What is needed for a good 'Decaf'?

Catheter-based pulmonary vein (PV) isolation is associated with good clinical outcomes in patients with recurrent atrial fibrillation (AF) and moderate heart disease. 1 , 2 Limited efficacy of PV isolation in patients with structural heart disease and persistent AF might be explained by non-PV-mediated AF or a predominance of a critical substrate maintaining AF. This substrate for AF—thought to consist of focal drivers/rotors, multiple wavelet re-entry, or a three-dimensional labyrinth of narrow waves showing discontinuous lateral conduction—is expected to show considerable inter-individual variation. 3 In order to optimize ablation outcomes in these patients, one could advocate either a uniform, anatomy-based ablation strategy with predefined linear lesions or an electrogram-based ablation approach guided by individual diagnosis of the critical arrhythmogenic substrate. This latter patient-specific strategy has been shown to be highly effective in many atrial and ventricular tachyarrhythmias with reported success rates up to ∼95%. Interestingly, in most of these cases, fractionated electrograms identify the critical substrate and as such the ablation target of the corresponding arrhythmia. 4 Likewise, complex fractionated atrial electrograms (CFAEs) during AF have been proposed as the electrocardiographic representation of the ‘critical fibrillatory substrate’. Konings et al.5 demonstrated that fractionation of extracellular unipolar electrograms during AF is caused by various types of conduction disturbances such as collision, block, pivot points, or slow conduction, all essential for the maintenance of AF. Other studies demonstrated that fractionated fibrillatory electrograms might be recorded in the vicinity of focal AF sources or ganglionated plexi. 6 , 7 Ablation targeting exclusively complex fractionated electrograms (defractionation or ‘de-CFAE’) was reported to be associated with a high success rate in persistent and permanent AF. 8 However, although many fractionated electrograms might point to regions critical for AF (‘active CFAEs’), some CFAEs might be present in regions of structural and anatomical heterogeneity not critical for AF (‘passive CFAEs’). Passive CFAEs can be explained by the complex three-dimensional histological architecture of the atria, resulting in summation and subsequent fractionation of electrograms at adjacent structures. Moreover, structural remodelling in patients with persistent AF (myolysis, cellular uncoupling, fibrosis, dilatation, hypertrophy, or anisotropy) leads to asynchronous and discontinuous propagation, resulting invariably in the fractionation of the bipolar electrogram in most parts of the atria. 9 Because both active and passive CFAEs coexist within the atria, indiscriminate ablation of all CFAEs may compromise safety (impairment of atrial function, collateral injury, atypical flutter) without enhancing efficacy. Therefore, it is essential to develop criteria to distinguish active from passive patterns. Only recently, Takahashi et al.10 identified continuous electrical activity as a significant electrocardiographic marker for active CFAEs (i.e. site-specific AF slowing or termination after local ablation). However, the predictive value of the percentage of continuous activity for favourable ablation regions was only modest. To further enhance the specificity of CFAEs, it would be very helpful to develop tools for an online, objective, and algorithmic description and quantification of electrogram fractionation and to use these tools to describe the spatial and temporal characteristics of different types of CFAEs. Berntsen et al.11 present a detailed analysis of fractionated electrograms in an effort to characterize CFAEs in patients with persistent AF. Their study is an important contribution in several ways. Most importantly, it offers new tools for an objective and automated detection of fractionation; secondly, the study shows region-dependent temporal instability in fractionation characteristics (such as continuous activity percentage). It is tempting to speculate that this temporal variation in fractionation characteristics might explain the moderate predictive value of ‘snapshot’ continuous fractionation as a marker for active CFAEs. Much more work of this type will be required to fully decipher CFAEs. Optimal recording settings including recording time, size of electrodes, unipolar vs. bipolar recording, contact force, adequate filtering, and so on will have to be determined. Correlation of fractionation with underlying structural and anatomical heterogeneity both during sinus rhythm and AF has to be investigated. Ablation studies based on algorithmic-guided CFAE detection are required to reveal the causal relation between the local electrogram morphology and the resulting effect of local ablation. Until then, electrogram-based defractionation will remain hampered by the low specificity of CFAEs to identify the critical AF substrate.