Remote AF screening with ECG patch nudges up diagnosis numbers

A remote screening method using a wearable ECG patch yields a modest increase in atrial fibrillation (AF) diagnoses among older adults at risk of stroke, according to the AMALFI trial.

The rate of AF detection at 2.5 years was 6.8 percent in the ECG patch arm vs 5.4 percent in the usual care arm, translating to a 26-percent increase (95 percent confidence interval [CI], 1.02–1.57) in the ratio of proportions in favour of the ECG patch (p=0.03). [JAMA 2025;doi:10.1001/jama.2025.15440]

Results were consistent across subgroups defined by age (p=0.78 for heterogeneity) and sex (p=0.06 for heterogeneity), said principal investigator Prof Louise Bowman from the University of Oxford in Oxford, UK, who presented AMALFI at the ESC annual meeting.

At 2.5 years, mean exposure to oral anticoagulation was significantly longer in the ECG patch arm than the usual care arm, at 1.63 vs 1.14 months (difference, 0.50 months, 95 percent CI, 0.24–0.75; p<0.001). “This suggests that general practitioners (GPs) were clearly responding appropriately to the patch findings and starting the preventative treatments that were suitable,” Bowman noted.

Patch-detected AF

Among the participants with ECG patch-detected AF, 57 percent had their arrhythmia detected on the very first day of wearing the patch. AF burden was below 10 percent for more than half of the detected cases (55 percent) and 100 percent for roughly a third of the cases (33 percent).

The longest AF episode was at least 24 hr in 42 percent of the participants, between 6 and 24 hr in 20 percent, between 6 min and 6 hr in 29 percent, and between 30 s and 6 min in 9 percent.

The patch detected several conditions that were more or less clinically relevant, including advanced AV block, pauses of >6 s, and supraventricular tachycardia, among others.

“Where there were findings that were clinically important, we made sure those were reported back to the GP urgently so that they could act on them appropriately,” Bowman said.

By 2.5 years, stroke occurred in 2.7 percent of participants in the ECG patch arm and 2.5 percent in the usual care arm (event rate ratio, 1.08, 95 percent CI, 0.77–1.51). A total of 4.1 percent and 5 percent of participants in the respective arms had died.

A simple remote trial

“One of the key aspects of AMALFI was making it exceptionally simple for the participants to take part,” according to Bowman.

The investigator, along with her colleagues, performed an electronic health record search across 27 UK primary care practices and identified a high-risk population of individuals aged ≥65 years with a CHA2DS2VASc score of ≥3 for men or ≥4 for women and who had no prior AF or atrial flutter.

Invitations to participate were mailed out to 22,044 individuals, of which 5,116 responded. A total of 5,040 participants (mean age 78 years, 47 percent female, median CHA2DS2VASc score 4) met the inclusion criteria and were randomly allocated to either the ECG patch arm (n=2,520) or the usual care arm (n=2,520).

Participants in the EGC patch arm received a noninvasive, single-lead ECG patch by mail. They were asked to self-apply the patch, undergo continuous monitoring over a 14-day period, and then return the patch. Those in the usual care arm received a letter confirming their allocation and were not required to undertake any further action.

Bowman highlighted that 84.4 percent of the participants in the ECG patch arm wore the patch as instructed and returned it after 14 days.

“This remote approach was clearly very acceptable, with the participants adhering to the intervention very well. It suggests that this approach to screening could be a highly scalable approach, possibly for a population level intervention,” she said.

No definitive answer

Bowman acknowledged that while remote screening with the ECG patch successfully found more cases of AF, it’s still unclear how much of an impact this screening will have on preventing strokes in this high-risk population over the long term.

“But we’re doing a longer-term follow-up in health economic analyses, and we hope to be back in a couple of years with updated information so that we can further contribute to this story,” she said.

“In the meantime, we’ve demonstrated an effective approach to screening that’s clearly scalable and potentially quite low cost for a healthcare system, using remote digital technologies and electronic health records to help us improve the care for our patients,” she concluded.