Dabigatran and myocardial ischemia: a pharmacologic perspective

In RE-LY, the landmark trial comparing dabigatran (Pradaxa^®) and warfarin in patients with atrial fibrillation, the investigators observed an increased risk of myocardial infarction (MI) among patients randomized to dabigatran [1].  Although the increased risk was only numerically different in patients randomized to the lower dose of dabigatran (110 mg), the higher dose (150 mg) group met the threshold for statistical significance (relative risk 1.38, CI 1.00 - 1.91, p = 0.048). Since the original publication of RE-LY, the subject of increased ischemic risk associated with dabigatran has been a topic of heated debate.  Two analyses published last month look at this issue in greater depth [2, 3].

In a sub-analysis of the original RE-LY trial [2], Hohnloser, et al found a numerical but not statistically significant difference in the annual risk of MI among patients randomized to dabigatran -- why this is different from the original trial is still a mystery to me, but I'll attribute it to how small the difference was in the original trial. Outcomes were similar among the subgroup of patients with a known history of coronary artery disease (CAD) and a pre-specified analysis of net clinical benefit -- a composite of ischemic, thrombotic, and hemorraghic events -- also favored dabigatran (p = 0.02).

However, in a meta-analysis of seven trials comparing dabigatran to standard therapy in atrial fibrillation, acute coronary syndromes, or venous thromboembolism [3], Uchino, et al observed a small but statistically significant risk of MI associated with dabigatran therapy (absolute risk 0.14% - 0.17%, p = 0.03). Results were similar when the revised data from RE-LY were included and when trials of shorter duration were excluded.

Most have contended that the increased risk attributed to dabigatran is due to the protective effects of warfarin rather than an adverse effect of dabigatran.  After all, warfarin improves outcomes post-MI and is more effective than aspirin for this indication [4].  However, given the difficulty of managing warfarin in such an extensive patient population, dual antiplatelet therapy has become the preferred standard of care in the majority of post-MI patients.

While the authors of these trials do not propose a pharmacologic basis for the potential risk of ischemic events associated with dabigatran therapy, I think a signal observed in an earlier trial may provide a few clues.  In PETRO [5], the first trial to evaluate dabigatran in patients with atrial fibrillation, patients were randomized to one of three doses of dabigatran (alone or in combination with aspirin) or adjusted-dose warfarin.  In the patients randomized to dabigatran, investigators observed an unexpected increase (17-31%) in the urinary excretion of 11-dehydrothromboxane B2 (DTB₂), a byproduct of thromboxane A₂ and a marker of platelet activation.  Thromboxane A₂ is a pro-inflammatory mediator responsible for platelet activation and aggregation and is implicated in the pathogenesis of acute coronary syndromes. The inhibition of thromboxane A₂ production is thought to be the mechanism by which aspirin exerts its beneficial effects in patients with ischemic heart disease.  Interestingly enough, the urinary excretion of DTB₂ in patients taking dabigatran was attenuated (by 40-57%) in those patients that were also on aspirin.

If the DTB₂ excretion observed in patients on dabigatran is responsible for the increased rates of ischemic events attributed to the drug, this could be responsible for the discrepancies observed between the RE-LY sub-analysis and the meta-analysis conducted by Uchino, et al.  Given the association between ischemic disease and atrial fibrillation, approximately 40% of the patients in RE-LY were also taking aspirin, a percentage that may have been sufficient enough to dilute the rates of ischemic disease that may have otherwise been observed.  If this is truly the case, it is not surprising that increased rates were not observed among patients with CAD, as the vast majority of them were likely on aspirin.

Although the incidence of CAD was not reported among patients enrolled in other trials included in the meta-analysis, I anticipate it was low given the heterogeneity of the patient population (with the exception of the single trial of patients with acute coronary syndromes). As a result, fewer of these patients would have been on concomitant aspirin therapy and would not have obtained a potential protective effect of aspirin.  Because more of these non-aspirin patients were included in the meta-analysis, it may have been enough to signal an increased risk of MI with dabigatran use.

Although this is entirely speculative, I do think it is important to at least recognize a plausible pharmacologic explanation for the increased risk of ischemic events attributed to dabigatran if such an effect actually exists. The issue is unlikely to be resolved without a prospective analysis -- a recommendation made by the authors of both publications. While I do not anticipate an investigation to be performed any time soon, I hope one will at least address the aspirin issue at some point in the future, as this could change how we manage patients on dabigatran therapy.  In other words, should we also be placing the majority of them on aspirin?

References

1. Connolly SJ, Ezekowitz MD, Wallentin L, et al; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009 Sep 17;361(12):1139-51.
2. Hohnloser SH, Oldgren J, Connolly SJ, et al. Myocardial Ischemic Events in Patients With Atrial Fibrillation Treated With Dabigatran or Warfarin in the RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy) Trial. Circulation. 2012 Feb 7;125(5):669-76.
3. Uchino K, Hernandez AV. Dabigatran Association With Higher Risk of Acute Coronary Events: Meta-analysis of Noninferiority Randomized Controlled Trials. Arch Intern Med. 2012 Jan 9.
4. Hurlen M, et al. Warfarin, aspirin, or both after myocardial infarction. N Engl J Med. 2002 Sep 26;347(13):969-74.
5. Ezekowitz MD, Reilly PA, Wallentin L, et al. Dabigatran with or without concomitant aspirin compared with warfarin alone in patients with nonvalvular atrial fibrillation (PETRO Study). Am J Cardiol. 2007 Nov 1;100(9):1419-26.