Journal reference: Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009; 361:1139-1151 ^[1]

Link: http://dx.doi.org/10.1056/NEJMoa0905561

Published: 17 September 2009

Evidence cookie says...

Dabigatran (Pradaxa) is non-inferior to warfarin for the prevention of strokes or systemic embolism, in non-valvular atrial fibrillation.  Its effectiveness and safety has not been demonstrated in patients with severe renal impairment or active liver disease (excluded). The benefits (↓ major bleeding and ↓ stroke) from dabigatran should be balanced against the harms (↑ myocardial infarction and ↑ gastrointestinal bleeding). The study design may have biased results favourably towards dabigatran. Note: dabigatran is not currently funded by the PBS for this indication.

Article details:

Study design:

multinational open label randomised controlled trial

Study aim:

To evaluate whether dabigatran (Pradaxa) is non-inferior to warfarin for the prevention of stroke or systemic embolism in patients with atrial fibrillation.

Study conclusion:

In patients with atrial fibrillation, dabigatran given at a dose of 110 mg was associated with rates of stroke and systemic embolism that were similar to those associated with warfarin, as well as lower rates of major hemorrhage. Dabigatran administered at a dose of 150 mg, as compared with warfarin, was associated with lower rates of stroke and systemic embolism but similar rates of major hemorrhage. ^[1]

Critical appraisal:

Methodology (PICO):

Note: the appraisal of this study was from the article ^[1], and also from a previous article describing the design fo the RE-LY trial ^[2].

Participants: who was studied?

• 18,113 patients
• enrolled between December 2005 and December 2007
• 951 clinical centres in 44 countries
• inclusion criteria:
  • ECG documented atrial fibrillation (AF), and
  • risk factor for stroke, and
  • age > 18 years at entry, and
  • written, informed consent
• exclusion criteria:
  • heart valve disorders
  • severe disabling stroke within previous 6 months, or any stroke within the previous 14 days
  • conditions associated with increased risk of bleeding
  • contraindication to warfarin treatment
  • reversible cause of atrial fibrillation
  • severe renal impairment (estimated creatinine clearance ≤ 30 mL/min)
  • active liver disease (persistent ALT, AST, alkaline phosphatase > 2 x ULN, or active viral hepatitis)
  • anaemia (Hb < 100g/L) or thrombocytopaenia (platelet < 100 x 10⁹/L)
  • patients considered unreliable by the investigator or have a life expectancy less than the expected duration of the trial because of concomitant disease
• participant characteristics:
  • mean age ~ 71 years, SD 8.7 years
  • mean weight ~ 83 kg
  • mean BP ~ 131/77
  • gender: 63.6% male
  • ~ one third of the participants had persistent AF, paroxysmal AF and permanent AF
  • ~ 20% with a previous stroke or TIA
  • ~ half were on long-term warfarin at baseline (prior to randomisation)
  • median duration of follow up was 2.0 years

Intervention: what was the exposure?

dabigatran 110 mg group:

• dabigatran capsules 110 mg twice daily (110 and 150 mg capsule groups were blinded to the dose)
• follow up visits at 14 days after randomisation, then 1 month, 3 months, and 3 monthly for the first year, and then every 4 months until study end

dabigatran 150 mg group:

• dabigatran capsules 150 mg twice daily (110 and 150 mg capsule groups were blinded to the dose)
• follow up schedule as above

Comparator: what was the control/alternative?

• warfarin (administered in an unblinded fashion)
• dose adjusted locally to an INR of 2.0 to 3.0
• INR measured at least monthly
• follow up visits otherwise as above
• the mean percentage of the study period during which the INR was within the therapeutic range was 64%

Outcomes: what was measured?

primary outcomes:

• study outcome: stroke or systemic embolism
• safety outcome: major haemorrhage

secondary outcomes:

• stroke
• systemic embolism
• death

Are the trial results valid?

Internal validity: Wikipedia

Was the assignment of patients to treatment randomised?

Yes.

• the patients were randomised by a central randomisation service

Were the groups similar at the start of the trial?

Yes.

• the patient characteristics were very similar between all three participant groups

Aside from the allocated treatment, were groups treated equally?

No, and unclear.

• the group randomised to warfarin had their INR locally monitored and doses adjusted; the dabigatran groups received no similar treatment
• it is unclear whether the open labelled nature in the treatment assignment between dabigatran and warfarin would have affected the follow up visits

Were all patients who entered the trial accounted for?

Yes.

• only 20 patients were lost to follow up (99.9% completed)
• analyses were performed by intention to treat principles

Were measures objective or were the patients and clinicians kept blind to which treatment was received?

Unclear.

• neither the patients nor clinicians were blinded to the allocation of dabigatran and warfarin
• given that this was a trial for a potential successor to warfarin, it is possible that there may have been an observer-expectancy effect (the clinician’s cognitive biases causes them to unconsciously influence the participants) and/or the Rosenthal effect (the greater expectation placed on the dabigatran group led to them performing better)
• it should be noted that the mean duration of the study period where the INR was not in the target range was over a third

What were the results?

Primary outcome

stroke or systemic embolism:

• dabigatran 110 mg group vs warfarin:
  • non-inferior to warfarin (P < 0.001)
  • relative risk: 0.91 (95% CI 0.74 to 1.11), P = 0.34
  • interpretation: no statistically significant difference was found between dabigatran 110 mg bd and warfarin for prevention of stroke and systemic emboli
• dabigatran 150 mg group vs warfarin:
  • non-inferior to warfarin
  • relative risk: 0.66 (95% CI 0.53 to 0.82), P < 0.001
  • interpretation: dabigatran 150 mg bd was associated with a 34% lower risk of stroke or systemic embolism (likely real value between 18% to 47%)

major bleeding:

• dabigatran 110 mg vs warfarin:
  • relative risk: 0.80 (95% CI 0.69 to 0.93), P = 0.003
  • interpretation: dabigatran 110 mg bd was associated with 20% fewer major bleeds compared to the warfarin group but the likely real value is in a relatively wide range (7% to 31%)
• dabigatran 150 mg vs warfarin:
  • relative risk: 0.93 (95% CI 0.81 to 1.07), P = 0.31
  • interpretation: there was no statistical significance in the rates of major bleeding between dabigatran 150 mg bd and warfarin

Other outcomes:

dabigatran as compared to warfarin:

• lower relative risk of haemorrhagic stroke
  • dabigatran 110 mg group: RR = 0.31 (0.17 to 0.56), P < 0.001
  • dabigatran 150 mg group: RR = 0.26 (0.14 to 0.49), P < 0.001
• higher relative risk of myocardial infarction
  • dabigatran 110 mg group: RR = 1.35 (0.98 to 1.87), P < 0.07
  • dabigatran 150 mg group: RR = 1.38 (1.00 to 1.91), P < 0.048
• no statistical significant difference in all cause mortality
• higher relative risk of gastrointestinal bleeding
  • dabigatran 110 mg group: RR = 1.10 (0.86 to 1.41), P < 0.43
  • dabigatran 150 mg group: RR = 1.50 (1.19 to 1.89), P < 0.001

Will the results help me care for my patient?

External validity: Wikipedia

Are the participants different to my patient?

• the participants in this study were older and had risks for stroke
• the exclusion criteria should be noted: there were no patients with valvular disease, severe renal impairment, persistent LFT dysfunction, or considered “unreliable” by the investigators

Is the treatment feasible?

• dabigatran is simpler to take and manage than warfarin in the community
• however, dabigatran is not funded (at the time of this article) by the PBS for the prevention of systemic embolism in patients with atrial fibrillation

Were all the clinically important outcomes considered?

Yes.

• the primary outcomes are those of most concern
• it should be noted that economic analyses were not a feature of this study

Are the treatment benefits worth the potential harms/costs?

Unclear.

• dabigatran 110 mg bd was associated with fewer bleeding complications than warfarin and non-inferior for strokes and systemic embolism
• dabigatran 150 mg bd was associated with superiority for strokes and systemic embolism and had a similar bleeding complication rate
• however, dabigatran was associated with a higher risk of myocardial and gastrointestinal bleeding and no benefit to all cause mortality

Study weaknesses (summary)

• a major weakness is that the study was open label, with regards to dabigatran and warfarin allocation
  • it is conceivable biases may have occurred that resulted in the dabigatran groups receiving better care
    • on average, INR was not within the target range over a third of the time; this was in a group where half were experienced in taking long term warfarin, and where “unreliable” participants were excluded
    • the rate of major bleeding with warfarin was higher than in other studies
  • biases in the design of the study are likely to benefit the dabigatran groups as compared to warfarin
  • note: it should be recognised that a double-blinded study would have been very difficult to perform as it would involve sham INR management
• little comment was made of the substantially higher discontinuation rate of dabigatran compared to warfarin (relative risk: 1.28)

Study strengths (summary)

• this was a very large trial over many centres and countries
• the statistical analyses were well designed and non-inferiority when compared to warfarin for strokes and systemic embolism seems well demonstrated

Biases and conflicts of interests

• This was a industry supported study.  It was funded by a grant from Boehringer Ingelheim, the manufacturer of dabigatran.
• All the main authors have received various fees and grants from Boehringer Ingelheim.
• The article describes results in a manner that minimises some of the adverse associations:
  • Major bleeding was characterised as the “primary safety outcome” in the section on “Outcomes” (p 1141) ^[1] but the relative risks of dabigatran compared to warfarin is not reported in the section on “Primary Outcome” (p 1142).  Rather, the relative risks of haemorrhagic stroke is given instead; this gives a misleading impression that dabigatran has a greater benefit to major bleeding than actually found.
  • There is little discussion on the higher discontinuation rate in the groups taking dabigatran compared to warfarin.
  • There is little discussion on the lack of benefit to all cause mortality.
  • The higher rates of myocardial infarction and gastrointestinal bleeding, though acknowledged, is not discussed in detail.

Clinical relevance to primary health care

Warfarin has been for decades the only oral anticoagulant available. It has clear benefits in patients with non-valvular atrial fibrillation; it reduces the risk of stroke by 68% ^[2].  However, management of warfarin involves substantial administrative time for the general practitioner in regular INR monitoring and dose titration.

Dabigatran (Pradaxa) is a direct thrombin (factor IIa) inhibitor ^[3] and is currently funded by the PBS on authority for prevention of venous thromboembolism in a patient undergoing total hip replacement or total knee replacement ^[4].  It has recently gained attention from the Pharmaceutical Benefits Advisory Committee’s (PBAC) positive recommendation for the authority to be extended to including the “prevention of stroke or systemic embolism in patients with non-valvular atrial fibrillation who are at moderate to high risk” ^[5].  Boehringer Ingelheim as been promoting dabigatran for this indication heavily.

This large study demonstrates that dabigatran 110 mg twice daily, or 150 mg twice daily is non-inferior to warfarin, for the prevention of stroke or systemic embolism.

The other potential (and advertised) benefits of dabigatran should be taken cautiously.  The lower dose of dabigatran was associated with less major bleeding and the higher dose was associated with lower risk of stroke or systemic embolism.  However, dabigatran was also associated with a higher risk of myocardial infarction and a higher risk of gastrointestinal bleeding.  It was associated with higher rates of discontinuation compared to warfarin.  There was no statistically significant effect on all cause mortality.  Potential study biases are likely to favour dabigatran over warfarin.

For the Australian general practitioner working in primary care, dabigatran appears to be an alternative to warfarin for the prevention of strokes in patients with non-valvular atrial fibrillation.

References

1. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009; 361:1139-1151
2. Ezekowitz MD, Connolly S, Parekh A, et al. Rationale and design of RE-LY: randomized evaluation of long-term anticoagulant therapy, warfarin, compared with dabigatran. Am Heart J 2009; 157: 805-10
3. Pradaxa.  MIMS Online Retrieved 18 June 2011.
4. Dabigatran etexilate. PBS Australian Government Website Retrieved 18 June 2011. http://pbs.gov.au/medicine/item/9319L-9323Q
5. March 2011 PBAC Outcomes – Positive Recommendations. PBS Australian Government Website Retrieved 18 June 2011. http://pbs.gov.au/info/industry/listing/elements/pbac-meetings/pbac-outcomes/2011-03/positive-recommendations