Diagnostic value of ECG characteristics in precordial leads V1-V3 for the diagnosis of the origin of outflow tract ventricular arrhythmias with a lead V3 precordial transition

Funding Acknowledgements Type of funding sources: None. Background and Purpose. To distinguish the origin of outflow tract ventricular arrhythmias (OTVAs) with a V3 precordial transition is still a challenge. To date, numerous diagnostic algorithms have been described, mainly by analysis of leads V1 and V2, while a complete study of lead V3 has never been conducted. Methods. The ECG characteristics of 45 patients (Pts) with a left bundle branch block pattern OTVAs (LVOT 51.1%, RVOT 48.9%) who underwent successful catheter ablation were retrospectively analyzed. The region where the radiofrequency delivery resulted in the persistent suppression of the arrhythmia was identified as the site of origin. For every duration or amplitude variable and for derivative indices, the comparison of means (U Mann Whitney or t-test), the determination of the AUC by ROC curves, and the correlation with BMI and BSA were performed. The odds ratio was calculated for every variable with an AUC ≥ 0.700. The three best results are described. Results. Pts with a LVOT origin were older (59.91 ± 10.48 vs 50.95 ± 15.48, p = 0.027) and predominantly males (82.6% vs 54.5%, p = 0.042) but they shared similar BMI (24.74 ± 2.76 vs 24.09 ± 2.94, p = 0.45) and BSA (1.83 ± 0.12 vs 1.77 ± 0.16, p = 0.157). In leads V1 – V2 the duration but not the amplitude of the R wave showed a significant correlation with the BSA; no variable, with the exception of the S wave amplitude, exhibited an AUC ≥ 0.700. Criteria based on the prevalence of the R wave in those leads were very specific but not sensitive as a V1 – V2 duration index ≥ 50% and a V1 – V2 amplitude ratio ≥ 30% were present only in 13.04 and 26.08% respectively of patients with a LVOT origin. No measurement in lead V3 showed any correlation with auxological characteristics. The three best ECG variables were: 1) the V3 R wave duration index (R wave \ QRS duration); AUC 0.905, LVOT origin if ≥ 50%, OR 74.80 95% CI [7.97 – 701.48], p < 0.001; 2) the V3 R wave duration; AUC 0.900, LVOT origin if ≥ 80 msec, OR 47.25 95% CI [7.73 – 288.82], p < 0.001; 3) the V3 R wave percentage (amplitude of the R wave with respect to the global amplitude of the QRS, expressed as a percentage); AUC 0.888, LVOT origin if ≥ 50%, OR 36 95% CI [6.19 – 209.06], p < 0.001. Based on the V2 transition ratio calculation formula, we calculated the V3 transition ratio (AUC 0.843, LVOT origin if ≥ 1) which was very sensitive but less specific; OR 30 CI [3.32 – 270.37], p = 0.002. To multivariate analysis, only a V3 R wave percentage ≥ 50% proved to be an independent predictor of LVOT origin; OR 9 CI [2.08 – 38.78], p = 0.003, even if the criterion with the highest accuracy was a V3 duration index ≥ 50% (88.89%). Conclusions. Although it has been poorly analyzed in previous studies on the origin of OTVAs, the morphological characteristics of lead V3 seem to provide valid elements for the creation of diagnostic algorithms. Abstract Figure. Comparison of ECG characteristics