Medical Grand Rounds

Sleep apnea ABCs: Airway, breathing, circulation

Author and Disclosure Information



Inflammatory and atherogenic effects

Increased levels of markers of systemic inflammation, prothrombosis, and oxidative stress have been observed in OSA and may be key pathophysiologic links between OSA and cardiovascular sequelae. OSA has been associated with up-regulation of a number of inflammatory mediators: interleukin (IL) 6, soluble IL-6 receptor, IL-8, tumor necrosis factor alpha, and C-reactive protein. Soluble IL-6 levels in particular are higher in people who have sleep-disordered breathing, as reflected by the apnea-hypopnea index independent of obesity, with relationships stronger in the morning than in the evening. This likely reflects the overnight OSA-related physiologic stress.11

Thrombotic potential is also enhanced, with higher levels of plasminogen activator inhibitor 1, fibrinogen, P-selectin, and vascular endothelial growth factor. Some of these factors normally have a diurnal cycle, with higher levels in the morning, but in OSA, increasing OSA severity is associated with increased prothrombotic potential in the morning hours. Of interest, levels of these substances showed a plateau effect, rising in people who had only mildly elevated apnea-hypopnea indices and then leveling off.12 Intermittent hypoxia followed by ventilatory overshoot hyperoxia, characteristic of sleep apnea, provides the ideal environment for augmentation of oxidative stress, with evidence of increased oxidation of serum proteins and lipids. Hypoxia and oxygen-derived free radicals may result in cardiac myocyte injury. Experimental data demonstrate that intermittent hypoxia combined with a high-fat diet results in synergistic acceleration of evidence of atherogenic lesions.

Patients with OSA also have evidence of endothelial dysfunction, insulin resistance, and dyslipidemia, all pathways that can facilitate the progression of atherosclerosis in OSA.13–15

Cardiac arrhythmias

In the Sleep Heart Health Study, a multicenter epidemiologic study designed to examine the relationships of OSA and cardiovascular outcomes, those who had moderate to severe OSA had a risk of ventricular and atrial arrhythmias two to four times higher than those without OSA, even after correction for the confounding influences of obesity and underlying cardiovascular risk.14 These findings were corroborated in subsequent work highlighting monotonic dose-response relationships with increasing OSA severity and increased odds of atrial and ventricular arrhythmia in a cohort of about 3,000 older men.11 Additional compelling evidence of a causal relationship is that the risk of discrete arrhythmic events is markedly increased after a respiratory disturbance in sleep.16

In patients who successfully underwent cardioversion for atrial fibrillation, those who had sleep apnea but were not treated with continuous positive airway pressure (CPAP) had a much higher rate of recurrence of atrial fibrillation during the subsequent year than those with CPAP-treated sleep apnea and than controls never diagnosed with sleep apnea. In the untreated patients with sleep apnea, the mean nocturnal fall in oxygen saturation was significantly greater in those who had recurrence of atrial fibrillation than in those who did not, suggesting hypoxia as an important mechanism contributing to atrial fibrillation.17

Since then, several other retrospective studies have shown similar findings after pulmonary vein antrum isolation and ablation in terms of reduction of atrial fibrillation recurrence with CPAP treatment in OSA.18

Walia et al19 described a patient with moderate sleep apnea who underwent a split-night study. During the baseline part of the study, the patient had about 18 ectopic beats per minute. During the second portion of the study while CPAP was applied, progressively fewer ectopic beats occurred as airway pressure was increased until a normal rhythm without ectopic beats was achieved at the goal treatment CPAP pressure setting.

Cardiovascular disease, stroke, and death

Marin et al20 followed about 1,500 men for 10 years, including some who had severe OSA, some with sleep apnea who were treated with CPAP, and controls. The risk of nonfatal and fatal cardiovascular disease events was nearly three times higher in those with severe disease than in healthy participants. Those treated with CPAP had a risk approximately the same as in the control group.

The Sleep Heart Study followed approximately 6,000 people with untreated sleep apnea for a median of nearly 9 years. It found a significant association between the apnea-hypopnea index and ischemic stroke, especially in men.21 Survival in patients with heart failure is also associated with the degree of OSA; patients with more severe disease (an apnea-hypopnea index ≥ 15) have a nearly three times greater risk of death than those with no disease or only mild disease (apnea-hypopnea index < 15).22

From the standpoint of health care utilization, findings that central sleep apnea predicts an increased risk of hospital readmission in heart failure are of particular interest.23

People with OSA are at increased risk of nocturnal sudden cardiac death.24 Sleep apnea is also associated with an increased overall death rate, and the higher the apnea-hypopnea index, the higher the death rate,25 even after adjusting for age, sex, body mass index, and underlying cardiovascular risk, with findings most pronounced in men under age 70.

Motor vehicle accidents

The need for caution during driving should be discussed with every patient, as motor vehicle accidents are an immediate danger to the patient and others. The association with motor vehicle accidents is independent of sleepiness, and drivers with sleep apnea often do not perceive performance impairment. Young et al26 found that men who snored were 3.4 times as likely to have an accident over a 5-year period, and that men and women with an apnea-hypopnea index greater than 15 were more than 7 times as likely to have multiple accidents over a 5-year period, highlighting the importance of discussing, documenting, and expeditiously diagnosing and treating OSA, particularly in those who report drowsiness while driving.


Risk factors can be divided into nonmodifiable and modifiable ones.

Nonmodifiable factors

Age. Bimodal distributions in OSA prevalence have been observed; ie, that the pediatric population and people who are middle-aged have the highest prevalence of OSA. A linear relationship between sleep apnea prevalence and age until about age 65 was identified in data from the Sleep Heart Health Study.27 After that, the prevalence rates plateau; it is unclear if this is secondary to natural remission of the disease after a certain age or because patients with more severe disease have died by that age (ie, survivorship bias), blunting an increase in prevalence.

Sex. Men develop sleep apnea at a rate three to five times that of women. Several explanations have been proposed to account for this.28,29 Sex hormones are one factor; women with sleep apnea on hormone replacement therapy have a significantly less-severe sleep apnea burden than other postmenopausal women,30 suggesting a positive effect from estrogen. Sex-based differences in fat distribution, length and collapsibility of the upper airway, genioglossal activity, neurochemical control mechanisms, and arousal response may also contribute to prevalence differences between men and women.

As with coronary artery disease, the presentation of sleep apnea may be atypical in women, particularly around menopause. Sleep apnea should be considered in women who have snoring and daytime sleepiness.

Race. Whites, African Americans, and Asians have a similar prevalence of sleep apnea, but groups differ in obesity rates and craniofacial anatomy.31–34 Asians tend to have craniofacial skeletal restriction. African Americans are more likely to have upper-airway soft-tissue risk and to develop more severe OSA. Whites tend to have both craniofacial and soft-tissue risk. For those with craniofacial anatomy predisposing to OSA, even mild obesity can make it manifest.

Syndromes that predispose to OSA can include craniofacial structural abnormalities, connective tissue problems, or alterations in ventilatory control (eg, Marfan, Down, and Pierre Robin syndromes).

Next Article:

A serious complication of a common stress test

Related Articles

  • Letters To The Editor

    Sleep apnea ABCs

    We believe that arousal contribute to the pathophysiology of obstructive sleep apnea.

  • Letters To The Editor

    In reply: Sleep apnea ABCs

    Whether cortical arousal during sleep is bad or good is controversial.