Development And Validation Of An Obstructive Sleep Apnea Syndrome Screening Score For Hypertensive Patients

Objective There are many screening tools of obstructive sleep apnea syndrome(OSAS).Their design are based on the community population.However,hypertension is closely related to OSAS.Whether these tools are applicable to the hypertensive population needs to be further verified.At present,none of screening tools are suit for this population.As we know,blood pressure always manifest itself as an increasing statement at night in patients with OSAS.In this study,we will test the blood pressure level of all participates by ambulatory blood pressure and monitor the sleep status by polysomnography.According to the results,we aim to design a score model which is used to screen OSAS in patients with OSAS.At the same time,verify the application value of the existing common scale in this population.Methods In this study,596 hypertensive patients were hospitalized in our hospital from September to November in 2018.According to the inclusion criteria,463 patients were finally included and provided anthropometric indexes,personal history,No-Apnea model,No SAS score,STOP-Bang questionnaire and Epworth sleep Scare(ESS).According to Apnea-hypopnea index(AHI),all subjects were classified two groups:non-OSAS groups(AHI < 5 times / h)and OSAS groups(AHI ? 5 times / h).The clinical score(new model)was established by multivariate analysis and logistic regression.According to the Youden index,we selected cutoff point to evaluate the patients' condition whether they were at the high risk of OSAS.We also evaluated the goodness of fit and stability of the model by Hosmer-lemeshow test and Bootstrap method.In order to evaluate the accuracy of the score model,we draw the receiveroperating characteristic(ROC)curve and calculated the Sensitivity,specificity,positive predictive value and negative predictive value and the Area Under Curve(AUC).Results The New Model is composed 5 factors,age,neck circumference,gender,snoring and highest nocturnal systolic blood pressure,which ranges from 0 to 20,and allocates 2 points for being old of 41 years to less 49 years or 3 points for being old of49 years old to less than 57 years or 6 points for being old of 57 years or more,1 point for having a neck circumference of 38 cm to less than 41 cm or 2 points for having a neck circumference of 41 cm to less than 43 cm or 4 points for having a neck circumference of 43 cm or more,3 points for male,3 points for snoring and 1 point for having maximum nocturnal systolic blood pressure of more than 135 mm Hg to 150 mm Hg or 2 points for having maximum nocturnal systolic blood pressure of more than150 mm Hg to 165 mm Hg or 4 points for more than 165 mm Hg.According to the Youden index,we select the 10 points or more as the cut-off points,and identify individuals at risk of clinically significant OSAS,with an area under the curve(AUC)of 0.775(95%CI 0.734-0.812).It performs significantly better than STOP-Bang questionnaire(AUC 0.702 [0.642-0.761]),No SAS score(AUC 0.720 [0.661-0.778]),No-Apnea model(AUC 0.730 [0.675-0.785])and ESS(AUC 0.607[0.544-0.670]).Conclusions The screening model established in this study has a better screening value for OSAS and perform better than STOP-Bang questionnaire,No SAS score,No-Apnea model and ESS,the screening model established in this study has a better screening value for OSAS.Therefore,the score is more accurate for screening OSAS in hypertensive patients.