不同神经病变评分系统在无症状糖尿病周围神经病变筛查中的临床价值比较  

A comparison of clinical effectiveness of different neuropathy scoring systems in screeningasymptomatic diabetic peripheral neuropathy

作　　者：胡泓 李红[1] 郑芬萍[1] 程怡[1] 缪菁[1] 张炜[1] 

机构地区：[1]浙江大学医学院附属邵逸夫医院内分泌科浙江大学邵逸夫临床医学研究所,杭州310016 [2]杭州市下沙医院肿瘤内科,310018

出　　处：《中华内科杂志》2012年第1期13-17,共5页Chinese Journal of Internal Medicine

摘　　要：目的比较密歇根神经病变筛查评分系统（MNSI）和多伦多临床评分系统（TCSS）在无症状糖尿病周围神经病变（ADPN）筛查中的临床应用价值。方法对2008年4月至2009年6月在浙江大学医学院附属邵逸夫医院住院的232例无糖尿病周围神经病变症状的2型糖尿病患者分别行MNSI、TCSS及神经电生理检查（NET）筛查，以NET为诊断ADPN的金标准，用受试者工作特性曲线（ROC）评价两种评分系统的诊断效能及分析它们不同切点诊断ADPN的敏感度、特异度、准确度、阳性预测值、阴性预测值、约登指数及K值；同时用logistic回归分析探讨两种评分系统与糖尿病周围神经病变（DPN）危险因素的相关性。结果MNSI及TCSS诊断ADPN的ROC曲线下面积分别为0．792、0．704，MNSI及TCSS诊断ADPN的最佳切点均为〉2，此时它们的敏感度、特异度、准确度、约登指数、K值分别为66．2％、90．4％、78．3％、0．566、0．588及73．3％、63．7％、68．5％、0．370、0．345。MNSI对ADPN的诊断效率及与NET的一致性较好且优于TCSS。Logistic回归分析得出DPN的大部分相关危险因素[年龄、糖化血红蛋白（HbAlC）、HbAlC×病程、胰岛功能、HDL—C]同样与MNSI评分相关。结论MNSI是一项相对快速、可靠性较好的筛查方法，因此可用于对ADPN的临床及流行病学的筛查和评估。Objective To evaluate the clinical effectiveness in screening asymptomatic diabetic peripheral neuropathy（ADPN） by the Michigan neuropathy screening instrument （MNSI） and the Toronto clinical scoring system（TCSS）. Methods MNSI,TCSS and neural electrophysiological test （NET） were conducted in 232 neurologically asymptomatic type 2 diabetes patients. By using the results of NET as the golden criteria for diagnosis of ADPN, we evaluated the effectiveness of the two different scoring system by the receiver operator characteristic curve. The sensitivity, specificity, positive and negative predictive values, accuracy, Youden indexes and kappa values on different diagnostic cut-off points of MNSI and TCSS were analyzed. The correlation between the two different scoring system and the risk factors of diabetic peripheral neuropathy （DPN） were also analyzed. Results The area under the ROC curve of MNSI and TCSS were 0. 792, 0. 704, respectively. The sensitivity, specificity, accuracy, Youden indexes and kappa values of MNSIover2 and TCSS over2 were 66. 2%vs 73.3%, 90. 4% vs 63.7%, 78.3% vs68.5%, 0. 566 vs 0. 370, and 0. 588 vs 0. 345, respectively. MNSI was better than TCSS in the effectiveness of diagnosing ADPN and consistence with the result of NET. Moreover, MNSI was associated with the most related risk factors of DPN including age, glycosylated hemoglobin （ HbA1 c ） , HbA1 c x disease duration, islet function and HDLC. Conclusions MNSI could be used as a relatively simple and reliable method for clinical and epidemiological screening and assessment of ADPN.

关 键 词：糖尿病神经病变 ROC曲线 糖尿病神经病变评分 

分 类 号：R587.2[医药卫生—内分泌]