Features of patients with upper gastrointestinal bleeding and factors affecting the re-bleeding risk

BACKGROUND: The risk of re-bleeding in upper gastrointestinal bleeding (UGIB) is a major complication that can be mortal. In this study, we aimed to determine the factors that can predict the risk of re-bleeding in UGIB patients. METHODS: We retrospectively screened UGIB patients admitted in adult intensive care. Along with age and gender, complaints of admission, vital signs, comorbidities, laboratory findings, medications, endoscopy status, and re-bleeding status were recorded. According to these findings, Glasgow-Blatchford and AIMS65 scores of the patients were calculated. All statistical tests were performed with the Predictive Analytics Software (PASW(®), version 18, SPSS Inc., Chicago, IL). RESULTS: A total of 241 patients were included in the study. Mean age of patients was 57.58±19.31, years and 176 (73.0%) of them were male. A total of 117 (48.5%) patients were Helicobacter pylori positive and re-bleeding occurred in 77 (32.0%) patients. Sclerotherapy was applied in 103 (42.7%) patients, while 5 (2.1%) underwent electrocoagulation, and 4 (1.7%) underwent hemoclips. There was a significant difference between patients with and without endoscopic intervention for re-bleeding (p<0.001). Hematocrit, urea values, Glasgow-Blatchford, and AIMS65 scores were statistically significant different between the groups with and without re-bleeding (p=<0.001, <0.001, <0.001, and 0.008, respectively). In the ROC analysis of Glasgow-Blatchford and AIMS65 scoring systems area, under the curve values were 0.700 (p=<0.001, 95% CI: 0.626–0.775), and 0.557 (p=0.194, CI 95%: 0.469–0.645), respectively. Mortality rate was 2.1% (n=5) among study population. CONCLUSION: Hematocrit and urea values seem beneficial among studied laboratory values, however, Glasgow-Blatchford scoring system performed better than AIMS65 in the prediction of re-bleeding risk in UGIB. The authors concluded that more specific predictive markers may be useful for clinicians.