Obstructive sleep apnea (OSA) is characterized by repeated complete or partial collapse of the pharyngeal airway during sleep, causing cessation of airflow (apnea) or shallow breathing (hypopnea). Persons with OSA may have repeated arousals from sleep (to reestablish breathing) with each episode of apnea or hypopnea. The resulting sleep disruption often leads to daytime somnolence and compromised neurocognitive function.
This pattern of sleep arousal, coupled with intermittent hypoxemia, is associated with serious adverse cardiovascular outcomes, including stroke. Among surgical patients, OSA is associated with postoperative complications and the need for increased medical intervention. This review discusses why OSA is important in the perioperative setting, preoperative screening for OSA risk, and perioperative management of patients with likely or confirmed OSA.
OSA AT A GLANCE
Prevalence in the general population
Four percent of middle-aged men and 2% of middle-aged women meet minimal diagnostic criteria for OSA, according to a landmark cohort study from the 1990s.1 This makes OSA more common than asthma among adults. Risk increases with age, as 24% of persons older than 65 years have OSA and up to 50% of nursing home residents have clinically significant OSA.2
Prevalence in the surgical population
The prevalence of OSA in the surgical population is higher than that in the general population, and it can vary widely according to the underlying medical condition. A study of 433 patients undergoing general surgery reported a 3.2% prevalence of OSA,3 but this study excluded patients undergoing cardiac surgery, in whom the risk of OSA is higher. In contrast, the prevalence of OSA among obese bariatric surgery patients has been reported at greater than 70%.4 Notably, the patients in the general surgery study3 who appeared to be at risk for OSA based on screening questions were invited to participate in a sleep study, whereas all patients in the bariatric surgery study4 were evaluated through sleep studies. It is likely that the prevalence of OSA among the general surgery study patients would have been higher if all patients had been evaluated with polysomnography.
OSA can occur when any part of the upper airway does not function normally. Upper airway patency is determined by muscle activity, craniofacial and soft tissue structure, and sleep state. During sleep, upper airway muscles are relaxed, which reduces airway patency. Sleep is associated with pharyngeal narrowing and substantially increased inspiratory resistance even among persons without sleep apnea. A person who is awake can compensate for abnormal pharyngeal function through increased muscle activity. During sleep this muscle compensation fails, resulting in partial collapse and subsequent snoring, and sometimes prolonged obstructive hypoventilation. Complete closure results in apnea.
WHY OSA MATTERS
Health consequences of OSA
OSA is associated with serious health consequences, such as increased risk of motor vehicle accidents, stroke, and a number of cardiovascular conditions—hypertension, coronary artery disease, and atrial fibrillation.
Accidents. The daytime hypersomnolence resulting from OSA contributes to reduced vigilance and is likely responsible for an increased incidence of motor vehicle accidents. One study found that among a sample of men and women with unrecognized OSA undergoing polysomnography studies, the likelihood of motor vehicle accidents during the prior 5 years was significantly correlated with the subjects’ apnea-hypopnea index (AHI) score, which reflects the number of apnea or hypopnea episodes per hour of sleep.5 Other studies have demonstrated similar associations.
Stroke. Numerous observational studies have demonstrated an elevated prevalence of OSA among patients with stroke as compared with the general population, but these studies did not adjust for other cerebrovascular risk factors. A recent observational cohort study aimed to address this evidence gap by using proportional hazards analysis to determine the independent effect of OSA on the incidence of stroke or death from any cause among persons with no history of stroke or myocardial infarction.6 Study participants were 1,022 consecutive patients who underwent polysomnography for evaluation of sleep-disordered breathing. OSA was identified in 68% of patients. During the 3.4-year follow-up period, 22 strokes and 50 deaths occurred among the 697 patients with OSA compared with 2 strokes and 14 deaths among the 325 patients without OSA. The probability of survival was significantly lower for patients with OSA compared with their counterparts without OSA (P < .003). After adjustment for other risk factors, OSA was significantly associated with stroke or death (hazard ratio = 1.97; 95% CI, 1.12–3.48).6
Hypertension. Four large studies involving a total of 10,708 patients evaluated for sleep-disordered breathing have established an association between OSA and hypertension risk.7–10 In each study, the risk of hypertension rose linearly with AHI scores. Clinically significant OSA, defined as an AHI score greater than 15, roughly doubled the risk of hypertension compared with the absence of apnea/hypopnea episodes, with odds ratios ranging from 1.37 to 2.89 across the four studies.7–10 Each apnea event per hour of sleep was estimated to increase the odds of developing hypertension by approximately 1%.8 Notably, the effects of OSA on blood pressure are most pronounced in patients younger than age 50.7
Coronary artery disease. The Sleep Heart Health Study evaluated the association between sleep-disordered breathing and cardiovascular disease in 6,424 community-dwelling adults undergoing home polysomnography.11 The population’s median AHI score was 4.4. At least one cardiovascular event was reported by 16% of participants. Sleep-disordered breathing was associated with self-reported heart failure, stroke, and, more modestly, coronary artery disease. A linear relationship was noted between AHI and cardiovascular risk.
Snoring, which is often an indicator for OSA, has also been associated with cardiovascular risk. The Nurses’ Health Study evaluated 71,000 women who completed medical questionnaires that included questions about snoring. Over 8 years of follow-up, the relative risks for cardiovascular disease were 1.46 among occasional snorers (95% CI, 1.23–1.74) and 2.02 among regular snorers (95% CI, 1.62–2.53) in comparison with nonsnorers. Snoring, even without a diagnosis of OSA, emerged as an independent risk factor for cardiovascular disease.12
Atrial fibrillation. OSA has been identified as a predictor of new-onset atrial fibrillation in a retrospective cohort study (hazard ratio = 2.18; 95% CI, 1.34–3.54).13 In a prospective study, patients with atrial fibrillation but normal left ventricular function were found to have significantly higher AHI scores than matched normal controls.14 After adjustment for relevant covariates, the odds ratio for an association between atrial fibrillation and significant sleep-disordered breathing (AHI score > 15) was 3.04 (95% CI, 1.24–7.46).14 In another prospective trial, patients with atrial fibrillation and OSA who underwent cardioversion were at increased risk for a recurrence of atrial fibrillation if OSA was untreated (82% for untreated vs 42% for treated OSA; P = .013).15