肝硬化食管胃底静脉曲张破裂出血患者经颈静脉肝内门体分流术后发生显性肝性脑病的列线图预测模型建立及评价

Establishment and evaluation of a nomogram prediction model for overt hepatic encephalopathy after transjugular intrahepatic portosystemic shunt in patients with liver cirrhosis and esophagogastric variceal bleeding

目的构建肝硬化食管胃底静脉曲张破裂出血后行经颈静脉肝内门体分流术(TIPS)患者发生显性肝性脑病的预测模型列线图,并对模型预测能力进行评价。方法选取苏州大学附属第一医院2020年1月—2022年12月收治的因肝硬化食管胃底静脉曲张破裂出血经药物治疗或内镜下止血失败后行TIPS的患者113例,所有患者术后随访6个月,根据TIPS术后是否发生显性肝性脑病,分为显性肝性脑病组(n=32)和非显性肝性脑病组(n=81),收集患者的临床资料、血常规、血清生化学及血凝检查等结果。计量资料两组间比较采用成组t检验或Mann-Whitney U检验。计数资料组间比较采用χ^(2)检验或Fisher精确概率法。利用多因素Logistic回归法分析TIPS术后肝性脑病发生的独立影响因素,并构建列线图预测模型,计算一致性指数(C-index)并绘制校准曲线,评价列线图预测能力。绘制临床决策曲线,分析模型临床净收益。通过受试者工作特征曲线(ROC曲线)对模型的预测能力进行验证。结果显性肝性脑病组与非显性肝性脑病组比较,年龄、糖尿病、ChildPugh分级、腹水、术前门静脉主干直径、GGT水平、凝血酶原时间及弗赖堡术后生存指数(FIPS)评分差异均有统计学意义(P值均<0.05)。多因素Logistic回归分析结果显示,Child-Pugh分级(OR=17.498,95%CI:2.561~119.548,P=0.004)、术前门静脉主干直径(OR=1.361,95%CI:1.057~1.752,P=0.017)、GGT(OR=1.032,95%CI:1.013~1.052,P=0.001)和FIPS评分(OR=2.838,95%CI:1.277~6.311,P=0.010)是TIPS术后发生显性肝性脑病的独立影响因素。基于以上4项指标建立列线图模型,C-index为0.875,校准曲线拟合良好;模型的ROC曲线下面积为0.875(95%CI:0.799~0.929,P<0.001);决策曲线分析结果显示,0.3~0.9阈概率间模型具有较良好的净获益。结论Child-Pugh分级、术前门静脉主干直径、GGT及FIPS评分对肝硬化食管胃底静脉曲张破裂出血患者TIPS术后显性肝性脑病的发生具有一定的预测价值,基于上述指标构建的列线图模型可以个体化预测肝硬化食管胃底静脉曲张破裂出血患者TIPS术后显性肝性脑病的发生率。

Objective To establish a nomogram prediction model for the development of overt hepatic encephalopathy(OHE)in patients with liver cirrhosis undergoing transjugular intrahepatic portosystemic shunt(TIPS)after esophagogastric variceal bleeding,and to evaluate the predictive ability of the model.Methods This study was conducted among 113 patients with esophagogastric variceal bleeding due to liver cirrhosis who were admitted to The First Affiliated Hospital of Soochow University from January 2020 to December 2022 and underwent TIPS after failed medical or endoscopic therapy.All patients were followed up for 6 months after surgery,and according to the presence or absence of OHE after TIPS,they were divided into OHE group with 32 patients and non-OHE group with 81 patients.Related data were collected from all patients,including clinical data,routine blood test results,serum biochemistry,and coagulation test results.The independent-samples t test or the Mann-Whitney U test was used for comparison of continuous data between two groups,and the chi-square test or the Fisher’s exact test was used for comparison of categorical data between two groups.The multivariate Logistic regression analysis was used to investigate the independent risk factors for the onset of OHE after TIPS,and then a nomogram prediction model was established.The index of concordance(Cindex)was calculated and the calibration curve was plotted to evaluate the predictive ability of the model,and the clinical decision curve was plotted to analyze the net clinical benefit of the model.The receiver operating characteristic(ROC)curve was used to validate the predictive ability of the model.Results There were significant differences between the OHE group and the non-OHE group in age,diabetes,Child-Pugh class,ascites,main portal vein diameter before surgery,gamma-glutamyl transpeptidase(GGT)level,prothrombin time,and Freiburg index of post-TIPS survival(FIPS)score(all P<0.05).The multivariate Logistic regression analysis showed that Child-Pugh class(odds ratio[OR]=17.498,95%confidence interval[CI]:2.561—119.548,P=0.004),main portal vein diameter before surgery(OR=1.361,95%CI:1.057—1.752,P=0.017),GGT(OR=1.032,95%CI:1.013—1.052,P=0.001),and FIPS score(OR=2.838,95%CI:1.277—6.311,P=0.010)were independent influencing factors for the development of OHE after TIPS.The nomogram model established based on the above four indicators had a C-index of 0.875,good fitting of the calibration curve,and an area under the ROC curve of 0.875(95%CI:0.799—0.929,P<0.001),and the decision curve analysis showed that the 0.3—0.9 threshold probability model had a good net benefit.Conclusion Child-Pugh class,main portal vein diameter before surgery,GGT,and FIPS score have a certain value in predicting the development of OHE after TIPS in patients with esophagogastric variceal bleeding due to liver cirrhosis,and the nomogram model established based on these indicators can be used to individually predict the onset of OHE after TIPS in patients with esophagogastric variceal bleeding due to liver cirrhosis.