# 16 - 308 Sleep Apnea

### 308 Sleep Apnea

■
■DIAGNOSIS
Respiratory symptoms associated with acute hyperventilation can be 
the initial manifestation of systemic illnesses such as diabetic ketoaci­
dosis. Causes of acute hyperventilation need to be excluded before a 
diagnosis of chronic hyperventilation is considered. Arterial blood gas 
sampling that demonstrates a compensated respiratory alkalosis with a 
near normal pH, low Paco2, and low calculated bicarbonate is neces­
sary to confirm chronic hyperventilation. Other causes of respiratory 
alkalosis, such as mild asthma, need to be diagnosed and treated before 
chronic hyperventilation can be considered. A high index of suspicion 
is required as increased minute ventilation can be difficult to detect on 
physical examination. Once chronic hyperventilation is established, a 
sustained 10% increase in alveolar ventilation is sufficient to perpetu­
ate hypocapnia. This increase can be accomplished with subtle changes 
in the respiratory pattern, such as occasional sigh breaths or yawning 
2–3 times per min.
■
■TREATMENT
There are few well-controlled treatment studies of chronic hyper­
ventilation owing to its diverse features and the lack of a universally 
accepted diagnostic process. Clinicians often spend considerable time 
identifying initiating factors, excluding alternative diagnoses, and 
discussing the patient’s concerns and fears. In some patients, reassur­
ance and frank discussion about hyperventilation can be liberating. 
Identifying and eliminating habits that perpetuate hypocapnia, such 
as frequent yawning or sigh breathing, can be helpful. Some evidence 
suggests that breathing exercises and diaphragmatic retraining may be 
beneficial for some patients. The evidence for using medications to 
treat hyperventilation is scant. Beta blockers may be helpful in patients 
with sympathetically mediated symptoms such as palpitations and 
tremors.
Acknowledgment
John F. McConville, Julian Solway, and Babak Mokhlesi contributed to 
this chapter in the 21st edition and some material from that chapter has 
been retained here.
■
■FURTHER READING
Anderson PM et al: EFNS guidelines on the clinical management of 
amyotrophic lateral sclerosis (MALS)–revised report of the EFNS 
task force. Eur J Neurol 19:360, 2012.
Benditt JO: Pathophysiology of neuromuscular respiratory diseases. 
Clin Chest Med 39:297, 2018.
Chung F et al: STOP-Bang questionnaire: A practical approach to 
screen for obstructive sleep apnea. Chest 149:631, 2016.
Douglas IS: Acute-on-chronic respiratory failure, in Principles of 
Critical Care, 4th ed. Hall JB et al (eds). New York, McGraw-Hill, 
2015, pp 482–495.
Gardner WN: The pathophysiology of hyperventilation disorders. 
Chest 109:516, 1996.
Macrea M et al: Long-term noninvasive ventilation in chronic stable 
hypercapnic chronic obstructive pulmonary disease. An official 
American Thoracic Society clinical practice guideline. Am J Respir 
Crit Care Med 202:e74, 2020.
Masa JF et al: Long-term clinical effectiveness of continuous posi­
tive airway pressure therapy versus non-invasive ventilation therapy 
in patients with obesity hypoventilation syndrome: A multicentre, 
open-label, randomised controlled trial. Lancet 393:1721, 2019.
Masa JF et al: Obesity hypoventilation syndrome. Eur Respir Rev 
28:180097, 2019.
Mokhlesi B et al: Evaluation and management of obesity hypoventila­
tion syndrome. An official American Thoracic Society clinical prac­
tice guideline. Am J Respir Crit Care Med 200:e6, 2019.
Rimmer KP et al: Home mechanical ventilation for patients with 
amyotrophic lateral sclerosis: A Canadian Thoracic Society clinical 
practice guideline. Canadian J Respir Crit Care Sleep Med 3:9, 2019.

Andrew Wellman, Daniel J. Gottlieb, 

Susan Redline

Sleep Apnea
Obstructive sleep apnea (OSA) and central sleep apnea (CSA) are both 
classified as sleep-related breathing disorders. OSA and CSA share 
some risk factors and physiologic bases but also have unique features. 
Each disorder is associated with impaired ventilation during sleep 
and disruption of sleep, and each diagnosis requires careful elicitation 
of the patient’s history, physical examination, and physiologic test­
ing. OSA, the more common disorder, causes daytime sleepiness and 
impaired daily function. It is a cause of hypertension and is strongly 
associated with cardiovascular disease in adults and behavioral prob­
lems in children. CSA is less common and may occur alone or in com­
bination with OSA. It can occur as a primary condition, as a response 
to high altitude, or secondary to a medical condition (such as heart 
failure) or medication (such as opioids). Patients with CSA often report 
frequent awakenings and daytime fatigue and are at increased risk for 
heart failure and atrial fibrillation.
Sleep Apnea
CHAPTER 308
■
■OBSTRUCTIVE SLEEP APNEA/HYPOPNEA 
SYNDROME
Definition 
OSA is defined on the basis of nocturnal and daytime 
symptoms as well as sleep study findings. Diagnosis requires the 
patient to have (1) either symptoms of nocturnal breathing distur­
bances (snoring, snorting, gasping, or breathing pauses during sleep) 
that impair quality of life or daytime sleepiness or fatigue that occurs 
despite sufficient opportunity to sleep and is unexplained by other 
medical problems; and (2) five or more episodes of obstructive apnea 
or hypopnea per hour of sleep (the apnea-hypopnea index [AHI], 
calculated as the number of episodes divided by the number of hours of 
sleep) documented during a sleep study. OSA also may be diagnosed in 
the absence of symptoms if the AHI is ≥15 episodes/h. Each episode of 
apnea or hypopnea represents a reduction in breathing for at least 10 s 
and commonly results in a ≥3% drop in oxygen saturation or a brain 
cortical arousal. OSA severity can be characterized by the frequency of 
breathing disturbances (AHI), the amount of oxyhemoglobin desatu­
ration with respiratory events, the duration of apneas and hypopneas, 
the degree of sleep fragmentation, and the level of reported daytime 
sleepiness or functional impairment.
Pathophysiology 
During inspiration, intraluminal pharyngeal 
pressure becomes increasingly negative, creating a “suctioning” force. 
Because the pharyngeal airway has no fixed bone or cartilage, airway 
patency is dependent on the stabilizing influence of the pharyngeal 
dilator muscles. Although these muscles are continuously activated 
during wakefulness, neuromuscular output declines with sleep onset. 
In patients with a collapsible airway, the reduction in neuromuscular 
output results in transient episodes of pharyngeal collapse (manifest­
ing as an “apnea”) or near collapse (manifesting as a “hypopnea”). The 
episodes of collapse are typically terminated when ventilatory reflexes 
are activated and cause arousal, thus stimulating an increase in neuro­
muscular activity and opening of the airway. The airway may collapse 
at different sites, such as the soft palate (most common), tongue base, 
lateral pharyngeal walls, and/or epiglottis (Fig. 308-1). OSA may be 
most severe during rapid eye movement (REM) sleep, when neuro­
muscular output to the skeletal muscles is particularly low, and in the 
supine position due to gravitational forces.
Individuals with a small pharyngeal lumen require relatively high 
levels of neuromuscular activation to maintain patency during wake­
fulness and thus are predisposed to airway collapse following the 
normal sleep-related reduction in pharyngeal muscle activity during 
sleep. The airway lumen may be narrowed by enlargement of soft tissue 
structures (tongue, palate, and uvula) due to fat deposition, increased 
lymphoid tissue, or genetic variation. Craniofacial factors such as

Palate
Lateral
pharyngeal
walls
PART 7
Disorders of the Respiratory System
Tongue
Epiglottis
FIGURE 308-1  The structures causing airway collapse in obstructive sleep apnea 
include the palate, the tongue, and/or the epiglottis. In addition, collapse can also 
occur due to the lateral pharyngeal walls.
mandibular retroposition or micrognathia, reflecting genetic variation 
or developmental influences, also can reduce lumen dimensions. In 
addition, lung volumes influence the caudal traction on the pharynx 
and consequently the stiffness of the pharyngeal wall. Accordingly, 
low lung volume in the recumbent position, which is particularly 
pronounced in the obese, contributes to collapse (less caudal traction). 
A high degree of nasal resistance (e.g., due to nasal septal deviation 
or polyps) can contribute to airway collapse by reducing intralumi­
nal pressure downstream in the pharynx. High-level nasal resistance 
also may trigger mouth opening during sleep, which breaks the seal 
between the tongue and the palate and allows the tongue to fall poste­
riorly and occlude the airway.
Pharyngeal muscle activation is integrally linked to ventilatory 
drive. Thus, factors related to ventilatory control, particularly ventila­
tory sensitivity, arousal threshold, and neuromuscular responses to car­
bon dioxide (CO2), contribute to the pathogenesis of OSA. A buildup 
in CO2 during sleep activates both the diaphragm and the pharyngeal 
muscles. Pharyngeal activation stiffens the upper airway and can 
counteract inspiratory suction pressure and maintain airway patency 
to an extent that depends on the anatomic predisposition to collapse. 
However, pharyngeal collapse can occur when the ventilatory control 
system is overly sensitive to CO2, with resultant wide fluctuations in 
ventilation, ventilatory drive, and upper airway stiffness. Moreover, 
increasing levels of CO2 during sleep result in central nervous system 
arousal, causing the individual to move from a deeper to a lighter level 
of sleep or to awaken. A low arousal threshold (i.e., awakening to a 
low level of CO2 or ventilatory drive) can preempt the CO2-mediated 
process of pharyngeal muscle compensation and prevent airway stabi­
lization. A high arousal threshold, conversely, may prevent appropriate 
termination of apneas, prolonging apnea duration and exacerbating 
oxyhemoglobin desaturation. Finally, any impairment in the ability of 
the muscles to compensate during sleep can contribute to collapse of 
the pharynx. The relative contributions of risk factors vary by age, sex, 
body mass index, and other factors. Approaches to the measurement of 
these factors in clinical settings, with consequent enhancement of “per­
sonalized” therapeutic interventions, are being actively investigated.
Risk Factors and Prevalence 
The major risk factors for OSA 
are obesity, male sex, and older age. Additional risk factors include 
mandibular retrognathia and micrognathia, a positive family history of 
OSA, sedentary lifestyle, genetic syndromes that reduce upper airway 

patency (e.g., Down syndrome, Treacher-Collins syndrome), adeno­
tonsillar hypertrophy (especially in children), menopause (in women), 
and various endocrine syndromes (e.g., acromegaly, hypothyroidism).
Approximately 40–60% of cases of OSA are attributable to excess 
weight. Obesity predisposes to OSA through the narrowing effects of 
upper airway fat on the pharyngeal lumen. Obesity also reduces chest 
wall compliance and decreases lung volumes, resulting in a loss of 
caudal traction on upper airway structures. Obese individuals are at 
a fourfold or greater risk for OSA than their normal-weight counter­
parts, although the association of OSA with obesity is weaker in the 
elderly than in younger adults. A 10% weight gain is associated with 
a >30% increase in AHI. Even modest weight loss or weight gain can 
influence the risk and severity of OSA. However, the absence of obesity 
does not exclude this diagnosis.
The prevalence of OSA is twofold higher among men than among 
women. Factors that predispose men to OSA include android pattern 
of obesity (resulting in upper-airway and abdominal fat deposition) 
and relatively greater pharyngeal length, which increases collapsibil­
ity. Premenopausal women are relatively protected from OSA by the 
influence of sex hormones on ventilatory drive. The decline in sex 
difference in older age reflects an increased OSA prevalence in women 
after menopause. The pathogenesis and presentation of OSA also differ 
in men and women: compared to men, women have a lower arousal 
threshold and less neuromuscular collapsibility. Women tend to have 
shorter duration of apneas and apneas that occur predominantly in 
REM sleep. Failure to recognize these differences can contribute to 
underrecognition of OSA in women.
Variations in craniofacial morphology that reduce the size of the 
posterior airway space increase OSA risk. The contribution of skeletal 
structural features to OSA is most evident in nonobese patients. Iden­
tification of features such as retrognathia can influence therapeutic 
decision-making.
OSA has a strong genetic basis, as evidenced by its significant famil­
ial aggregation and heritability. For a first-degree relative of a patient 
with OSA, the odds of having OSA are approximately twofold higher 
than that of someone without an affected relative. Several genetic vari­
ants have been associated with prevalence of OSA or with related traits, 
such as the frequency of apneas and hypopneas, the duration of respira­
tory events, and degree of overnight levels of hypoxemia.
OSA prevalence varies with age, from 5 to 15% among middle-aged 
adults to >20% among elderly individuals, although in a majority of 
affected adults, the disorder is undiagnosed. There is a peak due to 
lymphoid hypertrophy among children between the ages of 3 and 
8 years; with airway growth and lymphoid tissue regression during later 
childhood, prevalence declines. Then, as obesity prevalence increases 
in adolescence and adulthood, OSA prevalence again increases.
The prevalence of OSA is especially high among patients with 
certain medical conditions, including diabetes mellitus, hyperten­
sion, and atrial fibrillation. Individuals of East Asian ancestry appear 
to be at increased risk of OSA at relatively low levels of body mass 
index, reflecting the greater influence of craniofacial risk factors. In 
the United States, African Americans, especially children and young 
adults, are at higher risk for OSA than their white counterparts.
Course of the Disorder 
The precise onset of OSA is usually hard 
to identify. A person may snore for many years, often beginning in 
childhood, before OSA is identified. Weight gain may precipitate an 
increase in symptoms, which in turn may lead the patient to pursue an 
evaluation. OSA may become less severe with weight loss, particularly 
after bariatric surgery. In adults, there is a gradual increase in AHI with 
age, although marked increases and decreases in the AHI are uncom­
mon unless accompanied by weight change.
APPROACH TO THE PATIENT
Obstructive Sleep Apnea/Hypopnea Syndrome
An evaluation for OSA should be considered in patients with symp­
toms of OSA and one or more risk factors. Screening also should

be considered in patients who report symptoms consistent with 
OSA and who are at high risk for OSA-related morbidities, such as 
hypertension, diabetes mellitus, and cardiac and cerebrovascular 
diseases. 
SYMPTOMS AND HISTORY
When possible, a sleep history should be obtained with assistance 
from a bed partner or household member. Snoring is the most com­
mon symptom; however, its absence does not exclude the diagnosis, 
as pharyngeal collapse may occur without tissue vibration. Gasping 
or snorting during sleep may also be reported, reflecting termina­
tion of individual apneas with abrupt airway opening. Dyspnea is 
unusual, and its absence generally distinguishes OSA from parox­
ysmal nocturnal dyspnea, nocturnal asthma, and acid reflux with 
laryngospasm. Patients also may describe frequent awakening or 
sleep disruption, which is more common among women and older 
adults. The most common daytime symptom is excessive daytime 
sleepiness, identified by a history of difficulty maintaining alertness 
or involuntary periods of dozing. However, many women preferen­
tially report fatigue rather than sleepiness. Other symptoms include 
a dry mouth, nocturnal heartburn, diaphoresis of the chest and 
neck, nocturia, morning headaches, trouble concentrating, irrita­
bility, and mood disturbances. Insomnia, which is common in the 
general population, may coexist with OSA. Although difficulty fall­
ing sleep is rarely caused by OSA, awakening at apnea termination 
may cause difficulty maintaining sleep, a symptom more likely to be 
reported by women than by men, and often responds to treatment 
of OSA. Several questionnaires that evaluate snoring frequency, 
self-reported apneas, and daytime sleepiness can facilitate OSA 
screening. The predictive ability of a questionnaire can be enhanced 
by a consideration of whether the patient is male or has risk factors 
such as obesity or hypertension. 
PHYSICAL FINDINGS
Physical findings often reflect the etiologic factors for the disorder 
as well as comorbid conditions, particularly vascular disease. On 
examination, patients may exhibit hypertension and regional (cen­
tral) obesity, as indicated by a large waist and neck circumference. 
The oropharynx may reveal a small orifice with crowding due to an 
enlarged tongue, a low-lying soft palate with a bulky uvula, large 
tonsils, a high-arched palate, or micro-/retrognathia. Since nasal 
resistance can increase the propensity to pharyngeal collapse, the 
nasal cavity should be inspected for polyps, septal deviation, allergic 
rhinitis, and other signs of obstruction. Because patients with heart 
failure are at increased risk for both OSA and CSA, a careful cardiac 
examination should be conducted to detect possible left- or rightsided cardiac dysfunction. Evidence of cor pulmonale suggests a 
comorbid cardiopulmonary condition; OSA alone is not thought 
to cause right-heart failure. A neurologic evaluation is needed to 
evaluate for conditions such as neuromuscular and cerebrovascular 
diseases, which increase OSA risk. 
LABORATORY FINDINGS 
Diagnostic Findings  Since symptoms and signs do not accurately 
predict the severity of sleep-related breathing disturbances, specific 
diagnosis and categorization of OSA severity require objective 
measurement of breathing during sleep. The gold standard for diag­
nosis of OSA is an overnight polysomnogram (PSG). A negative 
in-laboratory PSG usually rules out OSA. However, false-negative 
studies can result from night-to-night variation in OSA severity, 
particularly if there was insufficient REM sleep or less supine sleep 
during testing than is typical for the patient. Home sleep tests that 
record only respiratory and cardiac channels are commonly used 
as a cost-effective means for diagnosing OSA. However, a home 
study may yield a false-negative result if sleep time is not accurately 
estimated or in individuals experiencing hypopneas with arousals 
rather than oxyhemoglobin desaturation. Therefore, if there is a 
high prior probability of OSA, a negative home study should be 
followed by PSG.

The key physiologic information collected during a sleep study 
for OSA assessment includes measurement of breathing (changes 
in airflow, respiratory excursion), oxygenation (hemoglobin oxygen 
saturation), body position, and cardiac rhythm. In addition, PSGs 
and some home sleep studies measure sleep continuity and sleep 
stages (by electroencephalography, chin electromyography, electrooculography, and actigraphy), leg movements, and snoring intensity. 
This information is used to quantify the frequency and subtypes of 
abnormal respiratory events during sleep as well as associated 
changes in oxygen hemoglobin saturation, arousals, and sleep stage 
distributions. Tables 308-1 and 308-2 define the respiratory events 
scored and the severity guidelines employed during a sleep study. 
Fig. 308-2 shows examples of sleep-related respiratory events. A 
typical sleep study report provides quantitative data such as the 
AHI (number of apneas plus hypopneas per hour of sleep) and the 
profile of oxygen saturation over the night (mean, nadir, time at low 
levels). Reports may also include the respiratory disturbance index, 
which includes the number of respiratory effort–related arousals 
in addition to the AHI. In-laboratory PSG also quantifies sleep 
latency (time from “lights off” to first sleep onset), the frequency 
of periodic limb movements during sleep, sleep efficiency (percent­
age of time asleep relative to time in bed), arousal index (number 
of cortical arousals per hour of sleep), and time in each sleep stage. 
These metrics can further characterize the severity of OSA, which is 
associated with an increased arousal index, low sleep efficiency, and 
a reduction of time in deep (stage N3) and REM sleep and increase 
in light (stage N1) sleep. The detection of autonomic responses to 
apneas and hypopneas, such as surges in blood pressure, changes 
in heart rate, and abnormalities in cardiac rhythm, also provides 
relevant information on OSA severity.
Sleep Apnea
CHAPTER 308
While the AHI is the chief disease-defining measurement derived 
from sleep studies, metrics that quantify respiratory event–related 
hypoxemia, heart rate response, and ventilatory reduction have 
been shown to predict adverse cardiovascular outcomes and mor­
tality and may soon be incorporated into clinical decision-making. 
Other Laboratory Findings  Various imaging studies, including 
cephalometric radiography, upper airway magnetic resonance 
imaging (MRI) and computed tomography (CT), and fiberoptic 
endoscopy, can be used to identify anatomic risk factors for OSA. 
While these may be useful for planning surgical interventions, they 
are not indicated in the routine evaluation of OSA. Cardiac test­
ing may yield evidence of impaired systolic or diastolic ventricular 
function or abnormal cardiac structure. Overnight blood pressure 
monitoring often displays a “nondipping” pattern (absence of the 
typical 10% fall of blood pressure during sleep compared to wake­
fulness). Arterial blood gas measurements made during wakeful­
ness are usually normal. Waking hypoxemia or hypercarbia suggests 
coexisting cardiopulmonary disease or hypoventilation syndromes. 
Patients with severe nocturnal hypoxemia may have elevated hemo­
globin values. A multiple sleep latency test or a maintenance of 
wakefulness test can be useful in quantifying sleepiness and helping 
to distinguish OSA from narcolepsy.
TABLE 308-1  Respiratory Event Definitions
• Apnea: Cessation of airflow for ≥10 s during sleep, accompanied by:
• Persistent respiratory effort (obstructive apneas, Fig. 308-2A), or
• Absence of respiratory effort (central apneas, Fig. 308-2B)
• Hypopnea: A ≥30% reduction in airflow for at least 10 s during sleep that is 
accompanied by either a ≥3% desaturation or an arousal (Fig. 308-2C)
• Respiratory effort–related arousal (RERA): Partial obstruction that does not 
meet the criteria for hypopnea but provides evidence of increasing inspiratory 
effort (usually through pleural pressure monitoring) punctuated by an arousal 
(Fig. 308-2D)
• Flow-limited breath: A partially obstructed breath, typically within a hypopnea 
or RERA, identified by a flattened or “scooped-out” inspiratory flow shape 
(Fig. 308-3)

TABLE 308-2  Obstructive Sleep Apnea/Hypopnea Syndrome 
(OSAHS): Quantification and Severity Scale
• Apnea-hypopnea index (AHI)a: Number of apneas plus hypopneas per hour of 
sleep
• Respiratory disturbance index (RDI): Number of apneas plus hypopneas plus 
RERAs per hour of sleep
• Mild OSAHS: AHI of 5–14 events/h
• Moderate OSAHS: AHI of 15–29 events/h
• Severe OSAHS: AHI of ≥30 events/h
aEach level of AHI can be further quantified by level of sleepiness and associated 
hypoxemia.
Abbreviation: RERAs, respiratory effort–related arousals.
PART 7
Disorders of the Respiratory System
Health Consequences and Comorbidities 
OSA is the most 
common medical cause of daytime sleepiness and negatively influences 
quality of life. It is also strongly associated with cardiac, cerebrovascu­
lar, and metabolic disorders and with premature death and increased 
risk for certain cancers. This broad range of health effects is attribut­
able to the impact of sleep fragmentation, cortical arousal, and inter­
mittent hypoxemia and hypercapnia on vascular, cardiac, metabolic, 
and neurohumoral functions. OSA-related respiratory events stimulate 
sympathetic overactivity, leading to acute blood pressure surges dur­
ing sleep and nocturnal as well as daytime hypertension. OSA-related 
hypoxemia also stimulates release of acute-phase proteins and reac­
tive oxygen species that exacerbate insulin resistance and lipolysis 
and cause an augmented prothrombotic and proinflammatory state. 
Inspiratory effort against an occluded airway causes large intrathoracic 
A
EEG
EOG
chin
EKG
snore
t. flow
n. p. flow
chest
abdomen
SaO2
C
Hypnogram
Stage
EEG
snore
flow
position
chest
abdomen
SaO2
FIGURE 308-2  Obstructive apnea. A. There are 30 s of no airflow, as shown in the nasal pressure (n. p. flow) and thermistor-measured flow (t. flow). Note the presence of 
chest-abdomen paradox, indicating respiratory effort against an occluded airway. B. Central apnea in a patient with Cheyne-Stokes respiration due to congestive heart 
failure. The flat chest-abdomen tracings indicate the absence of inspiratory effort during the central apneas. C. Hypopnea. Partial obstruction of the pharyngeal airway can 
limit ventilation, leading to desaturation (a mild decrease in this patient, from 93 to 90%) and arousal. D. Respiratory effort–related arousal (RERA). Minimal flow reduction 
terminated by an arousal (Ar) without desaturation constitutes an RERA. EEG, electroencephalogram; EKG, electrocardiogram; EOG, electro-oculogram.

Normal
Flow limitation
FIGURE 308-3  Example of flow limitation. The inspiratory flow pattern in a patent 
airway is rounded and peaks in the middle. In contrast, a partially obstructed airway 
exhibits an early peak followed by mid-inspiratory flattening, yielding a scooped-out 
appearance.
negative pressure swings, altering cardiac preload and afterload and 
resulting in cardiac remodeling and reduced cardiac function. Hypox­
emia and sympathetic-parasympathetic imbalance also may cause 
electrical remodeling of the heart and myocyte injury.
HYPERTENSION  OSA can raise blood pressure to prehypertensive and 
hypertensive ranges, increase the prevalence of a nondipping overnight 
blood pressure pattern, and increase the risk of uncontrolled and resis­
tant hypertension. Elevations in blood pressure are due to augmented 
sympathetic nervous system activation as well as alterations in the 
renin-angiotensin-aldosterone system and fluid balance. Treatment of 
OSA with nocturnal continuous positive airway pressure (CPAP) has 
been shown to reduce 24-h ambulatory blood pressure. Although the 
overall impact of CPAP on blood pressure levels is relatively modest 
(averaging 2–4 mmHg), larger improvements are observed among 
B
EEG
EOG
chin
snore
flow
chest
abdomen
SaO2
D
EEG
EOG
chin
EKG
Ar
Legs
snore
t. flow
n. p. flow
chest
abdomen
SaO2

patients who have a high AHI, report daytime sleepiness, or have resis­
tant hypertension.
CARDIOVASCULAR, 
CEREBROVASCULAR, 
AND 
METABOLIC 
DISEASES  Among the most serious health consequences of OSA may 
be its impact on cardiac and metabolic functions. Strong epidemiologic 
evidence indicates that OSA significantly increases the risk of coronary 
artery disease, heart failure with and without reduced ejection frac­
tion, atrial and ventricular arrhythmias, atherosclerosis and coronary 
artery disease, stroke, and diabetes. Treatment of OSA has been shown 
to reduce several markers of cardiovascular risk and improve insulin 
resistance and, in uncontrolled studies, is associated with a decreased 
recurrence rate of atrial fibrillation. Large randomized clinical trials, 
however, have failed to demonstrate that OSA treatment with CPAP 
reduces cardiovascular and stroke event rates or prolongs survival. 
These outcomes may reflect exclusion from these trials of patients with 
excessive sleepiness, as there is evidence that sleepy patients may have 
the greatest OSA-related cardiovascular risk. Limited adherence to 
treatment among trial participants or the widespread use of other effec­
tive secondary prevention measures, such as beta blockade, antiplatelet 
agents, and lipid-lowering therapy, may also limit the impact of CPAP 
on cardiovascular risk.
SLEEPINESS AND ASSOCIATED RISKS  More than 50% of patients with 
moderate to severe OSA report daytime sleepiness. However, there 
is not a strong association between AHI level and degree of sleepi­
ness. Patients with OSA symptoms have a twofold increased risk of 
occupational accidents. Individuals with elevated AHIs are involved 
in motor vehicle crashes approximately two to three times as often as 
persons with normal AHIs. Randomized controlled trials have shown 
that treatment of OSA with CPAP alleviates sleepiness as measured by 
either questionnaire or objective testing in patients with both mild and 
more severe disease. However, the degree of improvement varies widely. 
Residual sleepiness may be due to several factors, including suboptimal 
treatment adherence, insufficient sleep duration, other sleep disorders, 
or prior hypoxia-mediated damage in brain areas involved in alertness. 
Moreover, visceral adipose tissue, which is present in higher amounts 
in patients with OSA, releases somnogenic cytokines that may contrib­
ute to sleepiness. Thus, even after treatment, it is important to assess 
and monitor patients for residual sleepiness and to optimize treatment 
adherence, improve sleep patterns, and identify other disorders that 
may contribute to sleepiness. Careful and supervised use of alerting 
agents may be appropriate as adjunctive treatment in patients in whom 
sleepiness does not respond to CPAP alone.
QUALITY OF LIFE AND MOOD  Reductions in health-related quality 
of life are common in patients with OSA, with the largest decrements 
observed in scales that measure physical functioning and energy levels. 
Work-related productivity also has been shown to improve in patients 
with moderate to severe OSA treated with CPAP. Numerous studies, 
including a large-scale trial of minimally symptomatic patients, have 
shown that treatment with CPAP can improve these patient-reported 
outcomes. Depressive symptoms, in particular somatic symptoms 
(irritability, fatigue, lack of energy), are commonly reported in OSA 
and improve with CPAP.
TREATMENT
Obstructive Sleep Apnea
A comprehensive approach to the management of OSA is needed 
to reduce risk factors and comorbidities. The clinician should 
seek to identify and address lifestyle and behavioral factors as well 
as comorbidities that may be exacerbating OSA. As appropriate, 
treatment should aim to reduce weight; optimize sleep duration 
(7–9 h per night); regulate sleep schedules (with similar bedtimes 
and wake times across the week); encourage the patient to avoid 
sleeping in the supine position; treat nasal allergies; increase physi­
cal activity; eliminate alcohol ingestion (which impairs pharyngeal 
muscle activity) within 3 h of bedtime; and minimize use of opi­
ate medications. Sedative-hypnotic medications have inconsistent 

effects on OSA but should be avoided in most patients with moder­
ate to severe OSA. Patients should be counseled to avoid drowsy 
driving.

CPAP is the standard medical therapy with the highest level 
of evidence for efficacy. Delivered through a nasal or nasal-oral 
mask, CPAP works as a mechanical splint to hold the airway 
open, thus maintaining airway patency during sleep. An overnight 
CPAP titration study can determine the optimal pressure set­
ting that reduces the number of apneas/hypopneas during sleep, 
improves gas exchange, and reduces arousals; however, the use of 
“auto-titrating” CPAP (APAP) devices used in home settings has 
eliminated the need for titration sleep studies in many patients. 
Rates of adherence to CPAP treatment are highly variable (average, 
50–80%) and may be improved with support by a skilled health 
care team who can address side effects, help the patient “problem 
solve,” and provide motivational education (Table 308-3). Online 
CPAP support tools can provide the patient personalized support 
and feedback. Despite the limitations of CPAP, controlled studies 
have demonstrated its beneficial effect on alertness, mood, qual­
ity of life, work-related productivity, blood pressure, and insulin 
sensitivity. Uncontrolled studies also indicate a favorable effect on 
cardiovascular outcomes, cardiac ejection fraction, atrial fibrillation 
recurrence, and mortality risk.
Sleep Apnea
CHAPTER 308
Oral appliances for OSA work by advancing the mandible, thus 
opening the airway by repositioning the lower jaw and pulling the 
tongue forward. These devices generally work better when custom­
ized for patient use; maximal adaptation can take several weeks. 
Efficacy studies show that these devices can reduce the AHI by 
≥50% in two-thirds of individuals, although these data are based 
largely on patients with mild OSA. Some patients with moderate 
or severe OSA respond to oral appliances as well, although no con­
sistent predictors of success have been identified in these groups, 
and thus, follow-up sleep testing is recommended. Side effects of 
oral appliances include temporomandibular joint pain and tooth 
movement; thus, they require that the patient have adequate dental 
and periodontal structures. Oral appliances are most often used for 
treating patients with mild/moderate OSA or patients who do not 
tolerate CPAP. However, as some patients are more adherent to oral 
appliances than to CPAP, these devices are under investigation for 
treatment of more severe disease. In some patients, more consistent 
treatment of OSA may be obtained by alternating between CPAP 
and oral appliance therapy.
Upper airway surgery for OSA is less efficacious than CPAP and 
is mostly reserved for the treatment of patients who snore, have 
mild OSA, or cannot tolerate CPAP. Uvulopalatopharyngoplasty 
(UPPP, removal of the uvula and the margin of the soft palate) is 
the most commonly performed surgery for OSA. However, results 
vary greatly, and as a standalone procedure, UPPP often has lim­
ited efficacy, particularly in severe OSA and in obese patients. 
Thus, palatal surgery is often combined with other procedures 
(“multilevel surgery” involving more than one pharyngeal site/
structure) performed by an experienced surgeon, but the selection 
of patients is an important factor and relies on careful targeting of 
culprit areas for surgical resection. Bariatric surgery is an option 
TABLE 308-3  Side Effects of Continuous Positive Airway Pressure 
(CPAP) and Their Treatments
SIDE EFFECT
TREATMENT
Nasal congestion
Provide heated humidification, administer saline/steroid 
nasal sprays
Claustrophobia
Change mask interface (e.g., to nasal prongs), promote 
habituation (i.e., practice breathing on CPAP while 
awake)
Difficulty exhaling
Temporarily reduce pressure, provide bilevel positive 
airway pressure
Bruised nasal ridge
Change mask interface, provide protective padding
Aerophagia
Administer antacids

for obese patients with OSA and can improve not only OSA but 
also other obesity-associated health conditions. Other procedures 
that can decrease snoring but have minimal effects on OSA include 
injection of a hardening agent to the soft palate (resulting in stiffen­
ing), radiofrequency ablation, laser-assisted uvulopalatoplasty, and 
palatal implants.

Upper airway neurostimulation is a recently tested alternative 
treatment for OSA. Unilateral stimulation of the hypoglossal nerve 
through a surgically implanted device was shown to significantly 
decrease the AHI and improve a number of patient-reported out­
comes, such as sleepiness and quality of life, for a duration of at least 
5 years after treatment in carefully selected patients. This therapy 
is reserved for patients who cannot tolerate or fail CPAP therapy. 
Current inclusion criteria are moderate to severe OSA (AHI 15–65), 
body mass index <35 kg/m2, and absence of complete concentric 
collapse at the level of the velum documented by awake and druginduced endoscopy (a predictor of response to surgery). This ther­
apy is also approved for use in children ages 13 or older and adults 
with Down’s syndrome who have an AHI of 10–50. Additional 
research is underway to further evaluate longer-term effectiveness 
and potential utility of this treatment in other patient groups.
PART 7
Disorders of the Respiratory System
Vibrotactile positional therapy (devices that emit incremental 
vibratory stimuli in response to body position in the supine posi­
tion) can reduce AHI by as much as 50% in individuals with OSA 
occurring predominantly in the supine position.
Supplemental oxygen can improve oxygen saturation, but there 
is little evidence that it improves OSA symptoms or the AHI in 
unselected patients. There is conflicting evidence regarding the effect 
of supplemental oxygen on blood pressure in patients with OSA.
Orofacial myofunctional therapy (OMT), which involves exer­
cises designed to enhance the strength and coordination of orofacial 
muscles, has been tested in several small studies of OSA. These 
studies predominantly focused on individuals suffering from mildto-moderate OSA and showed modest improvements in snoring 
and AHI, as well as in overall sleep quality. However, larger and 
more comprehensive studies are necessary to evaluate fully the 
efficacy of myofunctional therapy in treating OSA.
Currently, there are no U.S. Food and Drug Administration 
(FDA)-approved medications for the treatment of OSA. However, 
ongoing research is exploring the efficacy of weight loss medica­
tions and drugs that enhance pharyngeal muscle activity during 
sleep. Among weight loss drugs, anorectics are generally not favored 
for OSA due to their side effects and potential for abuse. However, 
glucagon-like peptide 1 (GLP-1) receptor agonists have garnered 
recent interest because of their substantial impact on weight loss. 
Given the strong correlation between weight and OSA, drugs that 
facilitate weight reduction, such as GLP-1 agonists, which slow 
gastric emptying and suppress appetite, are being investigated for 
their potential in improving OSA symptoms. Liraglutide, a GLP-1 
agonist, has shown small effects on OSA severity, but tirzepatide, 
which targets both GLP-1 and glucose-dependent insulinotropic 
polypeptide (GIP) receptors, may offer more significant weight loss 
and is currently under trial for OSA. Sodium-glucose cotransporter 
2 (SGLT-2) inhibitors also promote weight loss and are under inves­
tigation for the treatment of OSA.
Concerning drugs that stimulate pharyngeal muscle activity, 
studies have shown that sleep-related withdrawal of norepinephrine 
from cranial motor neurons leads to pharyngeal hypotonia and 
consequent upper airway obstruction in patients with OSA. Pha­
ryngeal hypotonia is further exacerbated by muscarinic inhibition 
during REM sleep. Based on these mechanisms, drugs with norad­
renergic and antimuscarinic properties have been tested for their 
potential in OSA management. Early studies have reported notable 
reductions in AHI, and more advanced phase 3 trials of these drug 
combinations are in progress. 
CENTRAL SLEEP APNEA
CSA, which is less common than OSA, may occur in isolation or, 
more often, in combination with obstructive events in the form of 

“mixed” apneas. CSA is often caused by an increased sensitivity to 
Pco2, which leads to an unstable breathing pattern that manifests 
as hyperventilation alternating with apnea. A prolonged circulation 
delay between the pulmonary capillaries and carotid chemorecep­
tors is also a contributing cause; thus, individuals with conges­
tive heart failure are at risk for CSA. With prolonged circulation 
delay, there is a crescendo-decrescendo breathing pattern known as 
Cheyne-Stokes breathing (Fig. 308-2B). Other risk factors for CSA 
include opioid medications (which appear to have a dose-dependent 
effect on CSA) and hypoxia (e.g., breathing at high altitude). In 
some individuals, CPAP—particularly at high pressures—seems to 
induce central apnea; this condition is referred to as complex sleep 
apnea or treatment-emergent central sleep apnea. Rarely, CSA may 
be caused by blunted chemosensitivity due to congenital disorders 
(congenital central hypoventilation syndrome) or acquired factors. 
CSA is associated with increased risk for the development of both 
heart failure and atrial fibrillation. This is possibly related to eleva­
tions in sympathetic nervous system activity that accompany this 
disorder; alternatively, CSA may be an early marker of subclinical 
myocardial dysfunction. Patients with CSA may report symptoms 
of frequent awakenings as well as daytime fatigue. Treatment of 
CSA is difficult and depends on the underlying cause. As treatment 
of CSA has not been shown to improve long-term health outcomes, 
specific treatment is generally indicated only for treatment of symp­
tomatic CSA.
Limited data suggest that supplemental oxygen can reduce the 
frequency of central apneas, particularly in patients with hypox­
emia; however, the effectiveness of supplemental oxygen on clinical 
outcomes is unknown.
In patients with CSA or Cheyne-Stokes breathing associated 
with heart failure, treatment is directed at optimizing therapy for 
heart failure. Device-based therapies that adjust pressure support 
on a breath-by-breath basis (adaptive servoventilation [ASV]) can 
be effective for regularizing breathing; however, two large inter­
national trials showed that ASV resulted in no cardiovascular or 
mortality benefit. The first trial unexpectedly reported increased 
mortality, leading to an FDA warning to avoid ASV in patients 
with CSA who have a left ventricular ejection fraction <45%. The 
second trial tested an ASV device that delivered lower pressure 
and reported no evidence of harm along with small improvements 
in symptoms, raising the possibility that ASV may have a role as 
adjunctive therapy for patients with heart failure and CSA.
Moderate to severe CSA in adults can be treated with an FDAapproved transvenous phrenic nerve stimulator. A clinical trial 
testing this therapy in patients with CSA from various etiologies 
showed an ~50% reduction in AHI and a near elimination of cen­
tral apneas, along with improvements in sleep quality and quality 
of life. Ongoing research is addressing long-term safety and effec­
tiveness, as well as subgroups who may benefit the most from this 
intervention.
■
■FURTHER READING
Berry R, Wagner M: Sleep Medicine Pearls, 3rd ed. Philadelphia, 
Elsevier, 2015.
Gottesman RF et al: Impact of sleep disorders and disturbed sleep 
on brain health: A scientific statement from the American Heart 
Association. Stroke 55:e61, 2024.
Gottlieb DJ, Punjabi NM: Diagnosis and management of obstructive 
sleep apnea. A review. JAMA 323:1389, 2020.
Javaheri S et al: Sleep apnea: Types, mechanisms, and clinical cardio­
vascular consequences. J Am Coll Cardiol 769:841, 2017.
Kapur VK et al: Clinical Practice Guideline for Diagnostic Testing 
for Adult Obstructive Sleep Apnea: An American Academy of Sleep 
Medicine clinical practice guideline. J Clin Sleep Med 13:479, 2017.
Marklund M et al: Update on oral appliance therapy. Eur Respir Rev 
28:190083, 2019.
Redline S et al: Obstructive sleep apnoea heterogeneity and cardiovas­
cular disease. Nat Rev Cardiol 20:560, 2023.