Keele Research Repository

Objectives
This study aims to examine in-hospital gastrointestinal (GI) bleeding, its predictors and clinical outcomes, including long-term outcomes, in a national cohort of patients undergoing percutaneous coronary intervention (PCI) in England and Wales.

Background
GI bleeding remains associated with significant morbidity, mortality and socioeconomic burden.

Methods
We examined the temporal changes in in-hospital GI bleeding in a national cohort of patients undergoing PCI between 2007-2014 in England and Wales, its predictors and prognostic consequences. Multivariate analysis was performed to identify independent risk factors between GI bleeding and 30-day mortality. Survival analysis was performed comparing patients with, and without, GI bleeding.

Results
There were 480 in-hospital GI bleeds in 549,298 patients (0.09%). Overall, rates of GI bleeding remained stable over time but a significant decline was observed for patients with ST segment elevation myocardial infarction (STEMI). The strongest predictors of bleeding events were STEMI - odds ratio (OR) 7.28 (95% confidence interval (95% CI) 4.82-11.00), glycoprotein IIb/IIIa inhibitor use OR 3.42 (95% CI 2.76-4.24) and use of circulatory support OR 2.65 (95% CI 1.90-3.71). Anti-platelets/coagulants (clopidogrel, prasugrel and warfarin) were not independently associated with GI bleeding. GI bleeding was independently associated with a significant increase in all-cause 30-day mortality (OR 2.08 (1.52-2.83)). Patients with in-hospital GI bleed who survived to 30-days had increased all-cause mortality risk at 1 year compared to non-bleeders (HR 1.49 (1.07-2.09)).

Conclusions
In-hospital GI bleeding following PCI is rare but is a clinically important event associated with increased 30-day and long-term mortality.