By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: In this randomized trial of post-transcatheter aortic valve replacement patients without a separate indication for anticoagulation, a rivaroxaban-based approach was associated with a higher risk of death and thromboembolic complications compared with dual antiplatelet therapy.

SOURCE: Dangas GD, Tijssen JGP, Wöhrle J, et al. A controlled trial of rivaroxaban after transcatheter aortic-valve replacement. N Engl J Med 2020;382:120-129.

The optimal antithrombotic regimen for routine post-procedure management of transcatheter aortic valve replacement (TAVR) patients is unknown. Dual antiplatelet therapy was used in early TAVR trials, based on expert opinion rather than on comparative data. This remains the most common post-procedure pharmacologic treatment. Although clinically evident thrombosis affecting valve function is relatively rare, imaging studies dating back to 2015 have revealed a substantial rate of subclinical leaflet thrombosis in the months after TAVR: highly variable, but accounting for more than 15% of patients in most studies, resulting in leaflet thickening or reduced leaflet motion. The relationship of this imaging-defined leaflet thrombosis to either short-term thromboembolic events or to downstream valve degeneration is unknown.

Based on the observation that patients on oral anticoagulation show reduced rates of subclinical leaflet thrombosis, Dangas et al sought to study how a rivaroxaban-based approach compares with standard antiplatelet therapy in post-TAVR care. To this end, patients without an established indication for oral anticoagulation who had undergone successful TAVR were randomized one-to-one to either rivaroxaban-based treatment (aspirin plus rivaroxaban 10 mg/day for three months, followed by rivaroxaban monotherapy) or to an antiplatelet regimen consisting of aspirin plus clopidogrel for three months, followed by aspirin monotherapy.

The design of the trial designated a primary efficacy endpoint: the composite of death and all thromboembolic complications, including myocardial infarction, stroke, valve thrombosis, and both arterial and venous thromboembolism. The authors also reported major bleeding as a primary safety endpoint and used these data in combination with the efficacy endpoint to report a net benefit outcome.

Over the 30-month trial period, 1,644 patients were enrolled at 136 centers in 16 countries. The mean age was 80.6 years, just fewer than half were women, and most were recruited from centers in Western Europe. A total of 826 patients were randomly assigned to the rivaroxaban group and 818 to the antiplatelet group. At a median follow-up of 17 months, death or first thromboembolic event (the primary efficacy outcome) occurred in 105 patients in the rivaroxaban group and in 78 patients in the antiplatelet group (incidence rates, 9.8 and 7.2 per 100 person-years, respectively; hazard ratio [HR] with rivaroxaban, 1.35; 95% confidence interval [CI], 1.01-1.81; P = 0.04). Notably, stroke and myocardial infarction were not different between the groups. However, mortality alone was significantly more common in the rivaroxaban group: 64 vs. 38 (HR for rivaroxaban, 1.69; 95% CI, 1.13-2.53). The primary safety outcome of life-threatening, disabling, or major bleeding increased in the rivaroxaban group, consistent with expectations (46 vs. 31 patients; HR, 1.5, 1.50; 95% CI, 0.95-2.37; P = 0.08). The data safety monitoring board halted the trial early in response to this information, suggesting overall harm of the treatment approach.

The authors concluded that in post-TAVR patients without an indication for oral anticoagulation, treatment with a regimen including rivaroxaban was associated with higher risks of death and thromboembolic complications and an increased risk of bleeding compared with the current antiplatelet standard of care.


GALILEO started with a good idea. Subclinical leaflet thrombosis as defined by imaging is quite common after TAVR. Prevention of this development might reasonably be expected to decrease thromboembolic complications. This idea is so logical that many practitioners in recent years have discussed adding oral anticoagulation to the routine post-TAVR regimen, even in the absence of efficacy data. The fact that the results have shown harm for this particular approach is a cautionary tale and highlights the need for meticulous randomized, controlled trials.

The rare outcome of symptomatic valve thrombosis was numerically lower in the rivaroxaban group (3 vs. 7 events), although this did not meet statistical significance. In the imaging substudy of the trial that included 231 patients, rivaroxaban treatment was associated with a significant reduction in subclinical leaflet thrombosis, as defined by leaflet thickening and reduced leaflet motion. In this case, rivaroxaban at the dose prescribed was effective at preventing the target surrogate outcome, but led to worse clinical outcomes anyway. At this point, we do not understand the clinical significance of subclinical leaflet thrombosis. The idea that it may be associated with earlier valve degeneration remains a viable concept. Whether the natural history of this process, and the longevity of these valves, may be altered for the better through anticoagulant medications is unknown. Future trials may give more insight into this process.

Notably, most of excess deaths in the rivaroxaban arm of GALILEO did not also experience severe bleeding. Most of these deaths occurred long after the study drug was discontinued. Hence, the cause of the observed increase in death with the rivaroxaban approach is unknown and will require further study. For now, what is known after this trial is routine treatment of post-TAVR patients with direct-acting oral anticoagulants is not indicated in the absence of a traditional indication for these drugs. Even in those cases of a clear anticoagulation indication (e.g., the TAVR patient with atrial fibrillation), the best regimen remains unknown, and further trials in this arena are ongoing.