Dynamic evaluation of atrial fibrillation risk factors using multiple sensors of implantable cardioverter defibrillators

Aims: The HEAD2TOES schema was introduced to identify modifiable risk factors (RF) for atrial fibrillation (AF), including heart failure (HF), physical inactivity (PA), and sleep apnea (SA). Modern implantable cardioverter-defibrillators (ICDs) can detect SA, assess HF status, and measure daily PA. This study investigates the relationship between atrial high-rate episodes (AHREs) and ICD-detected surrogates for these RF in HF patients. Methods and results: Data were collected from 411 HF patients with ICDs. The HeartLogic Index measured HF status, Respiratory Disturbance Index (RDI) severe SA (≥30episodes/hour), and accelerometer detected reduced PA (≥1 h/day). Endpoints were daily AHRE burdens ≥ 5 min and ≥ 23 h. Over 26 months of median follow-up, IN-alert-HF state was 13 % of the total observation time, RDI ≥ 30 episodes/h occurred 58 % of time, and 2 % of weekly activity values were ≥ 1 h lower than usual. AHRE burden ≥ 5 min/day occurred in 139 (34 %) patients and ≥ 23 h/day in 68 (17 %). Both IN-alert-state and reduced activity were independently associated with AHRE burden (≥5 min/day and ≥ 23 h/day), while RDI ≥ 30episodes/hour was associated only with AHRE ≥ 5 min/day. We defined a score predicting AHRE burden ≥ 5 min (2*HeartLogic_IN-alert + 1*RDI ≥ 30episodes/hour + 4*Reduction_in_activity ≥ 1 h), and a score predicting AHRE burden ≥ 23 h (1*HeartLogic_IN-alert + 2*Reduction_in_activity ≥ 1 h). Lower score levels (AHRE-5 min < 4 and AHRE-23 h < 2) comprised the largest proportion of follow-up duration (98 %), with higher scores linked to higher incidence rate ratios for AHRE (6.75 [95 %CI:1.88-20.16] and 11.46 [95 %CI:3.34-31.72], respectively). Conclusions: In HF patients, AHRE occurrence is independently associated with ICD-detected HF status, severe SA, and decreased PA. These ICD-indexes might serve as surrogates for HEAD2TOES RF, aiding in continuous AF risk assessment.