急性肝衰竭并发急性肾损伤的影响因素及预测模型

Influencing factors for acute kidney injury in acute liver failure and establishment of a predictive model

目的 探讨急性肝衰竭(ALF)患者并发急性肾损伤(AKI)的预测因素,建立新型预测模型。方法 收集2015年1月—2021年10月郑州大学第一附属医院确诊为ALF的253例患者的临床资料,根据是否发生AKI分为非AKI组(n=170)和AKI组(n=83)。收集并分析两组患者的临床资料和实验室指标。计量资料不服从正态分布采用M(P_(25)~P_(75))表示,两组间比较采用Mann-Whitney U检验,计数资料用例(%)表示,两组间比较采用χ~2检验。采用二元Logistic回归分析ALF患者发生AKI的危险因素,通过受试者工作特征曲线(ROC曲线)分析所得指标对ALF患者发生AKI的预测效能。结果 AKI组高血压、糖尿病、肝性脑病、腹水、肺部感染的比率以及WBC、INR、CRP、PCT、NLR、MELD评分均高于非AKI组,PLT、LMR、PNI水平低于非AKI组,差异均具有统计学意义(P值均<0.05)。多因素Logistic回归分析结果显示,WBC(OR=1.267,95%CI:1.124~1.428,P<0.001)、INR(OR=1.663,95%CI:1.205~2.293,P=0.002)、PCT(OR=1.416,95%CI:1.137~1.764,P=0.002)、MELD(OR=1.098,95%CI:1.029~1.172,P=0.005)是ALF患者并发AKI的危险因素。ROC曲线分析显示,WBC+INR+PCT+MELD联合预测ALF患者并发AKI的ROC曲线下面积(AUC)最高(AUC=0.908),而WBC、INR、PCT、MELD单独预测的AUC 分别为0.776、0.771、0.746、0.780。结论 WBC、INR、PCT、MELD是ALF患者并发AKI的独立影响因素,四者联合建立的预测模型预测价值较高。

Objective To investigate the predictive factors for acute kidney injury(AKI) in patients with acute liver failure(ALF),and to establish a new predictive model.Methods Clinical data were collected from 253 patients who were diagnosed with ALF in The First Affiliated Hospital of Zhengzhou University from January 2015 to October 2021,and according to the presence or absence of AKI,these patients were divided into non-AKI group with 170 patients and AKI group with 83 patients.Related clinical data and laboratory markers were collected.Non-normally distributed continuous data were expressed as M(P_(25)-P_(75)),and the Mann-Whitney U test was used for comparison between two groups;categorical data were expressed as cases(%),and the chi-square test was used for comparison between two groups.The binary logistic regression analysis was used to investigate the risk factors for AKI in ALF patients,and the receiver operating characteristic(ROC) curve was used to evaluate the performance of the indices obtained in predicting AKI in ALF patients.Results Compared with the non-AKI group,the AKI group had a significantly higher proportion of patients with hypertension,diabetes,hepatic encephalopathy,ascites,and pulmonary infection,significantly higher levels of white blood cell count(WBC),international normalized ratio(INR),C-reactive protein,procalcitonin(PCT),neutrophil-to-lymphocyte ratio,and Model for End-Stage Liver Disease(MELD) score,and significantly lower levels of platelet count,lymphocyte-to-monocyte ratio,and PNI(all P<0.05).The multivariate logistic regression analysis showed that WBC(odds ratio [OR]=1.267,95% confidence interval [CI]:1.124-1.428,P<0.001),INR(OR=1.663,95%CI:1.205-2.293,P=0.002),PCT(OR=1.416,95%CI:1.137-1.764,P=0.002),and MELD score(OR=1.098,95%CI:1.029-1.172,P=0.005) were risk factors for the development of AKI in patients with ALF.The ROC curve analysis showed that the combination of WBC+INR+PCT+MELD had the largest area under the ROC curve(AUC) of 0.908 in predicting AKI in ALF patients,while WBC,INR,PCT,and MELD alone had an AUC of 0.776,0.771,0.746,and 0.780,respectively,in predicting AKI.Conclusion WBC,INR,PCT,and MELD score are independent influencing factors for AKI in patients with ALF,and the predictive model established based on these four indices has a relatively high predictive value.