Clinical predictors of sleep apnea in patients undergoing bariatric surgery

Background: Sleep apnea is a frequent and unappreciated condition of morbidly obese patients. If unrecognized it could lead to significant postoperative complications. A clinical tool to assess the severity of sleep apnea is not available. We prospectively determined whether the Epworth Sleepiness Scale (ESS) or body mass index (BMI) predict the severity of sleep apnea in morbidly obese patients. Methods: 66 consecutive patients evaluated for bariatric surgery from June to November 1999 were examined and prospectively administered a health questionnaire including the ESS. Patients with an ESS greater than or equal to6 were referred for polysomnography with calculation of Respiratory Disturbance Index (RDI). Sleep apnea was graded as mild (RDI 6-20), moderate (RDI 21-40) and severe (RDI>40). Clinical variables such as BMI and ESS score were compared using regression analysis. Data are mean +/- SEM. Results: 4 men and 23 women (27/66) who scored >6 on the ESS completed a sleep study. Mean ESS was 13 +/- 4.5. Sleep apnea was mild in 13 patients, moderate in 7, severe in 6, and absent in 1, Mean age was 43 +/- 9.5 years. BMI was 52 +/- 10 kg/m(2). Linear regression analysis did not demonstrate correlation between ESS score and severity of sleep apnea (r(2)=0.03, p>0.05). Multiple regression analysis demonstrated no correlation between BMI, patient snoring, and RDI score. Conclusions: Sleep apnea is frequent in candidates screened for bariatric surgery. ESS is a useful tool to investigate daytime sleepiness and other manifestations of sleep apnea. However, the ESS does not predict the severity of sleep apnea. Clinical suspicion of sleep apnea should prompt polysomnography.