[Primary evaluation of the simplified Chinese version of STOP-Bang scoring model in predicting obstructive sleep apnea hypopnea syndrome].

OBJECTIVE: To evaluate the predictive significance of the simplified Chinese STOP-Bang (S-B) scoring model in predicting obstructive sleep apnea hypopnea syndrome(OSAHS).

METHOD: A total of 114 cases with suspected OSAHS were included in this study. S-B questionnaire was filled before polysomnography (PSG) monitoring. According to the PSG monitoring results the patients were divided into simple snoring, mild, moderate and severe OSAHS groups. The average S-B scores and the ratio of patients with S-B score > or = 3 were compared among the four groups respectively. The accuracy of S-B scoring model was analyzed and predictive significance was evaluated.

RESULT: Eighty nine of 114 (78.1%) patients were classified as being at high risk of OSAHS by S-B scoring. The S-B scores showed an accrescent tendency as the severity of OSAHS increased. The scores in the four groups were 2.74 +/- 1.39, 3.25 +/- 1.16, 4.30 +/- 1.20, 4.79 +/- 1.41 respectively. Significant difference in S-B scores was found between severe OSAHS group and the other three groups respectively (P < 0.01), and also between moderate OSAHS group and mild OSAHS/simple snoring group (P < 0 01). The ratio of patients with high risk of OSAHS predicted by S-B scoring in OSAHS group (85.7%) was significantly higher than that in simple snoring group (47.8%) (P < 0.01). The sensitivities of S-B scoring model for AHI > or = 5 > or = > 15 and > or = 30 were 85.7%, 92 5% and 100%, respectively; the negative predictive values were 48.0% ,80.0% and 100%. The coincidence rate of S-B scoring model with golden standard to diagnose OSAHS was 78 9 . The area under the ROC curve at AHI cutoff of greater than or equal to 5 was 0 774.

CONCLUSION: S-B scoring model can be used to evaluate patients with high risk of OSAHS. It is a concise and easy-to-use tool to screen and predict moderate and severe OSAHS with relative high sensitivity and negative predictive value.