Abstract 14337: Relationship Between Long-Term Bleeding Risk and Choice Between Drug-Eluting and Bare Metal Stents: Insights From the PRISM Study
Introduction: In patients at high risk for long-term bleeding after PCI, bare metal stents (BMS) may be preferable to drug-eluting stents (DES) given a significantly shorter requirement for mandatory dual antiplatelet therapy (DAPT). Risk models that allow calculation of patients’ personalized risk for long-term bleeding at the time of percutaneous coronary intervention (PCI) have been described, and could be used to help tailor stent selection for individual patients, but the relationship between patients’ predicted bleeding risk and stent selection is not known.
Hypothesis: There is a significant association between patients’ predicted bleeding risk and stent selection.
Methods: In the 10-center OPS/PRISM study of consecutive PCI patients we calculated each patient’s long-term bleeding risk using a published risk model to predict BARC ≥1 bleeding if they were treated with a DES. Using modified Poisson regression, we then examined the association between predicted bleeding risk and selection of BMS vs. DES at the time of PCI.
Results: Among 3,295 patients treated by 157 interventional cardiologists, 81% were treated with DES. Patients who received DES were more likely to be younger (64± 11 years), male (72%), white (92%), have a history of prior PCI (42%), and diabetes (34%) as compared to those who received BMS. The average bleeding risk (as estimated by the BARC Bleeding Risk Score) was similar for patients who received DES (58 ± 11) to those who got a BMS (58 ± 13). In the adjusted analysis, there was no statistically significant association between predicted long-term bleeding risk and stent selection (Relative Risk = 1.00; C.I 0.98-1.02 per 10 point increase in the bleeding risk score, p-value = 0.61).
Conclusion: Although physicians could balance the benefits of DES against the risks of post-PCI bleeding, we observed no relationship between patients’ predicted risk for BARC ≥ 1 bleeding over the year after treatment and stent selection. Use of novel approaches to provide personalized bleeding risk predictions at the time of PCI may help clinicians tailor stent selection according to individual patients’ preferences and risk/benefit profiles.
Author Disclosures: A. Shafiq: Research Grant; Modest; T-32 grant (Award Number T32HL110837). K. Gosch: None. J.A. Spertus: Research Grant; Modest; NHLBI, AHA, ACCF, Lilly, EvaHeart, Amorcyte. Honoraria; Modest; United Healthcare, Genentech, Amgen. Ownership Interest; Significant; Health Outcomes Sciences, the commercial entity distributing and supporting ePRISM,. A.C. Salisbury: Research Grant; Modest; Boston Scientific. Honoraria; Modest; Abiomed, Medtronic.
- © 2016 by American Heart Association, Inc.