INTRODUCTION — Obstructive sleep apnea (OSA) is a disorder that is characterized by obstructive apneas, hypopneas, and/or respiratory effort-related arousals caused by repetitive collapse of the upper airway during sleep.
The epidemiology, clinical presentation, diagnostic approach, and complications of OSA are reviewed here. The pathogenesis and management of OSA are described separately. (See "Pathophysiology of upper airway obstruction in obstructive sleep apnea in adults" and "Management of obstructive sleep apnea in adults".)
EPIDEMIOLOGY — OSA is the most common sleep-related breathing disorder. OSA is most common among older males, but it can also affect females and children [1]. The incidence rises following menopause such that rates are similar in postmenopausal individuals [2,3].
The estimated prevalence in North America is approximately 15 to 30 percent in males and 10 to 15 percent in females, when OSA is defined broadly as an apnea-hypopnea index (AHI) greater than five events per hour of sleep [4,5]. When more stringent definitions are used (eg, AHI ≥5 events per hour plus symptoms or AHI ≥15 events per hour), the estimated prevalence is approximately 15 percent in males and 5 percent in females [4-6]. Global estimates using five or more events per hour suggest rates of 936 million people worldwide with mild to severe OSA, and 425 million people worldwide with moderate to severe OSA, between the ages of 30 and 69 years of age [7].
The prevalence of OSA also varies by race. OSA is more prevalent in African Americans who are younger than 35 years old compared with White Americans of the same age group, independent of body weight [8,9]. The prevalence of OSA in Asia is similar to that in the United States, despite lower rates of obesity [10].
The prevalence appears to be increasing and may relate to the increasing rates of obesity or increased detection rates of OSA. In one study, the estimated prevalence of OSA between 1990 and 2010 increased from 11 to 14 percent in adult males and from 4 to 5 percent in adult females [5]. Another study from the United Kingdom also demonstrates a significant increase in the rates of OSA and obesity between 1994 and 2015 [11].
RISK FACTORS AND ASSOCIATED CONDITIONS — Several clinical risk factors are associated with OSA and include the following:
●Older age – The prevalence of OSA increases from young adulthood through the sixth to seventh decade, then appears to plateau [4,12,13].
●Male sex – OSA is approximately two to three times more common in males than females, although the risk appears to be similar once females are peri- and postmenopausal [2,13-16].
●Obesity – The risk of OSA correlates well with the body mass index (BMI) [5,14]. In one study, a 10 percent increase in weight was associated with a six-fold increase in risk of OSA [17]. In another study, moderate to severe OSA (apnea-hypopnea index [AHI] ≥15) was present in 11 percent of males who were normal weight, 21 percent who were overweight (BMI 25 to 30 kg/m2), and 63 percent of those who were obese (BMI >30 kg/m2) [13]. Similarly, in females, OSA was present in 3 percent of patients who were normal weight, 9 percent of those who were overweight, and 22 percent of those who were obese. The majority of individuals with obesity hypoventilation syndrome (OHS) have OSA (90 percent); OHS is discussed separately. (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome".)
●Craniofacial and upper airway abnormalities – Craniofacial or upper airway abnormalities increase the likelihood of having OSA [14]. These factors are best recognized in Far-East Asian patients where obesity is not as major a risk factor compared with the United States [18]. Examples of abnormalities include an abnormal maxillary or short mandibular size, a wide craniofacial base, and tonsillar and adenoid hypertrophy, the latter being common in children. (See "Evaluation of suspected obstructive sleep apnea in children", section on 'Risk factors'.)
Less well-established risk factors include the following:
●Smoking – Smoking may increase the risk of or worsen OSA. In one study, current smokers were nearly three times more likely to have OSA than past or never smokers [19].
●Family history of snoring or OSA – While a family history of snoring or OSA could be due to shared behavioral or environmental factors, there may also be a genetic predisposition to OSA through factors such as craniofacial structure [20]. It has been suggested that about 40 percent of the variance of the AHI has a genetic basis [21]. In another study of rural Brazilians, the heritability of an AHI >5/hour was intermediate (25 percent) [22]. (See "Pathophysiology of upper airway obstruction in obstructive sleep apnea in adults".)
●Others – Nasal congestion confers an approximately two-fold increase in the prevalence of OSA compared with controls, regardless of the cause [14]. However, OSA may or may not improve with correction of nasal congestion. Exposure to high levels of environmental nitrogen dioxide and particulate matter may contribute to variations in OSA among patient populations [23].
While a variety of substances and medications, including alcohol, benzodiazepines, narcotics, and possibly gabapentinoids may exacerbate OSA, a causative link is unproven [24,25]. (See "Management of obstructive sleep apnea in adults", section on 'Behavior modification'.)
The prevalence of OSA is also increased in patients with a variety of medical conditions, including the following (table 1):
●OHS (see "Clinical manifestations and diagnosis of obesity hypoventilation syndrome")
●Congestive heart failure (see "Sleep-disordered breathing in heart failure")
●Atrial fibrillation [26]
●Pulmonary hypertension [27]
●Hypertension (particularly resistant hypertension), cardiovascular disease, atrial fibrillation, and pulmonary hypertension (see "Obstructive sleep apnea and cardiovascular disease in adults")
●End-stage kidney disease (see "Sleep disorders in end-stage kidney disease", section on 'Sleep apnea')
●Chronic lung disease, including asthma, chronic obstructive pulmonary disease (COPD), and idiopathic pulmonary fibrosis (see "Evaluation of severe asthma in adolescents and adults", section on 'Assessing comorbid conditions' and "Sleep-related breathing disorders in COPD")
●Stroke and transient ischemic attacks (see "Sleep-related breathing disorders and stroke")
●Pregnancy (see "Obstructive sleep apnea in pregnancy")
•Gestational diabetes [28]
•Pregnancy-induced hypertension [29]
●Acromegaly (see "Causes and clinical manifestations of acromegaly", section on 'Sleep apnea')
●Hypothyroidism (see "Clinical manifestations of hypothyroidism", section on 'Respiratory system')
●Polycystic ovary syndrome (see "Clinical manifestations of polycystic ovary syndrome in adults", section on 'Sleep apnea')
●Parkinson's disease [30] (see "Clinical manifestations of Parkinson disease", section on 'Sleep disorders')
●Floppy eyelid syndrome [31-33]
Other medical conditions that may have an increased association with OSA include fibromyalgia [34,35], Barrett's esophagus [36], gastroesophageal reflux disease (GERD) [33,37], secondary polycythemia [38], and Down's syndrome [39]. Whether there is an increased prevalence of OSA in posttraumatic stress disorder is unclear [40].
CLINICAL FEATURES
Signs and symptoms — Most patients with OSA complain of daytime sleepiness, or their bed partner reports loud snoring, gasping, choking, snorting, or interruptions in breathing while sleeping (table 1). These symptoms are often detected during the evaluation of another complaint, or during health maintenance or preoperative screening. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Screening with a questionnaire'.)
●Daytime sleepiness – Daytime sleepiness is a common feature of OSA. Sleepiness is the inability to remain fully awake or alert during the wakefulness portion of the sleep-wake cycle [41]. Daytime sleepiness may be underestimated because of its insidious onset and chronicity. The patient may use terms such as fatigue, tiredness, low energy, or poor focus [42]. Targeted questioning of the patient and in particular their loved ones or bed partner, however, typically reveals a pattern of feeling sleepy or falling asleep in boring, passive, or monotonous situations. As an example, the patient may admit to consistently falling asleep while reading, watching television, or even while operating a motor vehicle. In addition, embarrassing or inappropriate episodes of sleep may be reported (eg, at religious services, listening to lectures, or driving). Reviewing patient behavior away from the workplace is essential because daytime sleepiness can be masked by activity. Patients should also always be asked about behaviors that may mask sleepiness, such as caffeine consumption. Patients often experience nonrestorative sleep (ie, do not wake up feeling refreshed) and nocturnal restlessness in association with their complaint of daytime sleepiness.
Sleepiness should be distinguished from fatigue. Fatigue is defined as a subjective lack of physical or mental energy that is perceived by the individual or caregiver to interfere with usual and desired activities. To facilitate this distinction, a series of directed questions can be combined with the Epworth Sleepiness Scale (ESS) to quantitatively document the patient's perception of sleepiness, fatigue, or both (table 2) (calculator 1) [43,44]. An ESS score >9 indicates abnormal sleepiness and should prompt further testing. Since there is often overlap of both sleepiness and fatigue in patients with OSA, our center administers both the ESS and the Fatigue Severity Scale (FSS) to help identify and manage these complaints [45,46].
Additional details regarding the approach to a sleepy patient and the differential diagnosis involved (table 3) is provided separately. (See "Approach to the patient with excessive daytime sleepiness", section on 'History'.)
●Snoring, choking, gasping during sleep – Snoring and associated events (ie, resuscitative gasping or snorting, witnessed apneic periods, periods of silence followed by loud snoring, restless or fitful sleep) are common features of OSA. It is usually helpful to have the patient's bed partner or a family member present during the interview because they often have greater insight than the patient into the frequency and severity of these symptoms during sleep.
While snoring is associated with a sensitivity of 80 to 90 percent for the diagnosis of OSA, its specificity is below 50 percent. The actual percentage of snorers who have sleep apnea varies greatly and solid data are lacking. The complaint of snoring, while common in patients with OSA, was found to have no predictive value in one study (likelihood ratio 1.1) [47]. On the other hand, the absence of snoring (particularly in the absence of risk factors such as obesity) reduces the likelihood of a diagnosis of OSA. For example, patients with mild snoring and a body mass index (BMI) lower than 26 are unlikely to have moderate or severe OSA [47]. A study of 1643 subjects with habitual snoring found a significant positive correlation between the severity of the OSA and snoring intensity [48]. (See "Snoring in adults".)
Choking or gasping during sleep or wakening with a dry mouth are commonly reported, often in association with snoring. In a systematic review of six studies that examined the accuracy of the clinical examination in the diagnosis of OSA, nocturnal choking or gasping had a lower sensitivity than snoring (52 versus 80 percent) but greater specificity (84 versus 50 percent) [47].
●Morning headaches – Morning headaches are reported by 10 to 30 percent of patients with untreated OSA [49,50]. They are usually bifrontal and squeezing in quality, with no associated nausea, photophobia, or phonophobia. They typically occur daily or most days of the week and may last for several hours after awakening in the morning. The cause of the headaches is not well-established and may be multifactorial; proposed mechanisms include hypercapnia, vasodilation, increased intracranial pressure, and impaired sleep quality. Early morning headaches may indicate severe disease, although a consistent association with disease severity has not been found [50-52]. (See "Evaluation of headache in adults".)
●Others – Other less common manifestations of OSA include:
•Sleep maintenance insomnia – Sleep maintenance insomnia with repetitive awakenings should prompt consideration of OSA. Approximately one-third of patients with OSA complain of insomnia rather than daytime sleepiness [53]. This phenomenon is more common in females [53,54].
•Symptoms of associated conditions and complications – Some patients may present with the symptoms of associated conditions and complications including neuropsychiatric symptoms, postoperative hypoxemia, or nocturnal cardiovascular events such as chest pain due to angina pectoris or palpitations due to atrial fibrillation. (See 'Risk factors and associated conditions' above and 'Complications' below.)
•Nocturia – Nocturia is a common associated symptom of OSA [55]. When patients repeatedly awaken from apneic events, they experience the urge to urinate, although other physiological mechanisms may be important [56].
Physical examination — Common physical findings are the following:
●Obesity – While obesity (BMI ≥30 kg/m2) is the most common clinical finding in patients with OSA, some patients may be overweight (BMI 25 to 29.9 kg/m2) or their weight may be within the normal range. (See "Obesity in adults: Prevalence, screening, and evaluation".)
●Crowded oropharyngeal airway – Numerous craniofacial conditions can narrow the upper airway and contribute to the development of OSA (table 4). These include retrognathia, micrognathia, lateral peritonsillar narrowing, macroglossia, tonsillar hypertrophy, an elongated or enlarged uvula, a high arched or narrow palate, nasal septal deviation, and nasal polyps [57]. The modified Mallampati classification is commonly used to quantify airway narrowing, with classes 3 and 4 considered positive for airway narrowing (figure 1). Both the Mallampati classification and Friedman tongue position (figure 2) have been shown to correlate with OSA severity [58]. (See 'Risk factors and associated conditions' above.)
●Large neck and/or waist circumference – OSA is more strongly correlated with an increased neck size or waist circumference than general obesity [59,60]. OSA is particularly prominent among males who have a collar size greater than 17 inches and females who have a neck size greater than 16 inches [57]. In another study, cutoff values for waist circumference and for waist-to-height ratio for females with OSA were 95.5 cm and 0.595, respectively, whereas the values for males were 100.5 cm, and 0.575, respectively [61].
●Signs of associated conditions and complications – Patients with OSA may also have the signs of associated conditions and complications, most commonly systemic hypertension and heart failure, and less commonly pulmonary hypertension. (See 'Risk factors and associated conditions' above and 'Complications' below.)
DIFFERENTIAL DIAGNOSIS — Several conditions may mimic OSA, depending on the presenting symptoms.
●Excessive daytime sleepiness – A variety of conditions similarly present with excessive daytime sleepiness (table 3). In general, they can be distinguished from OSA via clinical history and polysomnography (PSG). However, in many circumstances, home sleep apnea testing is not useful in evaluating for more complex sleep disorders that present with daytime sleepiness (eg, periodic limb movement disorder, restless leg syndrome, narcolepsy, central sleep apnea, and non-OSA related conditions associated with sleep-disordered breathing). The differential diagnosis and approach to daytime sleepiness is discussed in detail separately. (See "Approach to the patient with excessive daytime sleepiness", section on 'Causes'.)
●Abrupt awakenings or abnormal sounds or sensations during sleep – There are conditions that should be considered in the differential diagnosis of OSA because they similarly cause abrupt awakenings or abnormal sounds or sensations during sleep (table 5).
These conditions can also be distinguished from OSA via PSG:
•Primary snoring – Snoring is far more common than OSA. Therefore, even though most patients who have OSA snore, most patients who snore do not have OSA. Sleep apnea testing is the only way to distinguish snoring in association with OSA from primary snoring. (See "Snoring in adults".)
•Gastroesophageal reflux disease – Gastroesophageal reflux disease (GERD) can mimic OSA by producing a choking sensation and dyspnea at night. In addition, GERD may improve with positive airway pressure therapy, further mimicking OSA [62]. Patients may have a history of dyspepsia or symptoms of reflux, although their GERD may be silent. (See "Clinical manifestations and diagnosis of gastroesophageal reflux in adults".)
•Miscellaneous – Other disorders that can mimic OSA include nocturnal asthma, swallowing disorders, nocturnal seizures [63], and psychiatric illnesses such as panic attacks, many of which should be evident clinically and can be distinguished from OSA by PSG (with the exception of nocturnal seizures, most of these disorders have a relatively normal PSG).
●Early morning headaches – Early morning headaches may also be associated with space occupying lesions of the brain and may need to be distinguished from OSA by brain imaging. Early morning headaches may also be a presenting manifestation of obesity hypoventilation and may be distinguished from OSA by the presence of hypercapnia/hypercarbia on laboratory examination. OSA is present in the majority of patients with obesity hypoventilation syndrome (OHS; >90 percent). (See "Evaluation of headache in adults".)
DIAGNOSTIC EVALUATION — OSA should be suspected whenever a patient presents with excessive daytime sleepiness, snoring, and choking or gasping during sleep (table 1), particularly in the presence of risk factors such as obesity, male sex, and advanced age. Less common manifestations are early morning headaches, or manifestations of associated disorders (eg, hypertension) or complications (eg, neuropsychiatric symptoms). (See 'Clinical features' above and 'Risk factors and associated conditions' above and 'Complications' below.)
Detailed questions that explore the etiologies of daytime sleepiness, snoring, and neuropsychiatric disease may help to tease out OSA from other conditions (table 3 and table 2) but sleep apnea testing is required to make the diagnosis of OSA. Details regarding taking a targeted history and performing a comprehensive examination in patients with daytime sleepiness is provided separately. (See "Approach to the patient with excessive daytime sleepiness", section on 'Initial evaluation'.)
There has been debate in the literature regarding whether non-sleep experts can adequately evaluate patients with suspected OSA [64-66]. While observational studies have suggested that evaluation by sleep and non-sleep experts may be similar, many of the non-sleep specialists in these studies have extensive training or experience in sleep medicine [64]. Accordingly, in support of the American Academy of Sleep Medicine (AASM) guidelines, a comprehensive evaluation with follow-up by a clinician who has some level of expertise in sleep medicine is appropriate [65].
Patient selection — OSA is not a clinical diagnosis and objective testing must be performed for the diagnosis [47,67,68]. Due to the wide differential associated with the symptoms of OSA, several clinical criteria and data from evaluation tools are used to select those who should be tested.
Clinical parameters — Diagnostic testing for OSA should be performed on patients with excessive daytime sleepiness (EDS) on most days and the presence of at least two of the following clinical features of OSA: habitual loud snoring, witnessed apnea or gasping or choking during sleep, and diagnosed systemic hypertension [65]. The rationale for these criteria is that these features predict an increased risk of moderate to severe OSA (ie, a population most likely to benefit from therapy).
In the absence of these criteria, many experts also perform diagnostic testing in the following:
●EDS alone
●Patients who have other clinical features of OSA (eg, obesity, fatigue, upper airway abnormalities, snoring) and conditions or complications associated with OSA (eg, refractory hypertension, atrial fibrillation, nocturnal angina or dysrhythmias, congestive heart failure, stroke, and transient ischemic attacks).
●Patients in whom OSA needs to be ruled in or out as an underlying cause or potential contributing factor to their symptoms (eg, unexplained pulmonary hypertension or polycythemia, or motor vehicle accident due to falling asleep while driving).
Evaluation tool parameters — Most sleep experts do not routinely use evaluation tools to select those at risk for OSA who need a sleep study since none of these tools have been shown to be superior to a history and physical examination and their poor accuracy make them imperfect diagnostic tools [59,65]. However, some non-sleep experts may find them useful in the outpatient setting. The most common evaluation tools used include the STOP-Bang questionnaire (table 6) or variants thereof (table 7) (calculator 2), and the Epworth Sleepiness Scale (ESS) (table 2) (calculator 1); for example, the STOP-Bang questionnaire is often used by physicians during preoperative evaluation to assess the risk of undiagnosed OSA. Other evaluation tools that are also available include the Berlin score (table 8 and table 9) and the sleep apnea clinical score (table 10). Importantly, none of these diagnostic tools should be used to replace a sleep apnea testing. (See 'Instruments' below.)
Selecting home or in-laboratory testing — In-laboratory polysomnography (PSG; either full-night [ie, diagnostic only] or split-night [ie, diagnostic and therapeutic with positive airway pressure]) is the gold standard diagnostic test. However, for patients in whom uncomplicated OSA is suspected and the pretest probability is estimated as moderate or severe, unattended home sleep apnea testing (HSAT) with a Type 3 device is a reasonable alternative to in-laboratory PSG. (See "Home sleep apnea testing for obstructive sleep apnea in adults", section on 'Type 3 devices (portable devices)'.)
Several factors influence the choice between home and in-laboratory testing (algorithm 1 and table 11). These include the following:
●The patient's profession (eg, mission-critical worker) (see 'Mission-critical workers with suspected OSA' below)
●Suspected nonrespiratory sleep disorders (table 12) [69-71] (see 'Suspected nonrespiratory sleep disorders' below)
●Non-OSA-related conditions associated with sleep-disordered breathing (table 13) (see 'Suspected OSA complicated by other respiratory sleep disorders' below)
●The severity of OSA (See 'Suspected uncomplicated OSA' below.)
●Others – Additional factors that influence the choice between performing a home or in-laboratory study include the following:
•The values and preferences of the patient (eg, access to a sleep laboratory, homelessness, space constraints, cost, wait time, concern for operator error for home monitoring)
•The practices of the institution and preferences of the clinician
•The preference of the payer (eg, many insurers in the United States require that patients receive home testing first)
Mission-critical workers with suspected OSA — Mission-critical workers should be tested using attended in-laboratory PSG. The rationale for this approach is that the risk to the patient and to others of missing a diagnosis of OSA is sufficiently high to justify laboratory testing. There is no definition of what "mission-critical" means but it should be self-evident to the clinician that if a patient has a job where falling asleep has a major negative impact (eg, airline pilots, bus drivers, taxi drivers, ride-sharing drivers, truck drivers, train operators, police, security, military posts, astronauts), then an in-laboratory sleep study is recommended. (See 'Polysomnography' below.)
Suspected nonrespiratory sleep disorders — When clinicians suspect that symptoms are due to a nonrespiratory sleep disorder or that a nonrespiratory sleep disorder coexists with or contributes to OSA, PSG should be performed. Examples of nonrespiratory sleep disorders include narcolepsy or other hypersomnia disorders, severe insomnia, parasomnias, movement disorders (eg, periodic limb movement disorder) (table 12). (See 'Polysomnography' below and "Clinical features and diagnosis of narcolepsy in adults" and "Approach to abnormal movements and behaviors during sleep" and "Polysomnography in the evaluation of parasomnias and epilepsy" and "Clinical features and diagnosis of restless legs syndrome and periodic limb movement disorder in adults" and "Classification of sleep disorders".)
Suspected OSA complicated by other respiratory sleep disorders — Complicated OSA refers to the presence of medical conditions that could potentially affect respiration during sleep and create additional respiratory abnormalities over and above those associated with OSA (table 13). This population should have PSG testing since the detection of nonobstructive events is underestimated and less accurate on most devices used for home testing. (See 'Polysomnography' below and "Home sleep apnea testing for obstructive sleep apnea in adults".)
Determining whether or not suspected OSA is "complicated" is practically achieved by clinically identifying conditions that increase the risk of nonobstructive sleep-disordered breathing. Examples of conditions associated with complex sleep-disordered breathing include chronic obstructive pulmonary disease (Global Obstructive Lung Disease [GOLD] stage 1, 2, 3) (table 14), heart failure (New York Heart Association [NYHA] class III or IV (table 15)), hypoventilation syndromes (eg, obesity hypoventilation syndrome), neuromuscular weakness, chronic opioid use, and stroke (table 13). (See "Clinical manifestations and diagnosis of obesity hypoventilation syndrome".)
Suspected uncomplicated OSA — Uncomplicated OSA is that which is not complicated by other conditions that could potentially affect respiration during sleep (see 'Suspected OSA complicated by other respiratory sleep disorders' above). Testing in this population depends upon the suspected degree of severity of OSA.
Pretest probability of moderate to severe OSA — We agree with the AASM, which states that patients at high pretest probability of moderate to severe OSA include those with daytime hypersomnolence and at least two of the following three criteria: habitual loud snoring, witnessed apnea or gasping/choking, or diagnosed hypertension [65]. Based on an apnea-hypopnea index (AHI) >15, the prevalence is estimated to be 64 percent in a high-risk population [59,65]. We consider this population suitable for HSAT, provided it is performed with an adequate device (eg, Type 3 device) and under adequate supervision with interpretation by a clinician knowledgeable in sleep medicine (algorithm 1 and table 11). The rationale for this approach is that HSAT has been shown in this setting to similarly detect obstructive events compared with PSG. (See 'Home sleep apnea testing' below and "Home sleep apnea testing for obstructive sleep apnea in adults".)
However, for those in whom a reliable assessment for the risk of OSA cannot be made using the above criteria, such as those who do not have hypersomnolence, in-laboratory PSG is preferred. (See 'Polysomnography' below.)
Importantly, for those in whom suspicion for OSA remains and HSAT is negative, inconclusive, or technically inadequate, PSG should be performed; this approach is based upon the observation that many home devices may miss nonobstructive events including arousals or limb movement that are seen in other complex sleep disorders, which could explain the patient's symptoms.
Pretest probability of mild OSA — For patients with suspected OSA who do not fit the definition of a high pretest probability of moderate to severe OSA per the AASM criteria, polysomnography is the preferred test. However, many insurers request that providers perform HSAT first, and if negative and the suspicion for OSA remains, then proceed to in- laboratory testing. (See 'Polysomnography' below.)
DIAGNOSTIC TESTS
Polysomnography — Attended, in-laboratory polysomnography (PSG) is considered the gold-standard diagnostic test for OSA. Diagnostic in-laboratory PSG (full-night or split-night) rather than home testing, is preferred in patients with the following (algorithm 1):
●Suspected complicated OSA (eg, conditions associated with nonobstructive sleep-disordered breathing such as chronic obstructive pulmonary disease (table 13)) (see 'Suspected OSA complicated by other respiratory sleep disorders' above)
●Suspected nonrespiratory sleep disorders other than OSA (eg, narcolepsy (table 12)) (see 'Suspected nonrespiratory sleep disorders' above)
●Uncomplicated OSA that does not fit the definition of suspected moderate or severe OSA by the American Academy of Sleep Medicine (AASM) criteria (see 'Pretest probability of mild OSA' above)
●Patients with suspected OSA who are mission-critical workers (eg, airline pilots) (see 'Mission-critical workers with suspected OSA' above)
●Patients with negative, inconclusive, or technically inadequate home testing in whom the suspicion for OSA remains high (see 'Home sleep apnea testing' below and "Home sleep apnea testing for obstructive sleep apnea in adults")
●Patients in whom home testing is not feasible (eg, physical impairment precluding HSAT)
Full-night study — Full-night PSG involves monitoring the patient during the patient's typical sleep period, which is generally nighttime. For patients who are diagnosed with OSA during a full-night study and who choose positive airway pressure (PAP) therapy, a return visit to the sleep laboratory is sometimes required for another study, during which PAP therapy is titrated. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea".)
Split-night study — Split-night PSG is similar to full-night testing, except the diagnostic portion of the study is performed during the first part of the night only. Those patients who are diagnosed with OSA during the first part of the night and choose PAP therapy can have PAP therapy titrated during the second part of the night.
Many experts consider the full-night test as the gold-standard but split-night studies are commonly used in practice.
●According to the AASM practice parameters for PSG and related procedures, a split-night study is a valid alternative to full-night diagnostic PSG (followed by a second full night of PAP titration) if the following criteria are met [65]: moderate to severe degree of OSA is observed during a minimum of two hours of recording time on the diagnostic polysomnogram, and at least three hours remain available for continuous positive airway pressure (CPAP) titration.
●Additional criteria set out by the Centers for Medicare and Medicaid Services (CMS) state that the apnea-hypopnea index (AHI) be >15 per hour for ≥2 hours of testing and ≥3 hours of sleep time is remaining for CPAP titration [72]. Per CMS, CPAP titration can be achieved in the majority of cases in one night and is considered the current standard approach. Despite CMS guidelines, insurance requirements for split-night studies are variable and need to be checked if such testing is contemplated. Moreover, many insurance companies will not issue a CPAP billing code for the first night of a diagnostic study, limiting coverage for a split-night protocol.
●Some experts also specify PSG-documentation of the elimination or near-elimination of obstructive events with PAP during rapid eye movement (REM) and non-REM (NREM) sleep. This should include REM sleep in the supine position, when apneas are most likely to occur [73]. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea".)
Further details on the therapeutic implications of split-night studies are provided separately. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea", section on 'Initial settings and titration during polysomnography' and "Titration of positive airway pressure therapy for adults with obstructive sleep apnea".)
Choosing between full- and split-night studies — Although full-night studies are the gold-standard, split-night studies are commonly chosen since, in most instances, they can provide an accurate appraisal of disease severity and establish the correct level of PAP in a single night in the majority of patients [71]. In general, the majority of patients, when educated to the options of testing, prefer a split-night protocol due to the convenience and an opportunity to start therapy after the test. Split-night PSG also decreases health care costs, minimizes scheduling delays, and does not appear to adversely affect compliance, despite decreased available time to educate the patient during the night [74]. Clinical practice guidelines also suggest that, if clinically appropriate, a split-night diagnostic protocol, rather than a full-night diagnostic protocol for PSG be used for the diagnosis of OSA [65].
However, while a split-night study is typically the default, the decision may be changed by the sleep technician supervising the study, depending upon his or her observations of sleep-related events as the study progresses. Alternatively, the sleep technician may proceed with full-night testing in a patient who does not meet the AASM criteria for a split-night study or in whom a complex sleep disturbance is noted during the initial phase of the study. The decision may also be affected by the clinician's suspicion that a full-night study is required. For example, based on the guidelines, patients who are suspected to be mild are not appropriate for split-night testing. Patients with associated complaints of insomnia also do not do well with a split-night protocol. Also, full-night studies may be preferred in patients in whom a more complex sleep disturbance is suspected (eg, central sleep apnea) or those in whom a split-night study has failed to elicit a clear diagnosis of OSA. Alternatively, some of the patients who are appropriate for split-night testing include severely symptomatic patients, older adults, those who are mentally and physically handicapped, shift-workers needing special time arrangements, and patients with difficult schedules due to family and work. Further details regarding CPAP titration are discussed separately. (See "Titration of positive airway pressure therapy for adults with obstructive sleep apnea" and "Mode selection for titration of positive airway pressure in adults with obstructive sleep apnea".)
Despite its reputation as the gold-standard test, negative in-laboratory polysomnography should be repeated if the clinical suspicion for OSA is high. This is supported by several studies that have demonstrated significant night-to-night variability in PSG results [75-77] and is supported by the AASM [65].
Polysomnography (including technique, measured variables, and derived information) is discussed in detail separately. (See "Overview of polysomnography in adults" and "Polysomnography in the evaluation of sleep-disordered breathing in adults".)
Home sleep apnea testing — In patients with a high pretest probability of moderate to severe uncomplicated OSA (ie, absence of another condition known to increase the risk of sleep-disordered breathing (table 13)), home sleep apnea testing (HSAT) with an adequate device is an appropriate alternative to PSG provided there is no suspicion for nonrespiratory sleep disorders (eg, narcolepsy (table 12)), the patient is not a mission-critical worker, and a sleep expert is available to interpret it (algorithm 1 and table 11) [65,78]. Importantly, should a single HSAT be negative, inconclusive, or technically inadequate, and the suspicion remains for OSA, HSAT should not be repeated (although some insurers prefer a second negative HSAT before proceeding with PSG). Rather, it is strongly recommended that an in-laboratory study be performed, due to the higher likelihood that a second test will also be negative, inconclusive, or technically inadequate. (See 'Pretest probability of moderate to severe OSA' above.)
There are a variety of devices that are used for in-home, unattended monitoring of cardiorespiratory parameters during sleep, few of which typically include electroencephalography. Many HSAT devices have been validated against standard PSG, typically by testing the same patient with both modalities in the sleep laboratory. In general, the sensitivity and specificity seem to be high in populations at high risk of moderate to severe OSA on the basis of clinical symptoms, assuming there are no significant comorbid medical disorders that affect sleep or nonrespiratory sleep disorders. Data that support HSAT as a viable alternative to in-laboratory PSG in this population include the following:
●One unblinded trial randomly assigned 373 patients with suspected moderate to severe OSA to either in-laboratory PSG with conventional CPAP titration or to HSAT with auto-titrating CPAP and subsequent transition to fixed CPAP [70]. Among patients who had moderate to severe OSA (ie, AHI >15 events per hour), there were no differences in the number of effective titrations, titration pressures, time to treatment, or indices of daytime sleepiness. Patients who underwent HSAT with auto-titrating CPAP were more likely to use CPAP for ≥4 hours at night.
●A randomized controlled trial compared HSAT for diagnosis followed by one night of auto-CPAP (APAP) titration against a traditional in-laboratory approach in managing 172 clinic patients with suspected OSA [79]. Modified intention-to-treat analysis of those with AHI ≥15/hour on CPAP reported that HSAT resulted in a similar Epworth sleepiness score and greater improvement in Sleep-Apnea-Quality-of-Life-Index at three months. In addition, the waiting time for all steps from diagnosis through treatment was significantly shorter than in the group with in-laboratory testing.
●A multicenter randomized noninferiority study compared limited data obtained from PSG with full polysomnographic data [80]. Data from PSG studies were given to sleep physicians at varying levels: level 1 (L1; full PSG data), level 3 (L3; information on airflow, thoracoabdominal bands, body position, electrocardiography, oxygen saturation which simulates data included from in-home studies), and level 4 (L4; oxygen saturation and heart rate). Diagnoses and treatment recommendations were made based upon the information provided and outcomes including the functional outcomes of sleep questionnaire score (FOSQ) were measured. There was no difference in the FOSQ score or in the distribution of initial diagnoses with L1, L3, or L4 testing. However, L4 testing was associated with less improvement in sleepiness, less CPAP use, and lower physician diagnostic confidence. While this study supports in-home-style testing, none of the studies were done at home, which was a major limitation. Further validation with in-home testing is needed.
There are several commercially available devices that can be used for home testing, not all of which compare adequately with PSG such that only devices that are considered adequate should be used (eg, Type 3 devices). Further details regarding devices are provided separately. (See "Home sleep apnea testing for obstructive sleep apnea in adults", section on 'Sleep monitoring devices'.)
In-hospital testing — In response to increased requests for sleep medicine consultation, many hospital-based centers and respiratory departments now provide bedside monitoring for inpatients with a high pretest suspicion for sleep apnea (see 'Pretest probability of moderate to severe OSA' above). However, most experts question the accuracy of data obtained in the hospital, given the higher rates of co-morbidities that can affect sleep and paucity of data to support its use. Diagnosis and management of OSA in the inpatient setting is reviewed in more detail separately. (See "Obstructive sleep apnea and other sleep disorders in hospitalized adults", section on 'Obstructive sleep apnea'.)
DIAGNOSIS — The diagnosis of OSA is based upon the presence or absence of related symptoms, as well as the frequency of respiratory events during sleep (ie, apneas, hypopneas, and respiratory effort-related arousals [RERAs])
Criteria — The diagnostic criteria and indices used on official sleep study reports vary according to whether the data are polysomnography (PSG)- or home sleep apnea testing (HSAT)-derived (table 16):
●PSG – The diagnosis of OSA is confirmed if either of the two criteria below is present [69]:
•There are five or more predominantly obstructive respiratory events (obstructive and mixed apneas, hypopneas, or RERAs) per hour of sleep in a patient with one or more of the following:
-Sleepiness, nonrestorative sleep, fatigue, or insomnia symptoms
-Waking up with breath holding, gasping, or choking
-Habitual snoring, breathing interruptions, or both noted by a bed partner or other observer
-Hypertension, mood disorder, cognitive dysfunction, coronary artery disease, stroke, congestive heart failure, atrial fibrillation, or type 2 diabetes mellitus
•There are 15 or more predominantly obstructive respiratory events (apneas, hypopneas, or RERAs) per hour of sleep regardless of the presence of associated symptoms or comorbidities.
PSG data can generate two indices as quantitative measures of sleep-related obstructive events per hour of sleep:
•The apnea-hypopnea index (AHI = [apneas + hypopneas] / total sleep time in hours)
•The respiratory disturbance index (RDI = [apneas + hypopneas + RERAs] / total sleep time in hours)
Because of the inclusion of RERAs, the RDI classifies more patients as having OSA than does the AHI, using the same threshold values. However, there is no consensus on whether the AHI or RDI should be the gold-standard index for diagnosis and their use varies across insurance carriers. In our laboratory, we report both the AHI and RDI. There is conjecture that the AHI may correlate better with cardiovascular outcomes, while the RDI may yield more information about daytime sleepiness and symptoms. We prefer the RDI since this index imparts more information about sleep fragmentation (RERAs) in addition to apneas and hypopneas and adds greater sensitivity to the diagnosis of OSA. Outcome studies are needed to determine which index is preferred. Interpretation of polysomnography-identified apneas, hypopneas, and RERAs are provided separately. (See "Polysomnography in the evaluation of sleep-disordered breathing in adults".)
●HSAT – Most HSAT devices do not include electroencephalogram (EEG) monitoring, and therefore RERAs and hypopneas characterized by arousals cannot be reliably identified. Accordingly, the number of respiratory events per hour of recording time rather than total sleep time is used to generate the respiratory event index (REI) (table 16). In validated HSAT devices, the REI correlates well with AHI and RDI, but is typically lower since the denominator (ie, total recording time) is larger than total sleep time used to calculate AHI and RDI. Moreover, outcomes in properly selected high risk patients tested by HSAT are similar to patients undergoing in-laboratory studies. Evidence does show that the length of the recording time used to generate the REI with HSAT should be at least four hours [65,81]. With these caveats in mind, cutoff values that are used for REI to diagnose OSA are similar to those in whom in laboratory sleep testing is performed. Thus, patients with a REI ≥15 events per hour and REI 5 to 14 and symptoms is supportive of a diagnosis of sleep apnea. Patients with a negative study, inconclusive results or a technically inadequate study should be evaluated in a laboratory setting. Differences among portable devices are discussed separately. (See "Home sleep apnea testing for obstructive sleep apnea in adults".)
Classification of severity — Patients who meet criteria for a diagnosis of OSA are traditionally classified as having mild, moderate, or severe disease on the basis of the AHI and symptoms [82]. This classification is based on consensus, and depending on definitions of "hypopnea," there can be great variance in the AHI [83]. With increasing use of HSAT, similar stratification is being used for REI. This is, however, also arbitrary. Moreover, the literature is confusing, often interchanging AHI for RDI.
Although the AHI alone is not entirely predictive of the presence and severity of symptoms and complications, the following is a general description of patients in each category.
●Mild – Patients traditionally classified as having mild OSA are those with an AHI/RDI/REI between 5 and 14 respiratory events per hour of sleep. Such patients may be relatively asymptomatic or report sedentary (ie, passive) daytime sleepiness, becoming noticeable once the patient is unstimulated. The daytime sleepiness often does not impair daily life, although it may be recognized by family members. Alternatively, daytime sleepiness may become apparent to the patient only after it improves due to weight loss, alcohol abstinence, or treatment of OSA. The sleep stages and slow wave sleep are generally preserved in mild OSA. Even when asymptomatic, mild OSA is associated with increased risk of hypertension, and this becomes a stronger association at younger ages [84]. However, using the latest AASM definition of hypopnea, symptomatic patients with mild OSA are without increased cardiovascular risk [85].
●Moderate – Patients traditionally classified as having moderate OSA are those with an AHI/RDI/REI between 15 and 30 respiratory events per hour of sleep. Such patients are typically aware of daytime sleepiness and take steps to avoid falling asleep at inappropriate times (eg, taking a nap or avoiding driving long distances). They are able to continue their daily activities, but at reduced levels, and they may have an increased incidence of motor vehicle violations or accidents. Systemic hypertension may coexist. Sleep fragmentation is observed in moderate OSA, but sleep architecture (ie, the timing and percentage of sleep stages) is better conserved than with severe disease.
●Severe – Patients classified as having severe OSA are those with an AHI/RDI/REI greater than 30 respiratory events per hour of sleep. Such patients more often have daytime sleepiness that interferes with normal daily activities. They tend to fall asleep often during the day (in a sitting posture) and are at risk for accidental injury from sleepiness. Patients with severe OSA are at increased risk for all-cause mortality and a variety of cardiovascular comorbidities, including hypertension, coronary artery disease, and arrhythmias. (See 'Complications' below.)
Upper airway resistance syndrome — Upper airway resistance syndrome (UARS) occurs when airflow limitation due to increased upper airway resistance (ie, RERAs) induces arousals from sleep, leading to excessive daytime sleepiness [86]. More commonly, the presence of prolonged partial upper airway obstruction is a common phenotype of sleep-disordered breathing, and is underreported [87]. In one study of patients referred for polysomnography, 29.9 percent presented with OSA, whereas 10.8 percent had prolonged partial obstruction with a normal AHI [88]. During recordings of nasal airflow, periods of flow limitation longer than a hypopnea (minimum one to three minutes) are often used as indicative of sustained upper airway resistance [87]. However, there is no consensus about the optimal detection, proper measurement, or degree of clinical impact. A PSG rather than HSAT is recommended for the detection of UARS, since it is better at identifying prolonged flow limitation via the nasal cannula.
While UARS was previously classified as an independent disorder, it is now considered a type of OSA. It presents similarly to classic OSA but it is common in thin females with certain craniofacial abnormalities [89,90], and may be more common in females referred for OSA than has been appreciated [87]. Unlike patients with classic OSA, patients with UARS have few discrete apneas or hypopneas or episodes of desaturation, but do have prolonged flow limitation and evidence of arousals on PSG, albeit with less overall sleep fragmentation than OSA. Thus, the study may be interpreted as absent or mild OSA, requiring no treatment. Partial upper airway obstruction is treatable with nasal continuous positive airway pressure (CPAP) and patients have good adherence to therapy [87]. We have also used dental appliances with good success.
COMPLICATIONS — Patients with OSA are at increased risk for several adverse clinical outcomes.
●Drowsy driving and motor vehicle crashes – Motor vehicle accidents are two to three times more common among patients with OSA than without OSA [91]. (See "Drowsy driving: Risks, evaluation, and management".)
●Neuropsychiatric dysfunction – OSA can induce or worsen inattention, memory, and cognitive deficits which, together, can result in impaired executive function and increase the likelihood of errors and accidents [4,91-100]. Additional neuropsychiatric manifestations include moodiness and irritability as well as depression, psychosis, and sexual dysfunction [95,101-104]. (See "Insufficient sleep: Definition, epidemiology, and adverse outcomes" and "Risk factors for cognitive decline and dementia", section on 'Obstructive sleep apnea'.)
●Cardiovascular and cerebrovascular morbidity – Patients with OSA, particularly when it is moderate or severe and untreated, are at increased risk for systemic hypertension, coronary artery disease, cardiac arrhythmias, heart failure, and stroke. (See "Obstructive sleep apnea and cardiovascular disease in adults" and "Sleep-related breathing disorders and stroke".)
●Pulmonary hypertension or right heart failure – OSA is classically associated with group 3 pulmonary hypertension, particularly when OSA coexists with either obesity hypoventilation syndrome or an alternative cause of daytime hypoxemia (eg, chronic lung disease). Severe hypoxemia may also cause secondary polycythemia. (See "Pulmonary hypertension due to lung disease and/or hypoxemia (group 3 pulmonary hypertension): Epidemiology, pathogenesis, and diagnostic evaluation in adults".)
●Metabolic syndrome and type 2 diabetes – Patients with OSA have an increased prevalence of insulin resistance as well as type 2 diabetes and diabetes complications [105-108]. While this association can be manifested through shared risk factors such as obesity [109,110], an independent association between OSA severity, insulin resistance, and type 2 diabetes has been reported in several studies [107,111-115]. In one study, about 12 percent of patients with OSA developed diabetes over a 67-month period; patients with severe OSA (apnea-hypopnea index [AHI] ≥30 events per hour) had an approximately 30 percent higher risk of incident diabetes compared with patients without OSA [107]. In patients with the metabolic syndrome, OSA has been independently associated with increased glucose and triglyceride levels as well as markers of inflammation, arterial stiffness, and atherosclerosis, suggesting that OSA may exacerbate the cardiometabolic risk attributed to obesity and the metabolic syndrome [116]. (See "Metabolic syndrome (insulin resistance syndrome or syndrome X)".)
●Nonalcoholic fatty liver disease (NAFLD) – Patients with OSA, particularly those with severe OSA, have a two- to threefold increased prevalence of NAFLD that is independent of shared risk factors such as obesity [117-120]. (See "Epidemiology, clinical features, and diagnosis of nonalcoholic fatty liver disease in adults" and "Pathogenesis of nonalcoholic fatty liver disease".)
●Miscellaneous – Patients with OSA may have an increased risk of developing gout compared with patients who do not have OSA (4.9 versus 2.5 percent) [121]. One large retrospective study from a French cohort suggested a possible association between cancer and nocturnal hypoxemia in patients being investigated for OSA [122].
SCREENING QUESTIONNAIRES — We agree with the American Academy of Sleep Medicine (AASM) that screening tools, including questionnaires should not be routinely used in asymptomatic patients in the community to screen for OSA [65]. However, these tools, in particular, the STOP-Bang questionnaire, are being increasingly used as preoperative evaluation tools to evaluate those without a known diagnosis of OSA who may be at risk of perioperative complications. Further details in this population are provided separately. (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Screening with a questionnaire'.)
In general, these diagnostic tools display poor accuracy in the sleep clinic for any level of selected apnea-hypopnea index (AHI) stratification and are rarely if ever used in clinical practice by sleep experts. However, while nonsleep experts may use these tools in practice, their validity in that setting still remains weak [59,65].
Efficacy — Several clinical prediction rules, questionnaires, and scores have been evaluated as screening tools using common signs and symptoms of OSA that are easily obtained and interpreted in the primary care setting. While these tools may have some value in detecting OSA in highly symptomatic patients in a high-risk setting (eg, preoperative evaluation), they cannot be used as true screening tools in asymptomatic patients based upon the following:
●Most tools have been derived from sleep center or preoperative referral-based populations (ie, symptomatic patients), which likely have a higher prevalence of OSA. Thus, the sensitivity and specificity of most instruments are likely overestimated in relation to their value in an unselected asymptomatic community-based population.
●There is marked heterogeneity in OSA definitions in studies testing the validity of screening questionnaires [47,123].
●The high false positive rate of questionnaires diminishes their diagnostic value such that when the score is high, patients do not always have OSA; in contrast, when the score is low, patients are unlikely to have OSA [124]. Thus, questionnaires are sensitive but not specific and are generally only helpful when they are negative (ie, low score).
Highlighting their limited use as a screening tool in asymptomatic patients, the United States Preventive Services Task Force (USPSTF) found no randomized trials comparing screening with no screening in asymptomatic individuals and no studies that reported improved outcomes in association with the use of screening questionnaires. The USPSTF reported that evidence was insufficient to assess the balance of benefits and harms of screening for OSA in the general adult population [125]. For similar reasons, the AASM also supports not using these tools to screen asymptomatic patients for OSA [65].
Instruments — Examples of such questionnaires include the following:
●STOP-Bang – The STOP-Bang questionnaire is an eight-item survey that incorporates information on snoring, tiredness, observed apneas, blood pressure, Body mass index (BMI), age, neck circumference, and gender (table 6). A score of three or higher has a sensitivity and specificity of 84 and 56 percent for the diagnosis of OSA using an AHI threshold of 5 to 14 events per hour, and a sensitivity and specificity of 93 and 43 percent for an AHI of 15 to 29 (ie, moderate to severe OSA) [124,126]. Among the questionnaires, the STOP-Bang questionnaire has the highest sensitivity (88 percent for mild OSA, 90 percent for moderate OSA, and 93 percent for severe OSA) but the specificity is poor (42 percent for mild OSA, 36 percent for moderate OSA, 35 percent for severe OSA) [124,127]. One meta-analysis of 47 studies from several geographic regions reported that a STOP-Bang score of at least 3 had sensitivity >90 percent and had the discriminative power to exclude moderate to severe and severe OSA, with negative predictive values of 77 and 91 percent, respectively [128]. The diagnostic accuracy of a STOP-Bang score of at least 3 to detect moderate to severe OSA was maintained in all regions except East Asia. A modified version is used by some experts as a preoperative evaluation tool (calculator 2) (table 7). (See "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Screening with a questionnaire'.)
●Sleep apnea clinical score (SACS) – The SACS is a four-item questionnaire that incorporates information on neck circumference, hypertension, habitual snoring, and nocturnal gasping or choking to generate a score ranging from 0 to 100 (table 10) [129,130]. Scores greater than 15 result in a probability of OSA (defined as an AHI >10 events per hour) of 25 to 50 percent.
●Berlin questionnaire – The Berlin questionnaire consists of 10 items relating to snoring, nonrestorative sleep, sleepiness while driving, apneas during sleep, hypertension, and BMI [131]. The results stratify patients as having a high or low risk for OSA. A high risk score on the Berlin questionnaire is associated with a sensitivity and specificity of 80 and 46 percent when OSA is defined as an AHI of 5 to 14 events per hour, and 91 and 37 percent when OSA is defined as an AHI ≥15 events per hour (table 9 and table 8) [124,131].
●The NoSAS score – The NoSAS score assigns points based upon five parameters (neck circumference, body mass index, snoring, age, and sex). In a derivation and validation analysis, a score ≥8 identified individuals at risk of clinically significant sleep-disordered breathing with an area under the curve (AUC) of 0.74, higher than that associated with the STOP-Bang (AUC 0.67) and Berlin questionnaires (AUC 0.64) [132]. While encouraging, this score requires further validation before it can be used in a clinical setting.
●Multivariable Apnea Prediction (MVAP) instrument – The MVAP instrument is based on a formula consisting of three questions about the frequency of symptoms of sleep apnea, along with BMI, age, and sex. MVAP values ranged from 0 to 1, with 1 representing the highest likelihood of sleep apnea. Two studies were performed using this instrument in primary care and community populations (older adult patients complaining of sleepiness, and hypertensive patients visiting an internist, respectively) [133,134]. In the Medicare patients with daytime sleepiness, this tool had a sensitivity of 91 percent and a specificity of 64 percent to predict severe OSA (AHI ≥30 and Epworth Sleepiness Scale [ESS] score >10) [133]. The MVAP tool had a sensitivity of 92 percent and a specificity of 44 percent to predict severe OSA in hypertensive patients [134]. However, both study populations likely carry a higher prevalence of OSA, creating a bias in these study results.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sleep-related breathing disorders in adults".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Sleep apnea in adults (The Basics)")
●Beyond the Basics topics (see "Patient education: Sleep apnea in adults (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Definition and epidemiology – Obstructive sleep apnea (OSA) is a disorder that is characterized by obstructive apneas, hypopneas, and/or respiratory effort-related arousals (RERAs) caused by repetitive complete or partial collapse of the upper airway during sleep. The prevalence of OSA in the general adult population varies depending on the definition but is approximately 15 to 30 percent in males and 5 to 15 percent in females (when OSA is defined as an apnea-hypopnea index [AHI] greater than five events per hour of sleep). (See 'Introduction' above and 'Epidemiology' above.)
●Risk factors – Well-defined risk factors for OSA include older age, male sex, obesity, and craniofacial and upper airway abnormalities. Potential risk factors include smoking, family history of snoring or OSA, and nasal congestion. Rates of OSA are also increased in association with certain medical conditions, most commonly obesity hypoventilation syndrome, pregnancy, end-stage kidney disease, congestive heart failure, chronic lung disease, type 2 diabetes mellitus, and stroke; others include acromegaly, hypothyroidism, polycystic ovary syndrome, and floppy eyelid syndrome. (See 'Risk factors and associated conditions' above.)
●Clinical manifestations
•History – Most patients with OSA complain of daytime sleepiness, or their bed partner reports loud snoring, gasping, choking, snorting, or interruptions in breathing while sleeping (table 1). Less common symptoms include nonrestorative sleep, nocturnal restlessness, morning headaches, sleep maintenance insomnia, and nocturia.
Key clinical questions (table 2) and the Epworth Sleepiness Scale (ESS) may facilitate the distinction between sleepiness and fatigue (calculator 1); an ESS score >9 indicates abnormal sleepiness and should prompt further testing.
•Examination – Examination findings include obesity, a crowded oropharynx, craniofacial abnormalities, and a large neck and/or waist circumference. Patients may also present with the manifestations of associated conditions and complications (eg, cognitive deficits, mood changes, sexual dysfunction, signs of pulmonary hypertension, motor vehicle crashes). (See 'Clinical features' above.)
●Differential diagnosis – The differential diagnosis of the manifestations of OSA is wide (table 5) and includes several conditions that present with daytime sleepiness (eg, periodic limb movement disorder, restless leg syndrome, narcolepsy, central sleep apnea, non-OSA related conditions associated with sleep-disordered breathing, and sedative drugs (table 3)), many of which can be distinguished from OSA via clinical history and polysomnography (PSG). Other conditions that similarly cause abrupt awakenings or abnormal sounds or sensations during sleep are in the differential and include primary snoring, gastroesophageal reflux disorder, swallowing disorders, nocturnal seizures, nocturnal asthma, insomnia, and panic attacks.
●Diagnostic evaluation – Guidelines for diagnostic testing include the following (see 'Diagnostic evaluation' above):
•Who to test – We recommend diagnostic testing in patients who have excessive daytime sleepiness and two out of three of the following:
-Habitual loud snoring
-Witnessed apnea or gasping or choking during sleep
-Diagnosed hypertension
We also frequently perform testing in patients who have excessive daytime sleepiness alone; patients who have other clinical features of OSA (table 1) and associated conditions or complications (eg, refractory hypertension); and patients in whom OSA needs to be ruled in or out as an underlying cause or potential contributing factor to their symptoms (eg, unexplained pulmonary hypertension or motor vehicle accident due to falling asleep).
Evaluation tools or questionnaires (eg, ESS, Berlin, STOP-Bang questionnaires) are not typically used to select patients for diagnostic testing since they are inaccurate and have not been shown to be superior to a good history and physical examination. (See 'Patient selection' above.)
•Type of test
-For patients with suspected mild OSA, suspected complicated OSA (eg, chronic obstructive pulmonary disease) (table 13), suspected nonrespiratory sleep disorders (eg, narcolepsy (table 12)), or patients with suspected OSA who have mission-critical jobs, we recommend full-night or split-night, attended, in-laboratory PSG rather than home sleep apnea testing (HSAT). This approach is based upon the inadequacy of most HSAT devices to detect complex or mild sleep-related events and the lack of evidence to support their use. (See 'Selecting home or in-laboratory testing' above and 'Mission-critical workers with suspected OSA' above and 'Suspected nonrespiratory sleep disorders' above and 'Pretest probability of mild OSA' above and 'Polysomnography' above.)
-For patients in whom there is a high likelihood of moderate or severe uncomplicated OSA (ie, without non-OSA sleep-related breathing disorders (table 13)) and who have no other suspected nonrespiratory sleep disorders (eg narcolepsy (table 12)), we suggest unattended, in-home portable monitoring with a technically adequate type 3 device, the results of which are interpreted by providers who have experience in clinical sleep medicine (algorithm 1 and table 11). (See 'Pretest probability of moderate to severe OSA' above and 'Home sleep apnea testing' above.)
-Several additional factors that influence the choice between performing a home or in-laboratory study include the values and preferences of the patient (eg, access to a sleep laboratory, homelessness, cost, space constraints, wait time, concern for operator error for home monitoring), the practices of the institution and preferences of the clinician, and the preference of the payer (eg, many insurers in the United States require that patients receive home testing first). (See 'Selecting home or in-laboratory testing' above.)
-Importantly, if either PSG or HSAT is negative and the suspicion for OSA remains, PSG should be repeated or performed, respectively. (See 'Diagnostic tests' above.)
●Diagnosis – The diagnosis of OSA is based upon the presence or absence of related symptoms, as well as the frequency of respiratory events during sleep (ie, apneas, hypopneas, and RERAs).
•PSG criteria – In patients who undergo PSG, the diagnosis of OSA is confirmed if either of the two criteria below are met (table 16) (see 'Diagnosis' above):
-There are 15 or more apneas, hypopneas, or RERAs per hour of sleep (ie, an AHI or respiratory disturbance index [RDI] ≥15 events per hour) in an asymptomatic patient.
-There are five or more obstructive apneas, obstructive hypopneas, or RERAs per hour of sleep (ie, an AHI or RDI ≥5 events per hour) in a patient with the following: sleepiness, nonrestorative sleep, fatigue, or insomnia symptoms; waking up with breath holding, gasping, or choking; habitual snoring, breathing interruptions, or both noted by a bed partner or other observer; and/or hypertension, mood disorder, cognitive dysfunction, coronary artery disease, stroke, congestive heart failure, atrial fibrillation, or type 2 diabetes mellitus.
•HSAT criteria – In patients who undergo HSAT, actual sleep time is not recorded. The respiratory event index (REI) is calculated based on recording time from the HSAT device. REI appears to correlate well with in-laboratory measured AHI and RDI. With a moderate or high suspicion of OSA, an REI ≥15 events per hour is supportive of a diagnosis of OSA, while those with an REI <15 events per hour or those with inconclusive results or a technically inadequate study should be evaluated with PSG.
●Complications – Adverse outcomes associated with OSA include drowsy driving and motor vehicle crashes, neuropsychiatric dysfunction, cardiovascular and cerebrovascular morbidities, pulmonary hypertension, metabolic syndrome and type 2 diabetes, and nonalcoholic fatty liver disease. (See 'Complications' above.)
●Screening for OSA – Screening tools such as questionnaires should not be routinely used in asymptomatic patients in the community to screen for OSA. However, these tools (in particular, the STOP-Bang questionnaire (table 6)) are being increasingly used as preoperative evaluation tools to evaluate those who may be at risk of perioperative complications due to undiagnosed OSA. (See 'Screening questionnaires' above and "Surgical risk and the preoperative evaluation and management of adults with obstructive sleep apnea", section on 'Screening with a questionnaire'.)
