Intended for healthcare professionals

Rapid response to:


Dabigatran, bleeding, and the regulators

BMJ 2014; 349 doi: (Published 23 July 2014) Cite this as: BMJ 2014;349:g4517

Rapid Response:

In some recent articles, the British Medical Journal (BMJ) raised questions about the reliability of the RE-LY study, and whether plasma concentration monitoring should be recommended for dabigatran therapy. We, the leaders of RE-LY-trial, would like to restate the strong evidence for the currently recommended use of dabigatran provided by the trial; but also emphasize that the available data are inadequate for recommending dose adjustments based on monitoring of dabigatran plasma levels.

Efficacy and Safety of Dabigatran

The RE-LY study was a collaboration between three academic institutions and the sponsor, enrolling more than 18,000 patients from more than 900 hospitals/clinics in over 40 countries around the world. Patients were randomized to unblinded warfarin or to one of two blinded doses of dabigatran. The design was extensively discussed with the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in advance, and numerous safeguards were incorporated to minimize potential biases related the open comparison of dabigatran vs warfarin. The results showed that the higher dose of dabigatran (150 mg BID) was superior to warfarin (P less than 0.0001) and the lower dose was non-inferior in reducing stroke or systemic embolism. There was a highly significant reduction in intracranial hemorrhage (the most feared complication of any anti-thrombotic therapy) with both doses of dabigatran compared to warfarin (P less than 0.0001) (1).

The recent BMJ articles argued that there are reasons to doubt the evidence regarding the efficacy and safety of dabigatran. One argument was that unblinded warfarin might bias the results? There is however a rationale for open trials using warfarin as this design better reflects the real world situation allowing optimized warfarin dosing and INR control and more adequate management of bleeding, especially concerning antidotes. In order to avoid bias in the open trial we used central blinded event adjudication and review of all hospitalizations, deaths and adverse event reports for potential missed events. We also checked whether there was any evidence of bias in reporting of bleeding by reviewing hemoglobin changes to detect unreported bleeds. These extensive efforts did not identify any events that would alter the conclusions of RE-LY either on efficacy or safety.

After initial filing of the RE-LY database, the FDA raised questions about data inaccuracies, related mainly to INR reporting and major bleeds. As reported previously a subsequent double entry of all INR values changed the mean study time in therapeutic range from 64.2 to 64.4 (2,3). An exhaustive additional review of data was also undertaken at that time to assure regulatory agencies that there was no bias in the reporting of events. The findings before and after extensive checks indicated little change in the differences in the primary outcome and bleeding between the 3 randomized groups. Recently the fatal events were again reevaluated because of a litigation process in the USA. This second review also turned up a few previously unreported events in all 3 groups, which did not affect the differences in outcome between the treatment groups. The results of the two repeated reviews have been fully published (2, 3). The fact that the results of RE-LY do not change despite multiple and repeated checks of the data, attest to the robustness of the RE-LY findings. Table 1 in the attached document shows the first published results of RE-LY (1) and the results with the additional events included (2,3).

The main findings of the RE-LY trial have recently been supported by an FDA sponsored analysis of clinical outcomes in greater than 134,000 Medicare patients aged 65 years and older, receiving dabigatran or warfarin (4). This analysis shows that the results in clinical practice mirror the results of the RE-LY trial (Table 2 in the attached document).

Plasma Concentration Monitoring

We recently published the results of retrospective measurements of dabigatran plasma concentration from samples obtained during RE-LY. Blood samples at trough and peak of dabigatran plasma concentration were obtained after a few weeks of therapy in more than 9,000 dabigatran patients to enable population pharmacokinetic analysis (5). As with most pharmacologic agents there were substantial variations in plasma concentration in patients receiving dabigatran, which were primarily due to differences between patients in kidney function, age, weight and gender – which together explained much of the variation in plasma concentration of dabigatran. There was a strongly significant association between dabigatran plasma concentration and major bleeding and a weaker relationship between plasma concentration and stroke. A multi-variate analysis indicated that the factors associated with increased major bleeding included dabigatran plasma concentration (p less than 0.001), age (p less than 0.001) and aspirin use (p less than 0.003). This relationship between plasma concentration and bleeding raises the possibility that adjusting dabigatran dose according to plasma concentration could be useful in clinical practice to reduce the risk of bleeding. It is worth noting however that dabigatran dose adjustment for increased age, reduced renal function, or for co-administration of platelet inhibitors is clinically sensible, and stated in labelling in several countries where dabigatran 110 mg is available. Additional dose adjustments based on measurement of dabigatran concentration, might theoretically provide incremental clinical utility. However prospective studies would be needed to properly assess the utility of monitoring of dabigatran plasma concentration. Until these studies are done it would not be appropriate to recommend routine plasma concentration measurement.

Looking Forward

All the new oral anticoagulants, developed as alternatives to warfarin, have used fixed doses (with or without dose reduction in selected cases based on clinical features). In clinical practice both dabigatran doses are widely used with dose reductions in older patients, at lower renal function, lower body weight, and at factors associated with both higher bleeding risk and/or higher dabigatran plasma concentrations. Can we improve the individualization of therapy of the novel oral anticoagulants with monitoring of the concentration in plasma? While data from dabigatran population pharmacokinetic studies raise this possibility, stroke and major bleeding in atrial fibrillation have complex etiology and correlate with multiple clinical factors (e.g. age, renal dysfunction, myocardial dysfunction, coagulation activity) as well as with plasma concentration. Several of these factors interact and also change over time. Before recommending plasma concentration measurements we need to better understand how this information would add to readily available clinical information for dose selection; and what would be the effects on clinical outcomes. We also need to know when, or how often to measure plasma levels; and how exactly to react to the information. The cost benefit and other impacts of such measurements also should be evaluated. At present, we should not lose sight of the advances already made. The currently recommended use of the novel agents, while not perfect, provides substantial advances in both efficacy and safety compared to warfarin. Thus, the pursuit of the perfect must not be an impediment to implementing the advances already made.

Stuart J. Connolly (1), Salim Yusuf (1), Lars Wallentin (2)

1. Population Health Research Institute, Hamilton Health Sciences, McMaster University, Hamilton Ontario, Canada

2. Uppsala Clinical Research Center, Uppsala University, Uppsala, Sweden.

1. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, Pogue J, Reilly PA, Themeles E, Varrone J, Wang S, Alings M, Xavier D, Zhu J, Diaz R, Lewis BS, Darius H, Diener HC, Joyner CD, Wallentin L; the RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009; 361(12):1139-51

2. Connolly SJ, Ezekowitz MD, Yusuf S, Reilly PA, Wallentin L; Randomized Evaluation of Long-Term Anticoagulation Therapy Investigators. Newly identified events in the RE-LY trial. N Engl J Med. 2010 Nov 4;363: 1875-6.

3. Connolly SJ, Yusuf S, Wallentin L; Letter to New England Journal of Medicine (in press) 2014.

4. FDA Drug Safety Communication: FDA study of Medicare patients finds risks lower for stroke and death but higher for gastrointestinal bleeding with Pradaxa (dabigatran) compared to warfarin

5. Reilly PA, Lehr T, Haertter S, Connolly SJ, Yusuf S, Eikelboom JW, Ezekowitz MD, Nehmiz G, Wang S, Wallentin L; RE-LY Investigators. The effect of dabigatran plasma concentrations and patient characteristics on the frequency of ischemic stroke and major bleeding in atrial fibrillation patients: the RE-LY Trial (Randomized Evaluation of Long-Term Anticoagulation Therapy). J Am Coll Cardiol. 2014 Feb 4;63(4):321-8.

Competing interests: Dr Wallentin has received grant support/honoraria/consulting fees from Boehringer Ingelheim, Regado Biosciences, Athera Biosciences, Astra-Zeneca,GlaxoSmithKline, Eli Lilly, Schering-Plough, and Bristol-Myers Squibb. Drs. Connolly and Yusuf have received grant support/honoraria/consulting fees from Boehringer-Ingelheim, Bristol-Myers Squibb, Bayer, Pfizer, Sanofi-Aventis, and Portola.

29 September 2014
Stuart J. Connolly
Salim Yusuf, Lars Wallentin
McMaster University
30 Birge St. Hamilton Ontario, L8L 0A6