The irregularity factor: Mechanistic insights into beat-to-beat irregularities in atrial fibrillation cycle length

Measures derived from the atrial fibrillation (AF) cycle length (AFCL) are used to identify regions crucial for AF maintenance or predict (ablation) treatment outcomes. However, mechanisms underlying beat-to-beat irregularities in AFCL are unknown.

We aimed to (1) investigate beat-to-beat changes in AFCL (Δ-AFCL) by relating temporal variation in AFCL with quantified beat-to-beat changes in activation patterns and conduction times (Δ-CTs) and (2) test whether the irregularity factor (IF) (standard deviation of the Δ-AFCL histogram) is indicative of the complexity of activation patterns.

Mapping was performed in 24 patients with long-standing persistent AF (LSPAF) and a reference group (n = 25, patients with induced [acute] AF). At the mapping area center, Δ-AFCL was calculated by subtracting each AFCL from the preceding AFCL and related to causes of Δ-CT (CT_(N) - CT_(N-1)) of fibrillation waves from the electrode array border toward the mapping area center.

The IF was larger during LSPAF (52 ± 17 vs 25 ± 10; P < .001) than during acute AF. Δ-CTs were also larger (25 ± 8 ms vs 19 ± 7 ms; P = .005) and caused by beat-to-beat changes in conduction velocity (36 ± 6 cm/s vs 29 ± 6 cm/s; P = .001) and beat-to-beat changes in direction (94° ± 13° vs 64° ± 22°; P < .001); beat-to-beat changes in path length (13 ± 4 mm vs 14 ± 3 mm; P = .333) were comparable. The IF correlated with conduction block [probability (ρ) = 0.887], lines of conduction block lengths (ρ = 0.900), number of fibrillation waves (ρ = 0.889), and prevalence of focal waves (ρ = 0.609) (P < .001 for all).

The IF indicates the underlying complexity of activation patterns, is higher during LSPAF than during acute AF, and is comparable throughout the atria during LSPAF. It is a novel, easy-to-calculate, objective electrical biomarker, which can be used for, for example, guiding patient-tailored treatment independently of the AF mechanisms.