Background: Upper gastrointestinal bleeding (UGIB) is a common clinical condition associated with significant morbidity and mortality.
Aim: The purpose of this study was to determine the etiology of UGIB and predict the adverse outcome of variceal and non-variceal UGIB by using the parameters of validated scores.
Methods: This is a prospective observational study in which 589 UGIB patients underwent upper gastrointestinal endoscopy from May 2010 to April 2013 in Nepal. The variables of Glasgow-Blatchford score (GBS) and Rockall score (RS) were used to predict adverse outcomes, which were defined as transfusion, rebleeding, readmission, surgery and death within 30 days.
Results: Among UGIB patients, 33.1 % were variceal and 66.9 % were non-variceal bleeding (peptic ulcers 23.9 %, gastric erosion 16.5 % and others). The adverse outcomes of variceal and non-variceal UGIB were 77.9 % and 72.6 %, respectively. The significant predictors of adverse outcome were hemoglobin and systolic blood pressure in overall UGIB (p < 0.001), blood urea (p = 0.002), melena (p < 0.001) and syncope (p < 0.001) in variceal UGIB, and heart rate, melena, syncope and malignancy in non-variceal UGIB (p < 0.001). The area under receiver operating curve for adverse outcome in overall UGIB was 0.76, 0.80 and 0.94 using clinical RS, complete RS and GBS, respectively (p < 0.001).
Conclusions: Variceal bleeding was the leading cause of UGIB in Nepal followed by peptic ulcers and gastric erosion. Those variables of GBS and RS, which were significant predictors of adverse outcome for UGIB, can be utilized in determining more aggressive endoscopic management during endoscopy, or more prolonged intravenous proton pump inhibitor use, or an additional day or two of hospitalization.