Reduced-Dose Rivaroxaban May Be an Option After LAA Occlusion, ADRIFT Hints

The NOAC reduced thrombin generation relative to DAPT, but larger trials are needed.

Reduced-Dose Rivaroxaban May Be an Option After LAA Occlusion, ADRIFT Hints

PARIS, France—To head off the risk of device-related thrombus after left atrial appendage (LAA) occlusion, a reduced dose of rivaroxaban may be a viable alternative to dual antiplatelet therapy (DAPT), the ADRIFT study suggests.

Rivaroxaban (Xarelto; Bayer/Janssen) given at a dose of 10 or 15 mg once daily significantly reduced thrombin generation in the first 10 days after the procedure compared with DAPT, Gilles Montalescot, MD, PhD (Centre Hospitalier Universitaire Pitié-Salpêtrière, Paris, France), reported last week at the European Society of Cardiology Congress 2019. That difference was sustained out to 3 months.

The only two cases of device thrombosis detected during the study were in patients taking DAPT.

“The combined antithrombotic regimens currently used after [LAA closure] may not be adapted to high-bleeding-risk patients undergoing this procedure,” Montalescot said during his presentation. He concluded that “a reduced dose of rivaroxaban appears clinically feasible and deserves now further evaluation in large clinical trials.”

Though device-related thrombus is an infrequent complication of LAA occlusion, it has been associated with greater risks of both mortality and ischemic stroke.

Antithrombotic therapy has been used for at least 3 months after implantation of an LAA occluder to protect against this risk. In the pivotal trials of the Watchman device (Boston Scientific), patients were treated with aspirin and warfarin for 45 days after the procedure. Then, if there was an adequate seal, warfarin was stopped and clopidogrel was started, with DAPT lasting until 6 months after the procedure; aspirin was continued indefinitely after that point.

In Europe, Montalescot noted, most patients who undergo LAA occlusion have contraindications to anticoagulation. Thus, the most common approach for preventing device thrombosis is aspirin plus clopidogrel for at least 3 months and aspirin monotherapy thereafter.

He pointed out, however, that the optimal antithrombotic regimen after LAA occlusion has not been evaluated in a randomized trial.


ADRIFT was designed as a pilot study to start to fill that evidence gap. Conducted at 10 French centers, the trial included patients who had undergone successful LAA closure. They were randomized into three groups:

  • Rivaroxaban 10 mg (37 patients)
  • Rivaroxaban 15 mg (34 patients)
  • DAPT with aspirin 75 mg and clopidogrel 75 mg (33 patients)

Roughly half of patients had a prior history of stroke and most (87%) had a permanent contraindication to oral anticoagulation. Average CHA2DS2-VASc scores ranged from 4.5 to 4.7 and average HAS-BLED scores ranged from 3.5 to 3.8. The most common indication—in two-thirds of patients—was a previous major ISTH major bleed.

About two-thirds of occlusions were performed with the Amplatzer Amulet device (Abbott), with the rest done with Watchman. The rate of procedure-related complications—which included four groin hematomas, a blood transfusion/groin hematoma, a pseudoaneurysm, and a transfusion—ranged from 6% to 8% across groups.

Montalescot said thrombin generation, as assessed by levels of prothrombin fragments 1 and 2, was selected as the primary endpoint because prior research has shown this to be associated with venous and arterial thrombotic events and mortality. Also, he explained, it is higher when antithrombotic treatment fails and thrombus is detected in the LAA.

Thrombin generation at 10 days was highest in the DAPT group, with significant reductions with both rivaroxaban 10 mg (P = 0.00009) and rivaroxaban 15 mg (P = 0.00002). There were no major differences between the rivaroxaban groups. Similar patterns—though not significant—were also seen for other markers of thrombin generation, including thrombin-antithrombin complex and D-dimers.

The study was not designed to evaluate the impact on clinical events, and by 3 months, there were very few: one death in the rivaroxaban 15 mg group plus one MI and one stroke/TIA in the rivaroxaban 10 mg group.

ISTH major or clinically significant bleeding was seen in 24% of those taking the lower dose of rivaroxaban, 11% of those taking the higher dose, and 27% of those taking DAPT; the across-group differences were not significant.

The two cases of device thrombosis observed during the study occurred in a 79-year-old woman implanted with the Amplatzer Amulet device and a 69-year-old women implanted with the Watchman device; both were taking DAPT.

Commenting for TCTMD, Geoffrey Barnes, MD (University of Michigan, Ann Arbor), said it was interesting that the 20-mg standard dose of rivaroxaban for stroke prevention in A-fib was not used in the study. “But even with the two reduced doses, either 10 mg or 15 mg of rivaroxaban, you see a pretty significant reduction in the thrombin generation,” he said. “And so to me it sort of raised this question of: is using an anti-Xa specific agent like rivaroxaban sufficient for inhibiting thrombus formation and some platelet activity that can occur when you put a foreign device inside the heart?”

ADRIFT is too small to induce a change in practice at this point, Barnes said. “But I do think it lays the groundwork for us to move forward with a larger prospective study comparing a low or intermediate dose of rivaroxaban against the standard antiplatelet regimen. And if in a larger cohort of patients where they’re randomized to these two doses we see good clinical outcomes, it might change the strategies that we use, especially if the bleeding profile is better with the single anticoagulant rather than dual antiplatelets.”

Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • Montalescot G. Low-dose rivaroxaban after left atrial appendage closure: the randomized ADRIFT study. Presented at: ESC 2019. September 2, 2019. Paris, France.

  • ADRIFT was funded by Bayer and ACTION.
  • Montalescot reports receiving research grants to his institution or consulting/lecture Fees from Abbott, Amgen, Actelion, the American College of Cardiology Foundation, AstraZeneca, Axis-Santé, Bayer, Boston Scientific, Boehringer Ingelheim, Bristol-Myers Squibb, Beth Israel Deaconess Medical Center, Brigham and Women’s Hospital, China Heart House, Daiichi Sankyo, Idorsia, Elsevier, Europa, Fédération Française de Cardiologie, ICAN, Lead-Up, Medtronic, Menarini, MSD, Novo Nordisk, Partners, Pfizer, Quantum Genomics, Sanofi, Servier, and WebMD.
  • Barnes reports having served as a consultant for both Pfizer/Bristol-Myers Squibb and Janssen.