Endoscopic intervention is associated with a lower risk of in-hospital mortality in patients with acute myocardial infarction complicated by gastrointestinal bleeding

Introduction It remains unclear whether endoscopic intervention affects the prognosis of patients with acute myocardial infarction (AMI) complicated by gastrointestinal bleeding (GIB). This study aimed to evaluate the association between endoscopic intervention and the risk of in-hospital all-cause mortality in patients with AMI complicated by GIB. Methods The study had a single-center retrospective cohort design and included consecutive adult patients with AMI complicated by GIB who were hospitalized at Beijing Anzhen Hospital between January 2020 and February 2026. The patients were divided into an endoscopic intervention group and a no-endoscopic intervention group according to whether or not they underwent endoscopic intervention during hospitalization. Baseline differences between the two groups were further addressed using propensity score methods. The primary outcome was in-hospital all-cause mortality. For the primary survival analysis, follow-up began at recognition of GIB, and endoscopic intervention was modeled as a time-varying exposure in Cox regression. Propensity score matching (PSM) was used as a sensitivity analysis to address measured baseline confounding. The secondary outcome was serious in-hospital clinical adverse events after GIB. Results A total of 155 hospitalized patients with AMI complicated by GIB were included in the study (endoscopy group, n = 57; non-endoscopy group, n = 98). In the adjusted time-varying Cox regression model using the recorded intraday times, endoscopic intervention was associated with a lower risk of in-hospital all-cause mortality (adjusted hazard ratio 0.478, 95% confidence interval 0.240–0.951, P = 0.035). In the adjusted 24-hour landmark analysis, the association was directionally similar but did not reach statistical significance (hazard ratio 0.433, 95% confidence interval 0.184–1.020, P = 0.056). In the propensity-score-matched cohort, endoscopic intervention remained associated with a lower risk of in-hospital all-cause mortality after further adjustment for residual imbalance (adjusted hazard ratio 0.346, 95% confidence interval 0.152–0.786, P = 0.011). No statistically significant between-group difference was observed in serious in-hospital clinical adverse events after GIB (odds ratio 1.340, 95% confidence interval 0.660–2.724, P = 0.406). Discussion In this retrospective cohort, endoscopic intervention was associated with a lower risk of in-hospital all-cause mortality in patients with AMI complicated by GIB, although residual confounding and the uncertainty observed across complementary analyses should be considered when interpreting this association.