Following a transcatheter aortic valve replacement (TAVR), direct oral anticoagulants (DOACs) are preferable to vitamin K antagonists (VKAs) in patients who are candidates for oral anticoagulants, according to data drawn from a large multicenter French TAVR registry.
When oral anticoagulation is appropriate following TAVR, such as in patients with atrial fibrillation, “DOACs are associated with improved survival and lower incidence of bleeding, compared to VKA,” reported a team of investigators led by Martine Gilard, MD, PhD, director of interventional cardiology, Brest (France) University Hospital Center.
The comparison, using propensity score matching, is not definitive, but it might be the best data currently available to support DOACs over VKA until a randomized trial is completed, according to Dr. Gilard, senior author of the newly published study.
Asked in an interview if DOACs should now be used preferentially after TAVR when patients are indicated for oral anticoagulation, Dr. Gilard replied, “My answer is yes.”
Of more than 24,000 TAVR patients in the French TAVI and FRANCE2 multicenter registries, which are linked to the French single-payer claims database (SNDS), 8,962 (36.4%) received an oral anticoagulant following their procedure. Of these, 2,180 (24.3%) received a DOAC and the remaining received VKA.
By linking data from the registries to the SNDS, outcomes were tracked. Propensity matching was employed to control for differences in baseline characteristics, including age, body mass index, functional class, diabetes, comorbidities, and past medical history.
On the primary endpoint of mortality at the end of 3 years, the rates were 35.6% and 31.2% for VKA and DOACs, respectively. This translated in a 37% greater hazard ratio for death among those treated with VKA (P < .005).
The rate of major bleeding, a secondary endpoint, was also higher (12.3% vs. 8.4%) and significantly different (HR, 1.65; P < .005) for VKA versus DOACs. The rates of ischemic stroke, acute coronary syndrome, and hemorrhagic stroke were all numerically higher in patients treated with VKA than DOACs, although none of these differences reached statistical significance.
Residual confounding cannot be discounted
“The large number of events allowed for taking into account a higher number of potential confounders with appropriate statistical power,” according to the authors. However, they acknowledged that residual confounding cannot be eliminated by propensity matching and conceded that prospective data are needed for a definitive comparison.
In an accompanying editorial, Daniele Giacoppo, MD, a cardiologist at Alto Vicentino Hospital, Santorso, Italy, enlarged on this point . In addition to the inherent limitations of retrospective data, he also noted that data from other studies addressing the same question have been inconsistent.
Of these studies, he pointed to the ATLANTIS trial, presented 2 months ago at the annual meeting of the American College of Cardiology. This study failed to show an advantage for the DOAC apixaban over VKA in TAVR patients for the primary composite outcome of time to death, myocardial infarction, systemic emboli, valve thrombosis, or major bleeding. Although this study was not limited to patients with an indication for oral anticoagulants, Dr. Giacoppo pointed out that there was no advantage, even among the subgroup of patients who did have an indication.