无症状糖尿病周围神经病变患者足底压力改变与神经传导速度的关系  

Relationship between Changes in Plantar Pressure and Nerve Conduction Velocity in Asymptomatic Diabetic Peripheral Neuropathy

作　　者：马君 吴燕林 王跃臻 杨燕飞 马翠柳 段建颖 MA Jun;WU Yan-lin;WANG Yue-zhen;YANG Yan-fei;MA Cui-liu;DUAN Jian-ying(Department of Endocrinology,Zhangjiakou First Hospital,Zhangjiakou 075000,China)

机构地区：[1]河北省张家口市第一医院内分泌一科,张家口075000

出　　处：《微循环学杂志》2023年第4期35-44,50,共11页Chinese Journal of Microcirculation

基　　金：河北省医学科学研究课题计划(20211120)。

摘　　要：目的:探讨无症状糖尿病周围神经病变(ADPN)患者足底压力改变与神经传导速度(NCV)的关系。方法:选取2019-09-2021-06本院门诊就诊并建档的1630例2型糖尿病(T2DM)患者,根据临床感觉检查(踝/膝反射、针刺痛觉检查、尼龙丝检查、音叉检查、温度觉检查)将患者分为ADPN组(815例)、糖尿病周围神经病变组(DPN组,480例)和无糖尿病周围神经病变组(NDPN组,335例),比较三组患者NCV和临床感觉异常率。将ADPN患者随机分为建模组(611例)和验证组(204例),通过NCV检测将建模组患者分为NCV异常组(490例)和NCV正常组(121例),比较两组患者的一般资料和代谢指标。多因素Logistic回归分析影响建模组患者NCV异常的危险因素;分析足底压力参数与NCV之间的关系;Spearman相关性分析足底压力(压力-时间积分、应力-时间积分)与代谢指标的相关性;R软件构建贝叶斯网络模型,Netica软件进行贝叶斯网络模型推理预测;通过建模组数据集作ROC曲线和校准曲线评价模型的区分度和准确度,并用验证组数据集进行检验。结果:DPN组患者NCV均明显低于ADPN组和NDPN组患者,而临床感觉异常率显著高于其余两组(P<0.05);多因素Logistic回归分析结果显示,高龄、病程时间长、低水平高密度脂蛋白胆固醇(HDL-C)、高水平糖化血红蛋白(HbAlc)和高水平尿白蛋白肌酐比(UACR)是建模组患者NCV异常的独立危险因素(P<0.05);NCV异常组足底接触时间、最大应力、压力-时间积分、应力-时间积分和足弓、第5跖骨、第3-5趾区域的峰值压力均明显高于NCV正常组患者,而足跟、第2跖骨、第3跖骨区域的峰值压力明显低于NCV正常组(P<0.05);Spearman相关性分析结果显示,足底压力与代谢指标之间存在显著相关性;建模组和验证组ROC曲线和校准曲线显示贝叶斯网络模型具有良好的区分度和准确度。结论:高龄、病程时间长、低水平HDL-C、高水平HbAlc和高水平UACObjective:To investigate the relationship between plantar pressure and nerve conduction velocity(NCV)in asymptomatic diabetic peripheral neuropathy(ADPN).Method:A total of 1630 patients with type 2 diabetes who were treated and documented in the outpatient department of our hospital from September 2019 to June 2021 were selected as the research objects.Patients were divided into ADPN group(815 cases),DPN group(480 cases)and NDPN group(335 cases)according to clinical sensory examination(ankle/knee reflex,acupuncture pain examination,Nylon wire examination,tuning fork examination and temperature perception examination).ADPN patients were randomly divided into training group(611 cases)and validation group(204 cases).Patients in the training group were divided into the abnormal NCV group(490 cases)and the normal NCV group(121 cases)by NCV test,and the general data and biochemical parameters of the two groups were compared.Multivariate logistic regression analysis was used to identify the risk factors for abnormal NCV in training group.The relationship between plantar pressure parameters and NCV was compared.Spearman correlation analysis was conducted to analyze the correlation between plantar pressure(pressure-time integral,stress-time integral)and metabolic indexes.R software was used to build Bayesian network model,and Netica software was used to perform bayesian network model inference and prediction.The differentiation and accuracy of the model were evaluated by ROC curve and calibration curve of the training group data set,and verified by the validation group data set.Results:NCV in DPN group was significantly lower than that in ADPN group and NDPN group,and clinical paresthesia rate was significantly higher than those in other two groups(P<0.05).Multivariate logistic regression analysis showed that high age,long course of disease,low HDL-C level,high HbAlc level and high UACR level were independent risk factors for abnormal NCV in training group(P<0.05).The contact time of plantar,maximum stress,pressure-time i

关 键 词：无症状糖尿病周围神经病变 足底压力 神经传导速度 贝叶斯网络模型 

分 类 号：R587.1[医药卫生—内分泌]