老年2型糖尿病患者急性缺血性脑卒中预测模型的建立与验证  

Establishment and validation of a nomogram model for prediction of acute ischemic stroke in elderly patients with type 2 diabetes mellitus

作　　者：武慧敏 何继瑞[1] 胡彩虹[1] 胡晓霞[1] 李雅欣 侯丹丹 冯亚莉 Wu Huimin;He Jirui;Hu Caihong;Hu Xiaoxia;Li Yaxin;Hou Dandan;Feng Yali(Department of Geriatrics,Second Hospital of Lanzhou University,Lanzhou 730000,China;Department of Emergency,Second Hospital of Lanzhou University,Lanzhou 730000,China)

机构地区：[1]兰州大学第二医院老年病科,兰州730000 [2]兰州大学第二医院急诊科,兰州730000

出　　处：《中华糖尿病杂志》2022年第11期1279-1286,共8页CHINESE JOURNAL OF DIABETES MELLITUS

摘　　要：目的建立并验证老年2型糖尿病(T2DM)患者急性缺血性脑卒中列线图预测模型。方法回顾性分析2018年1月至2021年8月在兰州大学第二医院老年病科及神经内科住院的T2DM患者(445例)。按3∶1比例将患者分为建模组和验证组。其中,选取2018年1月至2020年8月就诊的334例患者为建模组,采用最小绝对收缩和选择算法(Lasso)回归筛选老年T2DM患者急性缺血性脑卒中最佳预测因子,应用R软件构建列线图模型。采用Bootstrap法进行内部验证,绘制受试者工作特征(ROC)曲线,计算一致性指数(C-index)评估模型预测性能。选取2020年9月至2021年8月的老年T2DM患者共111例为验证组,对列线图模型进行外部验证。绘制校准曲线,计算C-index、Brier分值、校准截距、校准斜率综合评估模型区分度与校准度。结果在建模组中,Lasso结合logistic回归分析结果显示,糖尿病合并高血压病程(OR=1.05,95%CI 1.01~1.10)、收缩压(OR=2.40,95%CI 1.74~3.40)、单核细胞/高密度脂蛋白胆固醇比值(OR=1.34,95%CI 1.14~1.58)、游离三碘甲状腺原氨酸(OR=0.36,95%CI 0.20~0.64)、颈内动脉斑块面积(OR=1.45,95%CI 1.19~1.79)、颅内动脉狭窄程度(OR=4.77,95%CI 2.60~9.81)是老年T2DM并发急性缺血性脑卒中最佳预测因子。ROC曲线结果显示,建模组的C-index为0.862(95%CI 0.824~0.900),经Bootstrap法进行内部验,得出的C-index、特异度、灵敏度分别为0.852(95%CI 0.812~0.895)、76.1%、80.1%。验证组的C-index、特异度、灵敏度分别为0.867(95%CI 0.803~0.932)、75.4%、82.0%。校准曲线结果显示,Brier分值为0.149,校准截距为-0.149,校准斜率为1.096。结论该研究构建的列线图预测模型,具有良好的区分度与校准度,对于临床老年T2DM急性缺血性脑卒中的早期预测具有较好的价值。Objective To establish and validate a nomogram prediction model for acute ischemic stroke in patients aged 60 years and over with type 2 diabetes mellitus(T2DM).Methods A retrospective analysis was performed on 445 T2DM patients hospitalized in the Department of Geriatrics and Neurology,Lanzhou University Second Hospital from January 2018 to August 2021.Patients were divided into modeling group and validation group according to 3∶1 ratio.A total of 334 patients treated from January 2018 to August 2020 were selected as the modeling group.We screened for the best predictors of acute ischemic stroke in elderly T2DM patients by Lasso regression.The R software was used to construct the nomogram model.The Bootstrap method was used to internal verification,and to draw receiver operating characteristic(ROC)curve.The C-index was calculated to evaluate the predictive performance of the nomogram model.A total of 111 elderly T2DM patients from September 2020 to August 2021 were enrolled as the validation group,external verification of the nomogram model.The calibration curves were drawn and C-index,brier score,calibration intercept and calibration slope were calculated.Results In the modeling group,Lasso combined with logistic regression analysis showed that diabetes mellitus with hypertension course(OR=1.05,95%CI 1.01-1.10),systolic blood pressure(OR=2.40,95%CI 1.74-3.40),high monocyte to high-density lipoprotein-cholesterol ratio(OR=1.34,95%CI 1.14-1.58),free triiodothyronine(OR=0.36,95%CI 0.20-0.64),carotid artery plaque area(OR=1.45,95%CI 1.19-1.79),degree of intracranial arterial stenosis(OR=4.77,95%CI 2.60-9.81)were the best predictors of acute ischemic stroke in elderly T2DM patients.ROC curve results showed that the C-index of the modeling group was 0.862(95%CI 0.824-0.900),internal verification was carried out by Bootstrap method,the C-index corrected to 0.852(95%CI 0.812-0.895),the sensitivity was 80.1%,and the specificity was 76.1%.The C-index of external validation was 0.867(95%CI 0.803-0.932),the sensitivity w

关 键 词：糖尿病 2型 急性缺血性脑卒中 列线图 

分 类 号：R587.2[医药卫生—内分泌] R743.3[医药卫生—内科学]