Table_2_Exploration of a nomogram prediction model of 30-day survival in adult ECMO patients.pdf

ObjectiveTo investigate the factors of 30-day survival in ECMO patients, establish a nomogram model, and evaluate the predictive value of the model.MethodsA total of 105 patients with extracorporeal membrane oxygenation (ECMO) were admitted to the Department of Critical Care Medicine, The First Affiliated Hospital of Anhui Medical University, from January 2018 to March 2021. Cox regression analysis screened out the risk factors. Based on the results of multivariate analysis, the nomogram model was established by using R software, and the discrimination of the model was verified by bootstrap and calibration.ResultsThe results showed that sex, acute physiology and chronic health evaluation (APACHE) II score, disseminated intravascular coagulation (DIC) score before ECMO initiation and average daily dose of norepinephrine were independent risk factors for prognosis. Verify that the nomogram model is verified by bootstrap internally, and the corrected C-index is C-index: 0.886, showing a good degree of discrimination. The calibration curve (calibration) showed that the nomogram model had good agreement. The decision curve analysis(DCA) curve shows good clinical validity above the two extreme curves. Kaplan–Meier curves were drawn for patients in the tertile and compared with the first and second groups. The third group predicted the worst 30-day prognosis for ECMO patients.ConclusionThe nomogram prediction model constructed based on the sex, APACHE II and DIC score, average daily dose of norepinephrine can effectively screen out the factors affecting the prognosis and provide a reference for individualized treatment of ECMO patients.

旨在探究体外膜肺氧合（ECMO）患者30天存活率的影响因素，构建预测图模型，并评估该模型的预测价值。方法：自2018年1月至2021年3月，共有105例ECMO患者在安徽医科大学第一附属医院重症医学科接受治疗。采用Cox回归分析筛选出风险因素。基于多因素分析结果，利用R软件构建预测图模型，并通过自举和校准验证模型的区分度。结果：研究结果表明，性别、急性生理和慢性健康状况评价（APACHE）II评分、ECMO启动前的弥散性血管内凝血（DIC）评分以及平均每日去甲肾上腺素用量是预后独立的危险因素。内部验证表明，预测图模型经自举检验得到验证，校正C指数为C-index: 0.886，显示出良好的区分度。校准曲线（校准）显示预测图模型具有良好的一致性。决策曲线分析（DCA）曲线显示，在两个极端曲线之上，模型具有良好的临床有效性。绘制了患者分层的三分位数Kaplan-Meier曲线，并与第一、第二组进行比较。第三组预测ECMO患者的最差30天预后。结论：基于性别、APACHE II和DIC评分、平均每日去甲肾上腺素用量的预测图模型能够有效筛选影响预后的因素，并为ECMO患者的个体化治疗提供参考。