Abstract：
The STOP-Bang questionnaire is an established clinical screening tool to identify the risk of having mild, moderate or severe obstructive sleep apnoea using eight variables. It is unclear whether all eight variables contribute equally to the risk of clinically significant obstructive sleep apnoea. We analysed each variable for its contribution to detecting obstructive sleep apnoea; based on the results, we investigated whether the STOP-Bang questionnaire could be abbreviated to identify patients at high risk for severe obstructive sleep apnoea. We recruited patients with suspected obstructive sleep apnoea who were referred for overnight polysomnography. We used multivariable logistic regression to investigate the association of STOP-Bang parameters with severe obstructive sleep apnoea based on clinical and polysomnography data. Regression estimates were used to select variables to create the novel B-APNEIC score. We constructed receiver operating characteristic curves for the STOP-Bang questionnaire and B-APNEIC scores to identify patients with severe obstructive sleep apnoea and compared the areas under the curve using the DeLong method. Of the 275 patients enrolled, 32% (n = 88) had severe obstructive sleep apnoea. Logistic regression demonstrated that neck circumference (OR 2.20; 95%CI 1.10-4.40, p = 0.03) was the only variable independently associated with severe obstructive sleep apnoea. Observed apnoea during sleep, blood pressure and body mass index were the three next most closely trending predictors of severe obstructive sleep apnoea and were included along with neck circumference in the B-APNEIC score. Receiver operating curves demonstrated that the areas under the curve for STOP-Bang vs. B-APNEIC were comparable for identifying patients with severe obstructive sleep apnoea (OR 0.75; 95%CI 0.68-0.81 vs. OR 0.75; 95%CI 0.68-0.81: p = 0.99, respectively). Our results suggest that the B-APNEIC score is a simplified adaptation of the STOP-Bang questionnaire with equivalent effectiveness in identifying patients with severe obstructive sleep apnoea. Further studies are needed to validate and build on our findings.
摘要：
STOP-Bang问卷是一种既定的临床筛查工具，用于识别患有轻度，中度或重度阻塞性睡眠呼吸暂停使用八个变量。目前尚不清楚所有八个变量是否都对临床上显着的阻塞性睡眠呼吸暂停的风险有同等贡献。我们分析了每个变量对检测阻塞性睡眠呼吸暂停的贡献；基于结果，我们调查了STOP-Bang问卷是否可以缩写以识别重度阻塞性睡眠呼吸暂停高危患者.我们招募了疑似阻塞性睡眠呼吸暂停的患者，这些患者被转诊为过夜多导睡眠图。我们根据临床和多导睡眠图数据，使用多变量逻辑回归研究STOP-Bang参数与严重阻塞性睡眠呼吸暂停的相关性。使用回归估计来选择变量以创建新的B-APNEIC评分。我们构建了STOP-Bang问卷和B-APNEIC评分的受试者工作特征曲线，以识别患有严重阻塞性睡眠呼吸暂停的患者，并使用DeLong方法比较了曲线下的面积。在入选的275名患者中，32％（n=88）患有严重的阻塞性睡眠呼吸暂停。Logistic回归表明，颈围（OR2.20；95CI1.10-4.40，p=0.03）是与严重阻塞性睡眠呼吸暂停独立相关的唯一变量。在睡眠期间观察到的呼吸暂停，血压和体重指数是严重阻塞性睡眠呼吸暂停的三个最接近趋势的预测因子,并与颈围一起纳入B-APNEIC评分.接收器工作曲线表明，STOP-Bang曲线下的面积与B-APNEIC在识别严重阻塞性睡眠呼吸暂停患者方面具有可比性（OR0.75；95CI0.68-0.81与OR0.75；95CI分别为0.68-0.81：p=0.99）。我们的结果表明，B-APNEIC评分是STOP-Bang问卷的简化适应，在识别严重阻塞性睡眠呼吸暂停患者方面具有同等效力。需要进一步的研究来验证和建立我们的发现。
