QRS characteristics fail to reliably identify ventricular tachycardias that require epicardial ablation in ischemic heart disease.

OBJECTIVES: We tested proposed algorithms for idiopathic and nonischemic tachycardias for their ability to identify epicardial LV-VT origins.
BACKGROUND: Several ECG features have been reported to identify epicardial origins for left ventricular tachycardias (LV-VTs) in the absence of myocardial infarction. Only limited data exist in postinfarction patients.
METHODS: The QRS features of 24 VTs that were ablated from the epicardium and 39 left ventricular VTs ablated from the endocardium were retrospectively analyzed for various 12-lead ECG features previously reported.
RESULTS: No ECG feature consistently predicted an epicardial LV-VT origin in infarct-related tachycardias, with epicardial VTs showing slightly longer QRS durations (189 ± 32 ms in epicardial vs 179 ± 37 ms in endocardial, P = 0.28). Pseudo-delta duration was 38 ± 27 versus 47 ± 27 ms (P = 0.2), intrinsicoid deflection time 93 ± 35 versus 86 ± 32 ms (P = 0.4), shortest RS 97 ± 38 versus 99 ± 32 ms (P = 0.77), and median deflection index 0.82 ± 0.25 versus 0.87 ± 0.22 (P = 0.43). The finding of a Q wave in lead I and the absence of a Q wave in the inferior leads failed to predict an epicardial origin in superior LV-VT sites. Q waves in any inferior lead and aVR/aVL-ratio<1 were not specific for an epicardial origin in inferior sites (all P = ns). Furthermore, all inferior LV-VTs showed a Q wave in the inferior leads which correlated with pre-existing Q-waves in sinus rhythm (P = 0.045).
CONCLUSION: Proposed 12-lead ECG features for differentiation of epicardial versus endocardial sites for nonischemic LV-VTs do not reliably identify VTs that require ablation from the epicardium. Endocardial mapping should be the first approach to catheter ablation for VTs in patients with ischemic heart disease.