Etiology, management and outcome of acute upper gastrointestinal bleeding in patients seen at Moi Teaching and Referral Hospital, Eldoret, Kenya

Abstract

Background: Upper gastrointestinal bleeding is bleeding into the lumen of the gastrointestinal tract involving the esophagus, stomach and the duodenum. It is the most common form of gastrointestinal bleeding being four times more common than lower gastrointestinal bleeding. It represents approximately 1% of emergency room admissions with a mortality rate of 4 – 14% worldwide. Leading causes worldwide include peptic ulcer disease and esophageal varices. Management requires initial resuscitation and stabilization followed by endoscopic evaluation to make a definitive diagnosis. Ultimate management is medical, endoscopic surgery or open surgery. Although this type of bleeding is common, it has not been described in the Western Kenya region. Objective: To describe the etiology, management and outcome of upper gastrointestinal bleeding among patients seen at the Moi Teaching and Referral Hospital. Methods: This was a prospective descriptive census study conducted at the Moi Teaching and Referral Hospital, Eldoret, Kenya, between October 2015 and September 2016. A total of 63 patients aged 18 years and above presenting with upper gastrointestinal bleeding were recruited into the study. Data was collected using a structured data collection form by interviewing the patients and checking their files. The etiology, management and outcome were recorded. Data analysis was done using R statistical package version 3.4.2, year 2017. Frequency tables were generated for categorical variables. Factors associated with control of bleeding and re-bleeding were assessed using logistic regression model with odds ratio and 95% confidence intervals reported. Time to death was described using Kaplan-Meier survival curve. Predictors of death were assessed using Cox proportional hazards regression model. Results were presented using tables and graphs. Results: The leading etiologies of upper gastrointestinal bleeding at Moi Teaching and Referral Hospital were gastroduodenal erosions (27.6%), gastroesophageal tumors (25.6%), varices (14.9%) and ulcer (14.8%). Diagnostic endoscopy was done in 74.6% of the participants. Definitive management was medical (81.0%), endoscopic surgery (12.7%) and open surgery (6.3%). Control of bleeding was achieved for 81.0% of participants with a 46.0% re-bleeding rate reported. A higher pulse rate (>100bpm) had up to 86% reduced odds of bleeding control, OR: 0.14(95% CI: 0.03, 0.62). On the other hand, high SBP (>90mmHg) was associated with increased odds of controlling bleeding, OR: 10.63 (95% CI: 2.86, 39.49). However, participants with normal platelet count were less likely to re-bleed, OR: 0.29 (95% CI 0.10, 0.81). Mortality rate was 20.6% with death occurring faster within the first 5 days. Conclusion: At Moi Teaching and Referral Hospital, the main etiologies were gastroduodenal erosions, gastroesophageal tumors, varices and ulcers. Majority of the patients underwent medical management. Re-bleeding and mortality rates were high. Recommendations: High index of suspicion on etiology of upper gastrointestinal bleed being a tumor. Measures to be put in place to mitigate the high morbidity and mortality by using the mentioned predictors in clinical practice such as thrombocytopenia and appropriate primary diagnosis of bleeding