Spatial Characterization of Electrogram Morphology from Transmural Recordings in the Intact Normal Heart

PURPOSE Unipolar (UE) and bipolar electrograms (BE) are utilized to identify arrhythmogenic substrate. We quantified the effect of increasing distance from the source of propagation on local electrogram amplitude; and determined if transmural electrophysiological gradients exist with respect to propagation and stimulation depth. METHODS Mapping was performed on 5 sheep. Deployment of >50 quadripolar transmural needles in the LV were located in Cartesian space using Ensite. Contact electrograms from all needles were recorded during multisite bipolar pacing from epicardial then endocardial electrodes. Analysis was performed to determine stimulus distance to local activation time, peak negative amplitude (V-P), and peak-peak amplitude (VP-P) for (1) unfiltered UE, and (2) unfiltered and 30 Hz high-pass filtered BEs. Each sheep was analysed using repeated ANOVA. RESULTS Increasing distance from the pacing sites led to significant (p<0.01) attenuation of UEs (V-P = 7.0±0.5%; VP-P = 5.4±0.3% per cm). Attenuation of BE with distance was insignificant (Vp-p unfiltered = 2.2±0.5%; filtered = 1.7±1.4% per cm). Independent of pacing depth, significant (p<0.01) transmural electrophysiological gradients were observed, with highest amplitude occurring at epicardial layers for UE and endocardial layers for BE. Furthermore, during pacing, propagation was earlier at the epicardium than endocardial layer by 1.6±2.0 ms (UE) and 1.4±2.8 ms (BE) (all p>0.01) during endocardial stimulation, and 2.3±2.4 ms (UE) and 1.8±3.7 ms (BE) during epicardal stimulation (all p<0.01). CONCLUSIONS Electrogram amplitude is inversely proportional to propagation distance for unipolar modalities only, which affected V-P>VP-P. Conduction propagates preferentially via the epicardium during stimulation and is believed to contribute to a transmural amplitude gradient.