暴发性心肌炎体外膜肺氧合延迟拔管的危险因素及预测模型  

Risk factors and prediction model for delayed extubation of extracorporeal membrane oxygenation in fulminant myocarditis

作　　者：高婧婧 高永霞 GAO Jingjing;GAO Yongxia(ICU,The First Affiliated Hospital of Nanjing Medical University,Jiangsu Provincial People's Hospital,Nanjing,210019,China)

机构地区：[1]南京医科大学第一附属医院(江苏省人民医院)ICU,南京210019

出　　处：《临床心血管病杂志》2025年第1期44-49,共6页Journal of Clinical Cardiology

摘　　要：目的:探讨暴发性心肌炎体外膜肺氧合延迟拔管的危险因素,并建立其列线图预测模型。方法:选取2020年1月—2023年11月江苏省人民医院体外生命支持中心及分中心收治的242例暴发性心肌炎接受体外膜肺氧合的患者作为研究对象,根据体外膜肺氧合延迟拔管的发生情况将其分为延迟拔管组(64例)和早期拔管组(178例)。收集两组患者的性别、年龄、恶性心律失常、饮酒史、吸烟史、其他支持治疗、糖尿病、高血压、白细胞计数(WBC)、血小板计数(PLT)、C反应蛋白(CRP)、肾功能不全、低蛋白血症、睡眠质量及肺部感染等资料。采用LASSO和logistic回归分析筛选延迟拔管的预测因素,采用R软件建立延迟拔管的列线图预测模型,并对模型进行验证。结果:两组性别、年龄、饮酒史、其他支持治疗、糖尿病、高血压、WBC、PLT及睡眠质量等资料均差异无统计学意义;延迟拔管组恶性心律失常、吸烟、肾功能不全、低蛋白血症及肺部感染发生率及CRP水平均显著高于早期拔管组(均P<0.05)。Logistic回归分析显示,恶性心律失常(OR=4.672,95%CI:1.834~11.901,P=0.001)、吸烟史(OR=4.561,95%CI:1.409~14.765,P=0.011)、CRP(OR=1.155,95%CI:1.098~1.215,P<0.001)、肾功能不全(OR=3.695,95%CI:1.178~11.586,P=0.025)、低蛋白血症(OR=5.018,95%CI:1.896~13.285,P=0.001)及肺部感染(OR=4.467,95%CI:1.675~11.910,P=0.003)是延迟拔管的危险因素。基于上述危险因素建立列线图模型,预测延迟拔管的ROC曲线下面积为0.846(95%CI:0.789~0.904);校正曲线的实际值与预测值基本吻合;决策曲线显示阈值概率为1%~100%时,列线图预测延迟拔管的净获益值较高。结论:恶性心律失常、吸烟史、CRP、肾功能不全、低蛋白血症及肺部感染是暴发性心肌炎患者体外膜肺氧合延迟拔管的危险因素,基于此建立的列线图模型预测延迟拔管的准确性和临床实用性良好。Objective To investigate the risk factors of delayed extracorporeal membrane oxygenation(ECMO)extubation in fulminant myocarditis,and to establish a nomogram model for predicting delayed extubation.Methods Two-hundred and forty-two patients with fulminant myocarditis who received ECMO and were admitted to the Extracorporeal Life Support Center and its sub centers of Jiangsu Provincial People's Hospital from January 2020 to November 2023 were selected as the study subjects.Patients were divided into the delayed extubation group(n=64)and the early extubation group(n=178)according to the occurrence of delayed ECMO extubation.Data of gender,age,malignant arrhythmia,drinking history,smoking history,other supportive treatment,diabetes,hypertension,white blood cell count(WBC),platelet count(PLT),C-reactive protein(CRP),renal dysfunction,hypoproteinemia,sleep quality,and pulmonary infection were collected.LASSO and logistic regression analysis were used to screen the risk factors of delayed extubation.R software was used to establish a nomogram model for predicting delayed extubation.The model was validated.Results There were no statistically significant differences in gender,age,drinking history,other supportive treatment,diabetes,hypertension,WBC,PLT,sleep quality,and lung infection between the two groups.The incidence of malignant arrhythmia,smoking history,renal dysfunction,hypoalbuminemia,pulmonary infection,and CRP levels in the delayed extubation group were significantly higher than those in the early extubation group(all P<0.05).Logistic regression analysis showed that malignant arrhythmia(OR=4.672,95%CI:1.834-11.901,P=0.001),smoking history(OR=4.561,95%CI:1.409-14.765,P=0.011),CRP(OR=1.155,95%CI:1.098-1.215,P<0.001),renal dysfunction(OR=3.695,95%CI:1.178-11.586,P=0.025),hypoalbuminemia(OR=5.018,95%CI:1.896-13.285,P=0.001),and pulmonary infection(OR=4.467,95%CI:1.675-11.910,P=0.003)were risk factors for delayed extubation.A nomogram model was established based on the above risk factors,the area under the ROC curv

关 键 词：暴发性心肌炎 体外膜肺氧合 早期拔管 危险因素 动态列线图模型 

分 类 号：R542.2[医药卫生—心血管疾病]