br Material and methods br Results br Discussion

Material and methods
Results
Discussion
Conclusions
Conflict of interest
Acknowledgments
Introduction Matching the morphology of paced complexes to tachycardia complexes is useful for mapping arrhythmias. In the atrium, pace mapping is useful for focal atrial arrhythmias in order to identify the origin [1], and in macro-reentrant atrial arrhythmias to identify the critical isthmus [2,3]. However, pace mapping in the atrium is limited by its inability to accurately match P-wave morphology obtained on surface electrocardiograms (ECGs), resulting in poor spatial topoisomerase inhibitors [4]. The objective of this study was to determine if matching the morphology of bipolar electrograms recorded between a set of widely spaced electrodes in the atrium can improve the resolution of pace mapping.
Materials and methods Fifteen patients were included in the study (age, 48±13 years; nine men). Six patients underwent ablation of accessory pathways, while the other nine had atrioventricular nodal reentrant tachycardia and underwent slow pathway ablation. A quadripolar catheter in the right atrial appendage (RAA) and a decapolar catheter in the coronary sinus (CS) were usually used for the electrophysiology study, and these were retained in position. A decapolar catheter with 2–5–2mm spacing (Webster CS catheter, Biosense Webster, Diamond Bar, CA, USA) was positioned along the lateral right atrium with the distal electrode in the low lateral right atrium, as shown in Fig. 1. Care was taken to ensure that all the electrodes were in contact with the atrial wall, and that the catheter position was stable. A quadripolar catheter was placed in the inferior vena cava to provide a reference electrode for unipolar pacing. The distal electrode (low lateral right atrium) or the most proximal electrode (high lateral right atrium) of the decapolar catheter was randomly chosen to simulate the tachycardia focus. Unipolar pacing was performed from this electrode with the distal electrode from the catheter in the inferior vena cava as a reference. Pacing threshold was determined at a pulse width of 2ms. Pacing was performed at 1.5 times the threshold for 15s at a cycle length of 400ms and was recorded as a template. Unipolar pacing was subsequently performed from each of the 10 electrodes of the decapolar catheter in a computer-generated random order for 15s, each at 1.5 times the diastolic threshold. During each of these pacing trains, the 12-lead ECG and a novel set of three widely spaced bipolar electrograms (EGMs) were recorded on the electrophysiology recording system (Labsystem Pro, Bard) at a sampling rate of 1000Hz. The widely spaced bipolar EGMs were recorded in the following configurations: (a) distal RAA to proximal CS, (b) proximal CS to distal CS, and (c) distal CS to distal RAA. ECG signals were filtered between 0.5 and 250Hz, while bipolar EGMs were filtered between 30 and 250Hz. The P-wave morphology in the 12-lead ECG and the bipolar recordings were matched manually by a trained electrophysiologist and by automated computerized analysis. For manual matching, the observer was presented with the waveforms showing the 12-lead ECG and the three bipolar EGMs printed at a speed of 100mm/s, with adequate gain to properly visualize the P waves. The observer was blinded to the location of the electrode from which the recording was made. Pace matching was performed by assessing the amplitude of the P wave and comparing the different components of the waveform. The largest deflection was termed the major component, and smaller, additional components were labeled minor components. Accordingly, the match between the template and the recording from any location was classified as follows: Stored ECGs and EGMs were exported for automated analysis. The details of the algorithm are shown in Fig. 2. Briefly, P waves were separated, and all the P waves from the 15-s recording were averaged for each of the test and template recordings. Intra-sample correlation was calculated as the correlation of P waves within each group to the average P wave, while inter-sample correlation was calculated as the correlation of the average test and template P waves. Based on the inter-sample and intra-sample correlations, a match score was calculated as follows: