Methods
A derivation cohort of consecutive patients, in whom OTVA ablation was performed at our centre, were retrospectively identified. Patients were included if there was an unambiguous SOO, in whom ectopy was successfully suppressed with radiofrequency (RF) ablation at a single site, with no recurrence during a 30 minute post-ablation waiting period as well as at least 90% reduction in ectopy burden on follow up 3 month Holter monitor. Patients with multiple ventricular arrhythmias were excluded.
For each patient, a single pre-ablation 24-hour ambulatory ECG was retrospectively analysed. Where possible, recordings performed prior to the introduction of anti-arrhythmic drugs (AAD) were selected. For every complete one-hour recording, the ventricular ectopic (VE) total was calculated. Variability in VE across the 24 hour period was assessed using the coefficient of variation (CoV) = standard deviation divided by the mean. Total number of bigeminy/trigeminy episodes and the variability (CoV) of this was also calculated. Percentage ectopic burden, mean heart rate and presence of sustained VT were also analysed. Patients with left and right sided SOO were compared and ROC curve analysis used to determine the optimal predictive parameter values. Circadian patterns in variability were analysed by comparing each patient’s total VE CoV between four quartiles of the day (00:00-06:00, 06:00-12:00, 12:00-18:00 and 18:00-00:00).
Parameters which were found to accurately predict SOO were then applied prospectively to a validation cohort of consecutive patients undergoing their first OTVA ablation. A single 24-hour ambulatory ECG, prior to AAD therapy where possible, was chosen for the primary analysis.
In order to assess inter-Holter reproducibility of the newly derived parameters, a supplementary analysis on any additional ambulatory ECGs was also performed on patients in whom multiple ambulatory monitors had been performed prior to ablation.
The predictive value of the derived parameters in determining SOO was calculated and compared to two well established morphological ECG-based algorithms shown to have the highest accuracy in a recent prospective analysis (5); transitional zone index (TZI) and V2S/V3R index (6,7).
TZI is defined as the precordial chest lead at which R wave transition is observed in the VE minus the equivalent lead for a sinus beat, with a TZI > 0 predicting a right sided SOO. The V2S/V3R index is calculated by dividing the ectopic S-wave amplitude in V2 by the ectopic R-wave amplitude in V3, with a value > 1.5 also predicting a right sided SOO.