The Screening Of Patients With Obstructive Sleep Apnea Hypopnea Syndrome

Objectives:To evaluate the ability of visceral adiposity variables(LAP,VAI and Ty G),ESS,STOP and STOP-BANG questionnaires in predicting obstructive sleep apnea hypopnea syndrome(OSAHS),and estimate the impact of gender on the prediction.Methods: Part 1 included consecutive subjects referred for suspected OSAHS from January,2007 to July,2013.Anthropometric and biological measurements,Epworth sleepiness scale(ESS),polysomnography(PSG)were examined in all subjects.Part 2included consecutive subjects referred for suspected OSAHS from October 2010 to January 2013.All included subjects completed ESS,STOP-BANG questionnaire,anthropometric measurements and overnight PSG.Receiver operating characteristic(ROC)curve analyses were used to evaluate the efficiency of LAP,VAI,Ty G,ESS,STOP,STOP-BANG and prediction models in predicting OSAHS between males and females.Results: A total of 2983 subjects were included in Part 1.ROC curve suggested that LAP and Ty G had moderate diagnostic accuracy for OSAHS,within which LAP was better.Besides,we found differences of diagnostic efficiency between males and females.In males,the area under ROC curve(AUC)for LAP was 0.741(0.700-0.759),with a sensitivity of 77.8% and a specificity of 57.8%;in females,LAP had an AUC of0.764(0.724-0.804),a sensitivity of 75.9% and specificity of 68.1%.In addition,we constructed a prediction model based on anthropometrics and visceral adiposity variables,which showed the sensitivity and specificity of 77.3%,67.2% in males and86.5%,63.2% in females.A total of 1087 adults were included in Part 2.ROC curve demonstrated that both STOP and STOP-BANG questionnaires had better diagnostic accuracy than ESS.The STOP-BANG based prediction model had an AUC of0.920(0.900-0.937),a sensitivity of 78.0% and a specificity of 90.0% in males;an AUC of 0.942(0.912-0.971),a sensitivity of 95.0% and a specificity of 83.5% in females.In addition,analysis showed that STOP-BANG questionnaire had different cut-off values between males and females,which was 3 in males and 2 in females.Conclusions: LAP,TyG,STOP and STOP-BANG had moderate efficiency in screening OSAHS.The STOP-BANG-based prediction model had excellent diagnostic efficiency,and could be applied to clinical use.Additionally,gender difference in prediction of OSAHS should be considered in clinical practice.