Article Text
Abstract
Background Major bleeding is a recognised complication after PCI. This study aims to evaluate the incidence, trends over time, predictors and prognostic significance of intracranial, gastrointestinal and retroperitoneal bleeding after PCI.
Methods We analysed patients who underwent PCI in England and Wales between 2007 and 2014 in the British Cardiovascular Intervention Society Database. In-hospital intracranial, gastrointestinal and retroperitoneal bleeding events were recorded in addition to patient demographics, procedural characteristics and the outcomes in-hospital major adverse cardiovascular events and 30 day mortality. Overall and rates of events over the years of study were examined and multivariable logistic regression were used to identify independent predictors of bleeding and adverse outcomes associated with bleeding.
Results A total of 107 (0.03%, n=426,046) intracranial bleeds, 480 (0.09%, n=549,298) gastrointestinal bleeds and 291 (0.06%, n=511,106) retroperitoneal bleeds were captured in the dataset. We observed in decline over all in intracranial and retroperitoneal bleeds over time and a decline in gastrointestinal bleeding among the STEMI subgroup. Age, diagnosis of NSTEMI, diagnosis of STEMI, valvular heart disease, warfarin use, glycoprotein IIb/IIIa inhibitor use and thrombolysis were predictors of bleeding events. The adjusted odds of 30 day mortality and in-hospital MACE were OR 10.14 95% CI 5.34–19.25 and OR 9.49 95% CI 4.93–18.27, respectively for patients who developed intracranial bleeds, OR 2.13 95% CI 1.59–2.85 and OR 3.73 95% CI 2.81–4.96, respectively for patients who developed gastrointestinal bleeding and OR 3.59 95% CI 2.19–5.90 and OR 5.76 95% CI 3.71–8.95, respectively for patients who developed retroperitoneal bleeding.
Conclusions Major bleeding is rare occurring in less than 1% of patients who undergo PCI but when it occurs it is associated with adverse outcomes. Observed declines in bleeding may be related to decline in use of antithrombotic medications.