FDA Review Finds No Increased Risk of MI With Dabigatran (Pradaxa)

Shelley Wood


May 13, 2014

SILVER SPRING, MD ( updated , corrected ) — A massive study of Medicare patients at risk of stroke secondary to atrial fibrillation (AF) has found no signal of increased MI risk with dabigatran (Pradaxa, Boehringer Ingelheim) compared with warfarin[1]. The study, undertaken by the US Food and Drug Administration as part of its ongoing review of dabigatran safety, included more than 134 000 Medicare patients 65 years or older.

The findings, however, are not enough to convince some that the MI signal from the pivotal trials of dabigatran can be easily dismissed, with one expert saying he thinks the FDA is "turning its back to the fire."

The FDA announced the new results today. In the analysis, new dabigatran use was associated with a reduced risk of ischemic stroke, intracranial bleeding, and death compared with warfarin, although the risk of major gastrointestinal bleeding was higher with the newer oral anticoagulant. Risk of MI, however, a longtime concern linked with dabigatran use, was similar for the two drugs.

Incidence Rate per 1000 Person-Years for Dabigatran 75 mg or 150 mg* vs Warfarin for Nonvalvular AF Based on 2010–2012 Medicare Data

Event Dabigatran Warfarin (reference) Adjusted hazard ratio (95% CI)
Ischemic stroke 11.3 13.9 0.80 (0.67-0.96)
Intracranial hemorrhage 3.3 9.6 0.34 (0.26-0.46)
Major GI bleeding 34.2 26.5 1.28 (1.14-1.44)
AMI 15.7 16.9 0.92 (0.78-1.08)
Death 32.6 37.8 0.86 (0.77-0.96)
*Primary findings for dabigatran are based on analysis of both 75 and 150 mg together without stratification by dose

The FDA notes that its press release today is a follow-up to FDA safety announcements about dabigatran dating back to November 2012 and December 2011.

It also points out that the new findings from Medicare patients represent a much larger and older sample than the patient population included in earlier FDA reviews "and employed a more sophisticated analytical method to capture and analyze the events of concern," the FDA said today. Of note, the findings from this latest study mirror those of the clinical-trial results that formed the basis of dabigatran's approval, with the exception of the neutral MI findings. In RE-LY , as previously reported by heartwire , MIs were slightly increased in patients randomized to receive the newer drug than in those on warfarin, although this difference reached statistical significance only at the higher (150-mg) dose.

"As a result of our latest findings, we still consider Pradaxa to have a favorable benefit-to-risk profile and have made no changes to the current label or recommendations for use," the FDA press release notes. "Patients should not stop taking Pradaxa (or warfarin) without first talking with their healthcare professionals. Stopping the use of blood-thinning medications such as Pradaxa and warfarin can increase the risk of stroke and lead to permanent disability and death. Healthcare professionals who prescribe Pradaxa should continue to follow the dosing recommendations in the drug label."

Commenting on the FDA analysis for heartwire , Dr Ramin Artang (University of Nebraska Medical Center, Omaha), who has studied the association between dabigatran and MI, points out that what the FDA has provided online is just a summary that omits important methodological information and details and was not peer-reviewed. "The table would never pass a peer review, as there are so many details that are missing," he said. "What continues to be the weakness of registry data in general, regardless of its size, is they do not count for the selection bias among the physicians who are sitting with the patients and deciding whether this or that patient should go on dabigatran, warfarin, or other novel anticoagulants based in the risk profile of the patient. It should also be noted that some fraction of physicians, if not all of them, were exposed to the findings from the original RE-LY study in 2009, where the increased incidence of MIs was first mentioned."

Since then there have been other studies raising concerns about MI with this drug. "It is therefore possible that some physicians choose not to give dabigatran to their patients with known coronary artery disease or at high risk of coronary artery disease. So the patients who did received dabigatran were lower-risk patients. Hypothetically, this could bias the study in favor of dabigatran. . . . Therefore, in my opinion, the FDA study, although reassuring, does not completely rule out higher risk of coronary events with MI, as the patients were not matched and randomized for each arm of the study."

Likewise, Dr Ilke Sipahi (Acibadem University Medical School, Istanbul, Turkey), whose 2013 meta-analysis identified a 48% increased risk of MI with dabigatran, says he's not reassured by the FDA study.

"The gold standard for clinical evidence is randomized trials, and every researcher would agree with that," he told heartwire . Citing his study, which included the randomized RE-LY and RE-MEDY trials, Sipahi estimates that there is a less than 1% probability that dabigatran does not increase the risk of MI.

Like Artang, Sipahi points out: "The FDA here is reporting again an observational study," and observational studies "are inherently open to multiple bias" that can't be controlled for statistically.

"No matter how big the database is, you will still have confounders in observational data sets. Unfortunately, the FDA is omitting gold-standard evidence from the randomized trials submitted to the agency and is providing second-rate data to clear a drug. The FDA sees a huge fire in front of itself, and instead of warning people about the fire, they are turning their backs against the fire and then tell the public, 'don't worry, you don’t see any fire when you look the other way.' "

Of note, Sipahi pointed out, the FDA has previously done an observational study focused on gastrointestinal bleeding, concluding that GI bleeds are almost 50% less with dabigatran. "Now with this new Medicare analysis they do a 180-degree turn and accept the excess gastrointestinal bleeding risk with dabigatran. This shows how unreliable the observational studies that the FDA conducts about dabigatran are."

An earlier version of this story neglected to specify that the signal of increased MI with dabigatran in RE-LY was statistically significant only at the higher of the two doses studied.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.