Application value of joint NoSAS score and Epworth Sleepiness Scale for assessment of obstructive sleep apnea hypopnea syndrome

Objective: By comparing the predictive value of the NoSAS (Neck circumference, Obesity, Snoring, Age and Sex) score combined with the Epworth Sleepiness Scale (ESS), STOP-Bang Questionnaire (STOP Bang), STOP Questionnaire (STOP) and Berlin Questionnaire (Berlin), the application value of the NoSAS score combined with ESS in screening Obstructive sleep apnea hypopnea syndrome (OSAHS) in the population is evaluated.Method: 2560 suspected OSAHS patients visited the Sleep Medical Center of the First Hospital of Guangzhou Medical University between September 1, 2016 and October 31, 2020, and were monitored with a polysomnogram (PSG) after completing the NoSAS score, ESS, STOP-Bang, STOP and Berlin. The sensitivity, specificity, positive predictive value, negative predictive value and receiver operating characteristic (ROC) curve of each scale were calculated, and the accuracy in predicting OSAHS of the NoSAS score combined with ESS and each scale was analyzed.Results: The areas under the ROC curve scored by Berlin were higher than those of the other four questionnaires with Apnea Hypopnea Index (AHI) cutoffs of >5 and > 10 events/h, while the area under the ROC curve scored by the NoSAS score was the highest with AHI cutoffs of >15, >20, >25 and > 30 events/h. Among the five scales, the diagnostic odds ratio (DOR) of the NoSAS score was the highest. When a NoSAS score of >7 was used as the cutoff point for diagnostic NoSAS, it had higher sensitivity and specificity with a NoSAS score of >8 as the cutoff point for diagnostic NoSAS. A NoSAS score of >7 combined with ESS significantly improved its specificity for predicted OSAHS patients.Conclusion: The NoSAS score is a simple and effective new tool for screening patients for OSAHS, while a NoSAS score of >7 combined with ESS can further improve its specificity. Thus, we suggest further screening with ESS after a NoSAS score of >7 in suspected populations.(c) 2022 Elsevier B.V. All rights reserved.