A novel algorithm for determining endocardial VT exit site from 12-lead surface ECG characteristics in human, infarct-related ventricular tachycardia.

INTRODUCTION: Characteristics of the 12-lead ECG during VT are used to guide initial placement of mapping catheters in endocardial ventricular tachycardia (VT) ablation. Previously constructed algorithms for guidance in human infarct-related VT are limited to patients known to have anterior or inferior infarcts only. We hypothesized that 12-lead ECG characteristics could be used to determine VT exit site in patients with all types of infarction of unknown location. METHODS AND
RESULTS: From noncontact activation maps of 121 LV VT in 51 patients undergoing catheter ablation, VT exit sites were determined and correlated with ECG characteristics according to bundle branch block configuration, limb lead polarity and patterns of precordial R-wave transition. Eight ECG patterns were identified that accounted for 71% of all VT and gave a positive predictive value (PPV) > or =70% using the first two criteria. No correlation was found with patterns of R-wave transition. Using these criteria an algorithm was developed, which was then applied prospectively and blinded to a further 17 VT in 11 patients. Of the 15 VT (88%) to which the algorithm predicted an exit site location (with a PPV > or =70%), 14 VT (93%) were correctly predicted by the algorithm.
CONCLUSION: This algorithm can be used to predict endocardial LV VT exit site location in patients undergoing catheter ablation of VT without knowledge of or reference to infarct location, and can be applied to patients with posterior and/or multiple sites of infarction.