Introduction: Electro-anatomical mapping of the atria is used to identify the substrate of atrial fibrillation (AF). Targeting this substrate by ablation in addition to pulmonary vein ablation did not consistently improve outcome in clinical trials. Generally, the assessment of the substrate is based on short recordings (≤10 s, often even shorter). Thus, targeting the AF substrate assumes spatiotemporal stationarity but little is known about the variability of electrophysiological properties of AF over time. Methods: Atrial fibrillation (AF) was maintained for 3-4 weeks after pericardial electrode implantation in 12 goats. Within a single AF episode 10 consecutive minutes were mapped on the left atrial free wall using a 249-electrode array (2.25 mm inter-electrode spacing). AF cycle length, fractionation index (FI), lateral dissociation, conduction velocity, breakthroughs, and preferentiality of conduction (Pref) were assessed per electrode and AF property maps were constructed. The Pearson correlation coefficient (PCC) between the 10 AF-property maps was calculated to quantify the degree spatiotemporal stationarity of AF properties. Furthermore, the number of waves and presence of re-entrant circuits were analyzed in the first 60-s file. Comparing conduction patterns over time identified recurrent patterns of AF with the use of recurrence plots. Results: The averages of AF property maps were highly stable throughout the ten 60-s-recordings. Spatiotemporal stationarity was high for all 6 property maps, PCC ranged from 0.66 ± 0.11 for Pref to 0.98 ± 0.01 for FI. High stationarity was lost when AF was interrupted for about 1 h. However, the time delay between the recorded files within one episode did not affect PCC. Yet, multiple waves (7.7 ± 2.3) were present simultaneously within the recording area and during 9.2 ± 11% of the analyzed period a re-entrant circuit was observed. Recurrent patterns occurred rarely and were observed in only 3 out of 12 goats. Conclusions: During non-self-terminating AF in the goat, AF properties were stationary. Since this could not be attributed to stable recurrent conduction patterns during AF, it is suggested that AF properties are determined by anatomical and structural properties of the atria even when the conduction patterns are very variable.
Keywords: AF dynamics; atrial fibrillation; conduction patterns; mapping; recurrence quantification analysis; stationary patterns.