Meta-Analysis: How NOACs Compared to Warfarin in AFib

Clinicians for the past half-century have been managing the rising population of patients with atrial fibrillation with either aspirin or a vitamin K antagonist (most commonly warfarin), the latter being the preferred option to reduce stroke or systemic embolic events (SEE) in appropriately selected high-risk patients.1,2 This phrase implies the need for clinical insight, and a plethora of literature have reported on ways to balance influencing factors to identify patients at high-risk for stroke or SEE, who derive the most efficacy from warfarin, and limit its use in patients at high-risk for major bleeding.3-5 Unfortunately many of these risk factors overlap. Moreover, despite its great potential to reduce arterial (and venous) thrombotic events, the drawback of warfarin has been that it requires considerable time for onset and offset of therapeutic effect; has a narrow therapeutic window that necessitates frequent monitoring; has many potential drug interactions which modify its metabolism; and is modified by variation in dietary vitamin K intake.

As a result, four new oral anticoagulants that dose-dependently strategically inhibit either activated factor X (factor Xa) or thrombin have been developed and are either approved or under review for use to further reduce stroke or SEE compared with warfarin in patients with atrial fibrillation.6-9 The benefits of these new anticoagulants are their predictable anticoagulant effect, fewer drug interactions, and hence ease of administration and use in routine clinical practice. The current limitations (all likely to be overcome with time) are the lack of an effective antidote, inability to effectively determine compliance, and cost compared with warfarin. In order to better understand the balance of efficacy and hazard of the four new oral anticoagulants (NOACs) compared with warfarin for treating patients with atrial fibrillation, and to provide additional insight into the most appropriate patients to select to treat with a NOAC versus warfarin, Ruff and colleagues recently published a definitive meta-analysis of the four phase 3 randomized trials on the subject, RE-LY, ROCKET-AF, ARISTOTLE, and ENGAGE AF-TIMI 48.10

Comparisons were made pooling the high-dose or single-dose NOAC studied in each trial with warfarin for superiority for the primary outcome of stroke or systemic embolic events. Secondary efficacy outcomes included ischemic stroke, hemorrhagic stroke, myocardial infarction (MI), and mortality. Safety outcomes included major bleeding, intracranial hemorrhage (ICH), and gastrointestinal bleeding. Additional analyses were performed pooling the low-dose NOACs studied in two trials compared with warfarin. Efficacy analyses included the intention-to-treat population whereas safety evaluations appropriately limited the focus to patients on treatment.

A total of 71,683 patients were studied in this comprehensive analysis over a pooled median follow-up duration of 2.2 years. Overall, the population studied was mean 71.5 years of age, and a third were women. RE-LY and ARISTOTLE enrolled participants at approximately equal distributions across the range of CHADS2 scores (categorized as low risk [0-1], moderate risk [2], and high risk [3-6]), whereas ROCKET-AF and ENGAGE AF-TIMI 48 enrolled a higher proportion of high-risk patients and virtually no low-risk patients. Just over half of the population had prior exposure to warfarin or similar anticoagulants, while half were warfarin-naïve. Median time in the therapeutic range (TTR) of international normalized ratio (INR) varied between 58-68%, with an overall median time of 65%.

Results for the pooled event rates, absolute risk reductions, and numbers needed to treat over 2.2 years of follow-up for significant outcomes are summarized in the Table. The NOACs significantly reduced stroke or SEE compared with warfarin (3.11% vs 3.79%, RR 0.81, 95% CI, 0.73-0.91). As suggested in the individual trials, the risk reduction of NOACs was mainly driven by a reduction in hemorrhagic stroke (0.44% vs 0.90%, RR 0.49, 95% CI, 0.38-0.64), with a significant relative and absolute reduction in total mortality (6.90% vs 7.68%, RR 0.90, 95% CI, 0.85-0.95).

Subgroup analyses were performed to evaluate whether any differences in baseline patient or trial characteristics modified the pooled treatment effect of the NOACs compared with warfarin. Reassuringly, efficacy results were consistent across all subgroups, however there was a greater relative risk reduction in major bleeding when patients were treated in centers with TTR <66% (RR 0.69, 95% CI, 0.59-0.81) compared with ≥66% (RR 0.93, 95% CI, 0.76-1.13; p-interaction=0.022).

The analyses of low-dose NOACs confirmed they provided similar results compared with warfarin for stroke or SEE reduction (RR 1.03, 95% CI, 0.84-1.27), superior results regarding hemorrhagic stroke (RR 0.33, 95% CI, 0.23-0.46), ICH (RR 0.31, 95% CI, 0.24-0.41), and mortality (RR 0.89, 95% CI, 0.83-0.96), but inferior with respect to an increased risk of ischemic stroke (RR 1.28, 95% CI, 1.02-1.60) and MI (RR 1.25, 95% CI, 1.04-1.50).

In my view, there are four takeaway messages from this comprehensive analysis:

Table: Summary Efficacy and Safety Outcomes of New Oral Anticoagulants Compared with Warfarin
Table 1: Meta-Analysis: How NOACs Compared to Warfarin in AFib
  1. High-dose NOACs appear to derive their efficacy compared with warfarin in reducing hemorrhagic, not ischemic, stroke. This is likely because of their pharmacokinetic advantage over warfarin, i.e. they considerably reduce the hazard of the wide variability and toxicity of anticoagulation outside time in the therapeutic range (TTR) with warfarin.
  2. With that context, it was predictable and reassuring that patients derived similar stroke and SEE results across sites when categorized by TTR achievement, whereas reduction in major bleeding was most significant when patients were treated by centers achieving below average TTR. Patients treated at centers in the lower half of TTR (driven in part by treatment of more complicated patients, centers with lower quality control, or both) fair better with respect to bleeding complications with a NOAC. Thus patients, providers, insurers, and policy makers struggling to identify the most efficient way to introduce (and afford) NOACs may consider center-based TTR as one quality measure, in conjunction with patient risk and preferences, to decide whether treatment with a NOAC will provide a superior safety profile along with efficacy. Centers achieving excellent TTR may see less bleeding advantages with the NOACs, though again individual patient preferences and the economics of managing such centers may still drive decision-making.
  3. Impressively, the outcomes with the lowest number needed to treat (with statistical significance) were:
    1. Mortality (128 patients over 2.2 years to prevent one death) despite a modest relative risk reduction, and
    2. Intracranial hemorrhage, an infrequent but devastating complication strongly reduced with NOAC therapy (132 patients over 2.2 years to prevent one ICH)
    These results, in addition to reducing stroke or SEE, may be the best indications that will drive increased use of the high-dose NOACs overall. Given individual endpoints such as these were relatively underpowered in each trial, few if any of these medications' labels will have these outcomes indicated.

  4. Finally, data suggest that clinicians in many regions of the world, where a low-dose option is approved, are prescribing a low-dose NOAC to patients that otherwise are eligible for high-dose NOAC as a compromise approach to minimize harm. This meta-analysis confirms that low-dose NOACs do also convey impressive absolute risk reductions in mortality and ICH but clearly at the expense of an increased risk of ischemic stroke and MI events. Low-dose NOACs may be the appropriate choice in select patients at high-risk for bleeding, but the balance of efficacy and risk should appropriately be conveyed to patients. When eligible, the high-dose NOACs appear to be the preferred option from this meta-analysis.


  1. Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010;31:2369-429.
  2. Cairns JA, Connolly S, McMurtry S, Stephenson M, Talajic M. Canadian Cardiovascular Society atrial fibrillation guidelines 2010: prevention of stroke and systemic thromboembolism in atrial fibrillation and flutter. Can J Cardiol 2011;27:74-90.
  3. Hylek EM, Go AS, Chang Y, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med 2003;349:1019-26.
  4. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest 2010;137:263-72.
  5. Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJ, Lip GY. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest 2010;138:1093-100.
  6. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139-51.
  7. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883-91.
  8. Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981-92.
  9. Giugliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013;369:2093-104.
  10. Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. The Lancet 2013.

Keywords: Anticoagulants, Atrial Fibrillation, Drug Interactions, Hemorrhage, Safety, Warfarin, Factor Xa, International Normalized Ratio, Intracranial Hemorrhages, Stroke

< Back to Listings