乙型肝炎相关慢加急性肝衰竭患者急性肾损伤的影响因素和高危人群分析

Study of risk factors and high-risk population of acute kidney injury in patients with hepatitis B virus related acute on chronic liver failure

目的探讨基于分类树模型分析乙型肝炎相关慢加急性肝衰竭患者急性肾损伤的影响因素和高危人群的应用价值。方法本研究为回顾性观察性横断面研究,收集2010年1月至2018年6月在佛山市第一人民医院感染科、江门市中心医院感染科和南方医科大学顺德医院感染性疾病科住院治疗的乙型肝炎相关慢加急性肝衰竭患者的临床资料,利用分类树模型探索急性肾损伤的影响因素和高危人群。结果急性肾损伤组患者的MELD评分(31.6±9.0)和年龄(51.0±14.3)岁显著高于非急性肾损伤组患者的MELD评分(21.5±6.1)和年龄(44.1±13.0)岁,两组间差异有统计学意义(P<0.05)。多因素Logistic回归分析提示MELD评分(OR=1.209,95%CI:1.141~1.281)和年龄(OR=1.042,95%CI:1.012~1.072)可能为慢加急性肝衰竭患者发生急性肾损伤的独立影响因素。分类树模型提示慢加急性肝衰竭患者发生急性肾损伤的影响因素有MELD评分和年龄。基于MELD评分和年龄而建立的分类树模型的估计误差为0.127,提示该分类树模型的拟合效果较好。结论通过分类树模型发现慢加急性肝衰竭患者发生急性肾损伤和MELD评分、年龄关系密切,根据这2个指标建立的简单风险评估方法可用于评估慢加急性肝衰竭患者发生急性肾损的风险。

Objective To investigate the risk factors and high-risk population of acute kidney injury(AKI)in patients with hepatitis B virus related acute on chronic liver failure(HBV-ACLF)by classification tree method,and to establish a simple method for evaluating the risk of AKI in patients with HBV-ACLF.Methods The clinical data of patients with HBV-ACLF hospitalized in the Department of Infectious Diseases from 3 hospitals including the First People's Hospital of Foshan,Jiangmen Central Hospital and Shunde Hospital of Southern Medical University from January 2010 to June 2018 were collected and analyzed.The risk factors and high-risk population of AKI were investigated by classification tree method.Results The model for end-stage liver disease(MELD)score and age in AKI group were significantly higher than those in non-AKI group(31.6±9.0 vs 21.5±6.1,51.0±14.3 vs 44.1±13.0),with significant difference(P<0.05).Multivariate logistic regression analysis suggested that MELD score(odds ratio[OR]=1.209,95%confidence interval[CI]:1.141-1.281)and age(OR=1.042,95%CI:1.012-1.072)were independent factors of AKI in patients with HBV-ACLF.The classification tree model suggested that the factors affecting AKI in patients with HBV-ACLF include MELD score and age.The total agreement rate of the classification tree model based on MELD score and age was 0.873,indicating that the classification tree model had a better fitting effect.Conclusion AKI in patients with HBV-ACLF is related to MELD score and age.And a simple classification tree model based on MELD score and age can be used to assess the risk of AKI in patients with HBV-ACLF.