Variceal Upper Gastrointestinal Bleeding: A Retrospective Cohort of 98 Cases, Historical Comparison, and Updated...

Background: Variceal upper gastrointestinal bleeding (VUGIB) remains a major cause of short-term mortality in cirrhosis despite advances in endoscopic and pharmacological therapy. Prognostic factors and outcomes were evaluated in a historical cohort, and a guideline-aligned management algorithm is proposed. Methods: We conducted a retrospective study of 98 consecutive adults admitted with VUGIB to a tertiary surgical center in Romania (2009-2014). Demographics, etiology, admission hemoglobin (Hb), timing of presentation, endoscopic and surgical management, and outcomes were recorded. Survival was analyzed using Kaplan-Meier with log-rank tests; associations were tested with chi-square and t-tests; predictors of mortality were assessed with logistic regression. Receiver operating characteristic (ROC) curve analysis was used to evaluate the discriminatory ability of hemoglobin for in-hospital mortality. Results: Mean age was 57.8 ± 11.7 years; 60.2% were male. Cirrhosis etiology was alcoholic in 73%, viral in 18%, and metabolic in 9%. Endoscopy occurred within 48 h in 62% of patients, but only 4% underwent the procedure within 8 h. Overall mortality was 17.3%. Kaplan-Meier analysis showed no survival difference between alcoholic and viral cirrhosis (log-rank p = 0.39), but survival was markedly lower with Hb < 8 g/dL (p < 0.001). Admission delay was prognostic: >24 h was associated with worse survival (p < 0.05). On multivariable analysis (reference 1-2 days), admission at 3-4 days (OR 35.3, 95% CI 1.6-786, p = 0.024), >4 days (OR 71.0, 95% CI 2.16-2337, p = 0.017), and <6 h (OR 22.4, 95% CI 1.25-399.7, p = 0.035) independently predicted death. Admission Hb predicted mortality with an AUC of 0.79; the optimal cut-off was 4.3 g/dL (sensitivity 57%, specificity 95%). Limited use of pre-emptive TIPS likely contributed to outcomes. Conclusions: In this historical cohort, mortality from VUGIB was driven mainly by bleeding severity and admission delay, rather than by cirrhosis etiology. The study provides a historical benchmark for Eastern Europe, highlights gaps in adherence to contemporary standards, and supports restrictive transfusion, early vasoactive therapy, antibiotics, urgent endoscopy, and pre-emptive TIPS. The retrospective single-center design and limited therapies available during the study period remain important limitations.