Classic risk factors for obstructive sleep apnea (OSA) include obesity, male gender, and age. Recent studies have found that body mass index (BMI) and age correlate poorly with the prevalence and severity of the disease. We aimed to identify the clinical variables that predict OSA in a symptomatic population with emphasis on nonobese individuals (BMI < 30 kg/m2).
Retrospective review of demographic and polysomnographic data of consecutive patients referred for evaluation of OSA between March 2007 and July 2008. Data were analyzed using a multivariable logistic regression model.
611 patients were included. Among these, 452 (73.9%) had OSA. In the total study group, a 1 unit higher BMI corresponded to 8% (95% CI, 5% to 11%) higher odds of OSA and for each 10-year increase in age, the odds of OSA increased by 4% (95% CI, 2% to 5%). Males had 4.7 (95% CI, 2.9 to 7.6) times the odds of having OSA. Severe snoring was associated with a higher risk of OSA (p<0.001). Epworth Sleepiness Scale (ESS) score and race were not associated with the disease (p=0.63 and p>0.18, respectively). Of the 611 patients, 105 (17.2%) were nonobese. Among these, 54 (51.4%) had OSA. In this subgroup, none of BMI, snoring, race and ESS score was associated with an increased risk of OSA (all p>0.31). Males had 11.7 (95% CI, 3.8 to 35.7) times the odds of having OSA. For each 10-year increase in age, the odds of OSA increased by 44% (95% CI, 4% to 85%).
Age and male gender were the only independent predictors of OSA in both study groups. In nonobese individuals, severe snoring and BMI were no longer associated with a higher risk of OSA.
OSA is common in nonobese individuals. The preconception that OSA is restricted to older obese patients who snore leads to underdiagnosis of the disease in other populations at risk. OSA should be considered in patients reporting daytime hypersomnolence regardless of BMI or snoring history.
Juan Cadavid, No Financial Disclosure Information; No Product/Research Disclosure Information