Rivaroxaban noninferior to warfarin in patients with AF, bioprosthetic mitral valve: RIVER
Click Here to Manage Email Alerts
Among patients with atrial fibrillation and a bioprosthetic mitral valve, rivaroxaban was noninferior to warfarin for the primary outcome of mean survival time to death, major CV event or major bleeding at 1 year, researchers reported.
Patients who received rivaroxaban (Xarelto, Janssen/Bayer) experienced fewer strokes at 1 year, but the researchers urged caution in the interpretation of this finding, according to results presented at the virtual American Heart Association Scientific Sessions.
For the randomized RIVER trial, simultaneously published in The New England of Medicine, researchers enrolled 1,005 patients with AF and a bioprosthetic mitral valve. Patients were randomly assigned to once-daily rivaroxaban 20 mg or dose-adjusted warfarin (target INR, 2-3). The primary outcome was a composite of death, major bleeding or major CV events, defined as stroke, transient ischemic attack, systemic embolism, valve thrombosis or HF hospitalization, at 1 year.
A primary outcome event occurred at a mean of 347.5 days among patients assigned rivaroxaban vs. 340.1 days among those assigned warfarin.
Between-groups difference was calculated as restricted mean survival time. Patients assigned warfarin experienced 7.4 more days of event-free survival compared with assigned rivaroxaban, but the noninferiority margin was met (95% CI, −1.4 to 16.3; P for noninferiority < .001), according to the results.
Secondary efficacy endpoints did not significantly differ between the rivaroxaban and warfarin groups, including:
- CV mortality or thromboembolic events (HR = 0.65; 95% CI, 0.35-1.2);
- CV death (HR = 0.85; 95% CI, 0.38-1.9);
- all-cause death (HR = 1.01; 95% CI, 0.54-1.87); and
- valve thrombosis (HR = 1.68; 95% CI, 0.4-7.01).
Before RIVER, “we had no dedicated trial that [evaluated rivaroxaban] in this population,” Otavio Berwanger, MD, PhD, cardiologist, clinical epidemiologist and director of the HCor Research Institute and Heart Hospital in Sao Paulo, said during a press conference. “These results can potentially inform practice, and rivaroxaban may represent an attractive alternative for this patient population.”
For the secondary endpoint of stroke, patients assigned rivaroxaban experienced fewer events compared with those assigned warfarin (HR = 0.25; 95% CI, 0.07-0.88).
“This finding must be interpreted with caution since we have a low number of events, a wide confidence interval, and this difference was not adjusted for multiplicity,” Berwanger said during the press conference.
Risk for major bleeding was lower in the rivaroxaban group; however, the findings were not significant (HR = 0.54; 95% CI, 0.21-1.35). There were no incidents of intracranial or fatal bleeding in the rivaroxaban group.
In addition, risk for clinically relevant nonmajor bleeding (HR = 1.05; 95% CI, 0.6-1.87), minor bleeding (HR = 0.75; 9% CI, 0.49-1.15) and total bleeding (HR = 0.83; 95% CI, 0.59-1.15) did not differ significantly between the treatment groups.
In a subgroup analysis of patients who underwent randomization up to 3 months after bioprosthetic mitral valve implantation, the mean time until an event was 348.6 days among patients in the rivaroxaban group vs. 313.5 days in the warfarin group (difference, 35.1 days; 95% CI, 8.6-61.7). Moreover, risk for a primary outcome event among patients in this subgroup was lower in the rivaroxaban group compared with the warfarin group (HR = 0.31; 95% CI, 0.12-0.79).
“The event-free survival time was 35 days longer with rivaroxaban compared to warfarin,” Berwanger said during the presentation. “This is a subgroup analysis, and it is hypothesis-generating at best.”
Discussant Elaine M. Hylek, MD, MPH, professor of medicine at Boston University School of Medicine, said during the press conference that “if we look at the key components of the primary composite endpoint, there are a few numbers here that actually raise the question in an open-label trial whether or not there could have been any type of misclassification. But the blinded adjudication here was a real strength.
“The RIVER trial importantly adds to the field. It is the largest [randomized controlled trial] to date,” Hylek said. “There were too few patients enrolled within that 48 hours postop to 30 days post-[bioprosthetic valve] surgery period. There were only 95 patients in that critical period enrolled, and the results would have been fortified with the systematic assessment of the mitral valve with some type of an exit echocardiogram. Overall, it was a well-conducted study.”
Reference:
Perspective
Back to Top
Annabelle Santos Volgman, MD, FACC, FAHA
In the 2019 AF guidelines, the definition of nonvalvular AF was specifically addressed due to confusion. Verbatim, the guidelines state: “nonvalvular AF is AF in the absence of moderate-to-severe mitral stenosis or a mechanical heart valve” (January CT, et al. Circulation. 2019;doi:10.1161/CIR.0000000000000665). The trials comparing warfarin to various non-vitamin K anticoagulants enrolled a very small number of patients with mitral bioprosthetic valves (only 161). By contrast, the RIVER randomized trial included 1,005 patients. The study found that rivaroxaban is noninferior to warfarin with respect to mean time free from death, major CV events or major bleeding. This trial gives clinicians a welcome peace of mind when prescribing a novel anticoagulant to patients with AF and bioprosthetic mitral valves.Perspective
Back to Top
Julia Grapsa, MD, PhD, FACC
The results from the RIVER trial are consistent with previous research, including ROCKET AF, ARISTOTLE and ENGAGE AF-TIMI 48, and is important for patients with bioprosthetic mitral valves. Since rivaroxaban does not require monitoring of the INR and has an anticoagulant effect that is more consistent and less influenced by food or concomitant medications than warfarin, it represents an attractive alternative for this patient population. In clinical practice, we have few patients who compromise their quality of life when they need to take warfarin, and up to now, we did not have many options to feel reassured that they are safe. The RIVER trial is the largest to date to confirm safety of rivaroxaban.
However, the authors stress certain limitations: Findings cannot be extrapolated to patients with a bioprosthetic aortic valve in the aortic position, or for patients with mitral stenosis or with mechanical valves, because there are trials undergoing. Furthermore, the open-label design could have introduced bias in the ascertainment or reporting of events.
Despite the limitations, no one can doubt that these results are important for clinical practice and give us more options in the therapeutic spectrum.
Collapse
Berwanger O, et al. LBS.03: Current challenges in coronary and valve disease. Presented at: American Heart Association Scientific Sessions; Nov. 13-17, 2020 (virtual meeting).
Read more about
Click Here to Manage Email Alerts