中年女性脑卒中的危险因素分析及诊断模型的建立与验证

Analysis of risk factors of stroke in middle-aged women and establishment and verification of diagnostic model

目的通过挖掘数据库建模和独立数据验证的方法探索中年女性脑卒中发生的危险因素并建立诊断模型。方法提取美国妇女健康研究(SWAN)数据库中37例42~52岁女性脑卒中患者作为卒中组,提取数据库中1818例无卒中病史的女性作为对照组,收集2组的临床资料并采用logistic回归筛选卒中相关的危险因素并建立诊断模型。同时,回顾性分析溧阳市人民医院42~48岁女性脑卒中患者74例及87例非卒中对照女性的临床资料,验证诊断模型的有效性。结果卒中组患者的体质指数(BMI)、纤溶酶原激活物抑制物-1(PAI-1)、组织型纤溶酶原激活剂(t-PA)、载脂蛋白B(ApoB)及C-反应蛋白(CRP)水平高于对照组(P<0.05),雌二醇水平低于对照组(P<0.05)。同时,多因素分析结果提示BMI、ApoB和CRP是脑卒中的独立危险因素(P<0.05)。基于以上指标构建卒中诊断模型并定义为BELC,其ROC曲线下面积(AUC)为0.734。验证数据集中,上述指标在卒中组和对照组中的差异均有统计学意义(P<0.05),诊断模型对应的AUC为0.710。结论基于BMI、雌二醇、ApoB和CRP的多参数模型BELC对中年女性脑卒中具有较高的诊断价值,可用于协助临床诊疗。

Objective To explore the risk factors of stroke in middle-aged women and establish a diagnostic model by public database analysis and independent data verification.Methods 37 female stroke patients aged 42-52 years in the Study of Women′s Health Across the Nation(SWAN)database were extracted as the stroke group,and 1818 women without stroke history in the database were extracted as the control group.The clinical data of the two groups were collected,the risk factors of stroke were screened by logistic regression method,and the diagnostic model was established.In addition,the clinical data of 74 female stroke patients aged 42-48 years and 87 non-stroke control women in Jiangsu Liyang People′s Hospital were collected retrospectively to verify the effectiveness of the diagnostic model.Results The body mass index(BMI),plasminogen activator inhibitor-1(PAI-1),tissue plasminogen activator(t-PA),apolipoprotein B(ApoB)and C-reactive protein(CRP)in the stroke group were higher than those in the control group(P<0.05),while the estradiol level was lower than that in the control group(P<0.05).Meanwhile,multivariate analysis showed that BMI,ApoB and CRP were the independent risk factors for stroke(P<0.05).Based on the above indicators,the stroke diagnosis model was defined as BELC and the area under the ROC curve(AUC)was 0.734.In the validation data set,the differences of the above variables between the stroke group and the control group were statistically significant(P<0.05),and the AUC of the BELC model was 0.710.Conclusion The multi-parameter model BELC based on BMI,estradiol,ApoB and CRP has high diagnostic value for stroke in middle-aged women and can be used to assist clinical diagnosis and treatment.