Sleep-Disordered Breathing and CPAP

Updated: Mar 02, 2022
Author: Vittorio Rinaldi, MD; Chief Editor: Zab Mosenifar, MD, FACP, FCCP 


Practice Essentials

Upper-airway obstruction occurring during sleep—that is, sleep-disordered breathing (SDB)—was first demonstrated in the 1960s. SDB represents a group of physiopathologic conditions that are characterized by an abnormal respiratory pattern during sleep that can be isolated or can coexist with other respiratory, nervous, cardiovascular, or endocrine diseases. SDB is now known to be widely prevalent in the general population, and it is responsible for or contributes to numerous problems, ranging from fragmented sleep patterns to hypertension to traffic accidents.[1]

SDB includes obstructive sleep apnea (OSA),[2] which consists of breathing cessations of at least 10 seconds occurring in the presence of inspiratory efforts during sleep. Central sleep apnea (CSA) is a distinct condition that consists of similar apneas[3] ; however, these instead take place in the absence of inspiratory efforts.

OSA syndrome (OSAS) is a potentially disabling condition characterized by excessive daytime sleepiness,[4]  disruptive snoring, repeated episodes of upper-airway obstruction during sleep, and nocturnal hypoxemia. It is defined by an apnea-hypopnea index (AHI; the total number of episodes of apnea and hypopnea per hour of sleep), or respiratory disturbance index (RDI), of 5 or higher in association with excessive daytime somnolence.

Risk factors for sleep apnea include obesity, increased neck circumference, craniofacial abnormalities, hypothyroidism, and acromegaly. Daytime consequences include not only excessive sleepiness but also impaired cognitive performance and disturbed moods with a reduced quality of life. Excessive daytime sleepiness is reported to be associated with a higher risk of motor vehicle accidents and work place injuries or poor work performance.

In general, everyone with SDB snores, but not everyone who snores has SDB. Snoring in the absence of SDB is termed primary or simple snoring. However, some evidence indicates that snoring is one end of a clinical continuum, with severe OSA at the opposite end. Some health problems may be associated even with primary snoring. The relation between snoring sounds and severity of OSAS has been widely investigated.[5]

Upper-airway resistance syndrome (UARS) is characterized by snoring with increased resistance in the upper airway, resulting in arousals during sleep. This can disturb sleep architecture to the point of causing daytime somnolence. No distinct diagnostic criteria exist for this entity. Patients with UARS can be treated with continuous positive airway pressure (CPAP)—specifically, nasal CPAP (n-CPAP).[6, 7]

Treatment of SDB involves elimination of contributing factors and provision of n-CPAP. n-CPAP is effective in improving sleep quality and reducing daytime sleepiness. Long-term treatment with n-CPAP reduces both mortality and the acute blood pressure elevation that occurs with SDB.[8]  Over time, a trend develops toward baseline blood pressure reduction in hypertensive patients with SDB. Medical and surgical interventions may also be indicated.

Go to Obstructive Sleep Apnea, Childhood Sleep Apnea, Central Sleep Apnea Syndromes, Obstructive Sleep Apnea and Home Sleep Monitoring, Surgical Approach to Snoring and Sleep Apnea, Oral Appliances in Snoring and Obstructive Sleep Apnea, and Upper Airway Evaluation in Snoring and Obstructive Sleep Apnea for more information on these topics.

For patient education information, see the Sleep Disorders Center, as well as Snoring and Narcolepsy.


Any factors that decrease upper-airway size or patency during sleep can lead to intermittent obstruction during inspiration, despite inspiratory effort. If the obstruction is sufficiently prolonged, blood oxygen levels drop. Then, the patient arouses or awakens. The arousals disrupt normal sleep architecture. These, together with the oxygenation drops, are responsible for the more severe conditions that can accompany SDB, including hypertension, arrhythmias, and death.[9]

Factors affecting upper-airway size or patency include numerous anatomic variants and abnormalities (eg, nasal obstruction, retrognathia, macroglossia), obesity, alcohol or sedative intake, and body position during sleep.

Obesity contributes to SDB by changing pharyngeal size and shape. Fat storage in the neck may be particularly associated with risk for SDB, though a subset of patients with SDB are of normal body weight. Many of these patients have a family history of snoring or SDB.

Alcohol intake near bedtime can cause or worsen SDB by reducing the activity of the upper-airway dilating muscles. Alcohol increases both the number and the duration of apneic or hypopneic events.

Racial studies and chromosomal mapping, familial studies, and twin studies have provided evidence for a possible link between OSAS and genetic factors. Genetic factors associated with craniofacial structure, body fat distribution, and neural control of the upper-airway muscles likely interact to produce the OSAS phenotype.[10, 11, 12]

Although the role of specific genes that influence the development of OSAS has not been fully defined, some research, especially in animal models, suggests that several genetic systems may be important.

Human leukocyte antigen (HLA)-DQB1*0602 allele, a well-known genetic risk factor for narcolepsy, has been described as a potential genetic factor influencing sleep physiology in individuals diagnosed with OSAS.[13]

Polymorphisms in the serotonin (5-HT) receptor gene can alter its transcription, affecting the number of receptors in the serotoninergic system, contributing to OSAS.[14]

A number of inflammatory factors, such as interleukin (IL)-6, IL-8, and tumor necrosis factor alpha (TNF-α), can be found in high concentrations in persons with OSAS and may serve as biologic markers of this disease. The concentration of these cytokines contributes to weight gain in patients with OSAS and can also modify the risk of obesity-related metabolic disorders, especially insulin resistance.[15]


Important clinical risk factors for SDB are as follows:

  • Nasal obstruction
  • Craniofacial abnormalities
  • Mandibular retrognathia
  • Micrognathia
  • Narrowed, tapered, and short maxillary arch
  • Overbite
  • Long soft palate
  • Modified Mallampati grade III or IV
  • Macroglossia
  • Tonsillar hypertrophy
  • Neck circumference more than 17 in. for men and more than 16 in. for women
  • Obesity

Other problems that can contribute to or exacerbate SDB are sedative or alcohol use and poor sleep hygiene.

A very small percentage of patients with SDB have CSA rather than OSA. CSA can be caused by various neurologic disorders or can be idiopathic.


Epidemiologic studies indicate that sleep apnea is more common in men than in women (male-to-female ratio, 2-3:1). Sleep apnea occurs in 4% of men and 2% of women aged 30-60 years. A retrospective study on 830 patients with OSAS reported a male-to-female ratio (M:F) that increases with the gravity of the disease: 2.2:1 in mild OSAS and 7.9:1 in severe OSAS.[16, 17]  Hypersomnolence is reported with a percentage of 16% in men and 22% in women; 24% of men and 9% of women have an apnea-hypopnea index of at least 5.

The discrepancy between the lower prevalence of OSA, the greater frequency of obesity, and the smaller airway size in women compared with men suggests that a gender difference underlies this condition.

Men tend to have a larger but more collapsible airway during mandibular movement than women, and this may play a partial role in the positional dependency and severity of OSA in men.

Another possible reason for the lower prevalence of OSAS may be reluctance on the part of many women to report symptoms mostly considered inappropriate, like snoring; this reluctance may cause a clinical underestimation of the problem in females.

The gender-related protective effect decreases in females who are postmenopausal and not on hormone replacement therapy.[18, 19]

The association between age and OSA is complex. Several studies have shown a higher prevalence of OSA in elderly persons than in middle-aged persons, though daytime symptoms may be less common with advancing age.

The Sleep Heart Health Study demonstrated that the influence of male sex and body mass index (BMI) on OSA tends to wane with age. For unclear reasons, the overall prevalence of OSA plateaus after age 65 years.[20]

The prevalence of OSAS among African-American persons seems to be at least equal to and possibly greater than that among white persons. The prevalence among men in urban India and men and women in Korea is similar to that observed in Western countries. Some researchers have noticed an increased incidence of OSA in persons of Asian origin.


Excessive daytime sleepiness resulting from SDB can impact focus and concentration, causing decreased work effectiveness. Even mild-to-moderate SDB lengthens reaction time, causing performance decreases similar to alcohol intoxication. This can lead to motor vehicle accidents and other serious accidents in situations where alertness is required for safety (eg, heavy machinery operators).

Moderate-to-severe OSA is associated with earlier death. The cardiovascular sequelae of untreated OSA include hypertension, cor pulmonale, arrhythmias, and increased risk of myocardial infarction or stroke.[21, 22, 23, 24, 25, 26] SDB is associated with higher levels of IL-6, a marker of myocardial infarction risk and mortality.[27] Adiposity may mediate the increased levels of C-reactive protein (CRP), fibrinogen, intercellular adhesion molecule (ICAM)-1, and P-selectin observed in SDB.[27]

OSA is associated with difficult-to-control hypertension.[28] CPAP also reduces markers of hypercoagulability, and this is a potential mechanism by which it can reduce the rate of cardiovascular morbidity and mortality in OSAS patients.[29]

In heart failure patients with sleep apnea, studies have not shown the use of PAP to reduce the risks of cardiovascular outcomes or death; however, such therapy has been associated with some improvements in OSA symptoms.[30]

Treatment of OSA may reduce new first-time cerebrovascular events and recurrences.[31]  A study by Gupta et al suggested that in patients with stroke and OSA, CPAP treatment can yield significantly better stroke outcomes and statistically nonsignificant favorable outcomes in terms of recurrence of vascular events.[32]

Untreated OSA has been associated with cognitive deficits and changes in brain electrophysiology, and there is evidence to suggest that CPAP may mitigate these effects.[33]

Many of the studies examining the relation between OSA and glucose tolerance have shown a direct and independent relation between OSA and diabetes. The Wisconsin Sleep Study Cohort showed a greater prevalence of diabetes in subjects with increasing levels of OSA.[34]  Several studies have shown a beneficial effect of CPAP therapy on insulin resistance or glucose levels.[35, 36, 37]

The probable mechanisms connecting OSA with glucose tolerance and type 2 diabetes mellitus include increased sympathetic activity, sympathovagal dysfunction, alterations in neuroendocrine function (especially in growth hormone [GH] and cortisol levels), and a high inflammatory state with an increase in the release of proinflammatory cytokines.[38, 36, 37, 39, 40]




The first clue in the history of patients with sleep-disordered breathing (SDB) is loud snoring. This is accompanied by breathing cessation; gasping, choking, and snorting; frequent arousals from sleep; and respiratory effort with no air. Nocturnal arrhythmias and acute blood pressure increases may occur. Morning headaches that dissipate as the day goes on, excessive daytime sleepiness, and poor concentration affect daytime performance. The disorder has been linked to an increased risk of angina, myocardial ischemia, stroke and motor vehicle accidents.

Older men may report getting up numerous times during the night to urinate and are convinced that they awaken because of the urge to urinate. The truth is often the reverse—namely, that they first awaken as a result of SDB and only then notice the urge to urinate. These patients are often surprised at their decreased need for nocturnal urination after successful SDB treatment.

Laryngopharyngeal reflux can cause a patient to suddenly awaken from sleep, gasping for breath. A feeling of terror is often present.

Inadequate sleep time can cause excessive daytime sleepiness. This may be involuntary, as in insomnia, or voluntary. Insomnia is characterized by the inability to fall asleep or awakening during the night and being unable to fall back to sleep. Inadequate sleep time occurs for other voluntary reasons (eg, working more than one job, family responsibilities).

Patients with hypothyroidism can also present with fatigue, daytime somnolence, and obesity. SDB and hypothyroidism can coexist.[41]

Narcolepsy can also cause excessive daytime sleepiness.

Physical Examination

Most patients with SDB are overweight or obese. A patient with a short, thick neck may be predisposed to SDB. Scalloped indentations along the lateral tongue (from teeth) are a marker for relative tongue/mandibular arch size mismatch, which may predispose individuals to SDB.

Children with obstructive sleep apnea (OSA) syndrome (OSAS) are likely to present with normal body weight, tonsillar hypertrophy, and inattentiveness during school classes.[42]  However, OSAS in children is not discussed in this article.


If not adequately diagnosed and treated, OSAS is associated with severe complications such as hypertension, strokes, coronary disease, and neurobehavioral complaints and is probably a predictor of premature death. At least 50% of patients with heart failure have sleep respiratory apneas, and patients with moderate-to-severe OSAS have a threefold higher risk of developing hypertension.[21, 22, 23, 24, 25]



Diagnostic Considerations

The differential diagnosis of sleep-disordered breathing (SDB) includes the following:

  • Simple snoring
  • Other disorders that cause daytime sleepiness (eg, insufficient sleep, a circadian-rhythm abnormality, narcolepsy, periodic limb movement disorder).


Laboratory Studies

The relation between snoring, obstructive sleep apnea (OSA), and hypothyroidism has been confirmed by many authors. Thyroid-stimulating hormone (TSH) levels should be determined in patients who are newly diagnosed with sleep-disordered breathing (SDB) because SDB is relatively common among patients with hypothyroidism. Significant increases in homocysteine levels have been observed in OSA syndrome (OSAS) patients with cardiovascular disease.

Imaging Studies

Radiologic and diagnostic studies have been used to identify the obstruction site, to direct surgical intervention, and to predict outcomes of sleep apnea surgery. These studies include lateral cephalometric radiography, computed tomography (CT), magnetic resonance imaging (MRI), asleep fluoroscopy, asleep and awake endoscopy with Mueller maneuver, upper-airway manometry, and acoustic reflection techniques.[43]

Most of those techniques have limitations (dynamic and tridimensional evaluation) with respect to the investigation of the mechanism of occlusion. Ultrafast MRI provides a reliable and noninvasive method for static and dynamic evaluation of the soft-tissue structures surrounding the upper airway during the respiratory cycle in wakefulness and sleep.

Epworth Sleepiness Scale

The Epworth Sleepiness Scale is a questionnaire filled out by the patient that is used to provide a standardized semiquantitative subjective assessment of daytime sleepiness.

In this questionnaire, patients are instructed to rate the chance of dozing off in a number of different situations. They are to choose the most appropriate ranking for each of these situations, working out how they would probably respond if it is something they have not actually done recently. Scoring for the Epworth Sleepiness Scale is shown in Table 1 below.

Table 1. Epworth Sleepiness Scale Questionnaire (Open Table in a new window)


0 - Would never doze off

1 - Slight chance of dozing off

2 - Moderate chance of dozing off

3 - High chance of dozing off

Score situation

_____ Sitting and reading

_____ Watching TV

_____ Sitting inactive in a public place (eg, theater, meeting)

_____ As a passenger in a car for an hour without break

_____ Lying down to rest in the afternoon when circumstances permit

_____ Sitting and talking to someone

_____ Sitting quietly after a lunch without alcohol

_____ In a car, while stopped for a few minutes in the traffic

_____ Total*

*A total score of 0-5 is supernormal; 5-10 is normal; 10-15 is sleepy; 15-20 is very sleepy; and >20 is dangerously sleepy (arrange transportation for patient)


Polysomnography (PSG) is the criterion standard diagnostic test for OSAS. A respiratory event suggestive of OSAS is defined as a decrease in nasal and oral airflow, alone or with thoracoabdominal movements, of more than 90% (apnea) or of more than 50% but less than 90% (hypopnea) that lasted for at least 10 seconds. A decrease in arterial oxygen saturation of 4% or more is considered significant oxygen desaturation.[44]

Information from PSG is reported in the form of the respiratory disturbance index (RDI; also referred to as the apnea-hypopnea index [AHI]). The RDI is the number of apneas or hypopneas 10 seconds or longer occurring per hour of sleep. A normal RDI is less than 5. An RDI less than or equal to 5 is suggestive of simple snoring with no OSAS. An RDI greater than 5 and less than or equal to 15 is suggestive of mild OSAS. An RDI greater than 15 and less than or equal to 30 is suggestive of moderate OSAS. Finally, an RDI greater than 30 is suggestive of severe OSAS.

The loudness and persistence of snoring (constant versus intermittent) are usually reported. Body position is also recorded so one can determine what position (usually supine) and in what sleep phase (usually rapid-eye-movement [REM] sleep, when muscle tone is most relaxed) the patient is in when respiratory events occur.

In-laboratory PSG is the criterion standard for diagnosing OSAS. However, PSG has several limitations, including the necessity of performing the test in a sleep laboratory, high costs, the considerable technical expertise required, and the long analyzing time needed by the operator. In addition, the examination often must be repeated because of the interference of monitoring electrodes with the physiologic sleep of the patient (“first night effect”). Therefore, timely access to PSG is often a problem.

Screening Questionnaires

Because PSG is expensive and not widely available, there has been extensive interest in alternative diagnostic approaches, such as clinical prediction rules and portable monitors. A limited number of questionnaires are available to detect some sleep disorders, but those instruments do not achieve the reliability of PSG, which remains the recommended method of assessing patients with suspected sleep disorders. The role of those questionnaires is mainly to serve as a screening tool for identifying patients at risk for OSAS.[45, 46, 47, 48]

Many questionnaires have been developed for screening OSAS. The Rome Questionnaire (RQ) is a seven-item questionnaire useful in identifying adult patients at risk for OSAS. The RQ, together with the body mass index (BMI), is reported to be a useful tool for selecting patients at higher risk for moderate-to-severe OSAS, who need a prompt PSG evaluation.[49]

Home Sleep Testing and Ambulatory Monitors

Home sleep testing pursues the goal of simplify the diagnosis of sleep apnea while retaining the essential recording features of PSG. There is some evidence to suggest that home sleep studies have benefits in terms of time and cost, but for diagnostic reliability, an in-laboratory sleep study may be required in more than half of the cases.

Various types of ambulatory (to be used at home) monitors can measure parameters such as airflow, chest, and abdomen movements (as indicators of respiratory effort); oxygen desaturations; snoring; pulse; and body position.[50, 51]

Although the data from such studies are not as detailed or accurate as those obtained from an overnight PSG, these studies can often be used to differentiate primary snoring from snoring with apneas and can usually provide an indication of the frequency with which apneas are occurring. In contrast, techniques that measure only one parameter (eg, home oximetry alone) seem to be less accurate than those that track several measurements.

The EdenTrace portable monitor measures nasal and oral air flow using thermistors, chest wall impedance, oxygen saturation with finger pulse oximetry, heart rate, and movement detected by electrical comparison of the signals from electrocardiography and pulse oximetry.

The MESAM IV system evaluates SDB on the basis of analysis of snoring, heart rate, and saturation change. Even if in many studies there is a good agreement between the RDI measured in the laboratory and that measure with home sleep testing, there is a risk that ambulatory diagnostic procedures may alter patients' relationship with their disease, the medical staff, or both in such a way that their subsequent compliance with treatment may be decreased.

The Nightwatch system has the ability to calculate the RDI. It records eye movement (one channel, piezo electrode), leg movement (one channel, piezo electrode), (finger pulse oximeter), nasal oral airflow (thermistor), chest and abdominal movements (piezo electrodes), body position and movement (mercury gauge placed on the chest), and heart rate.

The Nightwatch system also has the ability to send 2-minute portions of the complete recording to the laboratory for analysis so that signal quality can be assessed and transducer function corrected if necessary. However, further studies are necessary before this technology can be put into widespread use.

With the aim of developing a simpler, cheaper, and more accessible method for the diagnosis of OSAS, the peripheral arterial tonometer (PAT) has been proposed for use in ambulatory diagnosis of OSAS. The portable monitoring device WatchPAT 200 detects obstructive events by identifying the changes in sympathetic activity associated with the termination of the events. The wrist-worn device WatchPAT 200, compared with standard PSG, has been reported to be able to detect OSAS on the basis of the RDI with comparable accuracy.[52]

Acoustic Snoring Analysis

Acoustic analysis of snoring sounds may help differentiate between primary snoring and OSA. Various researchers have investigated the combination of clinical variables such as neck and chest circumference, BMI, and resting room air oxygen saturation; however, none of these has been shown to differentiate consistently between primary snorers and patients with significant apneas and desaturations during sleep.

Sleep Endoscopy

Drug-induced sleep endoscopy (DISE) is a safe and reliable technique for determining the pattern of upper-airway obstruction and the contribution of specific structures to airway obstruction.[53]

Identifying the site and the dynamic pattern of obstruction is mandatory in therapeutic decision-making, especially if a surgical therapeutic option is being considered.[54]

The nose-oropharynx-hypopharynx-larynx (NOHL) classification, which could be applied during awake and sleep endoscopy, allows a simple, quick, and effective evaluation of grade and patterns of upper-airway collapse.[55]

Histologic Findings

The histology of the soft palate and uvula in snorers and patients with OSAS has been investigated by many authors.[56, 57]  Some authors observed muscular atrophy, dilatation and congestion of the blood vessels, lymphocytic infiltrations, and hypertrophy of superficial salivary glands localized between the muscle bundles and epithelium. Those histopathologic changes were related to the influence of the vibration on the soft palate and uvula and were considered responsible for the excessive flaccidity of these structures.

Other authors found contents of glands, muscle, fat, blood vessels, and the epithelium in the uvula and the soft palate to be similar in OSAS patients and control subjects.



Initial Treatment

Elimination of contributing factors

The first task in treating patients with sleep-disordered breathing (SDB) is to eliminate all possible contributing factors. This includes weight loss for patients who are obese and elimination of alcohol or sedative use, especially near bedtime. Benzodiazepines, narcotics, and barbiturates can worsen SDB, or sometimes they initiate it where it had not previously been present.

A 10% weight loss was associated with a 26% decrease in the respiratory disturbance index (RDI; also referred to as the apnea-hypopnea index [AHI]) in a population-based study. Weight loss should be recommended for all obese patients with sleep apnea; however, weight loss takes time, and only a minority of patients successfully maintain it.[58, 59]

Alteration of body positioning during sleep

Body positioning during sleep can improve SDB in some patients. Because lying supine can allow gravity to assist in pulling lax tongue muscles back toward the posterior pharyngeal wall, patients should sleep on their sides, on their stomachs, or propped up 60°. These positions can improve SDB in patients whose symptoms occur primarily while supine.

Avoidance of supine sleeping can easily be accomplished with a sock, tennis ball, and safety pins. The tennis ball in a sock is pinned to the back of the pajamas, positioning the tennis ball between the scapulae. When the patient rolls into the supine position during sleep, this lump is uncomfortable enough that the position is immediately shifted, usually without the patient awakening.

Thyroid hormone replacement therapy

In patients with hypothyroidism and SDB, thyroid hormone replacement therapy is usually accompanied by an improvement in the SDB.

Use of oral appliances

In some individuals, a mouthpiece may improve the anatomy of the airway to the point where snoring or mild obstructive sleep apnea (OSA) can be corrected. Oral appliances, or mandibular advancement devices (MADs), can be an effective alternative for mild and medium-to-moderate OSA syndrome (OSAS), but they require strict monitoring because of differences in individual response to this therapy.[60, 61]

Many types of oral appliances have been designed for the treatment of sleep apnea. Most are custom fitted to the teeth of both dental arches to reposition the mandible and to enlarge the retropalatal and retrolingual airway space. However, consistent patient tolerance for this treatment is relatively low, and it is less effective than continuous positive airway pressure (CPAP) in reducing the frequency of apnea and hypopnea.[62]

Restriction or elimination of alcohol use

Alcohol significantly worsens SDB. Eliminating use of alcohol, especially near bedtime, improves SDB.[63]

Nasal CPAP

When none of the above therapies are appropriate or helpful, nasal CPAP (n-CPAP) is the most effective method of managing OSAS.[64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 59, 78] n-CPAP provides a pneumatic stent for the upper airway, eliminating the airway collapse during inspiration. It is administered via a soft mask that covers the nose only. Sufficient pressure is introduced to eliminate apneas, hypopneas, and snoring. (See the image below.)

CPAP machine. CPAP machine.

Most physicians agree that patients with an RDI higher than 20 require treatment. n-CPAP can also be useful for patients with a lower RDI, especially if they experience daytime sleepiness or other symptoms. If the severity of the daytime symptoms and the Epworth Sleepiness Scale score are much greater than would be expected with a particular RDI, a trial of n-CPAP can help determine whether elimination of the SDB leads to improvement of the daytime symptoms or if other factors contribute to the daytime symptoms.

Patients who are unlikely to benefit from n-CPAP include those with such severe nasal obstruction that n-CPAP cannot be used, patients with such extreme claustrophobia that they cannot tolerate a nasal mask, and patients in whom n-CPAP does not reliably eliminate apneas, hypopneas, and snoring.

Determination of required positive airway pressure

The criterion standard for determining the amount of pressure required to restore upper-airway patency is traditionally determined during polysomnography (PSG) by trained technicians. In some centers, this is performed as a split-night study, with data from the first half of the night used for diagnosis of SDB. Once this diagnosis is made, if the RDI is high enough to suggest benefit from initiation of n-CPAP (usually ≥20), the second half of the night’s study is used to determine the optimal amount of pressure.

The disadvantage of the split-night approach is that the second half of a full-night study often reveals more severe sleep apnea, and thus a diagnostic study limited to the first half of the night can underestimate disease severity.

The amount of pressure delivered is reported in centimeters of water. An average starting point for CPAP would be 8-10 cm H2O. Patients report that pressures at these levels feel odd but are tolerable even at the beginning of treatment and become more tolerable as the patients become accustomed to treatment. Higher levels (>15 cm H2O) are often not well tolerated.

When a second overnight study is logistically difficult, some clinicians empirically start a patient on n-CPAP with a pressure of 8-10 cm H2O. Newer n-CPAP machines can sense, on the basis of the patterns of inspiratory airflow, the amount of pressure needed to overcome upper-airway resistance. Patients are sometimes started on these machines without a prior titration study.[79, 80]

Alternatively, an autotitrating machine can be used for several nights, the record of amount of pressure required to suppress apneas and hypopneas can be downloaded and studied, and a suitable nightly pressure can be determined in this fashion. Also, the amount of pressure required to suppress snoring can be used as an audible guide to appropriate pressures.

A patient who routinely takes sedatives or ingests alcohol during the evening and does not intend to change this should probably be tested after continuing their usual nightly routine. n-CPAP titration without sedatives or alcohol is likely to lead to undertreatment of the SDB at home, when such patterns are resumed.

Effects of n-CPAP

Most patients feel better during the daytime on the first day after beginning n-CPAP. During the first week of treatment, most experience rebound sleep with prolonged episodes of REM sleep. Sleep patterns become more normal after the first week. For these reasons, several weeks of n-CPAP use may be helpful for normalization of sleep patterns in patients with severe sleep apnea who plan to undergo surgery. Sleep patterns should be normalized before the planned surgical procedure.

Regular use of n-CPAP improves the quality of life for both patients and their bed partners.[81, 82, 83]  The treatment lessens depressive symptoms, and improves daytime functioning, blood pressure and insulin sensitivity. In patients with OSA who receive antihypertensive treatment, long-term CPAP was found to be responsible for a significant reduction of diastolic blood pressure. Asthmatic OSA patients have fewer nighttime symptoms.[84, 74, 85]

Other effects of using CPAP include increased vagal tone, increased cardiac output, increased stroke volume, decreased systemic vascular resistance, and reduced risk of cardiovascular mortality.[8]

Patients with OSA often have increased arterial stiffness and sympathovagal imbalance. CPAP therapy is reported to have beneficial effects on the vascular function in such patients: improvement of the sympathovagal balance by CPAP therapy may be significantly related to decreased stiffness of the central to middle-sized arteries, independent of the changes in the blood pressure and vascular endothelial status.[86]

Repetitive obstructive apnea produces acute impairment of left ventricular longitudinal function, suggesting the development of subendocardial ischemia. CPAP therapy not only decreases the severity of OSA but also ameliorates sleep-induced longitudinal left ventricular dysfunction.

Problems with n-CPAP

One problem with n-CPAP is that although this modality provides good improvement in symptoms and physiologic parameters, compliance with treatment is not good. The rate of regular use is sometimes estimated to be as low as 30% (46% in one study defining use as at least 4 hours/day, 5 days/week).

Noncompliance was classified by Zoula et al into the following categories[87] :

  • Tolerance problems
  • Psychological problems
  • Lack of instruction, support, or follow-up

Tolerance problems may be due to side effects (ie, dry mouth, conjunctivitis, rhinorrhea, skin irritation, pressure sores, nasal congestion, epistaxis), mask leaks, difficulty exhaling, aerophagia, chest discomfort, and bed-partner intolerance.[81]  Psychological problems include lack of motivation, claustrophobia, and anxiety. The suggestions below for dealing with some of these problems may assist the physician in improving treatment compliance.

Many patients report claustrophobia. They find that the sensation of covering the nose with a mask makes them so uncomfortable that they cannot tolerate wearing the n-CPAP device. Sometimes this can be helped with a smaller or more transparent mask design. Use of nasal pillows (inserted into the nostrils) instead of a formal nasal mask may allow such patients to tolerate n-CPAP.

Some patients have trouble tolerating the initial pressure. Especially when higher pressures (>12-13 cm H2O) are required for elimination of apneas and hypopneas, this level of pressure may be uncomfortable. Many n-CPAP machines have a built-in ramp or gradual increase in pressure. With this feature, the mask can be placed and pressure begun at a very low and easily tolerated level. Over 30 minutes, the pressure gradually builds to the full amount necessary. Often, the patient can fall asleep during this time. Full pressure is not used until the patient is actually asleep.

Patients may experience nasal obstruction. Evaluation by an otolaryngologist reveals whether this is predominantly a fixed skeletal obstruction or a soft-tissue obstruction potentially modifiable without surgery. Marked septal deviation or turbinate hypertrophy usually requires surgery for resolution. Alar collapse may be adequately treated by internal or external dilators (eg, Breathe Right strip, Nozovent). Surgery is sometimes required for repair of marked alar collapse.

Mucosal edema may be due to allergic rhinosinusitis or to vasomotor or irritative rhinitis. Allergy testing and treatment and pharmacotherapy trials (eg, topical steroids or antihistamines, oral antihistamines, or decongestants) may be beneficial.

One way of determining whether there is sufficient potentially reversible mucosal edema to warrant pursuit of that avenue of treatment is to perform the topical decongestant test. The patient uses a nasal topical decongestant (eg, oxymetazoline) at bedtime for several days, with the patient and bed partner observing for any improvements in snoring or apneas. A marked improvement suggests potentially reversible mucosal edema as a main contributor to the nasal obstruction. Failure to improve suggests a fixed skeletal obstruction that requires surgical correction.

Sometimes the dryness of the air or its temperature may be irritating to the patient. Use of inline humidification and warming of the inspired air may alleviate patient discomfort.[88, 89]

A number of patients report facial or nasal pain. Sometimes this pain can be related to a poorly fitting mask. With the many different types of masks available now, different styles and sizes can be tried to select the optimal fit for each individual anatomy. Because the mask is pulled tight against the face, an edentulous anterior maxilla may not provide the resistance necessary for a good fit. Leaving dentures in at night can help with this.

If the facial or nasal pain persists despite mask refitting, evaluation for nasal obstruction or chronic sinusitis may be helpful. The CPAP Pro delivery method anchors the tubing to a platform based on an upper retainer, obviating the need for a forehead strap.

Patients may experience dry eye or other eye discomfort. If the mask does not seal well, egress of pressurized air from the upper end of the mask toward the eye may occur, causing dry eye or even exposure keratitis. Mask refitting usually eliminates this problem.

Patients may sleep with the mouth falling open, awakening with dry mouth. Sometimes a chin strap is required to prevent the mouth from opening at night. A commercially available disposable adhesive bandage may used to pull the chin up toward the lower cheeks.[90]

Patients may experience epistaxis. This may be related to the high-flow dry air and may be helped by humidification and warming of the inspired air.

Some patients experience nasal drying. Forced dry air can be irritating to the nose, encouraging mucosal inflammation and crusting. Use of humidified air for n-CPAP usually eliminates this problem.

Other problems may also occur. Pneumopericardium has been reported with n-CPAP.[91]  Pneumocephalus has occurred when n-CPAP was used in a patient with cerebrospinal fluid rhinorrhea. Eustachian tube dysfunction, serous otitis media, bulging of the eardrums, and eardrum perforation have also been reported.

Rigorous patient education and early reinforcing follow-up may improve long-term use of n-CPAP.

Other considerations

Variations of air pressure delivery can sometimes make n-CPAP use more comfortable for patients.

Autotitrating positive airway pressure (APAP) continually adjusts the pressure to a level that barely overcomes the collapsing forces. Bilevel positive airway pressure (BiPAP) provides higher pressure during inspiration (when the pneumatic splint is needed to prevent obstructive airway collapse) and lower pressure during expiration. C-Flex is another autoadjusting delivery method that increases pressure toward the end of expirations, as collapse would usually begin, and decreases pressure during early expiration.

Patients who require higher pressures to overcome obstructive apneas may tolerate these devices better than they do the one-level n-CPAP device, which delivers the higher pressure throughout the entire respiratory cycle.

Following treatment with CPAP, some patients with OSA remain sleepy despite effective CPAP, and attention should be paid to other diagnoses that can be associated to sleepiness. So-called post-CPAP sleepiness, as a specific disorder, may not exist.

Oxygen Administration

Because some of the effects of SDB are due to hypoxia during sleep, the administration of oxygen would seem like a reasonable treatment. Although oxygen administration improves the lowest blood-oxygen saturation level during sleep and can improve some of the arrhythmias occurring during desaturation, repeated studies have not demonstrated sustained clinically significant improvement in SDB with oxygen administration. Some prolongation of apneas also occurs, particularly at the beginning of therapy.

Oxygen administration may be beneficial in a subset of patients. Some patients with other coexistent pulmonary disorders may also benefit from use of oxygen in conjunction with n-CPAP.



Protriptyline, a tricyclic antidepressant, is the medication most studied in the treatment of SDB and does yield improvement in patients with this condition. This effect, however, appears to be mainly due to suppression of rapid-eye-movement (REM) sleep. Because SDB is often most severe during REM sleep, less REM sleep can mean fewer apneas.


Modafinil is a wake-promoting medication used in association with CPAP to treat patients with OSAS. It has an action similar to that of sympathomimetic agents (like amphetamine and methylphenidate), though its pharmacologic profile is not identical to that of sympathomimetic amines. The precise mechanism through which modafinil promotes wakefulness is unknown.

Headache and nervousness are the only adverse events reported. Because there is no benefit to using modafinil in patients with OSA who are not compliant with CPAP, it should not be administrated in such cases.

Other drugs

Other drugs that have been investigated for treatment of sleep apnea include progestational agents, aminophylline, acetazolamide, L-tryptophan, naloxone, baclofen, bromocriptine, chlorimipramine, and prochlorperazine. For the most part, these have not shown a consistently helpful effect on SDB. However, a study of the use of acetazolamide in patients with OSA and comorbid hypertension found that this agent reduced blood pressure, vascular stiffness, and SDB in this setting.[92]

Surgical Therapy

Surgical care of SDB is discussed more fully in Surgical Approach to Snoring and Sleep Apnea.[93, 94, 61]  In the perioperative period, n-CPAP is often used to ensure good ventilation even in the presence of postoperative edema. Because of the use of analgesics and swelling of the soft tissues, the pressure needed to maintain a patent airway postoperatively may be greater than the patient had been using before surgery.


Diet and exercise counseling play a major role in the initial therapy for SDB.

Weight reduction in the patient with obesity can dramatically improve SDB. Even a modest weight loss can have quite a beneficial effect on the frequency of apneas and hypopneas. Bariatric surgery may be needed in some cases. When rapid weight loss occurs after bariatric surgery or successful dieting, the pressure for overcoming apneas and hypopneas is likely to decrease; thus, retesting is recommended.


Multidisciplinary sleep teams, including pulmonologists, otolaryngologists, neurologists, and oral-maxillofacial surgeons, may offer the most convenient and comprehensive treatment for these patients.

Long-Term Monitoring

It msut be remembered that n-CPAP does not cure or alter the underlying OSA but, rather, provides daily relief from the apneas, snoring, hypoxias, and consequent daytime symptoms. After long-term n-CPAP use, a carry-over effect is often noted; therefore, PSG results on the first day or two off n-CPAP look remarkably improved. However, this carry-over is short-lived, and it is usually the case that within 1 week, the snoring, apneas, hypoxias, and daytime symptoms return to their original level.

n-CPAP is highly successful in managing OSA, as long as it is used. Unfortunately, compliance with n-CPAP use is less than ideal: Only about half of the patients for whom it is prescribed use it for at least 4 hours a night on 5 of 7 nights. For this reason, regular follow-up visits are mandatory for ensuring continued successful treatment.

Some physicians see patients on a 3- to 4-month basis during their first year of n-CPAP use and yearly thereafter. Repeat sleep studies are obtained after major weight loss or gain or after major change in daytime symptoms. Many patients happily and successfully use n-CPAP for years. Others find sustained use impossible; these are the patients for whom surgery may be helpful.[61]

Even snorers whose PSG does not show SDB should be monitored periodically because they can progress to SDB with time, even without weight gain.


Questions & Answers


What is sleep-disordered breathing (SDB)?

What is the pathophysiology of sleep-disordered breathing (SDB)?

What causes sleep-disordered breathing (SDB)?

Which patient groups have the highest prevalence of sleep-disordered breathing (SDB)?

What is the prognosis of sleep-disordered breathing (SDB)?


Which clinical history findings are characteristic of sleep-disordered breathing (SDB)?

Which physical findings are characteristic of sleep-disordered breathing (SDB)?

What are the possible complications of sleep-disordered breathing (SDB)?


Which conditions are included in the differential diagnoses of sleep-disordered breathing (SDB)?


What is the role of lab tests in the workup of sleep-disordered breathing (SDB)?

What is the role of imaging studies in the workup of sleep-disordered breathing (SDB)?

What is the Epworth Sleepiness Scale and how is it used in the evaluation of sleep-disordered breathing (SDB)?

What is the role of polysomnography (PSG) in the workup of sleep-disordered breathing (SDB)?

What is the role of screening questionnaires in the workup of sleep-disordered breathing (SDB)?

What is the role of ambulatory monitors in the workup of sleep-disordered breathing (SDB)?

What is the role of acoustic snoring analysis in the workup of sleep-disordered breathing (SDB)?

What is the role of sleep endoscopy in the workup of sleep-disordered breathing (SDB)?

Which histologic findings are characteristic of sleep-disordered breathing (SDB)?


What is included in the initial treatment of sleep-disordered breathing (SDB)?

Which modifications of sleep positions are used in the treatment of sleep-disordered breathing (SDB)?

When is thyroid hormone replacement therapy indicated in the treatment of sleep-disordered breathing (SDB)?

What is the role of oral appliances in the treatment of sleep-disordered breathing (SDB)?

How does alcohol affect sleep-disordered breathing (SDB)?

What is the role of nasal CPAP (n-CPAP) in the treatment of sleep-disordered breathing (SDB)?

How is required positive airway pressure determined for the treatment of sleep-disordered breathing (SDB)?

What are the positive effects of nasal CPAP (n-CPAP) in the treatment of sleep-disordered breathing (SDB)?

What are the common reasons for noncompliance with nasal CPAP (n-CPAP) for the treatment of sleep-disordered breathing (SDB)?

What are the options for autotitrating positive airway pressure during the treatment sleep-disordered breathing (SDB)?

What is the role of oxygen administration in the treatment of sleep-disordered breathing (SDB)?

What is the role of protriptyline in the treatment of sleep-disordered breathing (SDB)?

What is the role of modafinil in the treatment of sleep-disordered breathing (SDB)?

What is the role of medications in the treatment of sleep-disordered breathing (SDB)?

What is the role of surgery in the treatment of sleep-disordered breathing (SDB)?

Which dietary modifications are used in the treatment of sleep-disordered breathing (SDB)?

Which specialist consultations are beneficial to patients with sleep-disordered breathing (SDB)?

What is included in the long-term monitoring of patients with sleep-disordered breathing (SDB)?