Background:
Platelet inhibitor (PI) use following coronary artery stent confers bleed risks, especially for atrial fibrillation (AF) patients on oral anticoagulants (OAC), a subgroup often excluded from randomized stent trials. We studied a national sample of elderly stent recipients to quantify bleed risks associated with combined OAC and PI, pharmacotherapy common in AF.
Methods:
Using a 40% Medicare random sample denominator file and associated inpatient, outpatient (2003-2012) and prescription (2006-2012) claims, we studied patients age 65 and older receiving a coronary artery stent. We measured OAC and/or prescription PI use after stent placement. Cox regression models estimated risk of bleed events adjusting for individual characteristics, morbidities, and time-varying medication use.
Results
Overall, 165,718 patients in our sample received a stent, 2006-2012; 20.4% had AF; mean age was 73.8 (SD 6.3). In the first month post-stent, among AF patients, 40.7% received OAC, 85.6% prescription PI, 35.9% both; among non-AF patients corresponding use was 3.5%, 90.3% and 3.1% respectively. At 13 months post stent, 61.0% of AF patients and 71.1% of non-AF patients continued using prescription PIs while 15.5% and 1.7% respectively remained on both PI and OAC therapy. Compared to those taking only prescription PIs, the risk of bleeding associated with warfarin plus PI was 2.05 (95% CI: 1.97 - 2.13); the risk associated with novel OACs (dabigatran or rivaroxaban) plus PIs was 2.60 (95% CI: 2.26 - 2.99).
Conclusions:
In a national, older population combined OAC and PI use following stent is common in AF patients and associated with bleeding risk. These findings should inform stent selection and antithrombotic strategies. The association between novel OACs and bleeding was unexpected and warrants further exploration as these products gain use.
What is the optimal antithrombotic therapy of atrial fibrillation patients after coronary artery stent deployment?