Enter Note
Subclinical AF and Stroke: Is There a Safe AF Burden?
View session detail
Author Block: S. kakarla1, J. Ariyaratnam2, H. Iwakawa3, A. Elliott2, J. Shroff2, M. E. Middeldorp4, J. V. Freeman5, A. M. Russo6, S. Mittal7, E. Fu8, K. Anne9, T. Baykaner10, B. A. Steinberg11, D. Benjamin12, L. Robin12, J. P. Piccini13, T. F. Deering14, P. Sanders2; 1Department of Cardiology, Centre for Heart Rhythm Disorders, Adelaide University, Adelaide, AUSTRALIA, 2Adelaide University, Adelaide, AUSTRALIA, 3Adelaide university, Adelaide, AUSTRALIA, 4Adelaide University, Adealaide, AUSTRALIA, 5Yale Univ School Medicine, New Haven, CT, 6Cooper Medical School of Rowan University, Moorestown, NJ, 7Valley Health System, Paramus, NJ, 8OhioHealth, Columbus, OH, 9Medical University of South Caroline, Charleston, SC, 10Stanford University, Palo Alto, CA, 11Denver Health Medical Center, Denver, CO, 12PaceMate, Clearwater, FL, 13Duke University, Chapel HIll, NC, 14Piedmont Heart Institute, Atlanta, GA.
Disclosure Block: S. kakarla: None.
Background: Subclinical atrial fibrillation(SCAF) forces a binary anticoagulation decision from a continuous measurement. Contemporary data use different thresholds, ranging from 5 minutes to 24 hours, and stroke risk across the full burden range remains uncharacterized. Randomised trials of oral anticoagulation (OAC) produced modest, discordant results, with no burden threshold identifying who would benefit most. Methods: Retrospective cohort study of PaceMateLIVE, a vendor-neutral US remote monitoring database (79 centres, 32 states, 2018-2024). Of 187,717 monitored patients, 75767 had device-confirmed SCAF. A fixed 180-day landmark design separated burden ascertainment from outcome accrual, yielding a cohort of 41,681 never-anticoagulated patients. AF burden was duration-corrected at the episode of transmission-interval level. Incident ischemic strokes were analyzed using restricted cubic spline Cox models, censoring-aware discrimination, and propensity-matched cut-off comparisons. Net clinical benefit compared the stroke rate reduction with the excess major bleeding observed with apixaban versus aspirin in a randomised trial of SCAF (ARTESiA).Results: Among 41,681patients (30% female, median CHA2DS2-VASc score 3.0 [IQR 2.0- 4.0], with median follow-up 1.8 years [IQR 1.0-2.9]), 1,023 strokes occurred. Adjusted stroke risk rose continuously with burden, with no inflection point (non-linearity P=0.006) consistently across CHA2DS2-VASc strata(Interaction P=0.19). Burden alone discriminated poorly (AUC, 0.59) and improved with the addition of the CHA2DS2-VASc score (AUC, 0.664). Propensity-matched hazard ratios rose with wider cutoff separation, from 1.54 (95% CI, 1.31-1.81) at a 1% cutoff to 2.53 (95% CI, 2.00-3.20) at a 5% cutoff (log-rank P<.0001). OAC initiation was associated with lower stroke risk (HR 0.5; 95% CI 0.41-0.61) and flattened the burden gradient. Net clinical benefit was positive only where both burden and clinical risk co-occurred. Conclusions: In SCAF, stroke risk rises continuously from zero AF burden. Neither burden nor clinical risk alone identifies who should be anticoagulated. Anticoagulation provides a net benefit at CHA2DS2-VASc ≥ 5 when burden is below 5% and >3 above it.

