Response to Letter Regarding Article, “Infarct Tissue Heterogeneity by Magnetic Resonance Imaging Identifies Enhanced Cardiac Arrhythmia Susceptibility in Patients With Left Ventricular Dysfunction”
Dr Goldberger highlights several considerations when using electrophysiological testing to stratify patients at risk for lethal ventricular arrhythmias. The comments about our study’s electrophysiological protocol are important; indeed, we had acknowledged them as potential limitations. The incremental value of adding a second site for programmed stimulation, even in patients presenting with monomorphic ventricular tachycardia (MVT), has been debated.1 We chose an accelerated protocol at a single site in patients not undergoing a full electrophysiological study to limit the implantable cardioverter-defibrillator implantation time, anticipating only a small tradeoff in the false-negative rate for inducibility. The reason for the difference in inducibility in the patients undergoing full electrophysiological study versus noninvasive programmed stimulation through the implantable cardioverter defibrillator is uncertain, but the numbers are small. In our study, all 9 patients who were inducible by electrophysiological study had MVT induced from the first stimulation site, the right ventricular apex. Seven of the 9 also underwent noninvasive programmed stimulation, and all were inducible. Although this does not exclude the possibility that some patients who were noninducible by noninvasive programmed stimulation were misclassified, we believe this number is limited. In our work, even after statistically controlling for the type of electrophysiological evaluation, gray zone extent remained predictive of inducibility for MVT (P=0.02 as noted in our report; odds ratio 1.6, 95% confidence interval, 1.1 to 2.5 for each 5-g increment in gray zone extent, after controlling for method of electrophysiological evaluation).
We chose to not include polymorphic ventricular tachycardia/ventricular fibrillation (VT/VF) as an end point because of its controversial predictive value. However, if analyzed as 3 groups, gray zone mass in grams was significantly different: noninducible 12±9 g (n=19) versus MVT inducible 19±8 g (n=20) versus polymorphic VT/VF inducible 14±9 g (n=8; ANOVA P=0.035). By post hoc comparison, the difference was primarily between the noninducible versus MVT groups (P=0.017), with a nonsignificant trend (P=0.08) between the noninducible and polymorphic VT/VF groups. In both our work and Dr Goldberger’s prior study, the number of patients inducible for polymorphic VT/VF was relatively small, precluding definitive conclusions about this end point. Differences in the results may reflect the characteristics of the 2 populations. For our study, regardless of whether or not polymorphic VT/VF was included as a separate end point, the gray zone mass remained highly associated with inducibility of MVT.
Dr Gupta is employed by GE Healthcare Technologies. Dr Foo is employed by GE Global Research. The remaining authors report no conflicts.