乙型肝炎肝硬化上消化道出血的ΔCT特征性表现及预测模型的建立

Characteristic manifestations of ΔCT in hepatitis B cirrhosis with upper gastrointestinal bleeding and establishment of a predictive model

目的本研究旨在寻找乙型肝炎肝硬化CT特征,建立肝硬化上消化道出血预测模型,预测出血风险。方法回顾性分析2015年1月—2021年6月天津市第一中心医院101例型肝炎肝硬化患者的数据,将其分为上消化道出血组(P=58)及非出血组(n=43)。比较两组间实验室检查以及强化CT检查测的平扫期、动脉期、门脉期以及静脉期的CT值,并计算各期间CT值的变化(ΔCT)。计量资料两组间的比较使用t检验或Mann-Whitney U检验;使用logistic回归分析方法,预测相关危险因素;通过计算受试者工作特征曲线下的面积评估模型辨别力,而模型校准则通过Hosmer-Lemeshow确定。在多变量logistic回归分析结果的基础上,使用Rstudio4.1.2软件的R包构建预测的列线图模型,并绘制相应的ROC曲线、校准曲线以及临床决策曲线。结果非出血组血清TBil、WBC、PLT水平与出血组比较,差异均有统计学意义(P值均<0.05);两组在肝-Plain、脾-P-Plain、脾-P-AΔCT值存在统计学差异(P值均<0.05)。单因素logistic分析结果显示,白细胞(OR=0.770,95%CI:0.624~0952,P=0.016)、血小板(OR=0.979,95%CI:0.965~0.994,P=0.006)、肝脏平扫期(OR=1.142,95%CI:1.058~1.233,P=0.001)、脾脏门脉期-平扫ΔCT值(OR=0.979,95%CI:0.959~1.000,P=0.050)、脾脏门脉期-动脉期ΔCT值(OR=0.979,95%CI:0.944~0.994,P=0.015)在乙型肝炎肝硬化患者发生上消化道出血与未出血两者之间差异具有统计学意义。多因素logistic分析结果显示血小板(OR=0.968,95%CI:0.944~0.993,P=0.011)、肝脏平扫期(OR=1.148,95%CI:1.047~1.259,P=0.003)、脾脏门脉期-动脉期ΔCT值(OR=0.951,95%CI:0.908~0.995,P=0.030)为上消化道出血的独立危险因素。基于多因素logistic分析结果,构建了乙型肝炎肝硬化上消化道出血的预测模型并绘制校准曲线。该模型的受试者特征曲线下面积为0.801,cut-off值为0.433,其对应的敏感度是81.4%,特异度是77.6%。模型的校准曲线与理想曲线贴合良好。结论乙型肝炎肝硬化肝脏具有特殊的ΔCT变化,通过ΔCT构建的预测模型对于乙型肝炎肝硬化上消化道出血具有良好的预测能力。

Objective To investigate the CT characteristics of hepatitis B cirrhosis,and to predict the risk of bleeding by establishing a predictive model for upper gastrointestinal bleeding in liver cirrhosis.Methods A retrospective analysis was performed for the clinical data of 101 patients with hepatitis B cirrhosis who were admitted to Tianjin First Central Hospital from January 2015 to June 2021,and these patients were divided into upper gastrointestinal bleeding group and non-bleeding group.The two groups were compared in terms of laboratory findings and CT values in plain scan,arterial phase,portal vein phase,and venous phase measured by contrast-enhanced CT,and the changes in CT values(ΔCT)across different phases were calculated.The t-test or the Mann-Whitney U rank sum test was used for comparison of continuous data between two groups,and the chi-square test was used for comparison of categorical data between two groups.Logistic regression analysis was used to predict the related risk factors;The discrimination of the model was evaluated by calculating the area under the working characteristic curve of the subjects,and the model calibration criteria were determined by Hosmer-lemeshow.Based on the results of multivariate logistic regression analysis,Rstudio4.1.2 R package was used to establish a predictive model,and draws the corresponding ROC curve,calibration curve and clinical decision curve.Results There were significant differences in serum TBil,WBC and PLT levels between the non-bleeding group and the bleeding group(all P<0.05).There were significant differences in liver-plain,spleen-P-plain and spleen-P-AΔCT(all P<0.05).The univariate logistic analysis showed that there were significant differences in leukocytes(odds ratio[OR]=0.770,95%confidence interval[CI]:0.624-0.952,P=0.016),platelets(OR=0.979,95%CI:0.965-0.994,P=0.006),liver plain scan(OR=1.142,95%CI:1.058-1.233,P=0.001),ΔCT value of the spleen from portal vein phase to plain scan(OR=0.979,95%CI:0.959-1.000,P=0.050),andΔCT value of the spleen from portal vein phase to arterial phase(OR=0.979,95%CI:0.944-0.994,P=0.015)between the hepatitis B cirrhosis patients with upper gastrointestinal bleeding and those without bleeding.The multivariate logistic analysis showed that platelets(OR=0.968,95%CI:0.944-0.993,P=0.011),liver plain phase(OR=1.148,95%CI:1.047-1.259,P=0.003),andΔCT value of the spleen from portal vein phase to arterial phase(OR=0.951,95%CI:0.908-0.995,P=0.030)were independent risk factors for upper gastrointestinal bleeding.A predictive model for upper gastrointestinal bleeding in hepatitis B cirrhosis was established based on the results of the multivariate logistic analysis,and a calibration curve was plotted.This model had an area under the receiver operating characteristic curve of 0.801 at the cut-off value of 0.433,with a sensitivity of 81.4%and a specificity of 77.6%.The calibration curve of the model fitted well with the ideal curve.Conclusion There are specialΔCT changes in hepatitis B cirrhosis,and the predictive model based onΔCT has a good predictive ability for upper gastrointestinal bleeding in patients with hepatitis B cirrhosis.