A Four-Stepwise Electrocardiographic Algorithm for Differentiation of Ventricular Arrhythmias Originated from Left Ventricular Outflow Tract.

Abstract

Several electrocardiographic algorithms have been proposed to identify the site of origin for the ventricular arrhythmias (VAs) from the left ventricular outflow tract (LVOT) versus right ventricular outflow tract. However, the electrocardiographic criteria for distinguishing VAs originated from the different sites of LVOT is lacking. We aimed to develop a simple and efficient ECG algorithm to differentiate LVOT VAs originated from the aortic root, AMC and LV summit. We analyzed 12-lead ECG characteristics of 68 consecutive patients who underwent successful radiofrequency catheter ablation of symptomatic VAs from LVOT. Patients were divided into RCC (right coronary cusp) group (n = 8), the L-RCC (the junction between the LCC and RCC) group (n = 21), the LCC (left coronary cusp) group (n = 24), the aortomitral continuity (AMC) group (n = 9) and the LV summit group (n = 6) according to the final ablation sites. Measurements with the highest diagnostic performance were modeled into a 4-stepwise algorithm to discriminate LVOT VAs. The performance of this novel algorithm was prospectively tested in a validation cohort of 43 consecutive patients undergoing LVOT VAs ablation. Based on the accuracy of AUC, a 4-stepwise ECG algorithm was developed. First, the QS duration in aVL > 134 ms was used to distinguish VAs from AMC, LV summit and VAs from aortic root (80% sensitivity and 76% specificity). Second, the R duration in II > 155 ms was used to differentiate VAs from LV summit and VAs from AMC (67% sensitivity and 56% specificity). Third, the ratio of III/II < 0.9 was used to discriminate VAs from RCC and VAs from LCC, L-RCC (82% sensitivity and 63% specificity). Fourth, the QS duration of aVR > 130 ms was used to discern VAs from LCC and VAs from L-RCC (75% sensitivity and 62% specificity). In the prospective evaluation, our 4-stepwise ECG algorithm exhibited a good predictive value. We have developed a novel and simple 4-stepwise ECG algorithm with good predictive value to discriminate the AVs from different sites of LVOT.