RE-LY (2009): Dabigatran versus Warfarin in Atrial Fibrillation

Clinical question: in patients with atrial fibrillation, is dabigatran non-inferior to warfarin for the prevention of stroke or systemic embolism?

Study design: multinational randomized unblinded controlled trial comparing dabigatran with warfarin. Two doses of dabigatran (150mg BD and 110mg BD) also compared in a blinded fashion.

Inclusion criteria:

• Documented atrial fibrillation on ECG
• Age > 18
• At least one of the following (most are components of the CHADSVASC score):
  • History of stroke, TIA or systemic embolism
  • EF < 40%
  • Symptomatic heart failure (NYHA class II and above)
  • Age ≥ 75
  • Age ≥ 65 and at least one of:
    • Hypertension requiring treatment
    • Diabetes mellitus on treatment
    • Documented coronary artery disease

Exclusion criteria:

• Valvular heart disease, including infective endocarditis and prosthetic heart valves
• Severe, disabling stroke in the last 6 months, or any stroke 14 days prior to enrollment date
• Increased risk of bleeding:
  • Bleeding diathesis
  • Major surgery within the last month, or surgery planned within the coming 3 months
  • Gastrointestinal haemorrhage within the last year
  • Previous intracranial, spinal, retroperitoneal or spontaneous intraarticular bleeding
  • Peptic ulcer disease within the last 30 days
  • Uncontrolled hypertension
  • Malignancy with prognosis ≤ 3 years
• Contraindication to warfarin
• Renal impairment (creatinine clearance < 30ml/min)
• Active liver disease
• Anaemia (Hb < 10) or thrombocytopaenia (Plt < 100 x 10⁹)

Intervention:

• Patients randomized to 3 groups, enrolled for 24 months, minimum follow-up 12 months
• Warfarin, target INR 2-3
• Low-dose dabigatran 110mg BD (LDD) – dabigatran administered in identical capsules
• High-dose dabigatran 150mg BD (HDD)

Patient characteristics:

• 18113 patients enrolled between 2005 and 2007
• Mean age 71
• 6% men
• Mean CHADS₂ score 2.1
• Concurrent antiplatelet use allowed (for aspirin, maximum dose of 100mg/day)
  • Aspirin use evenly distributed throughout 3 groups

Outcomes and results:

• Primary outcomes:
  • Stroke or systemic embolism
    • Both doses of dabigatran non-inferior to warfarin (p < 0.001)
    • HDD superior to warfarin (RR 0.66, 95% CI 0.53 – 0.82)
    • LDD not superior to warfarin (RR 0.91, 95% CI 0.74 – 1.11)
    • HDD versus LDD reduced risk of stroke or systemic embolism (p = 0.005)
  • Intracranial haemorrhage
    • Higher for warfarin (0.74%/yr) v LDD (0.23%/yr), RR 0.31, 95% CI 0.20 – 0.47
    • Higher for warfarin (0.74%/yr) v HDD (0.30%/yr), RR 0.40, 95% CI 0.27 – 0.60
    • LDD and HDD – no difference
  • Major bleeding
    • Higher for warfarin (3.36%/yr) v LDD (2.71%/yr), RR 0.80, 95% CI 0.69 – 0.93
    • Warfarin vs HDD – no difference
    • HDD vs LDD – no difference, but trend towards increased risk with HDD
  • Gastrointestinal bleeding
    • Warfarin vs LDD – no difference
    • Higher for HDD (1.51%/yr) v warfarin (1.02%/yr), RR 1.50, 95% CI 1.19 – 1.89
    • Higher for HDD v LDD, RR 1.36, 95% CI 1.09 – 1.70
• Secondary outcomes:
  • Death – no difference
  • Myocardial infarction
    • 53%/year for warfarin, 0.72%/year for LDD (RR 1.35, 95% CI 0.98 – 1.87)
    • 53%/year for warfarin, 0.74%/year for HDD (RR 1.38, 95% CI 1.00 – 1.91)
  • Pulmonary embolism – no difference
  • Hospitalization – LDD lower risk compared to warfarin or HDD
  • Rates of life-threatening bleeding, intracranial bleeding, major or minor bleeding higher with warfarin than either dose of dabigatran (p < 0.05 for all comparisons)

Limitations:

• Approximately 30% of patients had CHADS₂ scores of 0 or 1 – these patients have relatively low risks of stroke
• 64% of patients in the warfarin group had a therapeutic INR. If a large proportion were sub-therapeutic, it may confound non-inferiority findings. Similarly, if a large proportion were supra-therapeutic, it may confound inferiority findings with regards to bleeding
• Outcome measures required clinical manifestations; asymptomatic strokes, bleeds or systemic emboli may not have been detected (although patients were regularly administered with symptom questionnaires and available discharge summaries were scrutinized for possibly unreported outcomes)
• Low mean CHADS₂ score, still unsure of utility of dabigatran for patients with higher CHADS₂
• Warfarin use was open-label; possible reporting bias
• Reported rate of major bleeding with warfarin in this study is higher than that of other studies of warfarin in AF
• Verapamil and amiodarone are both used in AF and may interact with dabigatran, potentially affecting interpretation of dose requirement for thromboembolic prophylaxis

Conclusions:

• Dabigatran is non-inferior to warfarin for the prevention of stroke and systemic embolisation in non-valvular atrial fibrillation
• Dabigatran 150mg BD compared to warfarin is associated with lower risks of stroke and systemic embolisation in non-valvular atrial fibrillation
• The risk on intracranial haemorrhage, life-threatening haemorrhage are lower with both doses of dabigatran compared to warfarin
• There is no difference in the risk of gastrointestinal haemorrhage between low-dose dabigatran and warfarin, although high-dose dabigatran is associated with a higher risk of GI bleeding compared with warfarin

Reference: Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, Pogue J, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139-51