Good News, Bad News: Sleep Apnea and Cardiovascular Risk

Abstract & Commentary

Synopsis: Middle-aged men with sleep apnea have an increased risk of cardiovascular disease compared with controls, even when controlling for age, body mass index, and blood pressure. Effective treatment reduces that risk.

Source: Peker Y, et al. Am J Respir Crit Care Med. 2002;166:159-165.

This study is a prospective, controlled look at development of cardiovascular disease (CVD) in 182 men whose average age was 46.8 years at enrollment. Sixty of these men had obstructive sleep apnea (OSA), with a mean oxygen desaturation index (ODI) of 16.5 events per hour of sleep. (ODI is roughly equivalent to Apnea plus Hypopnea Index [AHI], or respiratory disturbance index [RDI]. Alas, these terms are used interchangeably in the sleep literature.) The OSA patients were slightly heavier and older, had higher blood pressures, and desaturated more than did the 122 controls, though the controls smoked more. The OSA patients were relatively thin by US standards (mean BMI, 27.9 kg/m2), and they were offered continuous positive airway pressure (CPAP), uvulopalatopharyngoplasty (UPPP), or an oral appliance. CPAP was titrated in-laboratory, and compliance was objectively assessed 3 and 12 months after initiation. Patients who received UPPP or oral appliances had follow-up studies to assess "efficiency." Efficiently treated patients were considered to be those who used CPAP at least 50% of sleep hours, or who had fewer than 30 oxygen desaturations of 4% (which was probably about an ODI or RDI of 5 events per hour of sleep or less) after UPPP or while using the oral appliance.

Of the 60 patients with sleep apnea, 14 chose CPAP, 22 chose UPPP, 4 chose an oral appliance, and 19 did not receive treatment (I know this only adds up to 59, but this is what the article says). The 19 who did not receive treatment were asymptomatic or had mild sleep-disordered breathing. Nine (64%) of those who received CPAP returned the machine, and 50% of those who had surgery still had an ODI over 15 at follow-up. Only 1 in 4 patients treated with an oral appliance was felt to be adequately treated.

More than 50% of those with incompletely treated OSA had at least 1 cardiovascular diagnosis (hypertension, angina, atrial fibrillation, MI, cardiomyopathy, cardiac failure, or stroke) during the follow-up period. None of the patients with effectively treated OSA patients did. In a logistic regression analysis, controlling for age, BMI, smoking, and blood pressure, both age and OSA remained as significant predictors of the risk of CVD. A separate analysis revealed the ODI and the baseline minimum oxygen saturation to be significant predictors of CVD, without regard to treatment. Furthermore, those subjects who were felt not to have significant OSA, but who had 15-29 oxygen desaturations during the diagnostic test had a highly significant increase in incident CVD compared with those with lower levels of oxygen desaturation.

Comment by Barbara A. Phillips, MD, MSPH

The bad news is that sleep apnea appears to be a definite risk factor for cardiovascular disease, even when controlling for blood pressure, obesity, age, BMI, and gender. The good news is that effective treatment appears to reduce that risk substantially. The Sleep Heart Health Study has already demonstrated a very strong relationship between sleep-disordered breathing and cardiovascular disease.1 However, the current study is the first prospective, controlled study to look at sleep apnea, its treatment, and cardiovascular risk in a well-described cohort without CVD or hypertension at baseline.

We already know that OSA is a risk factor for hypertension,2-5 and that CPAP treatment reduces blood pressure in sleep apnea patients.6-8 What has been less clear is whether OSA is a risk factor for cardiovascular disease independent of the associated hypertension. This study suggests that it is, and that the severity of oxygen desaturation is the prime determinant of CVD risk.

One reservation I had about this study was the finding that CPAP and UPPP were each about 50% effective in this population. These Swedish subjects were much thinner than the typical American sleep apneic, and we know that obesity predicts poor outcome with surgical treatment of sleep-disordered breathing.9,10 Further, the ODI for the surgically treated patient fell as a group to a statistically significant degree, and half were considered efficiently treated, but I don’t think that an overall fall in ODI from 15 to 12 events/hour is clinically satisfactory, especially since Peker and colleagues point out that oxygen desaturation indices in the moderate range were a risk for the development of cardiovascular disease.

Because Peker et al used the ODI as their primary measure of severity of OSA, they have strengthened the relationship between oximetry findings and sequelae of OSA. While purists (and those who own sleep labs) will continue to insist that in-lab sleep testing is the only way to make a diagnosis of OSA, this study and others like it lend support to the notion that a good clinician with an oximeter can predict the likelihood of OSA with some degree of confidence. Since CPAP is a benign, effective, relatively inexpensive treatment, and since OSA affects about 3% of us and is growing more prevalent (as we get fatter), it is likely that insurance companies will soon permit good clinicians to do just that! 

Dr. Phillips, Professor of Medicine, University of Kentucky; Director, Sleep Disorders Center, Samaritan Hospital, Lexington, KY, is Associate Editor of Internal Medicine Alert.


1. Shahar E, et al. Am J Respir Crit Care Med. 2001;163: 19-25.

2. Lavie P, et al. BMJ. 2000;320:479-482.

3. Nieto FJ, et al. JAMA. 2000;283:1829-1836.

4. Grote L, et al. Am J Respir Crit Care Med. 1999;160: 1875-1882.

5. Peppard PE, et al. N Engl J Med. 2000;342:1378-1384.

6. Pepperell JCT, et al. Lancet. 2002;359:204-214.

7. Dimsdale JE, et al. Hypertension. 2000;35:144-147.

8. Faccenda JF, et al. Am J Respir Crit Care Med. 2001; 163:344-348.

9. Sher AE, et al. Sleep. 1996;19:156-177.

10. Larsson LH, et al. Laryngoscope. 1994;104:1362-1368.

Readers are Invited. . .

Readers are invited to submit questions or comments on material seen in or relevant to Internal Medicine Alert. Send your questions to: Neill Larmore, Internal Medicine Alert, c/o American Health Consultants, P.O. Box 740059, Atlanta, GA 30374. For subscription information, you can reach the editors and customer service personnel for Internal Medicine Alert via the internet by sending e-mail to We look forward to hearing from you.