An Accord Between Man and Machine: Concordance Between Traditional and Novel Mapping Techniques for Atrioventricular Nodal Reentrant Tachycardia Ablation.

Introduction Catheter ablation has evolved rapidly, starting with conventional anatomic techniques, followed by electrogram mapping, and now isochronal late activation mapping techniques are currently in practice. Success rates of ablation were higher with electrogram mapping compared to conventional anatomic mapping. Conventional techniques performed by an experienced operator have not previously been compared to novel mapping techniques in this cohort. Methods A total of 14 consecutive patients underwent Atrioventricular Nodal Reentry Tachycardia (AVNRT) (supraventricular tachycardia with ventriculoatrial (VA) interval <70 ms) ablations, where the operator predicted slow and fast pathway collision points, and a sinus collision mapping was also obtained. Ablation was performed with the operator blinded to mapping. Criteria for successful prediction were an ablation point within 4 mm of machine prediction, with a post-ablation junctional response; slow pathway elimination, confirmed by the absence of an Atrio-His jump with or without an echo beat; and non-inducibility of AVNRT. Other secondary outcomes included age, sex, total radiofrequency (RF) ablation time, number of RF applications, total fluoroscopy time, dose, and other postoperative complications or death. Results Operator prediction of sinus collision location coincided with machine prediction in 85.7% of cases. Regarding patient demographics, 57% of the population were female, with a mean age of 60 years. The average distance from operator prediction to machine prediction was 1.75 mm. The percentage of junctional rhythm post-ablation in concordant patients was 83.3%. The mean ablation time was 97 seconds, with seven RF applications on average. Fluoroscopy was used in two patients, with minimal exposure. No post-procedure complications, such as pericardial effusion or atrioventricular (AV) block, were noted. Conclusion Conventional techniques were not previously compared with novel mapping techniques. In our retrospective cohort study, there was a concordance of 85.7% between an experienced operator and an algorithm-predicted model. The distance between predicted and actual ablation points was close. Although no concrete predictions can be made given our limited retrospective data, with many limitations, novel mapping techniques are useful tools that currently supplement AVNRT ablations and will likely play a crucial role in the future.