Risk of bleeding in patients with and without cancer on antiplatelet therapy following myocardial infarction: a study in primary care

Abstract

Abstract Background Antithrombotic medication and antiplatelet therapy (ATPL) are used in the treatment and prevention of thrombotic disorders, especially in patients following a myocardial infarction (MI). However, the risk reduction of future ischaemic events must be balanced against the potential bleeding risk. Co-morbidities may predispose individuals to bleeding events. Cancer influences risk of bleeding either directly through bleeding of fragile cancer-associated blood vessels or by indirectly increasing the risk of bleeding remote from the cancer site. In patients with cancer, bleeding risk may be increased due to ATPL given for concomitant cardiovascular disease. Purpose To investigate the association between pre-existing cancer and risk of bleeding in patients on ATPL after myocardial infarction in a primary care population. Methods We performed a matched cohort analysis using data from the Clinical Practice Research Datalink (CPRD) and linked Hospital Episodes Statistics (HES) in England. People with an MI between 2006–2017 were matched to patients without an MI at a 1:5 ratio. Four groups: patients with both MI and cancer (breast, digestive tract, lung, prostate at baseline [within 1-year prior to MI]), patients with only cancer or MI; and patients without either condition. Using flexible parametric methods, hazard ratios were calculated to determine the association between patient group and risk of bleeding, as well as between APTL-regimen after MI and bleeding risk during 1-year of follow-up after MI. All data were processed using STATA/SE16, using a significance level of ≤0.05. Results We included 58,951 patients with an average (mean) age of 74±10 years (66% male). Of all patients, 12,526 (21%) patients had cancer, and 9,942 (17%) had an MI; numbers per subgroup are presented in Figure 1. A bleeding event was observed in 1,193 (2%) patients. The vast majority (N=56,148 [95%]) of patients did not take ATPL. Of those who did receive ATPL, aspirin was the most common (N=2,359 [4%]). In a multivariable model (adjusted for age, gender, ethnicity, prior bleeding, diabetes, renal disease, smoking status, drug, hypertension, hypercholesteraemia and cancer), people with MI had a risk (hazard) of bleeding of 2.52 (95% CI 2.23-2.86), p<0.001. For people with cancer only, the risk of bleeding was 1.26 (95% CI 1.10-1.44), p<0.001. Conclusion(s) Bleeding risks are higher in people with cancer after an MI than those with MI or cancer alone. More research is needed on individual anti-platelet therapies for people with cancer who have a MI to minimize the risk of bleeding and future cardiovascular events.Figure 1