Activation signature valleys are predictive of macroreentrant atrial tachycardia bottlenecks where uniform low voltage and uniform slow conduction reside

BACKGROUND The activation signature, i.e., the number of recording sites activating per unit time, is useful to detect ventricular tachycardia isthmus location. OBJECTIVE We utilized the activation signature for atrial tachycardia (AT) analysis. METHOD In 40 index cases with macroreentrant AT, plus 4 repeat ablations, it was supposed that circuit bottlenecks, the locations of which are predicted ablation targets, could be defined by local activation time intervals in the cardiac cycle having minimized sites activating. Predicted and actual target parameter values were compared for voltage and electrical conduction, versus peripheral (non-target) sites. RESULTS Of 44 procedures, 35 had 1 predicted target, 7 had 2 targets, and 2 had 3 targets, as determined via the signature (N=55 total targets). In 26/44 procedures, a predicted target approximately overlapped the actual target. The mean peripheral voltage was 0.71±0.47mV, versus 0.29±0.27mV at the actual target and 0.28±0.21mV at the predicted target (p<0.001). The peripheral voltage standard deviation was 1.14±1.02mV, versus 0.20±0.15mV at the actual target and 0.26±0.22mV at the predicted target (p<0.001). The coefficient of variation of nearest neighbor activation time difference, with lower values indicative of uniform slow conduction, was 1.99±0.46msec peripherally, versus 1.50±1.08msec at the actual target and 0.73±0.23mV at predicted target (p<0.001). CONCLUSION The activation signature detects potential bottlenecks for predictive targeting in AT patients, using differing criteria and often varying locationally from actual targeting, yet identifying areas with similarly uniform low voltage and uniform slow conduction. Predicted targets often coincide with voltage fissures, where sluggish conduction reinforces the excitable gap.