Issue 42 / 10 November 2014

EMERGING post-approval evidence on the bleeding risks associated with dabigatran highlights the need for an individualised approach to prescribing oral anticoagulants, experts say.

Two large US observational studies published in Circulation and JAMA Internal Medicine both link dabigatran with a substantially lower risk of intracranial haemorrhage compared with warfarin (Circulation: hazard ratio [HR], 0.34; JAMA Internal Medicine: HR, 0.32) in patients being treated for non-valvular atrial fibrillation, but a higher risk of gastrointestinal (GI) bleeding (Circulation: HR, 1.28 for major GI bleeds; JAMA Internal Medicine: HR, 1.85 for all GI bleeds). (1), (2)
The JAMA Internal Medicine study also found an overall higher incidence of major bleeding regardless of anatomical site with dabigatran compared with warfarin (HR, 1.58).

The authors wrote that they found opposite results compared with a US Food and Drug Administration (FDA) investigation published last year, which did not adjust for differences in patient characteristics between the treatment groups. (3)

The JAMA Internal Medicine study followed 1302 dabigatran and 8102 warfarin users for a maximum of 14 months, and found the bleeding risk was highest among African Americans and patients with renal impairment.

The researchers said their study “has several important clinical and policy implications”, concluding that dabigatran should be “prescribed with caution, especially among high-risk patients”.

An accompanying editor’s note said the study raised “cause for concern, because it appears that the bleeding risk for dabigatran is higher than for warfarin and significantly greater than originally appeared at the time of the FDA approval”. (4)

However, the authors of the much larger study in Circulation, which included 67 494 patients taking dabigatran and 273 920 taking warfarin, concluded that their results were “similar in direction and magnitude” to those observed in the randomised RE-LY trial.

In addition to the findings on intracranial haemorrhage and gastrointestinal bleeding, the Circulation study found dabigatran was associated with a reduced the risk of ischaemic stroke (HR, 0.80) and death (HR, 0.86) compared with warfarin.

The associations were most pronounced for patients treated with dabigatran 150 mg twice daily, whereas the 75 mg twice daily dose was associated only with a reduced risk of intracranial haemorrhage.

The study also found the increased risk of major gastrointestinal bleeding with dabigatran appeared to be restricted to women aged 75 years and older and men aged 85 years and older. More than half the study population was aged over 75 years.

Professor Ross Baker, director of the WA Centre for Thrombosis and Haemostasis at Murdoch University, said the two studies suggested a need to individualise treatment with new oral anticoagulants based on the patient’s clinical assessment and perhaps measuring blood levels.

“When dabigatran and the other new oral anticoagulants first came out there was great excitement because their wide therapeutic margins seemed to mean patients did not require close monitoring of blood levels”, Professor Baker told MJA InSight.

“However, these two large postmarketing study results, taken together with other recent reports, show that the bleeding and thrombotic risk could relate to large individual variations in drug levels in otherwise suitable patients. To individualise treatment it appears measuring blood levels may be necessary”, he said.

Professor Baker said if a patient had a history of gastrointestinal bleeding, warfarin was probably the preferred drug. Warfarin should also be used in patients with significant renal impairment, as it was not renally excreted.

He told MJA InSight there was a crucial need for Australian postmarketing data, which would hopefully come through the ongoing Anticoagulant Reversal and Events Study. (5)

“American data doesn’t necessarily translate to Australia because warfarin control and management of haemorrhage may not be at the same standard there as it is here, which could bias the results”, he said.

Professor Graeme Hankey, Winthrop Professor of Neurology at the University of WA, said the results of the latest two studies were consistent with the findings of the RE-LY trial and also with the known pharmacology of dabigatran. “It is 80% excreted in the urine, so we have to be careful prescribing it to people with renal impairment, or who are at risk of declining renal function”, Professor Hankey told MJA InSight.

“The results are also consistent with the known interaction of bleeding risk with age for dabigatran, which is higher in the elderly, and with GI bleeding,” he said.


1. Circulation 2014; Online 30 October
2. JAMA Intern Med 2014; Online 3 November
3. N Engl J Med 2013; 368 1272-1273
4. JAMA Intern Med 2014; Online 3 November
5. APTIN: ARERS Collaborative Synopsis

(Photo: Alexander Raths / Shutterstock)

3 thoughts on “Dabigatran caution urged

  1. David Ramage says:

    Yet another reason for delaying widespread implementation of new drugs in my clinical practice.

    Sit back and wait….

  2. Timothy Kleinig says:

    People will do anything for a headline, and everyone has their biases. Why isn’t the headline ‘14% reduction in the risk or death with dabigatran confimed in massive real-world Medicare study’? Personally, I’d prefer the reduction in death and cop the slightly higher risk of GI bleeding any day.  

  3. Fraser Brown says:

    As a Radiologist we are finding a huge number of patients on novel agents who have no idea why they are on them. 


    We get them presented for deep interventions without the courtesy of the referral stating:


    1: they are on novel agents

    2: the reason for the agent


    This should be mandatory in ANY referral to radiology or surgery. These are referrals not requests!!



    Dr F Brown

Leave a Reply

Your email address will not be published.