Validation of an obstructive sleep apnea symptom inventory in persons with multiple sclerosis

Background: Improved screening for obstructive sleep apnea (OSA) could enhance multiple sclerosis (MS) clinical care; yet the utility of current screening tools for OSA have yet to be evaluated in persons with multiple sclerosis (PwMS). Objectives: The STOP-Bang Questionnnaire is an 8-item screening tool for OSA that is commonly used in non-MS samples. The aim of this study was to assess the validity of the STOP-Bang in PwMS. Methods: STOP-Bang and polysomnography data were analyzed from n = 200 PwMS. Sensitivity, specificity, positive-, and negative-predictive value (PPV and NPV) were calculated, with receiving operating characteristic (ROC) curves, for each STOP-Bang threshold score, against polysomnography-confirmed OSA diagnosis at three apnea severity thresholds (mild, moderate, and severe). Results: Nearly 70% had a STOP-Bang score of > 3% and 78% had OSA. The STOP-Bang at a threshold score of 3 provided sensitivities of 87% and 91% to detect moderate and severe OSA, respectively; and NPV of 84% and 95% to identify PwMS without moderate or severe OSA, respectively. Sensitivity to detect milder forms of OSA was 76%. The NPV to identify persons without milder forms of OSA was 40%. Conclusion: The STOP-Bang Questionnaire is an effective tool to screen for moderate and severe OSA in PwMS, but may be insufficient to exclude mild OSA.