Background: Acute upper gastrointestinal (GI) bleeding is a common and potentially life-threatening emergency frequently seen in emergency departments. Early and accurate risk stratification is essential for guiding clinical management. This study aimed to assess the effectiveness of various clinical scoring systems and laboratory parameters in predicting in-hospital mortality in patients with upper GI bleeding.
Methods: This retrospective study included patients aged 18 and older diagnosed with upper GI bleeding between January 1, 2016, and January 1, 2019. The Glasgow Blatchford Score (GBS), Pre-endoscopic Rockall Score (PRS), total Rockall Score, AIMS65, and NEWS+L scores were calculated. Laboratory parameters were also analyzed for their association with mortality.
Results: A total of 316 patients were included, with an in-hospital mortality rate of 13.3%. Among the scoring systems, AIMS65 and NEWS+L demonstrated higher predictive accuracy (AUC) for in-hospital mortality compared to PRS. GBS and total Rockall scores also performed better than PRS.
Conclusion: AIMS65 and NEWS+L scores were superior to PRS in predicting mortality. Additionally, low hemoglobin, elevated lactate, high neutrophil-to-lymphocyte (NLR) and platelet-to-lymphocyte (PLR) ratios, and increased urea-to-creatinine levels were associated with higher mortality risk.
Primary Language | English |
---|---|
Subjects | Medical Education, Health Services and Systems (Other) |
Journal Section | Original Articles |
Authors | |
Publication Date | June 20, 2025 |
Submission Date | December 11, 2024 |
Acceptance Date | April 16, 2025 |
Published in Issue | Year 2025 Volume: 52 Issue: 2 |