For non-valvular atrial fibrillation, physicians, patients, and guideline organizations all prefer direct oral anticoagulants (DOACs) over vitamin K antagonists such as warfarin for thromboembolic prophylaxis. For those with valvular atrial fibrillation, the data for safety and efficacy are limited. Randomized trials comparing DOACs to warfarin often excluded patients with valvular atrial fibrillationRetrospective, non-inferiority, and pooled data from randomized trials suggest DOACs may be safe and effective for patients with valvular heart disease (often defined in these studies as mitral stenosis or prior valve repair but excluding valve replacement) compared to warfarin.

The investigators of a recent retrospective cohort study examined a large commercial insurance claims database from 2010 to 2019. Most participants had either mitral or aortic valve disease and 17 percent had a CHA2DS2-VASc score of at least 3. The investigators excluded patients with bioprosthetic and mechanical valve replacement. After propensity score matching to adjust for differences in baseline characteristics, 28,168 participants were included in each group. Participants were followed for a median of 134 days in the DOAC group and 124 days in the warfarin group.

Prescription discontinuation and disenrollment from the insurance, which included the outcome of death, accounted for the majority of the censored events (74 percent and 21 percent of the total outcomes censored, respectively). The rate of the primary effectiveness outcome of stroke or systemic embolism was 3.9 events per 100 person-years among those prescribed DOACs and 6.0 events per 100 person-years among those prescribed warfarin (hazard ratio [HR] 0.64, 95% CI 0.59-0.70). The rate of major bleeding was 7.1 events per 100 person-years and 10.6 events per 100 person-years of those prescribed DOACs and warfarin, respectively (HR 0.67, 95% CI 0.63-0.72). Both apixaban and rivaroxaban were associated with a reduced risk of stroke or systemic embolism and major bleeding, while dabigatran was associated with only a reduced risk of major bleeding. Apixaban had the strongest performance for the primary effectiveness and safety outcomes (HR for stroke or SE 0.54, 95% CI 0.47-0.61; HR for major bleeding 0.52, 95% CI 0.47-0.57).

This study adds to the growing body of evidence supporting the safety and efficacy of DOACs for patients with atrial fibrillation and valvular heart disease such as mitral valve disorders. But these real-world data also have real-world problems – the reasons for prescription discontinuation or disenrollment, including mortality events, are unknown. Efficacy of thromboembolic prophylaxis for most patients with atrial fibrillation is measured in years; the median four-month follow-up period in this study should be extended in further research. Overall, the data suggest that for these patients, DOACs may outperform warfarin for both thromboembolic and bleeding events. However, the strongest argument these data make is that there is need for a randomized trial evaluating DOACs for valvular heart disease.

For more information, see the topic Thromboembolic Prophylaxis in Atrial Fibrillation in DynaMed.