Impact of duration to endoscopy in patients with non-variceal upper gastrointestinal bleeding: propensity score matching analysis of real-world data from Thailand

Abstract

Background: The findings on mortality, rebleeding rate, and hospital stay in patients who underwent early vs. late endoscopy are conflicting. We aimed to compare in-hospital outcomes and medical resource use in patients with acute non-variceal upper gastrointestinal bleeding. Methods: We retrospectively reviewed the medical records of patients with acute non-variceal upper gastrointestinal bleeding who underwent early or late endoscopy between 2016 and 2019. The primary outcome was in-hospital mortality. The secondary outcomes were the need for packed red blood cells and number of transfusions, the proportion of lesions with high-risk stigmata, endoscopic and additional hemostasis, in-hospital rebleeding, duration of stay, and admission cost. Statistical analysis was performed using Pearson’s chi-squared or Fisher’s exact test for categorical variables, Student’s t-test, and Wilcoxon rank-sum test for continuous variables. Results: Early and late endoscopies were performed on 451 and 279 patients, respectively. After 1:1 propensity score matching, 278 patients from each group were included, and patients’ baseline characteristics were similar in the matched groups. Compared with the late group, the early group had a significantly increased rate of endoscopic hemostasis (22.7% vs. 13.7%, P = 0.006) and a low rate of packed red blood cell transfusion (53.6% vs. 61.9%, P = 0.048). Duration of stay and admission costs were significantly higher in the late group than in the early group (all P < 0.05). After adjusting for confounding factors, early endoscopy was positively associated with ulcers with high-risk stigmata (adjusted odds ratio = 1.83, P = 0.023) and endoscopic hemostasis (adjusted odds ratio = 1.97, P = 0.004). It was negatively associated with the need for packed red blood cell transfusion (adjusted odds ratio = 0.62, P = 0.017) and duration of stay (adjusted coefficient=-0.10, P < 0.001) with no impact on in-hospital mortality, rebleeding, or radiological interventions. Conclusions: The timing of endoscopy does not affect in-hospital mortality or rebleeding rate. This study supports using early endoscopy in patients with acute non-variceal upper gastrointestinal bleeding based on the potential benefits and feasibility of medical resource use.