Antiplatelets Protective Post-TAVR in A-fib Patients: PARTNER II Data

For patients with atrial fibrillation who undergo TAVR, being discharged on antiplatelet therapy seems key in reducing the risk of stroke at 2 years, according to post hoc analysis of PARTNER II data. Oral anticoagulation (OAC) alone does not appear to be protective.

The results imply that stroke mechanisms in this population are multifactorial, lead author Ioanna Kosmidou, MD, PhD (NewYork-Presbyterian Hospital/Columbia University Irving Medical Center, New York, NY), and colleagues conclude in their paper published online ahead of the August 26, 2019, issue of JACC: Cardiovascular Interventions.

Post-TAVR antithrombotic regimens today are quite varied among A-fib patients at high risk for stroke. Little is known about how these strategies, which can employ antiplatelets, anticoagulants, both, or neither, affect clinical outcome.

Currently, Kosmidou told TCTMD, there is “extreme ambiguity as to the best antithrombotic therapy” in this specific population. Not only do patients with severe aortic stenosis have a very high rate of concomitant A-fib, this condition also puts them at an elevated stroke risk compared with the general population, “and obviously choosing the optimal antithrombotic therapy is very important for [them],” she explained.

The researchers sought to examine “whether just pure anticoagulant therapy with either Coumadin or novel anticoagulants, the latter being of very limited use in the PARTNER trials, by itself was sufficient to prevent the long-term risk of stroke, and I think we showed that truly just anticoagulant therapy by itself was not [enough],” she said. Yet, “we observed a very clear-cut benefit with antiplatelet therapy.”

Kosmidou stressed, though, that the message is not to avoid anticoagulants post-TAVR in patients with atrial fibrillation. Rather, it should be understood that TAVR-related strokes aren’t “necessarily due predominately to thrombotic events,” she said. Platelet-rich thrombus can be responsible, too, therein supporting a role for antiplatelet therapy.

“Antiplatelet therapy,” Kosmidou said, “should not be omitted when a patient is on anticoagulant therapy—they function differently.”

Surmounting Stroke Risk

Among 4,702 patients in the PARTNER II trial and registries, 1,690 (36%) had both prior A-fib and a CHA2DS2-VASc score ≥ 2. The current analysis focused on the 1,651 individuals within this group who survived to discharge and whose records contained information on OAC.

One in six patients (57.6%) were discharged on an oral anticoagulant (predominately warfarin) during the 5-year enrollment period from 2011 to 2015. Of these patients, 58.3% were on uninterrupted antiplatelet therapy for at least 6 months (or until an endpoint was reached). Approximately three-quarters of patients (77.5%) not on OAC maintained antiplatelet therapy for at least 6 months.

OAC use rose from 46.4% in 2011 to 69.6% in 2015, with warfarin continuing to be the most widely prescribed anticoagulant. Antiplatelet utilization, on the other hand, held fairly steady over the course of the study.

By 2 years after TAVR, the rate of stroke was similar between patients who did or did not take oral anticoagulants (6.6% vs 5.6%; P = 0.53), as was the death/stroke rate (29.7% vs 31.8%; P = 0.33). Compared with no antithrombotic medications, OAC plus antiplatelet therapy resulted in a lower rate of stroke (5.4% vs 11.1%; P = 0.03) and death/stroke (29.7% vs 40.1%; P = 0.01).

Major bleeding rates at 2-year follow-up were 6.9% with OAC plus antiplatelet therapy, 4.5% with OAC alone, 5.1% with antiplatelet therapy alone, and 4.4% with no antithrombotic medications.

After multivariable adjustment for disease history, comorbidities, age, sex, baseline stroke risk, and valve type, antiplatelet therapy alone was associated with lower likelihood of stroke (HR 0.32; 95% CI 0.16-0.65), as was OAC plus antiplatelet therapy (HR 0.44; 95% CI 0.22-0.87). Notably, OAC alone did not show such a benefit. Landmark analyses capturing events from 30 days to 2 years, done in an effort to weed out procedural factors, confirmed these results.

Patient and Valve Factors

Causes of stroke in TAVR may involve patient- and valve-related factors, Davide Capodanno, MD, PhD, and Antonio Greco, MD (both from University of Catania, Italy), point out in an editorial accompanying the paper. Patient-related factors include A-fib, venous thromboembolic disease, concomitant CAD/PAD, and aortic arch atheroma. Valve-related factors are stent-frame endothelialization, artificial surface exposure, tissue injury, subclinical leaflet thrombosis, and flow turbulence, among others.

Capodanno and Greco posit two main pathways: “According to the antiplatelet hypothesis, thrombi develop in conditions of high shear stress and are rich in platelets: in these circumstances, antiplatelet therapy outperforms anticoagulant therapy. According to the antithrombin hypothesis, thrombi develop in conditions of low shear stress and are rich in fibrin: in these circumstances, anticoagulant therapy outperforms antiplatelet therapy.”

On the flip side, as highlighted by a recent study covered in TCTMD, TAVR-related bleeding doesn’t appear to hinge on platelet reactivity alone but also on some other, as-yet-unknown mechanism.

Antiplatelet therapy should not be omitted when a patient is on anticoagulant therapy—they function differently. Ioanna Kosmidou

In terms of interpreting the current results, the editorialists are cautious. Bleeding was not classified according to VARC-2 criteria, nor are there specifics reported for antiplatelet regimens or for time in therapeutic range with OAC. Few of the patients received non-vitamin K antagonist oral anticoagulants, and all were at intermediate surgical risk.

As TAVR shifts toward lower-risk patients, this “evolution may alter the net benefit of an antithrombotic strategy if the rate of stroke will decrease as the consequence of younger and less fragile patients being offered TAVR,” they say, noting, “Similar concerns on external validity regard the unclear generalizability of these results to newer iterations of the balloon-expandable and self-expandable bioprostheses.”

Even so, the findings of this study “are timely and thought-provoking, because they imply that contrary to current recommendations in Europe, OAC might not be sufficient for stroke prevention in patients with atrial fibrillation undergoing TAVR,” Capodanno and Greco conclude.

European Society of Cardiology/European Association for Cardio-Thoracic Surgery guidelines from 2017 advise lifelong OAC for A-fib patients (class I), the editorialists note. American College of Cardiology/American Heart Association guidelines do not make any recommendations specific to A-fib about antithrombotic therapy after TAVR.

Kosmidou, too, said these issues will need to be watched closely as TAVR moves into lower-risk and potentially younger patients. However, she predicted that comorbidities like atrial fibrillation that put patients at risk for stroke will not disappear along the way. “That will not change, and as a topic I think it will become even more important” as treatments evolve for aortic stenosis as well as for mitral and tricuspid valve disease, Kosmidou commented. “We have to keep an eye out for those patients. Their management may be different from what we know in the general population.”

Both Kosmidou and the editorialists are looking forward to results of ongoing RCTs that may provide a clearer picture of the best approach. Among them are the ATLANTIS, ENVISAGE-TAVI AF, and AVATAR trials plus the OAC cohorts of POPular-TAVI and CLOE.

Note: Kosmidou and co-authors Susheel Kodali, MD, and Martin B. Leon, MD, are faculty members of the Cardiovascular Research Foundation, the publisher of TCTMD.