Automatic measurement of short-term variability of repolarization captures modulation by pacing on human electrograms

Background: Short-term variability of repolarization (STV) increases prior to ventricular arrhythmias in both humans and animal models, making it a promising tool for real-time arrhythmic risk monitoring. Objective: An automatic STV measurement algorithm was developed for intracardiac electrograms (EGMs) to enable integration into cardiac devices. This method previously demonstrated high accuracy in predicting life-threatening ventricular arrhythmias in animals. This study compared the performance of the automatic method to the gold standard on EGMs in humans. Methods: EGM signals were recorded in 14 patients with a dual-chamber implantable cardioverter-defibrillator during de novo implantation (n = 5) or replacement (n = 9) procedures. Recordings were obtained in sinus rhythm (SR), atrial pacing at 80 beats/min, and dual-chamber pacing at 80 beats/min. STV was determined on the EGM from the activation recovery interval with the automatic method (STV-ARI_auto) and with fiducial segment averaging (STV-ARI_FSA), the gold standard. STV-ARI_auto was compared with STV-ARI_FSA for all pacing modes. Results: STV-ARI_auto and STV-ARI_FSA decreased from 0.90 ± 0.51 ms and 0.99 ± 0.39 ms in SR (53 ± 9 beats/min) to 0.60 ± 0.37 ms (and 0.68 ± 0.39 ms in atrial pacing at 80 beats/min, and to 0.32 ± 0.15 ms and 0.59 ± 0.24 ms in dual-chamber pacing at 80 beats/min, respectively (all P <.05 compared with SR). STV-ARI_auto strongly correlated with STV-ARI_FSA (r = 0.80, P <.0001), with a small bias of 0.18 ms and limits of agreement between –0.35 and 0.70 ms. Conclusion: The novel automatic STV measurement method accurately reflects pacing-induced changes, comparable to the gold standard. Future integration of this technique in implantable cardioverter-defibrillators could furnish continuous monitoring of arrhythmic risk and initiate preventive strategies.