Site-specific twelve-lead ECG features to identify an epicardial origin for left ventricular tachycardia in the absence of myocardial infarction

Background: Identification of an epicardial origin for left ventricular tachycardia (LV-VT) based on electrocardiogram (ECG) criteria facilitates the approach to catheter ablation. Reported criteria, although helpful, may not apply uniformly to all LV regions. Objective: We hypothesized that unique region-specific ECG patterns identify epicardial LV-VTs in patients without myocardial infarction. Methods: The QRS morphologies during pace mapping from 402 epicardial and 234 comparable endocardial sites and 19 epicardial VTs were analyzed in 15 patients with respect to morphology and duration of all and components of the QRS. Results: Basal superior (N = 244) and apical superior (N = 141) pace mapping sites showed Q wave in lead I more commonly from epicardial vs corresponding endocardial sites (90% vs 16%, 88% vs 26% respectively; P <.001). The absence of Q wave in leads II-III-aVF identified epicardial basal superior sites, P = .002. Basal inferior (N = 140) and apical inferior (N = 76) epicardial sites showed Q wave in leads II-III-aVF (81% vs 37%, 92% vs 33%, P <.001). These morphologic criteria identified 16 of 19 VTs. The QRS duration was longer from the epicardium, 213 ± 45 ms vs 191 ± 41 ms, P <.001, although significant overlap existed. Reported criteria (pseudodelta wave ≥34 ms, intrinsicoid deflection time ≥85 ms, and shortest RS complex ≥121 ms) were region specific in their ability to identify epicardial origin, with some criteria not having value in specific regions and sensitivity/specificity varying from 14% to 99% and 20% to 94%. Conclusion: ECG features distinguishing epicardial LV-VT are site specific, including the presence or absence of a Q wave in leads that reflect local ventricular activation.