不同量表在筛查重症肌无力患者睡眠呼吸障碍中的效果评价

Evaluation of the effect of different scales on screening sleep-disordered breathing in patients with myasthenia gravis

目的本文旨在对比不同量表在筛查临床稳定期重症肌无力(myasthenia gravis,MG)患者睡眠呼吸障碍(sleep-disordered breathing,SDB)的效能,以期找出最优的筛查方案。方法本研究共顺序纳入2017年10月至2019年1月就诊于唐都医院神经内科的30例临床稳定期MG患者。使用爱泼沃斯嗜睡量表(Epworth sleepiness scale,ESS)评估嗜睡程度,Stop-Bang量表、柏林量表(Berlin questionnaire,BQ)、sleep-disordered breathing in neuromuscular disease questionnaire-5(SiNQ-5)、NoSAS量表评估睡眠呼吸障碍患病风险。采用单独诊断及联合诊断的方式,分析5种量表筛查MG患者SDB的效能。采用受试者工作特征曲线(receiver operating characteristic,ROC)分析不同量表预测MG患者SDB的区分度。采用Spearman相关性分析比较各量表筛查结果间的相关性。以双侧检验P<0.05为差异有统计学意义。结果REI≥10时,调整NoSAS量表(NoSAS-revised)的曲线下面积(areaunder the curve,AUC)最大,Stop-Bang量表和NoSAS量表AUC略低,三者差异均有统计学意义。NoSAS-revised的灵敏度和阴性预测值最高,Stop-Bang量表和NoSAS量表的特异度和阳性预测值最高。BQ和SiNQ-5的灵敏度最低,SiNQ-5的特异度、阳性预测值、阴性预测值最低。REI≥5时,BQ诊断价值最优,SiNQ-5最差。REI≥10时,NoSAS诊断价值最优,ESS最差。REI≥15及更高界值时,Stop-Bang量表诊断价值最优,ESS最差。Stop-Bang+NoSAS的Se最高(平行试验71.0%,系列试验21.3%),Sp最高(平行试验88.6%,系列试验99.7%)。SiNQ-5与其他量表联合诊断的诊断价值均较低。结论NoSAS+SiNQ-5的联合诊断策略,既可以筛查患者是否存在膈肌无力或OSA,又兼顾了联合灵敏度与联合特异度,是较为可行的联合诊断策略。

Objective This study aims to compare the effectiveness of different scales in screening clinically stable myasthenia gravis(MG)patients with sleep-disordered breathing(SDB),in order to find the optimal screening program.Methods A total of 30 clinically stable MG patients who were admitted to the Neurology Department of Tangdu Hospital from October 2017 to January 2019 were included in this study.The risk of sleep-disordered breathing in MG patients was assessed using the Epworth sleepiness scale(ESS),Stop-Bang scale,Berlin questionnaire(BQ),sleep-disordered breathing in neuromuscular disease questionnaire-5(SiNQ-5),and revised neck circumference,obesity,snoring,age,sex(NoSAS-revised)score.The effectiveness of the five scales in screening SDB in MG patients was analyzed by plotting the receiver operating characteristic(ROC)curve of a single assessment or combination assessment.Spearman correlation analysis was used to compare the correlation among the screening results of each scale.A two-tailed,P<0.05 was considered statistically significant.Results When REI≥10,the area under the curve(AUC)of the NoSAS-revised was the largest,and the AUC of the Stop-Bang scale and the NoSAS scale were slightly lower,and the differences among the three were statistically significant.NoSAS-revised had the highest sensitivity and negative predictive value,and the Stop-Bang scale and NoSAS scale had the highest specificity and positive predictive value.BQ and SiNQ-5 had the lowest sensitivity,and SiNQ-5 had the lowest specificity,positive predictive value,and negative predictive value.When REI≥5,the diagnostic value of BQ was the best,and that of SiNQ-5 was the worst.When REI≥10,NoSAS had the best diagnostic value,and ESS was the worst.When REI≥15 and higher threshold value,the diagnostic value of Stop-Bang scale was the best,and the ESS was the worst.Stop-Bang+NoSAS had the highest sensitivity(71.0%in parallel tests,21.3%in series tests)and the highest specificity(88.6%in parallel tests,99.7%in series tests).The diagnostic value of SiNQ-5 combined with other scales was low.Conclusion The combined assessment of NoSAS+SiNQ-5 is able to screen patients with diaphragmatic weakness or OSA,which is highly specific and sensitive.It is a feasible combined diagnostic strategy.