Validity of the Berlin Questionnaire in identifying obstructive sleep apnea syndrome when administered to the informants of stroke patients

Given the high prevalence of sleep-disordered breathing (SDB) in stroke and its importance as a vascular risk factor, a clinical instrument to assess its incidence would be useful. Acute stroke patients (n = 121) were stratified into high- and low-risk groups for SDB using a modified Berlin Questionnaire (BQ) administered to the informants who were living with the patient. After a minimum of 4 weeks from stroke onset, patients who were fit underwent overnight polysomnography (PSG). On stratifying risk of obstructive sleep apnea (OSA) in these patients based on the BQ, 53% belonged to the high-risk group and 47% belonged to the low-risk group. There was poor correlation between the clinical questionnaire results and PSG findings, with sensitivity of 66.7%, specificity of 55.6%, a positive predictive value of 63.4%, and a negative predictive value of 58.8%.