RESUMO
PURPOSE: After trauma and surgery, upper gastrointestinal bleeding (UGIB) is the most common condition that can require massive transfusion (MT). The present study aimed to analyze and compare the prognostic performance of the Glasgow-Blatchford (GB), pre-endoscopy Rockall (PER), and modified early warning (MEW) scores for predicting MT in patients with unstable UGIB. MATERIALS AND METHODS: This retrospective observational study included patients with UGIB from March 2016 to February 2018. Receiver operating characteristics analysis was performed to examine the prognostic performance of the GB, PER, and MEW scoring systems. Logistic regression analysis was used to identify independent risk factors for MT, after adjusting for relevant covariates. The primary outcome was MT. RESULTS: Of the 484 included patients with unstable UGIB, 19 (3.9%) received an MT. The areas under the curves (AUCs) of the GB, PER, and MEW scores for MT were 0.577 [95% confidence interval (CI), 0.531–0.621], 0.570 (95% CI, 0.525–0.615), and 0.767 (95% CI, 0.727–0.804), respectively. The AUC of the MEW score was significantly different from those of the GB and PER scores. In multivariate analysis, MEW score was independently associated with MT in patients with unstable UGIB (odds ratio, 1.495; 95% CI, 1.100–2.033; p=0.010). CONCLUSION: In unstable UGIB patients, MEW score had the best prognostic performance for MT among three scoring systems.