[Is the Berlin questionnaire an appropriate diagnostic tool for sleep medicine in pneumological rehabilitation?]

Abstract

Background: The Berlin Questionnaire is an explorative tool of 13 questions designed to identify patients with obstructive sleep apnea. The questions are targeted toward key symptoms of snoring, apneas, daytime sleepiness, hypertension and overweight.

Methods: 153 patients undergoing pulmonary rehabilitation were screened for obstructive sleep apnea via i) ApneaLink (ResMed, Munich) and ii) the Berlin Questionnaire. The Epworth Sleepiness Scale (ESS) was administered to grade daytime sleepiness. Results of ApneaLink screening and questionnaires were prospectively compared. The Respiratory Disturbance Index (RDI) was used to compute predictive performance of the Berlin Questionnaire for obstructive sleep apnea. Results were also compared to previously published data from a primary care patient cohort.

Results: At a cut-off RDI=10/h, the Berlin Questionnaire had a sensitivity of 62.5 %, a specificity of 53.8 %, a positive predictive value of 38.4 %, a negative predictive value of 74.2 %, a positive likelihood ratio of 1.35 and a negative likelihood ratio of 0.69. At a cut-off RDI=15/h, the Berlin Questionnaire had a sensitivity of 67.2 %, a specificity of 52.8 %, a positive predictive value of 25.1 %, a negative predictive value of 84.7 %, a positive likelihood ratio of 1.42 and a negative likelihood ratio of 0.62. There was evidence that the participants of this study suffered from an increased daytime sleepiness (ESS=8.8 +/- 4.8). The study group was more likely to snore, have apneas (men 6.7 % more often, women 12.6 % more often) and suffer from daytime sleepiness (men 19.0 % more often, women 14.6 % more often) than the control group. Also, hypertension and overweight were more prevalent (men 6.1 % more often, women 11.1 % more often).

Conclusions: The Berlin Questionnaire is a poor predictor of obstructive sleep apnea in a random group of patients undergoing pulmonary rehabilitation.