乙型肝炎病毒相关慢加急性肝衰竭患者预后预测模型的构建

Construction of prognosis prediction model for patients with hepatitis B virus-associated chronic and acute liver failure

目的探讨肝炎病毒相关慢加急性肝衰竭(HBV-ACLF)血清甲胎蛋白(AFP)、CXC型趋化因子配体1(CXCL1)、N-myc和STAT相互作用蛋白(NMI)水平,并构建疾病预后预测模型。方法选择柳州市人民医院2017年3月至2022年3月收治的151例HBV-ACLF患者作为研究对象,根据随访90 d预后情况分为存活组、死亡组;另选取同期体检的30例健康志愿者作为对照组。比较3组血清AFP、CXCL1及NMI水平。采用单因素及多因素Logistic回归分析HBV-ACLF预后的影响因素,并建立HBV-ACLF预后预测模型,受试者工作特征(ROC)曲线分析血清AFP、CXCL1及NMI联合预后预测模型及终末期肝病模型(MELD)、慢性肝衰患者序贯性器官功能衰竭评分(CLIF-SOFA)对HBV-ACLF患者预后的预测价值。结果死亡组、存活组AFP、CXCL1、NMI水平显著高于对照组(P<0.05);死亡组患者血清AFP显著低于存活组(P<0.05),CXCL1、NMI水平显著高于存活组(P<0.05)。多因素Logistic回归结果显示,较大的年龄、腹水、肝性脑病、细菌感染、较低的血清AFP水平、较高的血清CXCL1、NMI水平均与HBV-ACLF患者预后较差(死亡)相关(P<0.05)。ROC曲线分析显示,基于血清AFP、CXCL1、NMI联合检测构建的HBV-ACLF患者预后预测模型具有较高的灵敏度、特异度、准确度。结论血清AFP低水平、CXCL1、NMI高水平是HBV-ACLF患者预后较差的危险因素,血清AFP、CXCL1、NMI联合建立的模型可预测患者预后,对于指导临床治疗和预后评估有一定价值。

Objective To explore the levels of serum alpha-fetoprotein(AFP),CXC type chemokine ligand 1(CXCL1),N-myc and STAT interaction protein(NMI)in hepatitis virus-associated chronic and acute liver failure(HBV-ACLF),and to construct the disease prognosis prediction model.Methods A total of 151 HBV-ACLF patients admitted to Liuzhou People′s Hospital from March 2017 to March 2022 were selected as the study subjects,and were divided into survival group and death group according to the prognosis during 90 days of follow-up.Another 30 healthy volunteers who underwent physical examination during the same period were selected as the control group.The levels of serum AFP,CXCL1 and NMI in 3 groups were compared.Univariate and multivariate Logistic regression were used to analyze the prognostic factors of HBV-ACLF,and a prognosis prediction model for HBV-ACLF was established.Receiver operating characteristic(ROC)curve was used to analyze the predictive value of serum AFP,CXCL1 and NMI combined prognostic model,end-stage liver disease model(MELD)and sequential organ failure score(CLIF-SOFA)in HBV-ACLF patients.Results The levels of AFP,CXCL1 and NMI in death and survival groups were significantly higher than those in control group(P<0.05).The serum AFP of death group was significantly lower than that of survival group(P<0.05),and the levels of CXCL1 and NMI were significantly higher than those of survival group(P<0.05).Multivariate Logistic regression results showed that older age,ascites,hepatic encephalopathy,bacterial infection,lower serum AFP level,higher serum CXCL1 and NMI levels were associated with poor prognosis(death)in HBV-ACLF patients(P<0.05).ROC curve analysis showed that the HBV-ACLF prognosis prediction model based on serum AFP,CXCL1 and NMI combined detection had high sensitivity,specificity and accuracy.Conclusion Low serum AFP level,high CXCL1 and NMI levels are risk factors for poor prognosis in patients with HBV-ACLF.The combined model established by serum AFP,CXCL1 and NMI can predict the prognosis of patients with HBV-ACLF,which has certain value for guiding clinical treatment and prognosis evaluation.