Native valve disease in patients with non-valvular atrial fibrillation on warfarin or rivaroxaban

Abstract

Objective: To compare the characteristics and outcomes of patients with atrial fibrillation (AF) and aortic stenosis (AS) with patients with AF with mitral regurgitation (MR) or aortic regurgitation (AR) and patients without significant valve disease (no SVD).

Methods: Using Rivaroxaban Once-Daily, Oral, Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET AF) data, we analysed efficacy and safety outcomes, adjusting hazard ratios (HRs) for potential confounders using Cox regression analysis.

Results: Among 14 119 intention-to-treat ROCKET AF trial patients, a trial that excluded patients with mitral stenosis or artificial valve prosthesis, 214 had AS with or without other valve abnormalities, 1726 had MR or AR and 12 179 had no SVD. After adjusting for prognostic factors, the composite of stroke, systemic embolism or vascular death increased approximately twofold in patients with AS (AS 10.84, MR or AR 4.54 and no SVD 4.31 events per 100 patient-years, p=0.0001). All-cause death also significantly increased (AS 11.22, MR or AR 4.90 and no SVD 4.39 events per 100 patient-years, p=0.0003). Major bleeding occurred more frequently in AS (adjusted HR 1.61, confidence intervals (CI) 1.03 to 2.49, p<0.05) and MR or AR (HR 1.30, 1.07 to 1.57, p<0.01) than in no SVD, but there was no difference between AS and MR or AR (HR 1.24, 0.78 to 1.97). The relative efficacy of rivaroxaban versus warfarin was consistent among patients with and without valvular disease. Rivaroxaban was associated with higher rates of major bleeding than warfarin in patients with MR or AR (HR 1.63, 1.15 to 2.31).

Conclusions: We found that patients with AF and AS on oral anticoagulants may have distinctly different efficacy and safety outcomes than patients with MR or AR or no SVD.

Trial registration number: NCT00403767; Post-results.

Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/

Figures

Figure 1

Efficacy endpoints by SVD subtype and for patients with no SVD based on ITT patients. Efficacy endpoints (events per 100 patient-years, unadjusted) by SVD subtype and for patients with no SVD based on ITT patients. p Values for any difference among groups are based on Cox proportional hazards models. Patients in both treatment arms are combined. For results of detailed statistical analyses, see table 3. ITT, intention to treat; MI, myocardial infarction; pt-years, patient-years; SE, systemic embolism; SVD, significant valve disease; MR or AR, mitral or aortic regurgitation; vasc. death, vascular death.

Figure 2

Safety endpoints by SVD subtype and for patients with no SVD. Safety endpoints (events per 100 patient-years, unadjusted) by SVD subtype and for patients with no SVD. p Values for any difference among groups are based on Cox proportional hazards models. Patients in both treatment arms are combined. For results of detailed statistical analyses, see table 3. NMCR, non-major clinically relevant; pt-years, patient years; SVD, significant valve disease; MR or AR, mitral or aortic regurgitation.

Figure 3

Treatment comparisons between patients allocated to rivaroxaban and warfarin. Treatment comparisons between patients allocated to rivaroxaban and warfarin for safety endpoints among SVD subtypes and for patients with no SVD. Events per 100 patient-years, unadjusted. p Values for any difference among groups are based on Cox proportional hazards models. Patients in each treatment are shown separately. For results of detailed statistical analyses, see online supplementary table S1. NMCR, non-major clinically relevant; pt-years, patient years; SVD, significant valve disease; MR or AR, mitral or aortic regurgitation.