急诊老年患者衰弱预测模型的建立与验证  

Development and validation of a prediction model to estimate the probability of frailty in older emergency patients

作　　者：李俊玉 王国栋[2] 商娜 王娜[1] 郭树彬[3] 刘慧珍[1] Li Junyu;Wang Guodong;Shang Na;Wang Na;Guo Shubin;Liu Huizhen(Capital Medical University School of Rehabilitation Medicine,Department of Emergency Medicine,Beijing Bo'ai Hospital,China Rehabilitation Research Center,Beijing 100068,China;Capital Medical University School of Rehabilitation Medicine,Cardiovascular Department,Beijing Bo'ai Hospital,China Rehabilitation Research Center,Beijing 100068,China;Emergency Medicine Clinical Research Center,Beijing Chao-Yang Hospital,Capital Medical University,Beijing Key Laboratory of Cardiopulmonary Cerebral Resuscitation,Beijing 100020,China)

机构地区：[1]首都医科大学康复医学院,中国康复研究中心北京博爱医院急诊科,北京100068 [2]首都医科大学康复医学院,中国康复研究中心北京博爱医院心血管内科,北京100068 [3]首都医科大学附属北京朝阳医院急诊医学临床研究中心,心肺脑复苏北京市重点实验室,北京100020

出　　处：《中华急诊医学杂志》2025年第2期226-232,共7页Chinese Journal of Emergency Medicine

摘　　要：目的利用客观临床资料和生物标志物建立急诊老年患者发生衰弱的预测模型,并对模型的性能进行验证。方法本研究为横断面研究。连续收集2021年1月至2021年12月中国康复研究中心急诊科收治的老年患者,记录临床资料,采用Fried衰弱表型对患者进行衰弱评估,将患者分为衰弱组和非衰弱组,比较两组临床资料的差异。采用LASSO回归对单因素分析中有统计学意义的变量进行降维处理,将筛选出的变量进行多因素Logistic回归分析以建立预测模型,并绘制列线图将预测模型可视化。采用受试者工作特征曲线的曲线下面积、校准曲线和决策曲线评价模型的区分度、校准度和临床实用性,采用Bootstrap法进行内部验证。结果最终纳入患者348例,其中发生衰弱者187例,发生率为53.74%。受教育程度、合并冠心病、慢性阻塞性肺疾病、白蛋白、纤维蛋白原、N-末端脑钠肽前体、肌酐降低和体重过低是急诊老年患者存在衰弱的独立预测因子(P<0.05),基于上述预测因子构建急诊老年患者发生衰弱的列线图预测模型,本模型具有良好的区分度、校准度和临床实用性。结论本研究采用客观临床资料和生物标志物建立了急诊老年患者发生衰弱的预测模型,有助于对急诊老年患者进行危险分层和针对性的干预治疗,改善患者预后。Objective To develop and validate a prediction model by combining clinical data and biomarkers to evaluate the probability of frailty among older emergency patients.Methods A cross-sectional study was conducted.From January 2021 to December 2021,patients aged 60 years and older admitted to the emergency department of China Rehabilitation Research Center were enrolled.Data of patient's clinical information were collected.The patients were divided into frail group and non-frail group according to the Fried's frailty phenotype and clinical data were compared between the two groups.LASSO regression was used to deal with dimension reduction and multivariate logistic regression was employed to construct a prediction model based on variables selected by the LASSO regression.Nomogram was used to visualize the prediction model.The area under the receiver operating characteristic curve,calibration curve,decision curve analysis and bootstrap were used to evaluate the discrimination,calibration,clinical applicability,and internal validity of the model respectively.Results A total of 348 patients were enrolled,and the incidence of frailty was 53.74%(187/348).Education,coronary heart disease,chronic obstructive pulmonary disease,albumin,fibrinogen,N-terminal pro-brain natriuretic peptide,decreased creatinine,and underweight were independent predictors for frailty in older emergency patients(P<0.05).A nomogram model was built based on the above predictors and the model showed good discrimination,calibration and clinical applicability.Conclusions The study utilized objective clinical data and biomarkers to establish a predictive model for the occurrence of frailty in elderly emergency department patients.This model aids in risk stratification and targeted intervention for elderly emergency patients,thereby improving patient outcomes.

关 键 词：衰弱 预测模型 列线图 内部验证 急诊 老年 临床资料 生物标志物 

分 类 号：R47[医药卫生—护理学]