To assess the clinical value of STOP-Bang questionnaire (SBQ) and modified mallampati score (MM) in predicting obstructive sleep apnea (OSA).

A total of 40 cases (6 females and 34 males) hospitalized in the Respiratory Sleep Center of our hospital were enrolled in this study. All the patients were prospectively predicted for risk of OSA using SBQ and MM, and their sleep quality was monitored by polysomnography (PSG). According to the apnea-hypopnea index (AHI), the patients were divided into three groups: AHI>5/h group, AHI>15/h group, and AHI>30/h group. Then all the patients received SBQ and MM assessment. Finally, we calculated the sensitivity, specificity, positive predictive values and negative predictive values for all the patients of both scores separately and in combination.

In this study, 44 patients were screened and 40 eligible subjects were included, including 34 males (85%) and 6 females (15%). Their average age was (45.1±14.6) years, their average neck circumference was (43.5±5.9)cm, their average body mass index (BMI) was 28.87 (26.35-32.13) kg/m^2, and their average AHI was 50.6 (16.22-70.98)/h. No OSA was found in 3 cases (7.5%), mild OSA in 1 case (2.5%), moderate OSA in 6 cases (15%) and severe OSA in 30 cases (75%). The results showed that SBQ≥3 had higher sensitivities (91.89%, 91.67%, and 93.33%, respectively) in predicting OSA, but lower specificities (33.33%, 25.00%, and 20.00%, respectively) were found. The sensitivities of MM≥Ⅲ in predicting OSA were 78.39%, 80.56%, 83.33%, respectively, and the specificities were 33.33%, 50%, 40%, respectively. The sensitivity of SBQ≥3 or MM≥Ⅲ in predicting OSA was higher, but the specificity was not ideal. The specificities (66.67%, 75.00%, and 50.00%, respectively) in predicting OSA were significantly improved by the combination of the two methods, and the sensitivities (75.68%, 77.78%, and 80.00%, respectively) were only slightly reduced. Among them, only moderate and severe OSA (AHI>15/h) passed the Kappa consistency test with PSG (P<0.05).

The combination of SBQ≥3 and MM≥Ⅲ for the patients treated in our sleep center can significantly improve the specificity in predicting OSA and at the same time has good sensitivity, especially for the patients with moderate or severe OSA. It has good consistency with PSG, therefore, it can be used for OSA screening.