静止性脑梗死风险预测量表的构建和验证  

Derivation and Validation of the Silent Brain Infarction Risk Score in the Healthy

作　　者：郑华光 张蔚韡[3] 张龙友[2] 王瑞青[2] 孟庆颖[2] 刘晓楠[2] 段芸芸[4] 刘亚欧 王拥军 ZHENG Hua-Guang;ZHANG Wei-Wei;ZHANG Long-You;WANG Rui-Qing;MENG Qing-Ying;LIU Xiao-Nan;DUAN Yun-Yun;LIU Ya-Ou;WANG Yong-Jun(National Clinical Research Center for Neurological Diseases,Beijing 100070,China;Health Management Center,Beijing Tiantan Hospital,Capital Medical University,Beijing 100070,China;Beijing Physical Examination Center,Beijing 100070,China;Department of Radiology,Beijing Tiantan Hospital,Capital Medical University,Beijing 100070,China;Department of Neurology,Beijing Tiantan Hospital,Capital Medical University,Beijing 100070,China)

机构地区：[1]国家神经系统疾病临床研究中心,北京100070 [2]首都医科大学附属北京天坛医院健康管理中心 [3]北京市体检中心 [4]首都医科大学附属北京天坛医院影像科 [5]首都医科大学附属北京天坛医院神经病学中心

出　　处：《中国卒中杂志》2020年第9期999-1005,共7页Chinese Journal of Stroke

基　　金：国家重点研发计划(2018YFC1311200,2018YFC1311203);首都卫生发展科研专项(2020-2-2244)。

摘　　要：目的调查与静止性脑梗死(silent brain infarction,SBI)相关的独立影响因素,构建SBI风险预测量表并验证。方法在单中心横断面研究中,前瞻性连续纳入无神经系统疾病既往史的体检者,收集其人口学信息,高血压、糖尿病等血管危险因素,血脂、糖化血红蛋白、血浆同型半胱氨酸等化验结果录入数据库。采用标准影像学操作规范进行头颅MRI扫描,并由影像学医师盲法判读,将受试者分为SBI组和无SBI组。将所有受试者按照3∶1比例随机分为训练集和验证集,在训练集中采用单因素和多因素Logistic回归分析SBI的独立影响因素,构建SBI预测量表。在训练集和验证集中应用ROC曲线检验量表的区分度,应用Hosmer-Lemeshow分析检验量表的校准度。结果共有633例研究对象纳入研究,平均年龄52.0±10.5岁,女性272例(43.0%)。训练集(475例)和验证集(158例)两个样本集合的基线特征均衡。校正混杂因素后多因素分析显示,年龄≥45岁(OR 8.37,95%CI 1.12~62.80,P=0.039),高血压(OR 2.30,95%CI 1.08~4.90,P=0.032),同型半胱氨酸(Q2~Q3:OR 6.89,95%CI 0.89~53.10,P=0.064;Q4:OR 13.6,95%CI 1.74~105.87,P=0.013)与SBI风险独立相关。根据OR值构建SBI危险评分(SBI risk score,SBI-RS)量表,量表赋值为:年龄≥45岁赋值8分;有高血压赋值2分;同型半胱氨酸根据四分位分层分别赋值为0分、7分和14分。SBIRS在训练集和验证集中ROC曲线显示曲线下面积分别为0.77(95%CI 0.69~0.84,P<0.001)和0.76(95%CI 0.63~0.88,P<0.001),区分度良好。Hosmer-Lemeshow相关分析提示SBI-RS具有较好的校准度(P>0.05)。结论在健康体检人群中,SBI-RS具有较好的区分度和校准度,可以帮助识别SBI高危人群。Objective To investigate the independent risk factors of silent brain infarction(SBI)in the healthy people,and to develop and validate a predictive scale for SBI risk.Methods In a cross-section study from National Clinical Research Center for Neurological Diseases,the consecutive eligible participants from the Health Management Center of Beijing Tiantan Hospital were enrolled.The clinical data and laboratory tests results were collected.MRI examination was performed according to a standard protocol and evaluated by radiologists using a blind method.All the subjects were divided to SBI group and non-SBI group based on MRI results.All the subjects were randomly divided into a derivation set and a validation set at a ratio of 3:1.Univariable and multivariable logistic analysis were conducted in the derivation set to analyze the independent risk factors of SBI,and a SBI risk score(SBI-RS)was built according to the adjusted OR.ROC curve was used to evaluate the discrimination of the SBI-RS,and the calibration of the scale was evaluated using the Hosmer-Lemeshow test.Results A total of 633 eligible participants were enrolled,with a mean age of 52.0±10.5 years old and 272(43.0%)females.The clinical features were balanced between the derivation set(n=475)and validation set(n=158).After adjusting for the confounding factors,multivariable logistic analysis showed that age over 45 years old(OR 8.37,95%CI 1.12-62.80,P=0.039),hypertension(OR 2.30,95%CI 1.08-4.90,P=0.032),and homocysteine(Q2-Q3:OR 6.89,95%CI 0.89-53.10,P=0.064;Q4:OR 13.6,95%CI 1.74-105.87,P=0.013)were independently associated with SBI.The SBI-RS was derivated according to the adjusted OR,every variable was assigned a value as follows:8 points for≥45 years old,2 points for hypertension and 0,7 or 14 points for the quartiles of the homocysteine level.The discrimination of the SBI-RS was reasonable,the area under the curve(AUC)was 0.77(95%CI 0.69-0.84,P<0.001)in the derivation set and 0.76(95%CI 0.63-0.88,P<0.001)in the validation set.The Hosmer-Lemeshow corre

关 键 词：静止性脑梗死 风险预测量表 区分度 校准度 

分 类 号：R743.33[医药卫生—神经病学与精神病学]