Objective: To evaluate the clinical utility of the NoSAS score in the screening of patients with obstructive sleep apnea-hypopnea syndrome(OSAHS), and to compare the performance of the NoSAS score with other tools including Epworth Sleepiness Scale(ESS), STOP, STOP-Bang(SBQ) and Berlin questionnaires. Methods: A total of 444 consecutive patients(328 males and 116 females) with suspected OSAHS who underwent an overnight polysomnography(PSG) were recruited into this study. Five questionnaires including the NoSAS score, ESS, STOP, SBQ and Berlin were completed. Based on the severity of OSAHS which was determined by apnea-hypopnea index(AHI), the patients were classified into 4 groups: normal(<5 events/h), mild(5-15 events/h), moderate(15-30 events/h) and, severe (≥30 events/h) OSA.Sensitivity, specificity, positive predictive values, negative predictive values and the area under the receiver operating characteristics curve of 5 questionnaires were calculated. Results: With AHI≥5 events/h as the standard diagnosis of OSAHS, the NoSAS score and SBQ questionnaire showed a moderate performance, with the NoSAS score having the largest area under the ROC curve(0.753, P<0.001), followed by the SBQ questionnaire (0.727, P<0.001). The performance of the ESS, Berlin, and the STOP questionnaire was not high. Using mild moderate-severe(≥5 events/h), moderate-severe(≥15 events/h), and severe(≥30 events/h)OSAHS as cutoffs, NoSAS had the highest specificity and positive predictive values(80.2% and 88%, 72% and 69.8%, 66.3% and 50.5%), and the sensitivity and negative predictive values were (51.5% and 36.9%, 56.5% and 59.1%, 66.3% and 74.2%) .SBQ had the highest sensitivity and the negative predictive values(80.2% and 88%, 72% and 69.8%, 66.3% and 50.5%), and the specific and positive predictive values were (45.7% and 81.0%, 39.1% and 61.9%, 34.8% and 44.4%). The NoSAS score ≥ 7 had higher sensitivity and negative predictive value(75.0% and 47.1%, 78.1% and 66.5%, 82.7% and 81.9%)than the NoSAS socre ≥ 8. With AHI≥5 events/h as the standard diagnosis of OSAHS, the NoSAS score and the SBQ questionnaire had a higher accuracy than the other 3 questionnaires as screening questionnaires for diagnosing OSAHS, and the value of DOR were 4.298 and 3.758 respectively. Conclusions: The NoSAS score and the SBQ questionnaire have a moderate performance in diagnosing OSAHS. The NoSAS score is a new screening tool, and it is similar to the SBQ questionnaire, being also simple and effective. While the SBQ questionnaire is more widely used, it is necessary to further evaluate the diagnostic value of NoSAS score.
目的： 评价NoSAS评分在人群筛查OSAHS中的应用价值，比较NoSAS评分与Epworth嗜睡量表(ESS)、STOP、STOP－Bang(SBQ)和Berlin问卷的预测价值。 方法： 对2016年10月至2017年4月广州医科大学附属第一医院睡眠医学中心连续就诊的444例可疑OSAHS患者进行整夜多导睡眠图(PSG)记录，其中男328例，女116例，年龄19～87岁，平均(47±13)岁。所有患者填写NoSAS量表、ESS量表、STOP、SBQ问卷和Berlin问卷，依据呼吸暂停低通气指数(AHI)将患者分为正常组(<5次/h)、轻度OSAHS组(5～15次/h)、中度OSAHS组(15～30次/h)、重度OSAHS组(≥30次/h)。计算每个量表的敏感度、特异度、阳性预测值、阴性预测值，绘制受试者工作特征(ROC)曲线，分析5种量表评分预测OSAHS的准确性。 结果： 以AHI≥5次/h为诊断OSAHS的标准，NoSAS评分的ROC曲线下面积最大(0.753，P<0.001)，其次为SBQ量表的ROC曲线下面积为(0.727，P<0.001)，对OSAHS均有中度预测价值，而ESS评分、Berlin及STOP的预测价值均不高(ROC曲线下面积分别为：0.543、0.645和0.684)。当NoSAS≥8分时判定OSAHS的特异度和阳性性预测值最高(轻度：80.2%和88.0%，中度：72.0%和69.8%，重度：66.3%和50.5%)，而敏感度和阴性预测值分别为51.5%和36.9%，56.5%和59.1%、66.3%和74.2%；SBQ≥3分时敏感度和阴性预测值最高(81.7%和46.9%、86.5%和71.7%、90.7%和86.7%)，特异度和阳性预测值分别为45.7%和81.0%、39.1%和61.9%、34.8%和44.4%；NoSAS≥7分时敏感度和阴性预测值高于8分时，分别为75.0%和47.1%、78.1%和66.5%、82.7%和81.9%。以AHI≥5次/h为诊断OSAHS的标准，NoSAS评分和SBQ时在筛查问卷中诊断OSAHS的准确性高，DOR值分别为4.298和3.758，而Berlin、STOP和ESS评分的DOR值分别为3.319、2.839和1.324。 结论： NoSAS评分和SBQ量表对OSAHS均有中度的预测价值。NoSAS评分是一项新的筛查工具，同SBQ问卷类似，简单，有效。而SBQ问卷应用更广，故需要进一步评估NoSAS评分的诊断价值。.
Keywords: Berlin questionnaire; Epworth sleepiness scale; SBQ questionnaire; STOP questionnaire; Screening; Sleep apnea, obstructive; The NoSAS scores.