脑小血管病患者脑白质病变评分及体积与脑微出血的相关性  

Correlation between white matter lesion score and volume and cerebral microbleeds in patients with cerebral small vessel disease

作　　者：陈歆[1] 顾雨铖 邹君惠 姜永程 刘任远[1] 程越 徐运[1] Chen Xin;Gu Yucheng;Zou Junhui;Jiang Yongcheng;Liu Renyuan;Cheng Yue;Xu Yun(Department of Neurology,the Affiliated Drum Tower Hospital of Nanjing University,Nanjing 210008,China)

机构地区：[1]南京大学附属鼓楼医院神经内科,210008

出　　处：《国际脑血管病杂志》2018年第11期819-825,共7页International Journal of Cerebrovascular Diseases

基　　金：科技部“十三五”慢性病防控重点专项(2016YFC300504);江苏省科技厅医学重点项目(BE2016610);江苏省医学重点学科(ZDXKA2016020).

摘　　要：目的 探讨脑小血管病(cerebral small vessel disease,CSVD)患者脑白质病变(white matter lesions,WMLs)评分及体积与脑微出血(cerebral microbleeds,CMBs)的相关性.方法 纳入2017年1月至2017年12月南京大学附属鼓楼医院CSVD随访队列研究中的非急性腔隙性梗死患者.采集相关临床资料并进行3.0T头颅MRI检查,包括T1加权成像、T2加权成像、液体衰减反转恢复序列、磁敏感加权成像及弥散加权成像.采用W2MHS软件定量WMLs体积,使用Fazekas法分别对脑室周围及深部WMLs进行评分,视觉计数CMBs数量.应用多变量logistic回归分析确定CMBs的独立影响因素,应用多元线性回归方程(逐步法)分析CMBs数量的独立影响因素,运用Medcalc 18.6描绘受试者工作特征(receiver operating characteristic,ROC)曲线,评价WML Fazekas评分和体积对CMBs的预测价值.结果 共纳入82例患者,CMBs组31例(37.8%),非CMBs组51例(62.2%).两组比较显示,吸烟、使用抗血小板药、既往短暂性脑缺血发作(transient ischemic attack,TIA)或腔隙性梗死史、WMLs体积较大、三酰甘油水平较高及高密度脂蛋白胆固醇水平较低可能是CMBs的危险因素.多变量logistic回归分析显示,Fazekas评分较高(优势比1.908,95%可信区间1.210 ~ 3.009;P=0.005)和WMLs体积较大(优势比4.620,95%可信区间1.279 ~16.683;P=0.019)是CMBs的独立危险因素.多元线性回归分析显示,Fazekas评分(r=0.379,P=0.001)和WML体积(r=0.260,P=0.023)与CMBs数量呈独立正相关.ROC曲线分析显示,Fazekas评分预测CMBs的曲线下面积为0.768(95%可信区间0.654~0.881),最佳截断值为3分,敏感性为61.29%,特异性为90.20%;WMLs体积预测CMBs的曲线下面积为0.783(95%可信区间0.677~0.867),最佳截断值为WMLs体积达到2 137.96mm3,敏感性为73.33%,特异性为84.00%.结论 在CSVD患者中,WMLs与CMBs存在显著相关性,WMLs Fazekas评分和体积定量有望作为紧急评估CMBs的替代指标.Objective To investigate the correlation between white matter lesions (WMLs) score and volume and cerebral microbleeds (CMBs) in patients with cerebral small vessel disease (CSVD).Methods Patients with non-acute lacunar infarction from the CSVD Follow-up Cohort Study in the Affiliated Drum Tower Hospital of Nanjing University from January 2017 to December 2017 were enrolled.The relevant clinical data were collected and 3.0 T cranial MRI examinations were performed,including T1-weighted imaging,T2-weighted imaging,fluid attenuated inversion recovery (FLAIR) sequence,susceptibility-weighted imaging,and diffusion-weighted imaging.W2MHS software was used to quantify the volume of WMLs.Fazekas method was used to score periventricular and deep WMLs separately.The number of CMBs was counted visually.Multivariate logistic regression analysis was used to determine the independent influencing factors of CMBs.Multivariate linear regression equation (stepwise method) was used to analyze the independent influencing factors of the number of CMBs.Medcalc 18.6 was used to deseribe receiver operating characteristic (ROC) curve.The predictive value of WMLs Fazekas scores and volumes for CMBs was evaluated.Results A total of 82 patients were enrolled,including 31 (37.8%) in CMBs group and 51 (62.2%) in non-CMBs group.The comparison between the two groups showed that smoking,use of antiplatelet agents,history of transient ischemic attack (TIA) or lacunar infarction,larger WMLs volume,higher levels of triacylglycerol,and lower levels of high-density lipoprotein cholesterol might be the risk factors for CMBs.Multivariate logistic regression analysis showed that higher Fazekas score (odds ratio 1.908,95% confidence interval 1.210-3.009;P=0.005) and larger WMLs volume (odds ratio 4.620,95% confidence interval 1.279-16.683;P =0.019) were the independent risk factors for CMBs.Multiple linear regression analysis showed that Fazekas score (r =0.379,P =0.001) and WMLs volume (r =0.260,P =0.023) were independently and positively correlated wi

关 键 词：脑小血管疾病 脑出血 脑白质疏松 磁共振成像 危险因素 

分 类 号：R743.3[医药卫生—神经病学与精神病学]