Evaluation of Obstructive Sleep Apnea for Adults
Alexander Osborn, M.D, Ph.D.
December 7, 2006

Obstructive sleep apnea is the collapse of the upper airways during sleep, which results in a cessation or reduction of airflow to the lungs. This is often accompanied by arousal from sleep and causes patients to have interrupted and poor sleep. Three possible etiologies have been suggested to explain the mechanism of this upper airway collapse. The first is narrowing of the upper airway, mostly by increase in parapharyngeal fat pushing in on the lateral edges of the airway, causing the normally elliptical shape to become a more circular and narrower shape. This narrowed upper airway causes an increase in negative pressure with increased airflow, and predisposes the upper airway to collapse. Secondly, the dilators that normally keep the airway open have been thought to fail in the sleep apnea patient. There has even been a suggestion of a pathological constriction of the airway, a pathological reflex with decreased upper airway pressure - the airway constricts abnormally in sleep apnea patients when compared to normal patients.

An apneic episode is defined fairly universally as a cessation of airflow for 10 seconds. The definition of hypopnea, however, varies widely from study to study and also from sleep lab to sleep lab. It is described as anywhere from a 30% to a 70% reduction in airflow and often there is a concomitant qualification that there must be a desaturation or an arousal from sleep in order to define the event as hypopnea. Obstructive sleep apnea-hypopnea syndrome occurs in the patients that have apneas, hypopneas, and some symptoms typically characterized by daytime sleepiness. The apnea and hypopnea index is a score that grades the severity of sleep apnea. It is simply the number of apneas and hypopneas per hour of sleep. The respiratory distress index is sometimes used interchangeably with the apnea and hypopnea index, although the definition, as you can see, is different. The respiratory distress index is the number of arousals per hour that are due to respiratory events, whether they qualify as full apnea, full hypopneas, or some other respiratory events that does not actually meet either standard.

People with sleep apnea-hypopnea syndrome are common in our population. This information comes from an often cited study from 1993 that showed a cross-sectional sample of the population and demonstrated that sleep disordered breathing, with an apnea and hypopnea index greater than 5, was very common. Twenty-nine percent of men and 9% of women had this condition. When participants in the study were asked about their symptoms of daytime sleepiness, qualifying them for sleep apnea-hypopnea syndrome, the incidence of this syndrome was 4% for men and 3% for women. This study also identified a correlation of sleep apnea with snoring and higher body mass index. As the prevalence of obesity has increased in our population over the past 16 years, the incidence of sleep apnea might also have increased. People with sleep apnea do not only have daytime sleepiness. When asked about how they feel about their health or quality of life, they near uniformly say that their quality of life is poor and that they have a subjective decreased state of health.

The effects of sleep apnea are not only functional and psychological. There are also ways that obstructive sleep apnea modifies our physiology. People with sleep apnea have increased sympathetic activity, increased respiratory mediators, and endothelial and metabolic dysfunction. All this, as you might guess, leads to higher incidence of cardiovascular complications in sleep apnea patients. Obstructive sleep apnea patients are at increased risk for high blood pressure, ischemic heart disease, stroke, and congestive heart failure. Their chances of dying from a cardiovascular event are increased. It is also important to note that as the apnea/hypopnea index (AHI) increases, so too does the risk of cardiovascular event. So, the AHI is a good indicator of disease severity and the urgency to treat a patient increases with increasing AHI, since the cardiovascular risk also increases.

When you do treat patients, especially with the common first line treatment CPAP machine, this does decrease the prevalence of cardiovascular events. Given the importance of treating sleep apnea patients, I want to discuss the clinical assessment of these patients, in terms of data that we have to keep in mind. A recent study identified three key components of the patient’s history, namely: witnessed apneas, nocturnal choking, and unrefreshed sleep. These were the most specific historical points that suggested obstructive sleep apnea. This study also suggested that the patient be asked how severe they believe their sleep apnea to be: no sleep apnea, mild-to-moderate, or severe. Patients actually had a good sense of how severe their sleep apnea was: their own rating correlated very well with measured AHI.

There are several questionnaires that have been used to gauge the severity of sleep apnea. The most common is the Epworth Sleepiness Scale, from a cross-sectional study performed in 1999. They basically took the general population and measured their AHI, and administered the Epworth Sleepiness Scale questionnaire. They found a correlation between the Sleepiness Scale and the AHI. This chart shows the severity of sleep apnea in the patients side by side with the mean Epworth Scale score, with the standard deviation in parenthesis. These differences are small, but the sample size was very large and these differences are statistically significant. The median score on the Epworth scale also increases with increasing AHI. The percentage of patients that scored higher than 11 on the Sleepiness scale also increased with increasing AHI. As you can see on this slide, there is a large variation when you apply the Sleepiness Scale to patients across the whole population, but when you narrow the analysis to patients with known sleep apnea, there does not seem to be a very good correlation between the AHI and the Epworth Sleepiness Scale score. This is a second study, performed just on patients who have sleep apnea. With increasing AHI or severity of sleep apnea, the Epworth Sleepiness Scale does not vary very much. The Epworth Scale may be a good screening tool, as identified in the first study: an easy, quick, short questionnaire to screen the general population. But when a patient comes to your clinic and says, “I have sleep apnea,” their score on the scale may not necessarily correlate with the severity of their disease. The Epworth Sleepiness Scale may be useful for evaluating the patient’s response to treatment, but do not make the assumption that since their score is higher on the Epworth scale, that their sleep apnea must be worse. All it really means is that it is affecting their life more. One reason that the Epworth Sleepiness Scale may not correlate well with AHI is suggested in this study from Italy, where Violani and colleagues asked a number of sleep apnea patients and non-sleep-apnea patients to rank the activities presented on the Epworth Scale from most soporific to least soporific. This slide illustrates the average rank in the order that they found, and shows a large difference between these two groups of activities. This large difference diminishes the scale’s ability to distinguish between finer shades of sleepiness. This group proposed a Resistance to Sleepiness Scale, in which they added some additional components to smooth the scale between activities viewed as progressively more soporific. They thought that this might help identify mild-to-moderate sleepiness and distinguish the two, as well as distinguish those from severe sleepiness. They also included more specific instructions to the patient to consider the likelihood of unintentionally falling asleep rather than just how likely the patient is to fall asleep. Unfortunately, this scale has not really been tested or validated in the literature, and although it is something that I think deserves testing, most people simiply use the Epworth scale.

The other questionnaire that I would like to discuss is the Sleep Apnea Quality of Life Index. This is a rather long questionnaire, 40 items with 5 additional items if patients are receiving CPAP treatment. This questionnaire was developed specifically for sleep apnea patients in order to measure their quality of life. It can be used to monitor the effects of treatment: improvement on the Sleep Apnea Quality of Life Index means that the patient’s quality of life is being improved. It is interesting to note that when this questionnaire was being created, they asked a population of sleep apnea patients to describe how important various items were to them, and then assigned these different items combined scores comprised of the frequency with which the item was rated, and the importance score the patients gave the item. They found that snoring that disturbs or awakens the patient’s partner was the most important and most frequent of all the items on the questionnaire as rated by sleep apnea patients. We do not use this questionnaire as a screening tool, as it is used mostly to evaluate treatment. A lot of the points that come up on this questionnaire, however, are things that the patients might complain of in the clinic.

There has been a lot of research dedicated to the development of clinical algorithms to predict sleep apnea. This is a grading system developed by Friedman in 1999. The patients are rated in three categories: their modified Mallampati grade, tonsil grade, and their BMI. In each category, they are given a score. The scores are combined in all the categories, allowing people to have a score from 1 to 12. The Friedman grading systems shows that in 90% of patients who have a score greater than 8, their AHI was greater than 20. This is a simple way to evaluate patients and has a good positive predicted value. However, in order to achieve that high positive predicted value, the AHI cut-off is 20. There are a lot of people who would be missed with this algorithm that actually have clinically significant sleep apnea, AHI from 5 to 20.

It would be useful to have an algorithm that demonstrates a good positive predicted value with a lower AHI cut-off. Tsai and colleagues actually developed such an algorithm. They developed a grading scale that accounts for three factors: the pharyngeal grade as identified by this diagram, combined with a cricomental space of less than 1.5cm and an overbite, which was not really defined in their paper. Ninety-five percent of the patients who had these three qualities had an AHI of greater than 10, which makes this a little more useful in identifying patients with clinically significant sleep apnea.

In their paper, they made the rather bold statement, based on 50 patients, that a cricomental space of greater than 1.5cm absolutely excluded the presence of obstructive sleep apnea in their study. I do not know if anyone would be willing to dismiss a patient with complaints based solely on this measurement, but they propose this as a clinical decision rule.

I would like to mention one more grading system and morphometric model developed by Kushida and colleagues, which is based on a number of measurements including palatal height, the distance between the maxillary and mandibular molars, the overjet of the top lying teeth, the neck circumference and body mass index. The authors developed a formula into which these values were inserted. Patients in their clinics score between 40 and 170, and basically every patient above 70 had sleep apnea. They propose this rather complex formula as a decision rule for predicting sleep apnea. It is very good at doing so, but it is somewhat cumbersome to use, and this probably explains why it is not widely used in practice. What it does do, however, and the reason I show it, is that it highlights two important and separate contributions to sleep apnea. If you consider this equation to be a representation of the probability of the patient having sleep apnea, the first half of the equation consists of craniofacial measurements. The second half of the equation consists of measurements based on body habitus. If you have normal body habitus, a BMI less than 20, actually the second half of the equation, the contribution to the final result is zero. This means basically that in a thinner patient, the contribution of craniofacial measurements and craniofacial abnormalities to sleep apnea is greater, while in a heavier patient, or patients with higher BMIs, the body habitus tends to make a larger contribution to the patient’s sleep apnea. In severe sleep apnea, it is likely that both BMI and craniofacial abnormalities are at work. We should keep in mind that there are two types of sleep apnea patients: thin sleep apnea patients and heavier sleep apnea patients and there are therefore different things that we should look at or stress in our evaluation.

One additional clinical evaluation we could perform is the Mueller maneuver. This was originally described in order to assist in determining which patients would be most suitable for surgical treatment of sleep apnea. The Mueller maneuver was originally left out of Friedman’s classification scheme because it was deemed to be too subjective to be reliable. However, a study from Stanford demonstrated that if you categorize the Mueller maneuver results properly, mainly on a quartile scale, basically calling the amount of collapse on the Mueller maneuver either 0%, 25%, 50%, 75%, or 100%, there was actually very good inter-observer reliability. The observers in this study consisted not only of different attending faculty, but also all levels of ENT residents. So, if quantified correctly, the Mueller maneuver does have good inter-observer reliability. They also found that the degree of collapse did correlate with the severity of sleep apnea.

Svensson and colleagues highlighted the important difference between heavier and thinner sleep apnea patients, once again by demonstrating that if the Mueller maneuver is performed on thinner patients, they tended to have collapse at the soft palate, whereas heavier sleep apnea patients tended to have collapse at the level of the tongue base. There is controversy over whether the Mueller maneuver is able to accurately predict the level of collapse in sleep apnea, and the jury is still out on that point. There are studies on both sides of the issue. But the Mueller maneuver is important for evaluation in extreme cases where patients have very large lingual tonsils. This photograph makes the point that the Mueller maneuver is very important in evaluating the lingual tonsil. This sleep apnea patient is undergoing surgical reduction of the lingual tonsil, which you can see is wrapping around the endotracheal tube and virtually abutting the posterior oropharyngeal wall. So it is not hard to see how this enlarged lingual tonsil can create an obstruction and the Mueller maneuver is very useful in identifying this region as the potential trouble spot in obstructive sleep apnea patients.

The last clinical exam that I would like to talk about is cephalometrics. There have been a number of studies that have identified the correlation between obstructive sleep apnea and various craniofacial measurements and soft tissue measurements. But I would like to highlight this study because it attempts not only to establish a correlation, but also to quantify the contribution of different measurements to the AHI. They found that BMI and craniofacial measurements contribute equally to severity of obstructive sleep apnea, once again highlighting the role of these two characteristics in sleep apnea. When evaluating patients, we should keep in mind: craniofacial abnormalities in thinner patients and BMI factors in the heavier patients. This paper also identified, in terms of the contribution of craniofacial measurements, the distance from the anterior nasal spine to the vertical palatal line as a key player. This measurement contributed the most to the sleep apnea: when this distance was less than 97mm, there was 3- to 7-fold increase in a selected set of patients of developing sleep apnea episodes. Thus, craniofacial abnormalities, especially in thinner patients, do play a large role in contributing to sleep apnea and cephalometrics is not a bad way to evaluate this. This study also found that the position of hyoid bone as well as the width of the posterior airspace were both important in contributing to sleep apnea.

Obviously, there has been a lot of research on clinical predictors of sleep apnea. And, yet, as you can see, there does not seem to be a nice, simple rule that allows us to easily identify whether a patient has sleep apnea or not. The bottom line is that if we suspect a patient of having sleep apnea, a sleep study is still the best diagnostic tool. We are not going to forgo a sleep study based solely on our clinical impression. So why is there so much research aimed at the clinical predictors of sleep apnea? Well, I think that if we can keep all of these clinical models in mind and try to assimilate them for ourselves, we come up with a better picture of what influences sleep apnea, and we hone our clinical ability to suspect it. The other interesting point about clinical predictors of sleep apnea is that as more and more research is done on what surgical interventions are effective with various patients, I think these clinical models will be more useful in terms of predicting what surgery we should do on a patient or what treatment will work best.

Given that the sleep study still remains the gold standard for diagnosis, I would like to briefly talk about the sleep study. This is a sample sleep study strip in which you can see extraocular movements, muscle tone, brain-wave activity, cardiac activity, and airflow measures. You can see that the diagnosis of an apneic and hypopneic episode depends upon the interpreter of the sleep study accurately coordinating an arousal with a respiratory event. Arousals are often judged by the increase in muscle tone and change in brain wave activity, but determining what constitutes an arousal can be difficult. Also, what constitutes a respiratory event can be difficult. There is a lot of room for inter-observer or inter-reader variability. Consider this illustration of an arousal that is not associated with a respiratory event. You can see how distinguishing certain patterns of arousal, especially in light sleep where brain waves are very similar, can be rather difficult. There have been two studies that examined interobserver variability with the sleep study. One focuses on the definition of an arousal event based on the EEG readings. In this study, 90 potential arousal events were sent to 14 readers. They found that any two experts seem to agree about 75% of the time on what constitutes an arousal. This means that readers can differ in how they score a sleep study. The second study focused on the differences in the definition of hypopnea, since, as I already mentioned at the beginning of the talk, the definition of hypopneas are variable. The score, the AHI, from the sleep study will inherently be variable depending on what kind of definition is used. This study showed that if 11 studies were sent to 9 different sleep centers for interpretation, there was a large variation between the minimum score given on the test and the maximum score given on the test. You can see that even if we use a liberal definition of severe sleep apnea with an AHI of greater than 40, two of these tests, when scored by different readers, were returned as either no disease or severe sleep apnea. This does not mean to say that we should not use sleep, but it means that we have to keep in mind what definition our sleep labs are using for hypopnea and how people are scoring arousals. Also, we should keep in mind that two sleep studies from different labs are not necessarily comparable.

Lastly, I wanted to talk about CPAP, which is a common treatment for sleep apnea. It is the gold standard. No other measures have been shown to be as effective as CPAP machines for lowering the AHI. However, as we have seen in our patient in the case presentation, CPAP is often poorly tolerated. Some studies show compliance to be as low as 40% to 50%, while others can show a rather high compliance, 70% to 80%. Regardless, there appears to be a problem with compliance. Compliance patterns are established early in the course of treatment, during the first few weeks. Compliance is lower in people who are receiving CPAP solely for snoring, because they do not tend to feel any better as a result of their treatment. For some reason, compliance is higher in Europe.

I would like to suggest some ways to increase CPAP compliance in our patients. Studies have shown that appropriate education of patients, including information on the severity of their illness, what they can expect from their CPAP treatment, and the expected benefits of CPAP treatment, can improve compliance with CPAP. Another variable shown to increase compliance is the interface that is used to deliver CPAP. Nasal pillows are preferred by patients and compliance actually increases, in terms of the number of nights that the CPAP is used. Two other variables that have been examined are humidification and auto-titrating CPAP. Humidification does not seem to play much of a role in enhancing CPAP compliance. Automatically titrated CPAP machines, in select cases where patients report side effects from their CPAP machines, help increase compliance, but not in the general population as a whole. Finally, two studies identified the nasal passages as important factors, and even predictors in CPAP compliance. Smaller nasal passages create a more uncomfortable CPAP environment and these studies suggest that increasing the area of the nasal passages can improve CPAP compliance.

Case Presentation:

JC is a 29-year-old man with a 12-year history of obstructive sleep apnea. He describes a history of loud snoring, frequent nocturnal awakenings, unrefreshing sleep, and awakening throughout the night choking or gasping for breath. He reports that apneic episodes during sleep have been witnessed. His score on the Epworth Sleepiness Scale is 13.

His past medical history is negative. He does not have diabetes or hypertension. Past surgical history is negative. His current medication is Seroquel.

His social history consists of smoking three cigarettes a week. He drinks one to two drinks per night. He denies any drug use.

Physical Exam: Height is 66 inches, weight is 178 pounds, and his BMI is 28.7. Nasal passages patent to anterior rhinoscopy, no masses or mucosal thickening. His tongue appears normal in size, no scalloping of tongue edges, long uvula, 3+ tonsils. A fiberoptic flexible endoscopy was performed. His adenoid tissue not prominent, and no masses in nasal passages. Mueller maneuver demonstrated 75 percent collapse at level of the soft palate.

A sleep study was conducted which demonstrated AHI 88 with SaO2nadir of 68%. CPAP decreased his AHI to 0, but it was poorly tolerated by the patient.

Bibliography:

Ayas NT, Patel SR, Malhotra A, Schulzer M, Malhotra M, Jung D, Fleetham J, White DP. Auto-titrating versus standard continuous positive airway pressure for the treatment of obstructive sleep apnea: Results of a meta-analysis. Sleep 2004;27:249-53.

Dempsey JA, Skatrud JB, Jacques AJ, Ewanowski SJ, Woodson BT, Hanson PR, Goodman B. Anatomic determinants of sleep-disordered breathing across the spectrum of clinical and nonclinical male subjects. Chest 2002;122:840-51.

Doherty LS, Kiely JL, Swan V, McNicholas WT. Long-term effects of nasal continuous positive airway pressure therapy on cardiovascular outcomes in sleep apnea syndrome. Chest 127:2076-84.

Drinnan MJ, Murray A, Griffiths CJ, Gibson GJ. Interobserver variability in recognizing arousal in respiratory sleep disorders. Am J Respir Crit Care Med 1998;158:358-62.

Flemons WW, Reimer MA. Development of a disease-specific health-related quality of life questionnaire for sleep apnea. Am J Respir Crit Care Med 1998;158:494-503.

Friedman M, Tanyeri H, LaRosa M, Landsberg R, Vaidyanathan K, Pieri S, Caldarelli D. Clinical predictors of obstructive sleep apnea. Laryngoscope 1999;109:1901-7.

Gottlieb DJ, Whitney CW, Bonekat WH, Iber C, James GD, Lebowitz M, Nieto FJ, Rosenberg CE. Relation of sleepiness to respiratory disturbance index: The Sleep Heart Health Study. Am J Respir Crit Care Med 1999;159:502-7.

Haniffa M, Lasserson TJ, Smith I. Interventions to improve compliance with continuous positive airway pressure for obstructive sleep apnea. Cochrane Database Syst Rev 2004;Oct 18:CD003531.

Hia KM, Skatrud JB, Finn L, Palta M, Young T. The effect of correction of sleep-disordered breathing on BP in untreated hypertension. Chest 2002;122:1125-32.

Hui DS, To KW, Ko FW, Fok JP, Chan MC, Ngai JC, Tung AH, Ho CW, Tong NW, Szeto CC, Yu CM. Nasal CPAP reduces systemic blood pressure in patients with obstructive sleep apnoea and mild sleepiness. Thorax 2006;61:1083-90.

Gau Ss, Soong WT, Merikangas KR. Correlates of sleep-wake patterns among children and young adolescents in Taiwan. Sleep 2004;27:512-9.

Kushida CA , Efron B, Guilleminault C. A predictive morphometric model for the obstructive sleep apnea syndrome. Ann Intern Med 1997;127:581-7.

Le Bon O, Hoffmann G, Tecco J, Staner L, Noseda A, Pelc I, Linkowski P. Mild to moderate sleep respiratory events: One negative night may not be enough. Chest 2000;118:353-9.

Logan AG, Perlikowski SM, Mente A, Tisler A, Tkacova R, Niroumand M, Leung RS, Bradley TD. J Hypertens 2001;19:2271-7.

Mador MJ, Krauza M, Pervez A, Pierce D, Braun M. Effect of heated humidification on compliance and quality of life in patients with sleep apnea using nasal continuous positive airway pressure. Chest 2005;128:2151-8.

Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: An observational study. Lancet 2005;365:1046-53.

Massie CA, Hart RW. Clinical outcomes related to interface type in patients with obstructive sleep apnea/hypopnea syndrome who are using continuous positive airway pressure. Chest 2003;123:1112-8.

Massie CA , Hart RW, Peralez K, Richards GN. Effects of humidification on nasal symptoms and compliance in sleep apnea patients using continuous positive airway pressure. Chest 1999;116:403-8.

McNicholas WT, Ryan S. Obstructive sleep apnoea syndrome: Translating science to clinical practice. Respirology 2006;11:136-44.

Morris LG, Setlur J, Burchtin OE, Steward DL, Jacobs JB, Lee KC. Acoustic rhinometry predicts tolerance of nasal continuous positive airway pressure: A pilot study. Am J Rhinol 2006;20:133-7.

Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S, D’Agostino RB, Newman AB, Lebowitz MD, Pickering TG. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-bases study. Sleep Heart Health Study. JAMA 2000;283:1829-36.

Pang KP, Terris DJ, Podolsky R. Severity of obstructive sleep apnea: Correlation with clinical examination and patient perception. Otolaryngol Head Neck Surg 2006;135:555-60.

Pang KP, Dillard TA, Blanchard AR, Gourin CG, Podolsky R, Terris DJ. A comparison of polysomnography and the SleepStrip in the diagnosis of OSA. Otolaryngol Head Neck Surg 2006;135:265-8.

Peker Y, Hedner J, Norum J, Kraiczi H, Carlson J. Increased incidence of cardiovascular disease in middle-aged men with obstructive sleep apnea: A 7-year follow-up. Am J Respir Crit Care Med 2002;166:159-65.

Peppard PE , Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378-84.

Powell NB, Zonato AI, Weaver EM, Li K, Troell R, Riley RW, Guilleminault C. Radiofrequency treatment of turbinate hypertrophy in subjects using continuous positive airway pressure: A randomized, double-blind, placebo-controlled clinical pilot trial. Laryngoscope 2001;111:1783-90.

Series F, Marc I. Importance of sleep stage- and body position-dependence of sleep apnoea in determining benefits to auto-CPAP therapy. Eur Respir J 2001;18:170-5.

Svensson M, Holmstrom M, Broman JE, Lindberg E. Can anatomical and functional features in the upper airways predict sleep apnea? A population-based study in females. Acta Otolaryngol 2006;126:613-20.

Terris DJ, Hanasono MM, Liu YC. Reliability of the Mueller maneuver and its assocation with sleep-disordered breathing. Laryngoscope 2000;110:1818-23.

Tsai WH, Remmers JE, Brant R, Flemons WW, Davies J, Macarthur C. A decision rule for diagnostic testing in obstructive sleep apnea. Am J Respir Crit Care Med 2003;167:1427-32.

Tyrrell J, Poulet C, Pe Pin JL, Veale D. A preliminary study of psychological factors affecting patients’ acceptance of CPAP therapy for sleep apnoea syndrome. Sleep Med 2006;7:375-9.

Violani C, Lucidi F, Robusto E, Devoto A, Zucconi M, Ferini Strambi L. The assessment of daytime sleep propensity: A comparison between the Epworth Sleepiness Scale and a newly developed Resistance to Sleepiness Scale. Clin Neurophysiol 2003;114:1027-33.

Weaver EM, Kapur V, Yeuh B. Polysomnography vs. self-reported measures in patients with sleep apnea. Arch Otolaryngol Head Neck Surg 2004;130:453-8.

Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230-5.