Isochronal Versus Electrogram-Based Conduction Velocity Estimation to Predict Inducible Monomorphic Ventricular Tachycardia in Repaired Tetralogy of Fallot.

Abstract

Background: Isthmus conduction velocity (CV) predicts ventricular tachycardia risk in tetralogy of Fallot and traditionally is calculated using electrogram-based methods (electrogram CV index [EGM-CVi]). Isochronal activation mapping approximates CV in adults with ischemic and nonischemic cardiomyopathy but has never been tested in congenital heart disease populations. Our objectives were to (1) investigate the diagnostic performance of isochronal CV index (ISO-CVi) in predicting inducible ventricular tachycardia and (2) compare it to traditional EGM-CVi in a population of patients with repaired tetralogy of Fallot undergoing electrophysiology study before transcatheter pulmonary valve replacement.

Methods: Retrospective cohort subanalysis of the CATAPULT-TOF registry (Catheter Ablation of Ventricular Tachycardia Before Transcatheter Pulmonary Valve Replacement in Repaired Tetralogy of Fallot). The primary outcome was inducible monomorphic ventricular tachycardia (MMVT) at electrophysiology study. Predictor variables were ISO-CVi and EGM-CVi. Electroanatomic activation maps profiling anatomic isthmus 3 were centrally adjudicated to measure anatomic isthmus 3 ISO-CVi; real-world anatomic isthmus 3 EGM-CVi values reported by contributing centers per the parent registry protocol were used as the comparator. Multivariable logistic regression and receiver operating characteristic analyses quantified associations between predictors and outcomes.

Results: Seventy patients with tetralogy of Fallot underwent pretranscatheter pulmonary valve replacement electrophysiology study at a median age of 40 (interquartile range, 24.8-51) years. Inducible MMVT was observed in 24 (34%). Median ISO-CVi was lower in those with versus without MMVT (0.17 versus 0.30 m/s; P<0.001). Median EGM-CVi did not differ between groups (0.43 versus 0.70 m/s; P=0.055). Coefficients of variation were higher for EGM-CVi (79%) than ISO-CVi (46%). On multivariable analysis, ISO-CVi was associated with MMVT (odds ratio, 0.31 per 0.1 m/s increase, [95% CI, 0.13-0.63]; P=0.004) with a sensitivity and specificity of 88% and 70%, respectively, at a threshold of 0.22 m/s.

Conclusions: ISO-CVi was independently associated with inducible MMVT and is measured with less variability than legacy EGM-CVi methods in a real-world sample. If external validity is confirmed, ISO-CVi may represent a more standardized method of isthmus CV estimation that could be broadly implemented across centers.