PREDICTORS OF 30-DAY REBLEEDING AFTER SUCCESSFUL ENDOSCOPIC HEMOSTASIS FOR PEPTIC ULCERS: A SYSTEMATIC REVIEW AND EXPLORATORY META-ANALYSIS

Abstract

Rebleeding after successful endoscopic hemostasis remains a clinically important complication of peptic ulcer bleeding. We assessed prognostic factors for clinically confirmed rebleeding within 30 days and pooled effects only when predictor definitions, outcome timing, and effect measures were sufficiently comparable. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 statement. Two reviewers independently screened 1,523 deduplicated records, assessed the available full texts, extracted prognostic estimates, checked cohort overlap, and evaluated risk of bias with the Quality In Prognosis Studies tool. The primary synthesis prioritized adjusted odds ratios from independent cohorts reporting a 30-day outcome; random-effects pooling was prespecified only for genuinely comparable studies. Four independent contemporary cohorts contributed 10 verified adjusted associations, but no identically defined adjusted predictor was reproduced in at least two independent cohorts. The strongest individual associations were reported for shock (adjusted odds ratio, 12.691; 95% confidence interval, 5.129–31.399), estimated glomerular filtration rate <30 mL/min/1.73 m² (adjusted odds ratio, 4.770; 95% confidence interval, 1.168–18.211), higher Rockall risk, and procedure duration >13.5 minutes (adjusted odds ratio, 2.899; 95% confidence interval, 1.768–4.754). Contemporary studies identify several clinically relevant risk signals, but heterogeneity in predictor definitions and multivariable models precludes a reliable pooled adjusted estimate. Future prognostic studies should use standardized definitions and a uniform 30-day outcome.