High Burden of Post-CABG A-fib After Hospital Discharge: SEARCH-AF

Roughly one in five patients who underwent coronary artery bypass graft surgery, with or without valve replacement/repair, developed postoperative atrial fibrillation or flutter in the first month following the procedure, according to the results of the SEARCH-AF study.

Following surgery, atrial fibrillation/flutter with a duration of 6 minutes or longer, or documented by 12-lead ECG, occurred in 19.6% of patients randomized to enhanced cardiac rhythm monitoring (SEEQ; Medtronic or CardioSTAT; Icentia) versus 1.7% of patients discharged home and left unmonitored (P < 0.001).

Researchers say the findings, which was presented as a late-breaking clinical trial at the virtual American Heart Association 2020 Scientific Sessions, underscore the fact that if patients don’t develop atrial fibrillation in the hospital after cardiac surgery, it doesn’t mean they’re in the clear.

“We know that postoperative atrial fibrillation is exceedingly common after cardiac surgery—30% to 50% may develop postoperative atrial fibrillation—but usually people have thought that if patients don’t have it in the immediate post-operative period before discharge, then they’re probably OK,” senior investigator Subodh Verma, MD, PhD (University of Toronto/St. Michael’s Hospital, Canada), told TCTMD. “We’ve had no real guidance on whether such patients develop occult, unrecognized atrial fibrillation in the community after discharge.”

At present, the clinical guidelines provide little or no direction on the optimal duration of monitoring after cardiac surgery, particularly for those patients who are in normal sinus rhythm at the time of discharge, said Verma. If physicians could detect postoperative atrial fibrillation during the subacute phase—ie, in the early weeks after patients have been discharged—there is the potential to identify those who might benefit from oral anticoagulation for the prevention of stroke. While the study was not powered for clinical outcomes, SEARCH-AF is an important first step as it documents the scope of the problem, particularly in patients who are at higher risk of stroke, where it was not “obvious” to clinicians that they would develop atrial fibrillation, said Verma.

What we see here is that this is not something limited to the hospital stay, per se. Subodh Verma

The SEARCH-AF trial, as Verma noted, included an “enriched population” with an elevated risk for stroke, with the 336 randomized patients having a median CHA₂DS₂-VASc score of 4.0. Patients with a history of atrial fibrillation/flutter were excluded from the trial, as were those who developed the condition in hospital.

At 30 days, the rate of atrial fibrillation/flutter was higher in those randomized to cardiac monitoring. When investigators focused on episodes lasting 6 hours or longer, cardiac monitoring picked up events in 8.6% of patients compared with no documented cases in those unmonitored. For events lasting 24 hours or longer, atrial fibrillation/flutter was documented in 3.1% of patients but in no patients randomized to usual care.

“What we see here is that this is not something limited to the hospital stay, per se, and there is a substantial incidence of atrial fibrillation following cardiac surgery that persists into the first 30 days,” said Verma. “The study isn’t powered to determine if that translates into an increased risk of stroke or not, or whether this mandates anticoagulation. But the reality is that in someone with a high CHA₂DS₂-VASc score, if we had found atrial fibrillation postoperatively, I think, in my opinion, it would seriously warrant a consideration for anticoagulation in that setting.”

Need a Large Anticoagulation Trial, but…

Ben Freedman, OAM, MBBS, PhD (University of Sydney, Australia), who commented on the study following the late-breaking presentation, said the episodic duration of atrial fibrillation/flutter is important prognostically, noting that events lasting longer than 24 hours occurred much less frequently in SEARCH-AF than shorter bouts of atrial fibrillation.

In terms of the big picture, said Freedman, SEARCH-AF and other monitoring trials come down to the fundamental question of “to screen or not to screen.” He highlighted a 2014 published in JAMA showing that the cumulative rate of stroke 1 year after cardiac surgery was just 1.0% per year in those who developed new-onset perioperative atrial fibrillation compared with 0.83% in those who didn’t develop atrial fibrillation. That translated into a very small increase in the relative risk of stroke (HR 1.3; 95% CI 1.1-1.6).

If you look harder for [postoperative] atrial fibrillation, you’ll find it. Ben Freedman

To move the field forward, Freedman said, it will be important to determine the risk of stroke in patients with untreated postoperative atrial fibrillation and whether treating it improves outcomes. “If you look harder for [postoperative] atrial fibrillation, you’ll find it, but you’ll need to do a randomized, controlled trial to be sure you’re actually doing the people good by looking for postoperative AF,” he said.

Doing the math on the large number of patients required to show a benefit of this screening approach, Freedman said he thinks such a trial is unlikely to occur. Without a trial, cardiac monitoring for 30 days after cardiac surgery simply catches atrial fibrillation. Physicians, though, won’t have any idea what to do about it.     

For Verma, SEARCH-AF is more than simply seek and ye shall find. “We’re not creating patients,” he said. “We’re trying to find ways to identify people who may be at high risk if they were otherwise not identified.” The SEARCH-AF study, he stressed, identifies a significant, unrecognized burden of atrial fibrillation, and like Freedman, he highlighted that monitoring captured patients with longer durations of atrial fibrillation (≥ 24 hours), which is known to be associated with poor clinical outcomes. “It’s not just that we look more and find more. That’s not the case because even if you compare this to other strategies, like the Apple Watch strategy and other monitoring strategies, there is still a significantly higher burden of atrial fibrillation in this population.”   

Verma agreed about the need for a clinical trial to determine if anticoagulation is beneficial to reduce the risk of stroke. Still, he said that from a clinical perspective, it lowers his threshold for cardiac monitoring to watch for as yet unidentified atrial fibrillation.

Deepak Bhatt, MD, MPH (Brigham and Women’s Hospital, Boston, MA), one of the SEARCH-AF investigators, said cardiologists worry a lot about atrial fibrillation in cardiac surgery patients. “We give it an enormous amount of concern, because we don’t want our patients after heart surgery to have a stroke,” he told TCTMD. “Now some of that is related to the procedure itself, cross-clamping the aorta and mechanical issues, but some of it is atrial fibrillation and some of it occurs after hospitalization. I think the study nicely quantified how much atrial fibrillation was occurring silently in these patients if we monitor them.”

Both Verma and Bhatt stressed that SEARCH-AF is just the first step, which was identifying the scope of the atrial fibrillation burden after surgery, and the next will involve studying how to address it.

“Even now in terms of how actionable this is, if there is a patient that has a really high CHA₂DS₂-VASc score and they are having lots of silent atrial fibrillation, might I act on that? Maybe. That situation comes up right now when we serendipitously detect atrial fibrillation in patient with a pacemaker or an ICD who comes in for monitoring. . . . It’s not always clear what to do when you pick up atrial fibrillation like that, but it happens and we just try to deal with it.”