Association of ischemic and bleeding events with mortality in patients with a recent acute myocardial infarction

Abstract

Abstract Background/Introduction Duration and intensity of antithrombotic treatment after myocardial infarction should be individualized based on a patient's ischemic and bleeding risk [1,2]. While such strategies are typically based on calculations that give equal weight to both types of events, uncertainty remains regarding their relative importance. Purpose To describe the incidence of ischemic and bleeding events in patients with a recent myocardial infarction, to compare the association of an ischemic vs bleeding event with mortality and to assess whether this association had changed over the past two decades. Methods Patients with acute myocardial infarction enrolled in the SWEDEHEART registry and discharged alive with antithrombotic treatment (aspirin, P2Y12 inhibitor, or oral anticoagulant) from January 2012 to December 2017 were followed from discharge until an ischemic event (recurrent myocardial infarction or ischemic stroke) or bleeding event. Cox regression adjusted for demographic factors, comedications and comorbidities, was used to estimate hazard ratios (HR) for time to death after an ischemic and bleeding event as compared with no event (in a model using time-varying exposure definition) and for an ischemic vs bleeding event in a direct comparison. We then assessed whether the adjusted HR for mortality of an ischemic vs bleeding event had changed across three time-periods (1997–2000, 2001–2011 and 2012–2017) by using an interaction term between time period and type of event. Results From January 2012 until December 2017 86, 736 patients were discharged alive with antithrombotic treatment after a myocardial infarction. Of these, 4,039 patients experienced a first ischemic event (incidence rate 5.7 events per 100 person-years), and 3,399 a first bleeding event (incidence rate 4.8 events per 100 person-years). As compared with no event, both ischemic events (adjusted HR 4.16, 95% CI 3.91 to 4.43) and bleeding events (adjusted HR 3.43, 95% CI 3.17 to 3.71) were associated with an increased risk of death. In the direct comparison, ischemic events were associated with a higher risk of death than bleeding events (adjusted HR 1.27, 95% CI 1.15 to 1.40). There was no evidence of a change in the aHR across the three time periods (aHR; 1.17, 95% CI 1.02 to 1.35 in 1997–2000 and 1.18, 95% CI, 1.11 to 1.27 in 2001–2011, p for interaction between time period and type of event ≥0.646). Conclusion In this nationwide study of patients with a recent MI, post-discharge ischemic events were more common and associated with higher mortality risk as compared with bleeding events. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Swedish Heart and Lung FoundationSwedish Diabetes Foundation