Accuracy of Berlin questionnaire with different body mass index cutoff values in the screening of patients with obstructive sleep apnea

Abstract

Objective: To evaluate the accuracy of Berlin questionnaire in the screening of patients with obstructive sleep apnea (OSA), after using different body mass index (BMI) cutoff values. Methods: The data were collected from September, 2011 to May, 2012 at the Sleep Medicine Center of West China Hospital, Sichuan University (Chengdu, China). A total of 350 consecutive patients (302 males and 48 females, 19-68 years old, average (42.9±11.0) years old) with suspected OSA who underwent an overnight polysomnography (PSG) were recruited into this study. Berlin questionnaire was completed. Based on the severity of OSA which were determined by apnea-hypopnea index (AHI), the patients were classified into four groups of primary snoring (<5 events/h), mild (5-<15 events/h), moderate (15-<30 events/h) and severe OSA (≥30 events/h). Sensitivity, specificity, positive predictive values, negative predictive values and the area under the receiver operating characteristics curve of Berlin questionnaire with different BMI cutoff values were calculated. Results: In patients with BMI≥30 kg/m(2,) to diagnose mild-moderate-severe (AHI≥5 events/h), moderate-severe (AHI≥15 events/h), and severe (AHI≥30 events/h) OSA, the sensitivity of Berlin questionnaire were 69.2%, 71.6%, and 76.6%; the negative predictive values were 15.7%, 33.0%, and 56.5%, respectively. In patients with BMI≥24 kg/m(2,) questionnaire had the highest sensitivity (94.3%, 94.8%, and 96.7%, respectively) and negative predictive values (28.0%, 44.0%, and 72.0%, respectively). Conclusions: When BMI cutoff values gradually declines, the sensitivity and negative predictive values of Berlin questionnaire increase, but the specificity and positive predictive values decline. The accuracy of Berlin questionnaire possibly is related to BMI cutoff values.