Incidence and predictors of de novo atrial fibrillation in patients with heart failure undergoing continuous rhythm monitoring

Atrial fibrillation (AF) is associated with adverse clinical outcomes in patients with heart failure (HF). The true incidence of de novo AF in HF, however, remains uncertain. Moreover, the role of clinical variables and circulating biomarkers in AF risk stratification in this population is not well established. This study aimed to determine the incidence of AF in patients with HF using long-term continuous rhythm monitoring and to identify clinical, structural, and biomarker-based predictors. This multicenter, prospective study included patients with HF and reduced or preserved left ventricular ejection fraction (LVEF). Long-term rhythm monitoring was performed using implantable loop recorders or single-chamber defibrillators without atrial leads and with ventricular pacing < 20%. Serial blood samples were obtained at baseline, during follow-up, and during HF decompensation, and analyzed using a multiplex panel of selected biomarkers. Multivariable analyses were performed using both baseline biomarker levels and laboratory values closest to AF onset or HF decompensation. A total of 345 patients were enrolled (79% male; median age 68 years; 53% with LVEF< 40%. During a median follow-up of 756 days, 86 patients (24.9%) developed AF episodes lasting > 6 minutes, and 64 patients (18.5%) died. Hypertension, left atrial volume, beta-blocker use, CA-125, NT-proBNP, COL4A1, and AT1-AA were independently associated with AF, whereas aldosterone receptor antagonist therapy was associated with a lower risk of HF (C-index, 0.74). Patients with persistent AF as the initial manifestation had significantly higher mortality than those with paroxysmal AF or no AF. NT-proBNP and COL4A1 were associated with mortality. Clinically relevant bradyarrhythmias and ventricular tachyarrhythmias were also detected. De novo AF is a common complication in patients with HF. Integrating clinical variables, structural parameters, and biomarkers improves the identification of patients at increased risk of developing AF. Continuous monitoring enables early detection of rhythm abnormalities that may influence therapeutic management.