Different definitions of complex fractionated atrial electrograms do not concur with the clinical perspective

Complex fractionated atrial electrograms (CFAEs) have been suggested as potential targets for ablation for persistent atrial fibrillation (AF), but there is no consensus on their definition. This study aimed to quantify the impact of different CFAE definitions on automated CFAE detection based on NavX criteria, and compare them with visual assessment. Baseline and post pulmonary vein isolation continuously-recorded bipolar electrograms obtained using a decapolar catheter placed into the coronary sinus from six persistent AF patients were analysed using different time-windows. Automatic CFAE detection algorithm based on selected negative deflections was developed. Three different CFAE definitions, as described in the literature, were evaluated, and electrograms were labelled as fractionated or non-fractionated according to each definition. Fractionation interval and CFAE_mean were assessed based on each definition, followed by a blinded visual inspection of a clinician. The results show that the electrogram window size did not change overall CFAE_mean. However, different CFAE definitions impacted fractionation significantly on both pre-/post-ablation groups. If automated algorithms are to be used in clinical practice, differences in CFAE definitions might have significant impact on CFAEs quantification.