To detect subclinical AF, insertable monitor is unequaled

ORLANDO – An insertable continuous ECG monitor detected previously unidentified atrial fibrillation that would otherwise have gone unnoted with other intermittent ambulatory monitoring strategies, in a secondary analysis from the REVEAL AF study.

“If you really want to look for atrial fibrillation because your concern is that the patient has a high-risk profile, and if you saw it you would anticoagulate it and maybe prophylactically use rate control, nothing beats the implanted monitor,” James A. Reiffel, MD, said at the annual meeting of the American College of Cardiology.

^{Bruce Jancin/MDedge News}

Dr. James A. Reiffel

As previously reported (JAMA Cardiol. 2017 Oct 1;2[10]:1120-7), the primary outcome – an AF episode lasting for at least 6 minutes – occurred during the first 18 months of the study in 29% of participants. By 30 months, it was 40%. At ACC 2018, Dr. Reiffel presented a new analysis looking at how the insertable cardiac monitor (ICM) would have stacked up against other device-based strategies aimed at detecting silent AF, including a 30-day implantable memory loop, daily transtelephonic ECG monitoring, and one-time or periodic 24- or 48-hour Holter monitoring.

To accomplish this, he and his coinvestigators conducted modeling studies harnessing the REVEAL AF continuous monitoring data. They looked at how many of the real-world patients found to have AF in the study would have been identified had they instead undergone a one-time recording period lasting 1, 2, 7, 14, or 30 days beginning at the time they would have received their ICM. They also looked at the yield of repeated monitoring strategies, including monthly or quarterly 24- or 48-hour Holter monitoring sessions. They repeated the various simulated monitoring strategies 10,000 times each in order to beef up the sample size and stability of the results.

It was no contest, according to Dr. Reiffel, professor of clinical medicine at Columbia University in New York.

That’s because the median time to AF detection in REVEAL AF was 123 days. Thus, any monitoring strategy of 30 days duration or less was doomed to be of comparatively low yield. Indeed, the 12-month AF incidence rate as detected by ICM in REVEAL AF was 27.1%, compared with 1.1%-13.5% for the various modeled monitoring strategies.