Abstract
Objective To explore the impact of serological indicators on clinically significant portal hypertension(CSPH)in liver cirrhosis.Based on serum metabolomics analysis,to screen for the differential metabolites related to CSPH,and to explore the predictive effect of bile acid on CSPH.Methods Patients with liver cirrhosis,who admitted to the Fifth Medical Center of Chinese PLA General Hospital between Jan.2023 and Dec.2023,were prospectively enrolled.Based on the Baveno Ⅶ expert consensus,patients were stratified into the CSPH group(n=21)and the non-CSPH(n=27)group.Clinical characteristics were compared between the two groups,and Logistic regression analysis was applied to identify the serological indicators related to CSPH.Serum metabolites were analyzed by liquid chromatography-mass spectrometry(LC-MS).Differential metabolites were screened,and a decision tree classification model was developed to assess the predictive effect of differential metabolites on CSPH.The sensitivity and specificity were evaluated with ROC curve analysis.Results Baseline characteristics gender,age,body mass index(BMI),cirrhosis etiology,alanine aminotransferase(ALT),aspartate aminotransferase(AST),blood urea(BUN),serum creatinine(Scr),portal vein diameter showed no significant difference between the two groups(P>0.05).However,significant differences were observed in ascites depth,liver stiffness,white blood cell count,platelet count,hemoglobin,albumin(ALB),totaol bile acid(TBA),total bilirubin(TBIL),direct bilirubin(DBIL),Child-Pugh score,and spleen length(P<0.05).Univariate Logistic regression analysis showed that hepatitis B virus infection,hemoglobin>134 g/L,TBIL>19.8 μmol/L,TBA>11.5 μmol/L,and spleen length>125 mm were related to CSPH(P<0.05).Multivariate Logistic regression analysis identified that TBA>11.5 μmol/L was independent risk factor for CSPH(P<0.05).Metabolomics analysis revealed 8 differentially expressed serum metabolites(P<0.05),including cholic acid,p-toluenesulfonic acid,uridine,estrone glucuronide,phosphatidic acid(15:0/0:0),1-hydroxy-2-naphthoic acid,19-oxotestosterone,and phosphatidic acid[20:5(7Z,9Z,11E,13E,17Z)-3OH(5,6,15)/i-13:0].Cholic acid was significantly upregulated in the CSPH group(P<0.01).The ROC curve for cholic acid alone yielded an AUC of 0.762(sensitivity:1.000;specificity:0.444).A decision tree classification model incorporating cholic acid,19-oxotestosterone,and estrone glucuronide achieved an AUC of 0.919(sensitivity:0.810;specificity:0.926).Conclusion TBA is an independent risk factor for CSPH in cirrhosis(P<0.05),with cholic acid significantly elevated.The decision tree classification model incorporating cholic acid,19-oxotestosterone,and estrone glucuronide demonstrates strong predictive value for CSPH.关键词
临床显著性门静脉高压症/肝硬化/胆汁酸/代谢组学分析/决策树模型Key words
clinically significant portal hypertension/liver cirrhosis/bile acid/metabonomics analysis/decision tree classification model分类
医药卫生