LAA Occlusion During Surgery Cuts Stroke Risk

Patients with atrial fibrillation (Afib) who were already undergoing cardiac surgery for another indication benefited from having the left atrial appendage (LAA) occluded during that surgery, according to results from the LAAOS III trial.

In the multicenter, randomized trial involving participants with Afib and a CHA₂DS₂-VASc score of at least 2, stroke or systemic embolism (SE) occurred in 4.8% in the occlusion group and in 7.0% in the no-occlusion group for a hazard ratio of 0.67 (95% CI 0.53-0.85, P=0.001), reported Richard Whitlock, MD, PhD, of McMaster University in Hamilton, Ontario, Canada, and colleagues.

In addition, a total of 92.1% of the participants received the assigned procedure, and at 3 years (study follow-up was a mean 3.8 years), 76.8% of the participants continued to receive oral anticoagulation.

The main benefit of performing the LAA closure was that the reduction in stroke occurred down the road after the surgery, Whitlock noted in a presentation at the American College of Cardiology (ACC) virtual meeting. The trial results were simultaneously published in the New England Journal of Medicine.

He pointed out that, in the first 30 days post-procedure, 2.2% of the patients who underwent LAA occlusion experienced stroke or SE versus 2.7% of the patients who did not have LAA occlusion. Beyond 30 days, those percentages came in at 2.7% and 4.6%, respectively.

“Surgical left atrial appendage occlusion reduces ischemic stroke by 33%,” Whitlock said. “After the first 30 days, the reduction is 42%. We saw no evidence of adverse effects of doing the additional surgery, and we saw that the surgery benefit is additive to oral coagulation.” The investigators reported that the LAA occlusion required an extra 6 minutes to perform in the study.

However, he stressed that the trial findings should not be interpreted as justification for eliminating oral anticoagulation.

The key is to use anticoagulation — warfarin and the direct oral anticoagulants are highly effective at preventing strokes in patients with Afib, Whitlock also emphasized — in combination with LAA occlusion, as adding on the latter can reduce stroke risk. Whitlock stated that a subgroup analysis indicated that patients who were not on oral anticoagulation at baseline still seemed to derive a benefit of LAA occlusion, although the latter alone would still not match the benefits of oral anticoagulant plus LAA occlusion.

LAAOS III was done at 105 centers in 27 countries, and was stopped based on a recommendation from the data and safety monitoring board after a second planned interim analysis. The primary analysis population consisted of 2,379 patients in the LAA occlusion group and 2,391 in the no-occlusion group. Patients had a mean age of 71 and a mean CHA₂DS₂ -VASc score 4.2; >60% were men.

About two-thirds of the patients underwent a valve procedure, while about 20% had isolated coronary artery bypass grafting, and 23% had isolated valve surgery. About a third had Afib ablation.

The investigators reported that the mean bypass time in the occlusion group was longer at 119 versus 113 minutes in the no-occlusion group. Mean cross clamp time also was longer at 86 versus 82 minutes, respectively (P<0.001 for both).

Whitlock said the incidence of perioperative bleeding, heart failure (HF), or death did not differ significantly between the trial groups. Specifically, SE occurred in 0.3% of patients in each arm, death in 22.6% vs 22.5%, and hospitalization for HF in 7.7% vs 6.8% at follow-up.

Laurence M. Epstein, MD, of Northwell Health in Manhasset in New York, told MedPage Today that the results are likely to change clinical practice, even though LAAOS III did not show a major reduction in absolute numbers of stroke.

“I think the study shows that if you are doing surgery in patients with atrial fibrillation, you might as well perform the occlusion, as long as you are going to do it well, and by doing that you may very well reduce the risk of stroke,” he said.

“This was a ‘belts and suspenders’ type of study; patients not only got the occlusion, they also received anticoagulation,” noted Epstein, who was not involved in the trial. He added that he would have liked to have seen a third trial arm of occlusion without anticoagulation.

Whitlock and colleagues stressed that “LAAOS III did not compare left atrial appendage occlusion with anticoagulation, and it would be incorrect to conclude that occlusion at the time of surgery should be considered as a replacement for anticoagulation…the findings from LAAOS III apply primarily to surgical occlusion of the appendage performed as a concomitant procedure and not to stand-alone surgical or endovascular occlusion.”

Epstein highlighted that the “[LAA occlusion] procedure was done in a variety of ways, which showed that it wasn’t one prescribed method that worked. And whichever way it was done, it didn’t increase the risk of stroke in these patients.”

Whitlock explained that several LAA occlusion techniques were allowed in the trial, but the preferred approach was amputation of the appendage followed by suture closure of the stump.

Epstein said he hopes cardiac surgeons start adding LAA occlusion when appropriate, explaining that “in cases in which patients have a bleed for any reason, and we have to stop anticoagulation, it will be more comforting to know that the appendage has been occluded.”