Supine REM and stage-2 sleep play a crucial role in accurately determining the severity of OSA.

Obstructive sleep apnea (OSA) syndrome is a common, but frequently underdiagnosed condition. It is one of the most prevalent causes of dangerous daytime sleepiness. It is estimated that more than 20 million US residents experience this problem.1 Of these, approximately 23% have moderate-to-severe disease, making OSA no less prevalent than asthma.2 The incidence of significant OSA increases with age, and among males, 35% are believed to have the problem.2 Amazingly, it is reliably estimated that 93% of women and 82% of men with moderate-to-severe OSA are undiagnosed and, therefore, untreated.2

Increased premature mortality and morbidity have been demonstrated in people who experience 20 or more respiratory events per hour of sleep. This figure was determined from the study3 of a group of patients who were diagnosed with OSA, but declined any form of intervention or treatment. Arterial hypertension has been a reported complication in up to 50% of the patients in whom OSA has been diagnosed.4 The cause-and-effect relationship between hypertension and sleep-disordered breathing has not been clearly established. After nasal continuous positive airway pressure (CPAP) treatment, there was a marked reduction in patients’ arterial blood-pressure levels. What is not entirely clear, however, is whether OSA predisposes patients to hypertension and CPAP then ameliorates it by eliminating OSA or, alternatively, whether hypertension and OSA coexist coincidentally, with treatment of OSA by CPAP lowering blood pressure as a side benefit.

Tachycardic and bradycardic arrhythmias have also been confirmed5 in patients with OSA. Cause and effect are, in this case, more obvious. In patients with preexisting coronary artery disease, hypoxemia, decreased heart rate, and arousal-related induced hypertension result in transient myocardial ischemia and infarction during apneic events. Other research6 has indicated an increased risk of cerebrovascular accident (CVA) in patients with OSA (independently of either hypertension or heart disease). OSA, then, is clearly a risk factor for CVA, myocardial infarction, and hypertension. Untreated OSA patients have shorter lifespans than successfully treated patients.3

In addition to the cardiovascular consequences of OSA, patients with this disorder also suffer from severe, excessive daytime sleepiness. They fall asleep in inappropriate settings and at inopportune times when they should be alert. These settings may include driving, operating dangerous machinery, performing customary chores or work-related tasks, or even taking a shower or sitting down to eat a meal. People with OSA have up to seven times as many automobile accidents as people who do not have OSA.7 Sleepiness and fatigue due to OSA have been linked to oil-tanker accidents, train wrecks, and nuclear power plant mishaps.8 Suboptimal performance in demanding careers has been confirmed as one of the sequelae of OSA.9 In some cases, this problem leads to reliance on illicit stimulant drugs (as seen, for example, in professional athletes and long-distance truck drivers). Performance improves markedly after successful treatment.10

The accurate diagnosis of OSA should not be based on clinical markers alone. Objective, recorded tabulation of the number of apneas and hypopneas occurring during sleep—collected simultaneously with continuous readouts for pulse oximetry, electrocardiography, electroencephalography, electromyography, and electrooculography—enables the sleep specialist to make an accurate diagnosis of the condition. For titration of CPAP, whether performed during a split night or during a second night of testing, the same values are the most accurate means for determining the level of positive pressure that completely reverses OSA. Body positioning is an important part of this testing.

The role of body position in the etiology of OSA is tied to the complex pathology of the problem. In describing the condition to patients, it is useful to tell them, in simple terms, that the tissues in their throats tend to relax and collapse, intermittently blocking their breathing while they are asleep. It is necessary to explain why this occurs when a person is asleep, rather than awake, and why some body positions make the condition worse than others.

Pathophysiology of OSA and Body Position
The pathologic process of OSA has central and anatomical components. OSA, by definition, begins at sleep onset, so it is evident that something happens at the beginning of sleep to predispose some people to this condition. The trigger is the sudden loss of the wakefulness stimulus. During wakefulness, neurons in the brain send a constant signal, which eventually reaches the spinal cord. This impulse causes the maintenance of a certain degree of muscle tone. In sleep, these neurons are not active, and muscle tone is lost. The muscles that maintain a patent airway in the posterior part of the oral cavity suddenly become flaccid.

As the airway muscles become more relaxed, collapsing into the airway entrance, airway resistance increases and air flow through both oral and nasal passages decreases. The respiratory system must compensate by increasing the work of the respiratory pump in order to maintain adequate tidal volume and minute ventilation. This response causes increased intrathoracic negative pressure; this, in turn, pulls the airway closed and further contributes to increased upper-airway resistance.

If a patient has very few events (or none at all) but has not been evaluated in the supine position during the rapid–eye-movement (REM) stage of sleep, the sleep specialist cannot determine whether significant OSA is present. Respiratory tracings and pulse oximetry during supine REM are the diagnostic gold standard for OSA. Sleep technicians are instructed to encourage patients to lie supine for this reason. If a patient experiences obstruction, it is more likely to occur during supine REM and stage-2 sleep. Lying supine increases the likelihood that the flaccid muscle tissue will fall towards the oropharynx, worsening obstruction. Most severe OSA is found in stage-2 sleep, as well as in REM sleep. It is less likely to occur in stages 3 or 4 of sleep, although the reasons for this are not entirely clear. Stage 1 sleep usually contains few (or no) obstructive events due to the fact that it is transitional between waking and deeper sleep; central neural output has not yet decreased fully during this stage. REM sleep is divided into tonic and phasic portions. Eye movements are considered phasic events, whereas the total muscle paralysis seen during this stage of sleep is the tonic phase of REM. The greatest amount of muscle relaxation (and, therefore, the greatest number of obstructive respiratory events) occurs during REM sleep in the supine position. If a patient passes through one or more REM stages with no demonstrable respiratory events or desaturations, it is likely that the patient does not have OSA. If a patient is undergoing CPAP titration for previously documented OSA, supine REM sleep without evidence of obstruction indicates that an optimal CPAP pressure has been reached.

Anatomical Causes of OSA
The anatomical structures of the oral cavity also play a role in the pathology of OSA syndrome. Moreover, the more crowded the oropharyngeal cavity, the more likely it is that OSA will develop. A low-hanging soft palate, narrow pharyngeal pillars, enlarged tonsils, a retrognathic jaw, and an enlarged tongue all contribute to the process.

Gravity affects the potential for OSA in a number of ways. If the patient is lying prone, gravity would have a tendency to pull flaccid muscle tissue out of the oropharynx and into the mouth. Sleeping in the supine position allows gravity to pull the tissues into the oropharynx, thus worsening the degree of obstruction.

Measuring or Marking Body Position
It is essential that the body position of any patient undergoing polysomnography be carefully documented by the recording technician. Without such documentation, it is impossible for the interpreting physician to determine what influence body position has had on the results of the test. If the patient had few or no events, but was never able to sleep supine, obstruction may be the reason. If the patient had severe desaturations and numerous events in one position but not another, it is important to document this. The night technician is thus required to watch the patient very carefully (usually on closed-circuit television) and to note any change in body position when it occurs, or as soon after it occurs as possible. Computerized polysomnography systems usually have clickable tags for body position that can be placed directly on the study as soon as a body-position change occurs.

A means of detecting body-position changes and automatically placing the position tag directly on the screen or paper tracing has also been developed. It involves a small pouch (containing a pellet) that can be attached to the thoracic belt. As the patient moves, the pellet falls to the dependent position, causing an electrical signal to be generated. The source of the signal describes the position to which the patient has changed.

Automated position-marking devices are capable of detecting and tagging shifts within 1 second. Computerized scoring systems take position tags into consideration when tabulating the number of respiratory events and the positions in which they occur. Some provide a histogram of respiratory events plotted against position, sleep stage, and oxygen saturation. In severe cases of OSA, such graphics provide dramatic evidence that respiratory events and desaturations occur more frequently in the supine position during REM and stage-2 sleep.

Unattended and/or home studies suffer from a lack of trained on-site monitoring personnel, so automated position indicators are necessary for systems designed to perform these kinds of tests. If an unattended study’s results are negative for OSA, it is extremely important to know whether the test included time spent in the supine position. Patients who have severe OSA when lying on their backs may have few (or no) respiratory events when lying on their sides; this creates the potential for a false-negative result, which could have grave consequences for the patient.

Positioning as a Therapeutic Intervention
Neill11 conducted a study designed to determine the effects of three different body positions on OSA. They developed a special CPAP mask through which they could measure upper-airway occlusion pressures. In this fashion, they could estimate the effect of body position on the stability of the upper airway in patients known to have OSA. The results demonstrated that elevation to 30° from the horizontal position creates the greatest amount of change in occlusion pressure. These results were dramatic in that there was a reduction of almost 50% in mean upper-airway occlusion pressure in the 30° upright position. Changing from the supine to the lateral recumbent (English) position did not greatly alter the occlusion pressure. In summary, the authors suggested that the 30° upright position can be used as a therapeutic measure and may allow CPAP pressures to be reduced. Other studies11 have suggested that, in some cases, OSA can be ameliorated by sleeping in a 60° upright position.

In a study of patients being treated for OSA using an oral appliance, Itasaka12 was able to demonstrate further improvement in the apnea-hypopnea index by shifting the patient’s body position from supine to lateral recumbent. In a very simple study, Berger13 examined the blood-pressure levels of patients who had been previously documented as having predominantly positional OSA. The experimental intervention involved tennis balls sewn into a pouch constructed on the back of the patient’s sleepwear. This is designed to help prevent the patient from rolling into the supine position. The experimenters found a significant reduction in blood pressure among patients who slept on their sides. They suggested that sleeping in the supine position worsened OSA, which then increased blood pressure. In the lateral recumbent position, OSA severity declined and blood pressure subsequently fell.

Oksenberg14 conducted a detailed retrospective study of the effect of position on OSA. For 666 consecutive patients, he determined whether the apnea-hypopnea index in the supine position was at least double the index seen in the lateral recumbent position, and results of this grouping clarified the overall effect of body position on OSA. Sleep quality was better in patients with positional OSA; they had increased sleep efficiency as well as greater amounts of deep sleep. They also demonstrated a lesser amount of light sleep. Likewise, the number of short arousals was significantly lower in the positional-OSA group (indicating a more solid sleep). The group of patients who had significant OSA irrespective of position had a higher apnea-hypopnea index, as well as significantly lower oxygen levels. Patients in the nonpositional-OSA group were also significantly more sleepy during the day, based on the results of multiple sleep-latency testing. The author concluded that several characteristics predict whether a patient will have positional OSA, but obesity does not always dictate the presence of positional OSA.

Low Tech Positional Therapy
Using a device such as a tennis ball to prevent supine sleeping or using an alarm to signal that a patient has turned onto his or her back has been advocated as positional therapy. The problem with such techniques is that they may, in some patients, result in a momentary disruption of sleep, creating an arousal that could serve to fragment sleep. The more often a patient turns onto his or her back, eliciting the positional-therapy stimulus, the more times he or she is apt to suffer an arousal. After several weeks of conditioning, however, many patients apparently are not roused from sleep as a result of being provoked by the positional-therapy stimulus.

Some patients sleep in recliner chairs or propped up on numerous pillows or bolsters. Others use bolsters or special pillows to force themselves into the lateral recumbent position; others may obtain restful sleep laterally by sleeping on a narrow, high-backed sofa.

Most patients learn on their own that supine sleeping causes more problems than other positions. The spouses of patients (who may spend much of the night forcing their mates to change positions or resume breathing) state that their partners breathe more easily in the lateral recumbent, prone, or semi-Fowler positions.

Candidates for positional therapy are typically those whose OSA is limited exclusively to the supine position. Such patients often have relatively mild disease that is, unfortunately, severe enough to cause daytime sleepiness (but not so severe as to allow them to tolerate therapies such as nasal CPAP easily).

The simplest form of positional therapy is elevating the head and chest during sleep. Studies11 have demonstrated that a 30° to 45° upright angle is more effective in reducing snoring or OSA than is rotation to the lateral position. A recliner chair easily accomplishes this; in addition, most of these chairs have elevated footrests, which allow more comfort. If a patient insists on sleeping in bed, however, the tennis-ball technique can be used. A single ball (sewn into the center back of a garment) and two tennis balls (with one located over each of the scapulae) have been tried. Unfortunately, many patients are made so uncomfortable that they refuse to continue using this technique.

If a patient will not tolerate nasal CPAP, and if oropharyngeal surgery or an oral appliance fails to correct the problem, the last option (short of tracheostomy) in the arsenal available against OSA is positional therapy, if the individual’s obstructive events occur mostly or entirely while the patient is supine.

Thomas M. Kilkenny, DO, is associate director of the Sleep Apnea Center, Staten Island University Hospital, Staten Island, NY. Steve Grenard, RRT, is clinical coordinator of the center.

1. National Commission on Sleep Disorders Research. Executive Summary and Executive Report. Bethesda, Md: National Institutes of Health; 1993.
2. Young T, Palta J, Dempsey J, Skatrud S, Weber S, Badr S. The occurrence of sleep disordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230-1235.
3. He J, Kryger MH, Zorick FJ, Conway W, Roth T. Mortality and apnea index in obstructive sleep apnea. Chest. 1988;94:9-14.
4. Shepard JW Jr. Hypertension, cardiac arrhythmias, myocardial infarction, and stroke in relation to obstructive sleep apnea. Clin Chest Med. 1992;13:437-458.
5. Becker HF, Koehler U, Stammnitz A, Peter JH. Heart block in patients with sleep apnea. Thorax. 1998;53:S29-32.
6. Hung J, Whitford EG, Parsons RW. Association of sleep apnoea with myocardial infarction in men. Lancet. 1990;336:261-264.
7. Findley LJ, Unverzadt ME, Suratt PM. Automobile accidents involving patients with obstructive sleep apnea. Am Rev Respir Dis. 1988;138:337-340.
8. Findley LJ, Fabrizio MJ, Thommi G, Surratt PM. Severity of sleep apnea and automobile crashes. N Engl J Med. 1989;320:868-869.
9. Bearpark H, Grunstein R, Touyz S. Cognitive and psychological dysfunction in sleep apnea before and after treatment with CPAP. Sleep Res. 1987;16:303.
10. Lamphere J, Roehrs T, Wittig R. Recovery of alertness after CPAP in apnea. Chest. 1989;96:1364-1367.
11. Neill A-M. Effects of posture on upper airway stability in apnea. Am J Respir Crit Care Med. 1997;155:199-204.
12. Itasaka Y. Effectiveness of prosthetic mandibular advancement for obstructive apnea. Psychiatry Clin Neurosci. 1998;
13. Berger M. Avoiding the supine position during sleep lowers 24 hour blood pressure in obstructive sleep apnea patients. J Hum Hypertens. 1997;11:657-664.
14. Oksenberg A. Positional vs nonpositional obstructive sleep
apnea: anthropomorphic nocturnal, polysomnographic and multiple sleep latency test data. Chest. 1997;112:629-639.