LAUSR

In this issue of the Indian Pacing and Electrophysiology Journal, Jason Tri and coll. report their interesting experience in mapping ventricular fibrillation (VF) in dogs [1]. In this model, they demonstrate some intramural differences in activation rate and organisation, as well as between insulated (proximal) and non-insulated (distal) regions of the His-Purkinje network. In addition, some areas have been shown to exhibit periods of regular electrical activity despite continuing fibrillatory pattern on surface ECG [1]. VF is the end-stage of the electro-physiological adventure. After deciphering all kind of supraventricular arrhythmias, after the huge workload dedicated to atrial fibrillation, VF remains the ultimate challenge. Knowledge about the intime mechanisms underlying VF in humans may ultimately lead to a definitive solution for most sudden cardiac deaths, thus representing a major public health issue. Unfortunately, to date, studies about VF mostly involved animal models or human hearts far from usual clinical conditions. Thus, their conclusions may not apply to clinical VF. The study by Jason Tri does notmake exception to the rule, andmoreover, even if interesting, many findings had been in fact already noted in the past. Historically, VF had been classified in type 1 VF (predominantly caused by multiple wandering wavelets) and type 2 VF (when myocardial activation is mainly driven by a mother rotor), which were associated with differences in conduction and repolarisation restitution [2,3]. The fact that VF was sometimes not a random disorganized rhythm came from optical mapping in explanted perfused hearts [2,3]. Then various endocardial and epicardial VF mapping studies have been conducted in humans [4e9], often revealing large sustained and repeated activation wavefronts, together with the existence of a limited number of rotors or localized reentry. However, in most of these studies, the experimental conditions made these unsuitable to human clinical VF in the real life. Only a few studies involved short-living « natural » human VF as occurring in clinical practice [10,11]. According to some of these works, endocardial activation during the initial steps of VF was already not found to be random,more consisting of a few largewavefronts [10]. Non invasive « panoramical » view of the clinical spontaneous human VF process was more recently reported [12], showing rather consistent and fixed drivers and rotor waves. However, these results were achieved after complex mathematical data processing of surface recordings, and it is not clear whether this is a true translation of what is really going on in cardiac activation during VF. In human clinical VF, using simple multi-electrode endocardial mapping, we also previously found fast and regular discrete activation covering the whole duration of every intracardiac recording,