Direct Oral Anticoagulants Can Continue During Cardiac-Device Surgery

Patrice Wendling

November 12, 2017

ANAHEIM, CA — Continuing direct oral anticoagulants (DOACs) did not reduce hematoma rates in patients at moderate to high risk of thromboembolism undergoing cardiac-device surgery, according to results of BRUISE CONTROL-2[1].

The rate of clinically relevant hematoma was identical at 2.1% with continued or interrupted anticoagulation with dabigatran (Pradaxa, Boehringer Ingelheim), rivaroxaban (Xarelto, Bayer/Janssen Pharmaceuticals), or apixaban (Eliquis, Bristol-Myers Squibb/Pfizer).

Further, there was no difference between the two strategies in all secondary outcomes including stroke/transient ischemic attack (TIA).

"Operating with continued DOAC should not be considered specifically as a strategy to reduce hematoma rates, but I think the take-home message for clinicians is that either strategy may be reasonable depending on the clinical scenario," co–principal investigator Dr David H Birnie (Ottawa Heart Institute, ON) said here at the American Heart Association 2017 Scientific Sessions.

 Dr David H Birnie

Although counterintuitive, the investigators hypothesized that performing device surgery without DOAC interruption would result in fewer hematomas, having shown in the original BRUISE CONTROL trial that continuing warfarin was safer than heparin bridging and reduced the relative risk of pocket hematoma by 81% in patients receiving a pacemaker or implantable cardioverter defibrillator (ICD).

"Actually, I thought it was a positive study in the sense that you can continue the [novel oral anticoagulants] NOACs. That's a big deal," Dr Timothy J Gardner (Christiana Care Health System, Newark, DE) told | Medscape Cardiology. "The reason they're on NOACs is because of the risk of thromboembolic complications, and what the study showed is that it didn't result in bleeding. So it's safer for the patient, and the bridge therapy was problematic because of the heparin-induced bleeding."

Dr Frank Sellke (Brown University Medical School, Providence, RI), who also was involved in the study, said a substantial number of patients remain on warfarin in the US and that there's real concern about performing surgery on or with interrupted DOACs, especially as a reversal agent is only available for dabigatran.

"It was a negative trial, but it can change the way we practice," he said in an interview. "There's always concern about stopping anticoagulant drugs in patients who have an artificial heart valve or atrial fibrillation, but if it shows no difference between groups, that can streamline the way we treat these patients or treat them more effectively."

During his presentation, Birnie also remarked on the uncertainty of DOAC management at device surgery and balancing the risk of thromboembolism against perioperative bleeding. Even brief, temporary interruptions have been associated with a threefold increase in stroke. But on the other hand, pocket hematomas may require prolonged interruption of anticoagulation, which increases the risk of thromboembolism and in extended follow-up of BRUISE CONTROL was associated with a more than sevenfold increased risk of serious device infection.

BRUISE CONTROL-2 randomized 662 patients with a CHA2DS2-VASc score ≥2 (mean 3.9) to continued DOAC throughout the surgical period or interruption of rivaroxaban or apixaban for 2 days before surgery or dabigatran based on their glomerular filtration rate (GFR). All three drugs were resumed within 24 hours after surgery.

Use of an intrapocket prohemostatic agent and postoperative pressure dressing was more common in the continued DOAC group (both P=0.035).

The procedure was not modified because of use of the DOACs, but surgeons were allowed to use Bovie cauterization if that was normal practice, although "we don't think that introduced any bias," Birnie said in an interview.

The incidence of clinically relevant hematoma, the primary outcome, was low at just seven patients in each group, with no significant differences in its individual components.

Results, Continued vs Interrupted DOACs

End Point Continued DOAC (n=328) Interrupted DOAC (n=334) P
Clinically significant hematoma 7 (2.1) 7 (2.1) 0.973
Prolonged hospitalization 1 (0.3) 2 (0.6) 1.00
Interrupted anticoagulation 7 (2.1) 7 (2.1) 0.973
Required reoperation 2 (0.6) 1 (0.3) 0.621
All-cause mortality 2 (0.6) 1 (0.3) 0.621
Cardiac tamponade 1 (0.3) 0 0.496
Pneumothorax 2 (0.6) 0 0.245
Stroke 1 (0.3) 1 (0.3) 1.00
TIA 0 0

The trial was terminated early after a second prespecified interim analysis of data from 590 patients showed no difference in outcomes.

Although it is a negative trial, "I think it's good news in that the [hematoma] event rates in both arms were much lower than in BRUISE CONTROL-1, and it's an important complication to prevent. Second, it gives physicians and patients options for management," Birnie said in an interview.

He suggested physicians may want to operate with continued DOACs in patients, for example, with complete heart block or unstable temporary pacing or in those who've recently undergone atrial ablation, are scheduled for concomitant cardioversion or defibrillation testing, or have a high CHA2DS2-VASc score.

During a press conference on the late-breaking clinical trial, discussant Dr Mina K Chung (Cleveland Clinic, OH) said, "This actually is a very timely randomized trial because now we have DOACs that have come on the scene and are very frequently used."

She added, "Either continued or interrupted NOACs are acceptable for CIED procedures because there's no increase in [cerebrovascular accident] CVA or TIA, and no increase in clinically significant hematomas."

She noted that the absence of differences in stroke/TIA incidence with briefly holding NOACs "may have implications for other procedures," although study power was likely limited—and further limited by premature stopping of the study. The study also did not address periprocedural NOAC management for other higher-risk procedures, patients with active device infection or need for lead extraction, a GFR<30 mL/min, or with rheumatic valve disease.

The study was supported by a grant from the Heart and Stroke Foundation of Canada, with additional funding from Boehringer Ingelheim, Bayer HealthCare AG, and Bristol-Myers Squibb. Birnie reports receiving research grants from Bayer, Bristol-Myers Squibb/Pfizer, and Boehringer Ingelheim.

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology,  follow 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.