2D Isochronal Correlation Method to Detect Pacing Capture during Ventricular Fibrillation

Abstract

During ventricular fibrillation (VF), a portion of myocardial tissue can be captured by pacing at a rate near the fibrillation rate. Interruption of ventricular fibrillation can be favored by achieving stable myocardial capture, lowering the energy required for electrical cardioversion. Existence of myocardial capture during electrical stimulation is determined by visual inspection of electrograms (EGM) by an experienced observer. The objective of this work is the development of a semi-automatic method for the detection of myocardial capture based on the 2D correlation of isochronal maps. In 4 isolated rabbit hearts VF was induced by ventricular pacing with an increasing rate. An array of 128 sensing electrodes plus a central electrode for pacing was used. For each experiment, an isochronal capture template (TImap) was computed by using isochronal maps corresponding to pacing before the induction of VF. After VF induction, epicardial pacing was delivered during 15 seconds at the same rate of the spontaneous fibrillatory rate and 10% higher. An experienced observer visually inspected the recordings and identified which stimulations resulted in capture. Isochronal maps were generated by automatically detecting the myocardial activation during VF. Detections of the algorithm were compared with captures previously identified by the observer. Our algorithm discarded 73.35% of non-capture cases and missed only one capture case reducing the time required for analyzing myocardial capture experiments. This is the first method that makes use of 2D correlation to detect myocardial capture during VF.