Atrial Fibrillation Begets Atrial Fibrillation Revisited:Early Entrenchment of AF Episodes

The 7-day threshold defining persistent atrial fibrillation (AF) is clinically entrenched but biologically arbitrary. Whether episode termination follows scale-free dynamics or time-dependent entrenchment remains uncertain. To quantify the probability of spontaneous termination and assess its dependence on episode duration using large-scale pacemaker monitoring. Continuous pacemaker monitoring data from the ACaSA study were analyzed. Device-detected AF episodes were reconstructed from daily burden using predefined continuity rules. Termination was modelled as the conditional probability that an ongoing episode would terminate within the subsequent 24 hours. Duration dependence was assessed using a shared frailty Weibull survival model, and episode duration distributions were compared using the Vuong likelihood-ratio test. After excluding patients with single-chamber devices or 100% baseline AF burden, 251 of 281 pacemaker recipients contributed 204,775 patient-days. Device-detected AF occurred in 154 patients, who contributed 7002 reconstructed episodes. Termination probability declined from 91.1% at 1 hour to 72.1% at 12 hours, 33.3% at 24 hours, and 4.3% at 7 days. Episode durations were better described by a lognormal distribution than by a scale-free power-law model. Device-detected AF episodes show marked early entrenchment, with spontaneous termination becoming progressively less likely as episode duration increases. These findings support the concept that "AF begets AF" even at short timescales and suggest that persistence beyond 24 hours identifies a state with a substantially reduced likelihood of near-term spontaneous termination. NCT05127720. NCT05127720.