Back to the First Factor? Predictors of Atrial Fibrillation in Heart Failure

Background: Heart failure is one of the strongest predictors of atrial fibrillation (AF); however, some patients resist the development of AF in the setting of this extensive and established substrate. Methods: 263 consecutive patients implanted with defibrillators were screened for a clinical history of AF. Patients with single chamber (n = 106) or CRT devices (n = 59) and preserved ejection fraction (EF > 45%; n = 28) were excluded. Of the 70 remaining patients, 30 did not have a clinical history or device evidence of AF > 2 min for a mean monitoring period of 327 ± 166 days, whilst 40 had a clinical history of AF, with device confirmation over a mean of 327 ± 287 days. Results: Sixty percent, 25% and 15% of the AF patients displayed paroxysmal, persistent and permanent sub-types of atrial fibrillation. Left atrial diameter was larger in AF patients as compared to controls (Table), but had only moderate predictive ability to segregate AF and control groups (Area Under ROC curve = 0.65). In contrast, other conventional substrate based risk factors for AF such as age, LVEF, underlying patho-physiology of heart failure (dilated or ischemic), percentage ventricular pacing burden and co-morbid disease risk profile were not different between AF patients and controls (Table). Conclusion: Our findings suggest that, in the setting of established substrate, the first factor of electrophysiological change or autonomic balance/triggered activity may persist in the development and sustenance of AF. In addition, regardless of extensive ventricular and atrial substrate, 60% of device implanted heart failure patients are still able to maintain paroxysmal arrhythmia.Tabled 1Heart Failure AF+Heart Failure SRP%CHF1001001.0LA diam. (mm)4.7 ± 0.84.2 ± 0.70.02Age (years)69 ± 968 ± 110.4LVEF (%)29 ± 929 ± 80.7% dilated origin28200.7% >30% V pace36240.3% hypertension63730.3% valvular disease20100.3 Open table in a new tab