Do NOACs Carry a Higher Risk of Acute MI? New Study Reopens Old Questions

Patients who start taking either rivaroxaban or dabigatran instead of a vitamin K antagonist (VKA) after getting diagnosed with A-fib may have a greater risk of acute MI, according to a retrospective study that adds to an already conflicting evidence base.

Current use of either non-VKA oral anticoagulant (NOAC) by patients treated in a primary practice setting was associated with about twice the risk of acute MI than that seen among patients taking a VKA (adjusted HR 2.11; 95% CI 1.08-4.12), Leo Stolk, PharmD, PhD (Maastricht University Medical Center, the Netherlands), and colleagues report. Their results, derived from a large database, were published online March 23, 2017, ahead of print in the British Journal of Clinical Pharmacology.

VKAs probably have a greater effect on reducing acute MI compared with NOACs, the authors say, adding that more research is warranted because of increasing use of the newer agents.

Commenting for TCTMD, Geoffrey Barnes, MD (University of Michigan, Ann Arbor), said the medical community should not rush to judgement based on one study. Even though there are conflicting results from various analyses, he observed, the totality of the data suggests that MI risk is not higher with the NOACs.

A conclusive answer to the question is difficult to nail down, he said, because no studies have been specifically designed and powered to evaluate MI as an outcome and the analyses that have been completed are subject to inherent biases. He said risk adjustment in observational studies cannot fully account for differences between patients who are given a NOAC instead of warfarin, for example.

So studies like this are “very hypothesis-generating and very thought-provoking, but they’re not definitive in their answers,” Barnes said.

Mixed Messages

There has been a question about whether NOACs increase MI risk ever since the RE-LY trial showed that the rate of MI was higher in patients taking the direct thrombin inhibitor dabigatran (Pradaxa; Boehringer Ingelheim)—the first such agent approved by the US Food and Drug Administration (FDA)—than in those treated with warfarin.

But subsequent randomized and observational studies and meta-analyses of dabigatran and the other approved NOACs—rivaroxaban (Xarelto; Janssen Pharmaceuticals), apixaban (Eliquis; Bristol-Myers Squibb), and edoxaban (Savaysa; Daiichi Sankyo)—have provided conflicting results. Some meta-analyses, including  one from 2012, have found no greater risk of MI in patients taking NOACs versus warfarin, whereas others, including one from 2012 and one from 2013, have shown that patients treated with dabigatran or another direct thrombin inhibitor had an elevated risk. In contrast, a registry study from 2013 did not find a difference in MI risk between dabigatran-treated and warfarin-treated patients.

The FDA weighed in on the issue in 2014 as part of a review of dabigatran, announcing that a large study of Medicare beneficiaries failed to show an excess risk of MI with the drug.

For the current study, Stolk and colleagues examined the Clinical Practice Research Datalink, which contains records from 674 primary care practices in the United Kingdom. The analysis included 30,146 patients who were diagnosed with A-fib and starting taking a VKA (43.4%), low-dose aspirin (51.1%), rivaroxaban or dabigatran (4.2%), or a combination of those drugs (1.3%) between 2008 and 2014. Among the NOAC users, 71.6% were given rivaroxaban and the rest dabigatran.

In addition to current users of NOACs, those taking aspirin had a higher risk of acute MI compared with VKA-treated patients after accounting for potential confounders (adjusted HR 1.91; 95% CI 1.45-2.51). Past use of aspirin also was associated with an elevated risk (adjusted HR 1.69; 95% CI 1.29-2.22).

There were not enough events to evaluate rivaroxaban and dabigatran separately, the authors note.

Barnes pointed out that sample size is a limitation of the study. Although there were more than 30,000 patients included, only about 1,200 were taking rivaroxaban or dabigatran. And there were only 10 acute MIs among the NOAC-treated group. “That’s a pretty small number to be drawing these comparisons on,” he said.

The Debate Continues

Even though there is no definitive proof that NOACs carry a higher risk of MI versus warfarin, Barnes explained why it is possible. Warfarin acts on several clotting factors involved in the coagulation cascade, whereas the NOACs act on one. Thus, some researchers have hypothesized that those additional factors affected by warfarin may be important in determining risk of ischemic events like MI, Barnes explained, adding, however, that that hasn’t been shown definitively.

The issue is worthy of further study, but MI risk is not the primary concern among clinicians prescribing oral anticoagulants for patients with A-fib, he said. More important issues to consider, he said, are renal function, bleeding risks, adherence, and cost.

“There is a chance that these newer drugs may have a slightly increased risk of heart attack compared to warfarin, but that’s not been definitively proven,” Barnes said. “We do know that they are as effective and potentially more effective [than warfarin] at preventing stroke in atrial fibrillation and cause similar or less bleeding.” The NOACs also reduce the risk of intracranial hemorrhage and are easier to take compared with warfarin, he said.

“So when I look at all of the pros and cons of the new drugs versus warfarin, many of my patients will elect to choose the new drugs,” he said. “They can be aware that there’s a little question about the MI risk, but there’s certainly nothing definitive, nothing that I would personally hang my hat on.”