Background: Gastrointestinal bleeding is a hemorrhagic complication after primary percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction (STEMI).
Objectives: To determine predictors of gastrointestinal bleeding and the impact of gastrointestinal bleeding on outcomes in STEMI patients undergoing primary percutaneous coronary intervention.
Methods and results: Gastrointestinal bleeding occurred in 18 (3.5%) of 519 consecutive patients with STEMI undergoing primary percutaneous coronary intervention. Univariate predictors of gastrointestinal bleeding were previous gastrointestinal bleeding (33% vs 4%, P < .001), impaired renal function (89% vs 37%, P<.001), Killip class IV at presentation (61% vs 18%, P<.001), higher peak creatinine kinase level (mean [SD], 3801.6 [3280.2] vs 2721.3 [2286.6] IU/L, P=.05), and mechanical ventilator support (44% vs 12%, P<.001). Coprescription of proton-pump inhibitors did not reduce the risk of gastrointestinal bleeding (22.2% vs 13.4%, P=.22). Multivariate analysis showed an odds ratio (95% confidence interval) for gastrointestinal bleeding of 22.1 (5.6-86.89, P<.001) for previous gastrointestinal bleeding, 6.74 (1.30-34.89, P=.02) for impaired renal function, and 4.68 (1.35-16.2, P=.01) for Killip class IV at presentation. Gastrointestinal bleeding was associated with longer intensive care unit stay (mean [SD], 5.4 [6.7] vs 3.6 [3.6] days, P=.04), and higher in-hospital (44% vs 9%, P<.001) and overall (44% vs 13%, P<.001) mortality rate.
Conclusions: Although rare, gastrointestinal bleeding in patients with STEMI significantly prolongs intensive care unit stay and increases mortality. Previous gastrointestinal bleeding, impaired renal function, and Killip class IV at presentation are associated with higher incidence of gastrointestinal bleeding.