Rapid risk stratification for obstructive sleep apnea, based on snoring severity and body mass index

OBJECTIVE: It is unclear whether all snoring patients require polysomnography. and there are no highly sensitive clinical predictors of sleep apnea. Our objective was to develop a simple clinical screening test for OSA in snoring patients. STUDY DESIGN: Prospective, IRB-approved study at a university sleep disorders center. SUBJECTS AND METHODS: In 211 patients undergoing polysomnography. snoring severity, Epworth sleepiness scale, body mass index, demographic, and sleep study data were collected. Receiver operating characteristic (ROC) analysis and Pearson correlation were used to develop a sensitive screening test for OSA. RESULTS: Snoring severity score (SSS) and BMI were the two most accurate predictors of OSA on the ROC Curve. A bipartite threshold of SSS 4 or BMI = 26 carried sensitivity of 97.4%, specificity of 40%. positive predictive value of 82.3%, and negative predictive value of 84.2% for moderate/severe OSA. Patients at high risk were those with BMI >= 32 (89% PPV) or SSS >= 7 (92% PPV). CONCLUSIONS: The statistic most predictive of OSA was snoring severity, Combining this with BMI yielded a highly sensitive screening test for mode rate/severe OSA. This clinical assessment may be useful in risk-stratifying patients for polysomnography and therapy. facilitating deferred work-up in low-risk patients and expedited therapy in high-risk patients. (C) 2008 American Academy of Otolaryngology-Head and Neck Surgery Foundation. All rights reserved.