powder). Adults and children are encouraged to eat, and breastfeeding can continue as there is no scientific basis for resting the gut. Antibiotics can shorten the illness and decrease diarrhoeal purging. One-​ to 3-​day courses of tetracycline, co-​trimoxazole, or ciprofloxacin have been effective but there is increasing resistance. Azithromycin has been used more recently, but growing macrolide resistance may limit its use as well. Antibiotic sensitivity testing is, therefore, recommended during outbreaks. Antibiotics are not in- dicated for asymptomatic contacts. Prophylactic use of antibiotics increases the risk of the development of resistance and it is not indi- cated to prevent cholera. Prognosis Case fatality should be 1% or less if adequate oral rehydration therapy is used early in the illness, even at the community level. Adequate fluid and electrolyte replacement reverses or prevents complications such as acute renal failure or hypoglycaemia even in moderate or severe cholera. Cholera may well persist in its brackish marine res- ervoir, but improved water and sanitation and increasingly available vaccines promise to control this dreaded disease. Other issues (health economics, areas of uncertainty or controversy, and likely developments ahead) Areas of uncertainty or controversy include the mechanisms and importance of natural reservoirs of cultivable and even non-​ culturable vibrios and marine organisms from plankton to shellfish in the ecology of cholera. Despite the remarkable advances in under- standing the pharmacological mechanisms of cholera toxin action, reliable, effective, and inexpensive means of blocking the effects of the toxin remain elusive. With molecular genetic understanding of virulence and pro- tective immunity, likely developments in the near future include the promise of new and better vaccines, toxin-​blocking or absorption-​ enhancing drugs, and continued improvements in oral rehydration therapy, perhaps with nutrients, micronutrients, or probiotics that compete with vibrio colonization or deliver proabsorptive drugs or nutrients. FURTHER READING Ali M, et al. (2015). Updated global burden of cholera in endemic countries. PLoS Negl Trop Dis, 9, e0003832. Arifuzzaman M, et al. (2011). Individuals with Le(a+b–​) blood group have increased susceptibility to symptomatic Vibrio cholerae O1 in- fection. PLoS Negl Trop Dis, 5, e1413. Chin CS, et  al. (2011). The origin of the Haitian cholera outbreak strain. N Engl J Med, 364, 33–​42. Ivers L (2016). Eliminating cholera transmission in Haiti. N Eng J Med, 376, 101–​3. Kim EJ, et  al. (2015). Whole-​genome sequence comparisons
reveal the evolution of Vibrio cholerae O1. Trends Microbiol, 23, 479–​89. Mutreja A, et al. (2011). Evidence for several waves of global trans- mission in the seventh cholera pandemic. Nature, 477, 462–​5. Qadri F, et al. (2015). Feasibility and effectiveness of oral cholera vac- cine in an urban endemic setting in Bangladesh: a cluster random- ised open-​label trial. Lancet, 386, 1362–​71. Qadri F, et al. (2016). Efficacy of a single-​dose, inactivated oral cholera vaccine in Bangladesh. N Eng J Med, 374, 1723–​32. Sack DA, et al. (2004). Cholera. Lancet, 363, 223–​33. Saha DS, et al. (2006). Single-​dose azithromycin for the treatment of cholera in adults. N Engl J Med, 354, 2452–​62. Son MS, et  al. (2011). Characterization of Vibrio cholerae O1 El Tor biotype variant clinical isolates from Bangladesh and Haiti, including a molecular genetic analysis of virulence genes. J Clin Microbiol, 49, 3739–​49. Sur D, et al. (2009). Efficacy and safety of a modified killed-​whole-​ cell oral cholera vaccine in India: an interim analysis of a cluster-​ randomized, double-​blind, placebo-​controlled trial. Lancet, 374, 1694–​702. Sur D, et al. (2011). Efficacy of a low-​cost, inactivated whole-​cell oral cholera vaccine: results from 3 years of follow-​up of a randomized, controlled trial. PLoS Negl Trop Dis, 5, e1289. World Health Organization (2015). Cholera 2015. Weekly Epidemiol Rec, 38, 433–​40. 8.6.13  Haemophilus influenzae Esther Robinson ESSENTIALS Haemophilus influenzae is a Gram-​negative bacillus that is an ex- clusively human pathogen and commensal. There are six capsular serotypes (a–​f), of which type b (Hib) is a major cause of childhood infectious disease. Transmission occurs by close bodily contact, the main source being other children. Carriage of the organism may be followed by disease in susceptible individuals. Clinical features In infants, Hib causes symptoms ranging from a mild non​specific fe- brile illness (occult bacteraemia) to full blown sepsis with meningitis, epiglottitis, pneumonia, septic arthritis, or cellulitis. Non​typeable H. influenzae are common nasopharyngeal commensals and cause otitis media and conjunctivitis in children. In adults, non​typeable Acknowledgement: The authors and editors gratefully acknowledge the inclu- sion in this chapter of material contributed to previous editions of the Oxford Textbook of Medicine by Professor Derrick W. Crook.

1067 8.6.13  Haemophilus influenzae H. influenzae cause exacerbations of chronic bronchitis, sinusitis, and pneumonia. Other Haemophilus species, including H. parainfluenzae, are common commensals and rare causes of infective endocarditis and other infection syndromes. Diagnosis and treatment Gram staining of cerebrospinal, synovial, or pleural fluid is a key in- vestigation, but definitive diagnosis requires culture or detection of H. influenzae DNA by polymerase chain reaction methods. Treatment requires good supportive care, appropriate antibiotics and, for men- ingitis, adjunctive corticosteroids. Antibiotic resistance leads to the agent of choice for invasive Hib dis- ease being a third-​generation cephalosporin with good cerebrospinal fluid penetration (e.g. ceftriaxone or cefotaxime). Chloramphenicol with or without ampicillin remains effective in some developing countries. Antibiotic treatment of otitis media, sinusitis, and chronic bron- chitis associated with non​typeable H. influenzae is widely practised, but largely unsupported by evidence. Corticosteroids reduce mortality, severe hearing loss, and neuro- logical sequelae of Hib meningitis, except in children in low-​income countries. Prevention Conjugate Hib vaccines are given as part of the routine infant im- munization schedule and have virtually eliminated invasive Hib disease from North America, Europe, and some other countries. Vaccines for non​typeable H. influenzae are in development. Introduction Haemophilus influenzae is a human-​adapted pathogen and com- mensal with no other reservoir. It is found in the nasopharynx and also the genital and intestinal tracts. It was first isolated in 1890 by Pfeiffer, who mistakenly thought it was the cause of a concurrent influenza pandemic. Description of the organism The genus Haemophilus includes H. influenzae, H. ducreyi, which causes chancroid, the sexually transmitted infection (Chapter 8.6.14), and other human-​adapted species that are commensals: H. parain- fluenzae and H. haemolyticus. Other human commensals, and occasional pathogens, that were previously included in this genus, namely H. aphrophilus, H. para- phrophilus and H. segnis have been reclassified as Aggregatibacter spp. H.  influenzae is a small (0.2–​0.3 × 0.5–​0.8 μm), Gram-​negative, non​motile coccobacillus that grows well on rich media (e.g. choc- olate agar) incubated in 5% CO2. It is fastidious and requires the growth supplements X factor (haemin) and V factor (NAD). H. parainfluenzae requires only V factor. H. influenzae produces 2-​ to 3-​mm-​diameter grey translucent colonies after 18–​25 h incuba- tion. Precise speciation requires bacterial genome sequence analysis (e.g. 16s ribosomal DNA). A minority of H. influenzae strains have polysaccharide capsules and can be serologically classified in six serotypes (a to f). These strains are relatively non​diverse, consisting of few lineages, suggesting that the genes encoding the capsule were acquired relatively recently. Most strains are non​encapsulated and are non​typeable (NTHi) by serological methods. These strains are genetically heterogeneous. Pathogenicity H.  influenzae expresses several cell surface features essential for colonization of the nasopharynx. These are virulence factors, of which the capsule is the most important. Of the six antigenically distinct structures (types a–​f), type b accounted for virtually all in- vasive H. influenzae disease in children prior to Hib vaccination. The serotype b capsule allows the organism to resist phagocytosis by interfering with binding of serum complement. The capsule also resists desiccation, perhaps promoting host-​to-​host transmission. Serum antibody directed against serotype b capsular polysaccharide is protective. This observation stimulated the development of the highly successful H. influenzae type b (Hib) vaccine, now used rou- tinely in national childhood immunization programmes. Modern vaccines contain capsular polysaccharide covalently conjugated to a protein carrier, such as tetanus toxoid. Other cell surface structures involved in pathogenesis, particu- larly in NTHi, include lipopolysaccharide, pili, and other adhesion proteins. Epidemiology Haemophilus influenzae type b Haemophilus influenzae type b (Hib) is a major cause of childhood infectious disease. Acquisition occurs by close bodily contact, usu- ally from other children, and is usually followed by asymptomatic carriage. The organism dwells harmlessly for months in the naso- pharynx. In a few susceptible individuals, acquisition immediately precedes invasive disease. Carriage rates increase from birth until 4  years and are higher in developing countries, especially where there is crowding, day care attendance, and contact with siblings. The incidence of Hib infections varies with age. Neonates are pro- tected, after which disease peaks by 9 months of age and declines to very low levels by 4 years. Age-​specific disease incidence is inversely related to serum antibodies to Hib. Risk factors for H. influenzae in- fection include complement deficiency, hypogammaglobulinaemia, hyposplenism, sickle cell anaemia, malignancy, and HIV infection. Ethnicity is also important, with higher rates of invasive disease ob- served in Native American and Aboriginal populations. Co-infection with respiratory viruses is also an important risk factor for disease. The main diseases caused by Hib are meningitis, bacteraemia, pneumonia, epiglottitis, and arthritis. Before the introduction of im- munization, Hib was the most important cause of childhood men- ingitis in the United States of America, accounting for 80% of cases and in the United Kingdom, accounting for approximately 50% of cases (Fig. 8.6.13.1). In contrast, it was a much less prominent cause of meningitis than Neisseria meningitidis in the ‘meningitis belt’ of Africa. In Western countries, the case fatality of Hib meningitis was about 5% and long-​term morbidity (deafness and neurological and learning deficits) occurred in at least 10% of cases. Hib immunization virtually eliminates carriage and pro- duces a marked herd effect, protecting against disease. Since the

section 8  Infectious diseases 1068 implementation of Hib conjugate vaccination, the disease has al- most disappeared from North America and Europe. There has been a similar dramatic decline in the majority countries with good vac- cine coverage, including the Gambia and Uganda. A striking but temporary re-​emergence of Hib disease in the United Kingdom was attributed to the introduction, in 2000, of a combined Hib-​ acellular pertussis vaccine that induced lower Hib antibody levels (Fig. 8.6.13.2). In 2014, the World Health Organization (WHO) estimated that vaccination against Hib protected 56% of children in 100 countries. However, there were an estimated 199 000 deaths in children under age 5 from Hib that year (down from 363 000 in 2000), of which nearly 80% were attributed to pneumonia. In some parts of Asia, particularly China, Hib is far less prevalent and so the health benefits of mass vaccination may be insufficient to justify a national vaccination programme. Non​typeable H. influenzae (NTHi) Non​typeable H.  influenzae is acquired soon after birth:  20% of children are colonized in the first year of life and over 50% by 5 years. Colonization persists throughout adulthood. NTHi causes mainly non​invasive infections in children and older adults. It is an important cause of otitis media, sinusitis, bronchitis, and post-​ traumatic meningitis. Since the introduction of the conjugate pneumococcal vaccine, NTHi has become the most common cause of otitis media. 0 0.1 0.15 0.4 0.8 1.5 2.5 7 15 25 40 60 0.55 10 20 30 40 50 60 70 Age in years Cases/100 000/year H. influenzae N. meningitidis S. pneumoniae Fig. 8.6.13.1  Incidence of meningitis in the United States of America caused by H. influenzae, N. meningitidis and S. pneumoniae before the introduction of the Hib conjugate vaccine. Data derived from various sources. Hib vaccine catch-up Hib combined vaccine Hib vaccine introduced 1000 500 0 1990 1995 2000 2005 2010 2015 Year Cases Cases of invasive Haemophilus infection by year and strain, England Hib Not typed NTHi Other capsular serotypes Total Fig. 8.6.13.2  Cases of invasive Haemophilus infection by year and strain in England. Data from Public Health England/​Health Protection Agency.

1069 8.6.13  Haemophilus influenzae NTHi can cause community-​acquired pneumonia in adults, which can be severe. Bacteraemia is increasingly recognized in neo- nates and elderly populations. With the decline in Hib disease fol- lowing introduction of vaccination programmes, NTHi now cause most cases of invasive H. influenzae disease in countries with high vaccination coverage (Fig. 8.6.13.2). Antibiotic resistance H.  influenzae resistant to antibiotics was first reported in the early 1970s. Since then, the prevalence of ampicillin-​resistant β-​ lactamase-​producing strains has risen rapidly in most parts of the world and strains resistant to tetracycline, chloramphenicol, and tri- methoprim, or multiresistant to these antibiotics have emerged. In recent years, ampicillin resistance rates have remained reasonably constant in Europe at 10 to 20%. β-​Lactamase-​negative ampicillin-​resistant H.  influenzae is an emerging pathogen and high rates have been reported in certain countries, for example, Japan (c.30–​50%) and Spain (c.56%). Clonal outbreaks have been reported in Scandinavia and Korea, although prevalence elsewhere remains low. Possible explanations for this in- clude low vaccine coverage, underdosing of oral ampicillin, or fre- quent use of cephalosporins. Clinical features Hib invasive disease (See Table 8.6.13.1.) Bacteraemia and meningitis In infants, clinical features vary from a mild non​specific febrile illness, reflecting occult bacteraemia, to full blown sepsis with men- ingitis. Severe cases show typical features of meningitis including al- tered mental status, stiff neck, and sepsis. There may be disseminated intravascular coagulation with purpuric rash and septicaemic shock, reminiscent of meningococcaemia. Diagnosis is by blood culture and cerebrospinal fluid (CSF) microscopy and culture. Lumbar puncture should not be performed if there is a suspicion of raised intracranial pressure, but antibiotic treatment must not be delayed. Epiglottitis Epiglottitis is an acute, life-​threatening, medical emergency. Symptoms are of systemic sepsis (fever, tachycardia, and tachypnoea), with rapidly evolving local signs and symptoms. Sore throat, drooling, dysphagia, hoarseness, barking cough, and stridor are characteristic. Attempts to examine the throat may precipitate acute airway obstruction; if visu- alized, the epiglottis is inflamed and swollen, looking like a red cherry. Pneumonia Pneumonia is common in Hib disease. The main features are fever with signs of respiratory distress including tachypnoea, nasal flaring, and intercostal recession. Chest examination and radiography are diagnostic of pneumonia. Septic arthritis The child with septic arthritis shows features of sepsis, is unwilling to use the affected limb, and resists movement of the painful joint. Examination and culture of joint fluid are diagnostic. Cellulitis Cellulitis is rare in children, but classically peri-​orbital. Cellulitis of the neck is seen occasionally in adults. These presentations have been increasingly associated with H. influenzae type f infection, es- pecially since the introduction of Hib conjugate vaccine. Non​typeable H. influenzae NTHi causes several clinical syndromes (Tables 8.6.13.1 and 8.6.13.2). Acute otitis media is common in childhood: up to 75% of children suffer at least one episode before the age of five. It usually presents with irritability, with or without fever. The tympanic membrane is inflamed and may perforate, discharging pus. Although not used routinely, tympanocentesis is the most reliable means of aetiological diagnosis. NTHi is a common cause of conjunctivitis, especially in chil- dren. Historically, H. influenzae biotype aegyptius has also been associated with conjunctivitis. Brazilian purpuric fever, a ful- minant septicaemic illness with high case fatality, is also caused by H. aegyptius. NTHi sinusitis presents with local pain, a sense of pressure in the head, local facial oedema, and visible pus draining from the ostia of the sinuses. Diagnosis is by skull radiography, or computed tomog- raphy (CT) scan. Sinus aspiration provides an aetiological diagnosis. In adults, exacerbations of chronic bronchitis are commonly as- sociated with NTHi. H. influenzae and S. pneumoniae are cultured from sputum of up to 50% of cases although their precise aetio- logical role is uncertain. Table 8.6.13.1  Clinical manifestations of Haemophilus influenzae disease: percentage of cases in which syndrome manifests Disease Haemophilus influenzae type b (Hib) Non​typeable Haemophilus influenzae (NTHi) Meningitis 52% 34% Pneumonia 12% 21% Epiglottitis 10% Not described Isolated septicaemia 8% 28% Cellulitis 5% Not described Other 13% 17% Data derived from various sources. Table 8.6.13.2  Clinical syndromes caused by non​typeable Haemophilus influenzae Syndrome Percentage of overall disease burden Otitis media

50% of childhood cases Exacerbations of chronic bronchitis 50% of patients positive for Haemophilus during exacerbation Bacterial conjunctivitis 25% of adult and 50% of childhood cases Sinusitis 40% of childhood cases Pneumonia Up to 20% of childhood cases Neonatal sepsis 5–​7% of neonatal sepsis Data derived from various sources.

section 8  Infectious diseases 1070 NTHi can cause severe invasive disease such as neonatal sepsis, resembling group B streptococcal neonatal sepsis, and pneumonia in adults, particularly older people. It has also been implicated in meningitis associated with skull fracture. Laboratory diagnosis Culture or detection of specific DNA is essential for aetiological diagnosis. Direct examination of cerebrospinal, pleural, or synovial fluids by Gram’s stain may reveal organisms with the morphological features of H. influenzae. Blood culture using most commercial systems yields excellent growth in both anaerobic and aerobic bottles. However, growth on agar and differentiation of Haemophilus spp. requires special con- ditions (see above). Antibiotic susceptibility testing using antibiotic discs requires supplemented media to support the growth of Haemophilus spp. This may be inaccurate and should be supplemented by measure- ment of β-​lactamase activity. Chloramphenicol disc susceptibility is frequently inaccurate and should be supplemented by an assay for chloramphenicol acetyltransferase activity. Capsular type b antigen can be rapidly detected in CSF, sterile site fluid, or urine. Polymerase chain reaction (PCR) of CSF was devel- oped for diagnosing H. influenzae meningitis. Capsular typing of H. influenzae can be achieved serologically but a PCR-​based method is more reliable. Treatment Antibiotics A third-​generation cephalosporin with good CSF penetration is the first-​line antibiotic treatment for invasive Hib disease. High-​ dose ceftriaxone or cefotaxime are effective for treating H. influen- zae meningitis and septicaemia, but cefuroxime must not be used. Alternatives include chloramphenicol alone (depending on the prevalence of chloramphenicol resistance) or in combination with ampicillin. β-​Lactamase-​negative ampicillin-​resistant H.  influen- zae with reduced cephalosporin susceptibility have been described; carbapenem resistance remains rare. Antibiotics are commonly prescribed for otitis media, sinus- itis, and chronic bronchitis, but large meta-​analyses have failed to demonstrate convincing efficacy; some subgroups may benefit. Oral amoxicillin is first-​line. Amoxicillin/​clavulanate, trimetho- prim, tetracycline (adults only), and quinolones (adults only) can also be used. Local sensitivity patterns should be used to guide empiric therapy. Corticosteroid treatment in meningitis Corticosteroids significantly reduce mortality, severe hearing loss, and neurological sequelae in Hib meningitis. In community-​ acquired bacterial meningitis, corticosteroid therapy should be started with the first antibiotic dose for adults. In children, data support the use of adjunctive corticosteroids in children only in high-​income countries. Prevention and control Polysaccharide conjugate vaccines are the best preventive measure for controlling Hib disease. Highly effective vaccines contain cap- sular antigen polyribosyl-​ribitol-​phosphate (PRP) conjugated to tetanus toxoid (PRP-​T), outer membrane protein (PRP-​OMP), or mutant diphtheria toxoid (PRP-​CRM, HbOC). Three doses are given at intervals between the ages of 2 and 6 months. In many countries, a booster dose is given at age 1–​2 years. Carriage rates are decreased to very low levels, with invasive disease being virtually eliminated. Rifampicin can be used to eradicate carriage and may prevent sec- ondary cases among close contacts. This is appropriate only where they have not received Hib vaccine. Significant efforts are now being focused on producing a NTHi vaccine; however, there are major challenges still to be overcome. Other nasopharyngeal Haemophilus spp. H. parainfluenzae H. parainfluenzae is a well-​adapted commensal that colonizes vir- tually everyone soon after birth but is rarely associated with disease. It has been isolated in cases of infective endocarditis, neurosurgical meningitis, prosthetic device infection, and brain and liver ab- scesses. It is treated in the same way as H. influenzae. H. haemolyticus A common commensal, H. haemolyticus has also been occasionally associated with bacteraemia and joint infection. FURTHER READING Barbour ML, et al. (1995). The impact of conjugate vaccine on carriage of Haemophilus influenzae type b. J Infect Dis, 171, 93–​8. Kim KS (2010). Acute bacterial meningitis in infants and children. Lancet Infect Dis, 10, 32–​42. Jalalvand F, Riesbeck K (2018). Update on non-typeable Haemophilus influenzae-mediated disease and vaccine development. Expert Rev Vaccines, 18, 1–10. Morris SK, Moss WJ, Halsey N (2008). Haemophilus influenzae type b conjugate vaccine use and effectiveness. Lancet Infect Dis, 8, 435–​43. Prasad K, Karlupia N, Kumar A (2009). Treatment of bacterial menin- gitis: an overview of Cochrane systematic reviews. Respir Med, 103, 945–​50. Ulanova M, Tsang RS (2009). Invasive Haemophilus influenzae dis- ease: changing epidemiology and host-​parasite interactions in the 21st century. Infect Genet Evol, 9, 94–​605. Van Eldere J, et al. (2014). Non-​typeable Haemophilus influenzae, an under-​recognised pathogen. Lancet Infect Dis, 14, 1281–​92. Watt JP, et al. (2009). Burden of disease caused by Haemophilus influen- zae type b in children younger than 5 years: global estimates. Lancet, 374, 903–​11. Zwahlen A, et  al. (1989). The molecular basis of pathogenicity in Haemophilus influenzae:  comparative virulence of genetically-​re- lated capsular transformants and correlation with changes at the capsulation locus cap. Microb Pathog, 7, 225–​35.