肝移植术后急性肾损伤的危险因素探究及列线图模型构建

Acute Kidney Injury after Liver Transplantation: Risk Factors and Nomogram Model

摘要: 目的：探讨肝移植(LT)术后急性肾损伤(AKI)的危险因素，构建列线图预测模型。方法：回顾性研究2017年1月至2019年8月在青岛大学附属医院接受LT手术244例患者临床资料。Logistic回归分析筛选术后AKI的危险因素，并以此构建列线图预测模型。结果：根据是否发生LT术后AKI将244例患者分为AKI组(n = 163)和非AKI组(n = 81)。与非AKI组相比，AKI组患者女性患者所占比例较高，BMI、MELD评分较高，术中失血量、术中输注红细胞较多，术前血肌酐(SCr)较低(P < 0.05)。多因素Logistic回归分析显示性别、BMI、MELD评分、术前肾小球滤过率、术前血肌酐、术中输注红细胞、无肝期时间是LT术后AKI的独立危险因素。Hosmer-Lemeshow检验显示预测模型拟合优度较高(X² = 3.62, P = 0.89)。受试者特征曲线分析显示预测模型曲线下面积为0.77 (95% CI: 0.71~0.83)。结论：基于围术期危险因素构建的LT术后AKI发生风险的列线图预测模型校准度和区分能力较好，便于临床医师进行危险分层和早期识别，进而采取有效干预措施改善预后。

Abstract: Objective: To explore the risk factors of acute kidney injury (AKI) after liver transplantation (LT) and establish a nomogram to predict the incidence of AKI after LT individually. Methods: The clinical data of 244 patients who underwent LT in the Affiliated Hospital of Qingdao University between January 2017 and August 2019 were retrospectively analyzed. Logistic regression methods were used to screen the risk factors of AKI after LT, and then to establish the nomogram to individually predict the incidence of AKI after LT. Results: According to the episode of AKI (yes/no), patients were divided into the AKI group (n = 163) and the non-AKI group (n = 81). Compared with patients without AKI, patients with AKI had higher proportion of female gender, higher body mass index, higher MELD score, higher intraoperative blood loss, higher intraoperative RBC transfusion and lower preoperative serum creatinine (P < 0.05). Multivariate Logistic regression analysis showed that gender, body mass index, MELD score, preoperative glomerular filtration rate, preoperative serum creatinine, intraoperative RBC transfusion, anhepatic time were independent risk factors for AKI after LT. Hosmer-Lemeshow test showed that the goodness of fit of the prediction model was good (X² = 3.62, P = 0.89). Receiver operating characteristic curve showed that the area under the curve predicted by prediction model for AKI after LT was 0.77 (95% CI: 0.71~0.83). Conclusion: Based on the perioperative risk factors for AKI after LT, a nomogram model for predicting the incidence of AKI after LT is established, which has a good calibration and discrimination, and can help clinicians with risk stratification and early recognition, thus enabling effective intervention to improve outcomes.