Bridging the Death Gap: Residual Mortality in AF

January 17, 2017

A study from Spain identified causes of residual death in patients with AF on anticoagulation. The results may come as a surprise.

Tremendous advances have been made in the reducing the risk of stroke and systemic embolism in patients with atrial fibrillation (AF). The introduction of the easy-to-use thrombotic risk score CHA2DS2-Vasc as well as the adoption of the novel oral anticoagulants (NOACs) has increased the number of individuals being treated with anticoagulation and reduced the risk of mortality. However, AF continues to remain associated with a significantly higher mortality.

To help highlight the persistent “death gap,” a group from Spain sought to identify the causes of death in AF patients already being treated with anticoagulation. Results were published in a recent issue of the Journal of the American College of Cardiology.1 With a systematic search of randomized controlled trials, they identified 71,683 patients from 4 trials with a total of 134,046 patient-years of follow-up. The death rate was 9% with an adjusted mortality rate of 4.7%/year. Almost half (46%) of the deaths were comprised of cardiac death (heart failure, MI, sudden cardiac death) with only 5.7% stroke/systemic embolism deaths and 5.6% hemorrhage-related deaths. Many clinical factors were associated with death, including heart failure, permanent/persistent AF, diabetes, male gender, worse renal function, and older age. These factors were also most strongly associated with cardiac death in the RE-LY2 and ROCKET-AF3 clinical trials that led to the approval of dabigatran and rivaroxaban, respectively. The use of NOACs was associated with a very slightly lower risk (-0.42%) of death compared with warfarin that was attributed to lower fatal bleeding (and specifically, intracranial bleeding).

The results indeed are compelling evidence of ongoing excess mortality among AF patients. Its limitations include a heterogeneous population (with variable CHA2DS2-Vasc  scores) and limitation to data from randomized trials only. However, it does highlight that although we have reduced mortality with the use of oral anticoagulants, there is a large proportion of cardiac death that continues to plague patients with AF. Perhaps this is a result of similar risk factors for both conditions, or as we are learning, perhaps AF and MI or CHF have a more complex causal relationship. Regardless, ongoing efforts need to be made by physicians to continue to address this residual mortality risk.


1. Gómez-Outes A, Lagunar-Ruíz J, Terleira-Fernández AI, et al. Causes of death in anticoagulated patients with atrial fibrillation. J Am Coll Cardiol. 2016;68(23):2508-2521. 

2. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883–91.

3. Pokorney SD, Piccini JP, Stevens SR, et al. Cause of death and predictors of all-cause mortality in anticoagulated patients with nonvalvular atrial fibrillation: data from ROCKET AF. J Am Heart Assoc 2016;4:e002197.