脑小血管病脑白质高信号危险因素相关性分析

Correlation analysis of risk factors for white matter hyperintensities in cerebral small vessel disease

目的:评估脑小血管病(CSVD)中脑白质高信号(WMHs)的危险因素,并进一步建立早期预测模型。方法:选取湖南省人民医院住院的233例患者进行回顾性分析,根据头部磁共振平扫T2序列辨别WMHs,分为WMHs的观察组187例,无WMHs的对照组46例,随机选取170例作为训练集,63例作为验证集。训练集中两组患者的一般资料、相关生化指标进行组间比较,对差异因素进行多因素Logistic回归分析,对独立危险因素进行单项及多项联合绘制受试者工作特征(ROC)曲线,同时构建列线图,建立预测模型,并评价模型预测效能。结果:训练集两组患者间年龄、高血压、冠心病、单核细胞计数、血尿素氮(BUN)、血清肌酐(SCr)、肾小球滤过率(eGFR)具有统计学差异,这7项指标进行多因素回归分析显示高血压、SCr水平是WMHs的独立危险因素。绘制高血压、SCr及2项指标联合预测WMHs的ROC曲线,曲线下面积分别为0.655、0.680、0.762。通过构建高血压、SCr的列线图,绘制预测模型并与单独的高血压、SCr诊断验证集发生WMHs的ROC曲线进行效能评价,其曲线面积分别为0.763、0.708、0.649。结论:高血压、SCr是WMHs的独立危险因素,且对WMHs具有一定的预测价值,联合高血压与SCr预测WMHs风险要优于单一指标,对WMHs的早期诊断具有一定的指导价值。

Objective To evaluate the risk factors of white matter hyperintensities(WMHs)in Cerebral small vessel disease(CSVD),and to further establish an early prediction model.Method A total of 233 patients hospitalized in Hunan Provincial People's Hospital were selected for retrospective analysis.According to the T2 sequence of head magnetic resonance plain scan to identify WMHs,they were divided into the observation group of 187 cases with WMHs and the control group of 46 cases without WMHs.The general data and related biochemical indicators of the two groups of patients in the training set were compared between the two groups,and the difference factors were analyzed by multivariate Logistic regression analysis.The receiver operating characteristic(ROC)curve was drawn for single and multiple independent risk factors,and a nomogram was constructed to establish a prediction model,and the predictive efficacy of the model was evaluated.Results There were significant differences in age,hypertension,coronary heart disease,monocyte count,blood urea nitrogen(BUN),serum creatinine(SCr),and glomerular filtration rate(eGFR)between the two groups of patients in the training set(P<0.05).Multivariate regression analysis of 7 indicators showed that hypertension and SCr level were independent risk factors for WMHs.The area under the ROC curve of hypertension,SCr and the two indexes combined to predict WMHs was 0.655,0.680 and 0.762,respectively.By constructing a nomogram for hypertension and SCr,the efficacy of the prediction model and the ROC curve of WMHs in the hypertension and SCr diagnostic validation set alone was evaluated,and the curve area was 0.763,0.708,and 0.649,respectively.Conclusions Hypertension and SCr are independent risk factors for WMHs,and have predictive value for WMHs.The combination of hypertension and SCr is better than single index in predicting the risk of WMHs,which has guiding value for the early diagnosis of WMHs.