Abstract & Commentary
Antiplatelet Agents Plus Oral Anticoagulants in Atrial Fibrillation
By Michael H. Crawford, MD, Editor
Source: Lamberts M, et al. Antiplatelet therapy for stable coronary artery disease in atrial fibrillation patients taking an oral anticoagulant: A nationwide cohort study. Circulation 2014;129:1577-1585.
In patients with stable coronary artery disease and atrial fibrillation (AF) on oral anticoagulants, adding antiplatelet agents is common and recommended in guidelines, especially during the first year after an acute coronary event or revascularization. However, concern has been expressed about the risk of bleeding. Thus, these investigators from Denmark evaluated nationwide Danish administrative registries to identify all AF patients hospitalized for myocardial infarction (MI) or a percutaneous coronary intervention (PCI) over 10 years between 2001-2011. Patients who were stable (without another event) after 360 days were entered into the analysis of the risk of cardiovascular or serious bleeding events in relation to ongoing antithrombic therapy, which included antiplatelet agents and oral anticoagulants. A total of 8700 patients were followed for a mean of 3 years. The crude rates of events per 100 person years are as follows: MI/coronary death 7.2, thromboembolism 3.8, and serious bleeding 4.0. Compared to oral anticoagulants alone (OAC), the risk of MI/death for OAC plus aspirin was 1.12 (hazard ratio [HR], 1.12; 95% confidence interval [CI], 0.94-1.34) and for OAC plus clopidogrel was 1.53 (HR, 1.53; 95% CI, 0.93-2.52). The risk of thromboembolism was comparable for all OAC conditions, but the risk of bleeding increased for OAC plus aspirin (HR, 1.5; 95% CI, 1.23-1.82) and for OAC plus clopidogrel (HR, 1.84; 95% CI, 1.11-3.06). The authors concluded that in stable CAD patients with AF, adding antiplatelet therapy (APT) to OAC treatment increased the risk of bleeding without any reduction in the risk of coronary events or thromboembolism.
This has been a bad month for aspirin. The FDA has withdrawn approval for its use in primary prevention and now it adds nothing to OAC therapy in patients with stable CAD and AF. In fact, it seems to cause harm in both populations. In some ways, these new data are not surprising because they support previous smaller studies, and the European Society of Cardiology has recommended dual antiplatelet therapy (DAPT) after an acute coronary event or PCI, plus OAC if the patient has AF, until 1 year and then they recommend OAC only if the patient’s CAD is stable. These data support the latter.
Other interesting data were exhibited by this study, but only included in the discussion. They showed that single-agent APT only increased the risk of death and DAPT increased the risk of bleeding, confirming the results of other studies. Also, they confirmed that OAC was superior to single or DAPT for preventing thromboemboli, and the addition of APTs added no benefit. Other studies have shown that the benefits of APT are greatest early after CAD events, which may help explain these data.
The major weakness of this study is that it is observational, so you can’t be certain about cause and effect relationships. The authors argue that the strength of the study is the large number of real-world patients. However, since drug therapy wasn’t randomized, it is possible that selection biases could have influenced the results. For example, the patients kept on triple therapy (OAC + DAPT) may have been sicker than those kept on fewer drugs. In addition, in this type of administrative database study, it is difficult to control for unmeasured confounders such as smoking, body mass index, coronary anatomy, the type of AF (paroxysmal, persistent, permanent), the INR, the type of stents, and over-the-counter medications. Finally, the OACs used were warfarin or phenprocoumon. Whether novel OACs would have performed differently is unknown.
At this time, it seems reasonable to follow the recommendation of the European Society of Cardiology and use OACs only in CAD patients with AF who have been stable for 1 year. What to do prior to the 1-year point is still debatable, so individual judgment about the particular patients’ risks and potential benefits of single, double, or triple therapy needs to be weighed. Also, patient preference may play a role in the decision.