Long-term aspirin adherence and risk of cardiovascular events and death after myocardial infarction: A nationwide cohort study

Abstract

Abstract Background Aspirin is mandatory in the acute phase after myocardial infarction (MI); however, the efficacy of aspirin for long-term secondary prevention is less established and may have changed with the addition of modern therapies and diagnostics methods. Purpose To investigate the effectiveness of adherence to long term aspirin therapy at different timepoints following MI. Methods Using Danish nationwide registries, we included patients ≥40 years with a first time MI undergoing percutaneous coronary intervention from 2004 through 2017, who were alive, not treated with anticoagulants, and had stayed adherent to aspirin at one year after the index event (Figure 1). We evaluated adherence to aspirin therapy as proportion of days covered (PDC) ≤80% and >80% at four landmarks (2, 4, 6, and 8 years after MI, respectively). At each landmark, patients were excluded from the analysis if they had suffered a recurrent MI or stroke, died, emigrated or were on anticoagulant or P2Y12-inhibitor therapy. The absolute and relative risks of a composite of death, recurrent MI, or ischemic stroke at 2 years from each landmark were calculated through multivariable logistic regression analysis with average treatment effect modelling standardized to the distribution of age, sex, diabetes, hypertension, hypercholesterolemia, chronic kidney disease, cancer, peptic ulcer, former bleeding, and chronic obstructive pulmonary disease (as a proxy for smoking). Non-CVD death was used as a neutral comparator. Results A total of 40,114 individuals were included in the analysis. Suboptimal adherence to aspirin (PDC≤80%) increased from 10% at first landmark 2 years after MI to 19% at fourth landmark 8 years after MI (Figure 1). The standardized absolute risk of death, recurrent MI, or ischemic stroke was highest at the first landmark 2 years after MI for both PDC-groups (PDC>80%: 8.34%, 95% confidence interval [CI]: 8.04-8.63% and PDC≤80%: 10.75%, 95% CI: 9.80-11.69%) (Figure 2). The standardized relative risk of the composite endpoint was significantly higher for patients with PDC≤80% at all landmarks. A trend towards a diminished protective effect of long-term aspirin therapy appeared from 4 years after the index MI and onwards, decreasing from 1.40 (95% CI: 1.26-1.53) at second landmark 4 years after MI to 1.20 (95% CI: 1.04-1.35) at fourth landmark 8 years after MI. Overall, we found no difference in the relative risk of non-CV death between PDC-groups. Conclusion Suboptimal adherence to long-term aspirin therapy following MI was associated with an increased risk of recurrent MI, ischemic stroke, or death, but the relative risk appeared to decrease slightly over time from 4 years after the index MI and onwards.Figure 1Figure 2