Association between platelet reactivity and long-term bleeding complications following percutaneous coronary intervention in patients with and without diabetes

Background The association between diabetes mellitus (DM) and bleeding complications following percutaneous coronary intervention (PCI) is controversial. We hypothesized that on-treatment platelet reactivity may have a role in the bleeding risk stratification of such patients. Purpose To investigate the role of low platelet reactivity (LPR) in the long-term bleeding risk stratification among patients with and without diabetes undergoing PCI. Methods In this observational, retrospective single-center study, 472 patients undergoing PCI for stable coronary artery disease were included. All patients were treated with dual antiplatelet therapy with aspirin and clopidogrel. Platelet reactivity was assessed using the VerifyNow P2Y(12) assay and LPR was defined by values of platelet reactivity unit (PRU) ≤178. Primary endpoint was the occurrence of all bleeding events at 5 years stratified by DM status and LPR. Results Out of the study population included, 30.5% had DM (N=144). LPR was numerically less frequent in patients with DM compared to those without (29.2% vs 37.6%; p=0.077). Overall, 11.9% of patients experienced a bleeding complication at 5-year follow-up; 44.6% of events were classified as major bleedings. The incidence of bleeding events per 1000 patients-year was 34.5 (95% CI 22.3–53.5) in DM and 24.2 (95% CI 17.3–33.8) in no DM (p=0.24). A stepwise increase in the crude rates of bleeding complications was observed across patients with and without DM and LPR (log-rank p=0.004), with those having both conditions being at the highest risk of events (Figure). LPR had a similar value for stratifying the risk of bleeding in patients with and without diabetes (p value for interaction between diabetes and LPR status=0.45). Conclusions Approximately 1 out of 3 patients undergoing PCI for stable coronary artery disease on clopidogrel has LPR. The assessment of LPR provides a significant incremental value for the prediction of bleeding events irrespective from DM status. While the presence of DM per se does not increase the risk of bleeding complications, the coexistence of DM and LPR identifies the subgroup at the highest risk of events who could benefit from a short-term and less intensive antiplatelet regimen.