Department of Cardiology, The Affiliated Panyu Central Hospital of Guangzhou Medical University, Guangzhou, China.
Department of Cardiothoracic ICU, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China.
Ren Fail. 2024 Dec;46(2):2398711. doi: 10.1080/0886022X.2024.2398711. Epub 2024 Sep 5.
The prognosis-predicting factors for non-surgical patients receiving continuous renal replacement therapy (CRRT) and extracorporeal membrane oxygenation (ECMO) remains limited. In this study, we aim to analyze prognosis-predicting factors in the non-surgical patients receiving these two therapies.
We retrospectively analyzed data from non-surgical patients with ECMO treatment from December 2013 until April 2023. Hospital mortality was primary endpoint of this study. The area under the curve and receiver operating characteristic curves were used to assess the sensitivity and specificity of mortality. The independent risk factors were identified by multivariate logistic regression. The prediction model was a nomogram, and decision curve analysis and the calibration plot were used to assess it. Using restricted cubic spline curves and Spearman correlation, the correlation analysis was performed.
The model that incorporated CRRT duration and age surpassed the two variables alone in predicting hospital mortality in non-surgical patients with ECMO therapy (AUC value = 0.868, 95% CI = 0.779-0.956). Older age, CRRT implantation, and duration were independent risk factors for hospital mortality (all < 0.05). The nomogram predicting outcomes model containing on CRRT implantation and duration was developed, and the consistency between the predicted probability and observed probability and clinical utility of the models were good. CRRT duration was negatively associated with hemoglobin concentration and positively associated with urea nitrogen and serum creatinine levels.
Hospital mortality in non-surgical ECMO patients was found to be independently associated with older age, longer CRRT duration, and CRRT implantation.
接受连续肾脏替代治疗（CRRT）和体外膜肺氧合（ECMO）的非手术患者的预后预测因素仍然有限。本研究旨在分析接受这两种治疗的非手术患者的预后预测因素。
我们回顾性分析了 2013 年 12 月至 2023 年 4 月接受 ECMO 治疗的非手术患者的数据。本研究的主要终点为住院死亡率。使用曲线下面积和受试者工作特征曲线评估死亡率的敏感性和特异性。使用多变量逻辑回归确定独立危险因素。预测模型为列线图，并使用决策曲线分析和校准图进行评估。使用限制立方样条曲线和 Spearman 相关分析进行相关性分析。
在接受 ECMO 治疗的非手术患者中，纳入 CRRT 持续时间和年龄的模型在预测住院死亡率方面优于单独的两个变量（AUC 值=0.868，95%CI=0.779-0.956）。年龄较大、CRRT 植入和持续时间是住院死亡率的独立危险因素（均<0.05）。建立了包含 CRRT 植入和持续时间的预测结局模型，模型的预测概率与观察概率之间的一致性和模型的临床实用性均较好。CRRT 持续时间与血红蛋白浓度呈负相关，与尿素氮和血清肌酐水平呈正相关。
非手术 ECMO 患者的住院死亡率与年龄较大、CRRT 持续时间较长和 CRRT 植入有关。
