Interpretable prediction model for short- and long-term hepatic encephalopathy and variceal rebleeding after TIPS: a real-world study

{"INTRODUCTION:":[0],"Transjugular":[1],"intrahepatic":[2],"portosystemic":[3],"shunt":[4],"(TIPS)":[5],"relieves":[6],"portal":[7,190,197],"hypertension,":[8],"but":[9],"post-TIPS":[10,231],"overt":[11],"hepatic":[12],"encephalopathy":[13],"(OHE)":[14,141],"and":[15,31,40,43,89,102,117,121,129,142,148,150,166,212,229,233,241],"variceal":[16],"rebleeding":[17],"(VRB)":[18,144],"remain":[19],"outcomes":[20],"that":[21],"significantly":[22],"impair":[23],"survival.":[24],"We":[25],"aimed":[26],"to":[27,127],"identify":[28],"risk":[29,243],"factors":[30],"develop":[32],"an":[33,76],"interpretable":[34,221],"machine":[35,78],"learning":[36,79],"model":[37,73,223],"predicting":[38],"2-":[39,228],"6-year":[41,230],"OHE":[42,116,158,211,232],"VRB.":[44],"METHODS:":[45],"In":[46],"this":[47],"multicenter":[48],"retrospective":[49],"study,":[50],"636":[51],"patients":[52],"undergoing":[53],"TIPS":[54],"(2014-2023)":[55],"with":[56,91],"2-6":[57],"years":[58,165],"follow-up":[59],"were":[60,100,107],"analyzed.":[61],"A":[62],"detailed":[63],"perioperative":[64],"dataset":[65],"was":[66,85,119],"curated.":[67],"LASSO":[68],"selected":[69],"key":[70],"predictors":[71],"for":[72,109,157,176,179,210,215,227,239],"training":[74],"in":[75,189,196],"integrated":[77],"voting":[80],"ensemble":[81],"(ML-VE)":[82],"framework.":[83],"Efficacy":[84],"assessed":[86],"by":[87],"AUC":[88],"compared":[90],"conventional":[92,172],"clinical":[93],"scores":[94,173],"(Child-Pugh,":[95],"MELD,":[96],"MELD-Na).":[97],"Subgroup":[98],"analyses":[99],"conducted,":[101],"SHapley":[103],"Additive":[104],"exPlanations":[105],"(SHAP)":[106],"used":[108],"interpretation.":[110],"RESULTS:":[111],"The":[112,220],"cumulative":[113],"incidence":[114],"of":[115,139,161],"VRB":[118,194,216,234],"22.2%":[120],"8.2%":[122],"at":[123,131,145,152,163,168,217],"2":[124,146,164],"years,":[125,133,147,170],"increasing":[126],"24.8%":[128],"10.1%":[130],"6":[132,153,169],"respectively.":[134],"Internally,":[135],"ML-VE":[136,222],"achieved":[137],"AUCs":[138,160],"0.960":[140],"0.966":[143],"0.849(OHE)":[149],"0.828(VRB)":[151],"years;":[154],"external":[155],"validation":[156],"yielded":[159],"0.914":[162],"0.924":[167],"outperforming":[171],"(AUC":[174],"0.528-0.563":[175],"OHE;":[177],"0.528-0.555":[178],"VRB).":[180],"All":[181],"prespecified":[182],"subgroups":[183],"had":[184],"2-year":[185],"AUCs≥0.93":[186],"(OHE":[187],"prediction":[188,195],"vein":[191],"thrombosis:":[192],"0.993;":[193],"pressure":[198],"gradient":[199],"9-12":[200],"mmHg:":[201],"0.987).":[202],"SHAP":[203],"identified":[204],"\\"diabetes\\"":[205],"as":[206],"the":[207],"strongest":[208],"predictor":[209],"\\"recent":[213],"bleeding\\"":[214],"2-years.":[218],"CONCLUSIONS:":[219],"showed":[224],"good":[225],"discrimination":[226],"risk,":[235],"suggesting":[236],"potential":[237],"utility":[238],"short-":[240],"long-term":[242],"stratification":[244],"after":[245],"TIPS.":[246]}

Authors

Institutions

Publication Details

Journal
Clinical and Translational Gastroenterology
Published
2026-09-18
DOI
https://doi.org/10.14309/ctg.0000000000001096
Primary Topic
Liver Disease and Transplantation
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

Interpretable prediction model for short- and long-term hepatic encephalopathy and variceal rebleeding after TIPS: a real-world study

Min Zhang, Bin Song, Xiao Zhang, Jianyu Li et al.
Clinical and Translational Gastroenterology
Liver Disease and Transplantation
article

Interpretable prediction model for short- and long-term hepatic encephalopathy and variceal rebleeding after TIPS: a real-world study

Min Zhang, Bin Song, Xiao Zhang, Jianyu Li, Longbin Zhang, Shang Wan, Hongxi Chen
article en

Abstract

INTRODUCTION: Transjugular intrahepatic portosystemic shunt (TIPS) relieves portal hypertension, but post-TIPS overt hepatic encephalopathy (OHE) and variceal rebleeding (VRB) remain outcomes that significantly impair survival. We aimed to identify risk factors and develop an interpretable machine learning model predicting 2- and 6-year OHE and VRB. METHODS: In this multicenter retrospective study, 636 patients undergoing TIPS (2014-2023) with 2-6 years follow-up were analyzed. A detailed perioperative dataset was curated. LASSO selected key predictors for model training in an integrated machine learning voting ensemble (ML-VE) framework. Efficacy was assessed by AUC and compared with conventional clinical scores (Child-Pugh, MELD, MELD-Na). Subgroup analyses were conducted, and SHapley Additive exPlanations (SHAP) were used for interpretation. RESULTS: The cumulative incidence of OHE and VRB was 22.2% and 8.2% at 2 years, increasing to 24.8% and 10.1% at 6 years, respectively. Internally, ML-VE achieved AUCs of 0.960 (OHE) and 0.966 (VRB) at 2 years, and 0.849(OHE) and 0.828(VRB) at 6 years; external validation for OHE yielded AUCs of 0.914 at 2 years and 0.924 at 6 years, outperforming conventional scores (AUC 0.528-0.563 for OHE; 0.528-0.555 for VRB). All prespecified subgroups had 2-year AUCs≥0.93 (OHE prediction in portal vein thrombosis: 0.993; VRB prediction in portal pressure gradient 9-12 mmHg: 0.987). SHAP identified "diabetes" as the strongest predictor for OHE and "recent bleeding" for VRB at 2-years. CONCLUSIONS: The interpretable ML-VE model showed good discrimination for 2- and 6-year post-TIPS OHE and VRB risk, suggesting potential utility for short- and long-term risk stratification after TIPS.

Clinical and Translational Gastroenterology
Xiamen University (CN), Sichuan University (CN), West China Medical Center of Sichuan University (CN), Leshan Normal University (CN), First Affiliated Hospital of Xiamen University (CN), West China Hospital of Sichuan University (CN), Zhongshan Hospital of Xiamen University (CN), Chongqing Three Gorges Central Hospital (CN), Chengdu Second People's Hospital (CN), Chongqing Three Gorges University (CN)
Peace, Justice and strong institutions, Reduced inequalities
Openalex Percentile: Top 12%
Liver Disease and Transplantation
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.