Mixed Effects of Early Anticoagulation on Bioprosthetic Valves

Patrice Wendling

August 28, 2019

Early anticoagulation (AC) after bioprosthetic surgical or transcatheter aortic valve replacement (SAVR/TAVR) does not appear to affect valve hemodynamics in the ensuing year, according to a new analysis of pooled PARTNER 2 data.

Although aortic valve mean gradients were lower in the AC group after TAVR at 30 days, there was no independent association with overall improved aortic valve gradients or aortic valve area after either SAVR or TAVR.

AC, however, was independently associated with fewer strokes after SAVR (hazard ratio [HR], 0.17; = .006) and more minor bleeding after TAVR (HR, 1.72; P = .01).

"AC may not affect valve hemodynamics but offers a significant benefit in rates of stroke after SAVR," lead author Tarun Chakravarty, MD, Cedars-Sinai Medical Center, Los Angeles, and colleagues write in the study, published online August 26 in the Journal of the American College of Cardiology.

Although bioprosthetic valve thrombosis is less prevalent in patients on AC, little is known about how routine AC affects valve hemodynamics and clinical outcomes, the authors note. This is despite increasing use of bioprosthetic surgical valves, expanding indications for TAVR, and recent reports of subclinical leaflet thrombosis in transcatheter and surgical bioprosthetic valves.

The recent GALILEO trial also highlights the pressing need to clarify postprocedure anticoagulant and antithrombotic therapy after it was prematurely stopped because of increased bleeding, thromboembolic events, and death with routine rivaroxaban-based anticoagulation (Xarelto, Bayer/Janssen), compared with dual antiplatelet therapy (DAPT) after TAVR.

The current analysis takes advantage of core lab-assessed echocardiograms and adjudicated clinical events in the PARTNER 2 randomized trials and nonrandomized registries to evaluate the impact of adjunct pharmacotherapy among 4832 intermediate- and high-risk patients after bioprosthetic AVR (Sapien 3 or Sapien XT valves). Of these, 3889 patients underwent TAVR and 943 underwent SAVR. The average follow-up was 1.7 years and 2.1 years, respectively.

AC was not routinely initiated at discharge, but administered based on clinical indications. Significantly more patients after SAVR, compared with TAVR, were discharged on AC (36.3% vs 29.2%). Similar trends were observed at 30 days (35.5% vs 28.0%) and 1 year (30.8% vs 27.3%). Monoantiplatelet therapy alone was more likely to be prescribed at discharge after SAVR (42.0% vs 19.4%), whereas patients after TAVR were more likely to be discharged on DAPT alone (47.2% vs 15.2%).

There was no significant difference in aortic mean gradients or valve areas between patients discharged on or off AC after SAVR or TAVR after adjustment for baseline mean annular diameter, valve size, baseline atrial fibrillation (AF), and left ventricular ejection fraction at the time of the end point. Similar results were seen in a sensitivity analysis that added end point observed at discharge to the model.

No patient had a severe increase in gradients that would be suggestive of valve thrombosis or stenosis, the authors said. Overall, 1.1% of patients discharged on AC and 2.3% not on AC had an increase in mean gradient of more than a 10 mm Hg from 30 days to 1 year (P = .03).

Valve Hemodynamics for Patients Discharged On vs Off Anticoagulation (AC)
Time Point AC No AC Unadjusted Value Adjusted P Value
TAVR: mean gradients (mm Hg)
30 days 10.6 11.4 <.0001 .83
1 year 11.1 12.0 .0001 .85
TAVR: aortic valve area (cm2)
30 days 1.64 1.60 .004 .049
1 year 1.61 1.56 .03 .10
SAVR: mean gradients (mm Hg)
30 days 10.4 11.2 .009 .89
1 year 11.1 11.8 .05 .56
SAVR: aortic valve area (cm2)
30 days 1.50 1.46 .23 .30
1 year 1.42 1.42 .88 .43

"Our findings do not support the hypothesis that the decreased incidence of subclinical leaflet thrombosis of transcatheter heart valves observed in people on AC translates into decreased aortic valve gradients or improved aortic valve areas, at least in the short term," Chakravarty and colleagues say.

AC after TAVR was independently associated with more minor bleeding (HR, 1.72) and a higher rate of the composite of death, stroke, and rehospitalization (HR, 1.26; P = .05). AC after SAVR was independently associated with significantly fewer strokes but a similar rate of death or rehospitalization. The lower stroke risk after SAVR is possibly due to a greater incidence of new-onset AF or to a decreased incidence of subclinical leaflet thrombosis, the authors suggest.

The study is not "groundbreaking" but "confirmatory," Philippe Pibarot, DVM, PhD, Université Laval, Québec City, Canada, who coauthored a related editorial, told theheart.org | Medscape Cardiology.

"It's good to know that while anticoagulation will reduce the risk of thrombosis and therefore the risk of outcomes that may be associated with that, on the other end, it definitely increases the risk of bleeding, and potentially life-threatening bleeding. So I'm not sure if the benefit/risk ratio is positive," he said.

The study's 1-year post-AVR rate of valve hemodynamic deterioration of 1.9% is consistent with data reported in previous studies, the editorialists note. However, the rate was based only on a mean gradient increase of at least 10 mm Hg and not on multiple criterion, as suggested in the Valve-in-Valve International Data (VIVID) position statement. As such, this "may have resulted in false positives or false negative cases of valve hemodynamic deterioration."

Also, the study only examines short-term effects of early AC and cannot compare warfarin with direct oral anticoagulants (DOACs), given that only 5% of patients were discharged on DOACs. Recent work suggests warfarin may promote the calcification of native and bioprosthetic valve leaflets by inhibiting activation of the Matrix Gla protein, a powerful protector against ectopic calcification. Thus, warfarin may have a protective effect after the procedure by preventing valve thrombosis, but a negative effect in the long-term by accelerating bioprosthetic leaflet calcification and thus valve deterioration.

Pibarot said the results do not support the recent guidelines recommending 3 months of AC after SAVR and TAVR. "The final message is that systematic anticoagulation is probably not the way to go and that an individualized approach according to risk factors for thromboembolism or valve thrombosis is the way to go."

Chakravarty reports serving as a proctor and consultant for Edwards Lifesciences and Medtronic. Pibarot reports having received funding from Edwards Lifesciences for echocardiography core lab analyses in several TAVR trials with no personal compensation, and is director of the echocardiography core lab for the PARTNER 2–SAPIEN 3 nonrandomized registry.

J Am Coll Cardiol. 2019;74:1190-1200 and 1201-1204. Abstract, Editorial

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: