The infarct characteristics on magnetic resonance imaging and ventricular tachycardia: do we see what we need to see?

The underlying mechanism of most ventricular tachycardias (VTs) after myocardial infarction (MI) is reentry involving areas of ventricular scar. Dense fibrous scars alone cannot cause arrhythmias; however, when surrounded by bundles of surviving myocytes in the peri-infarction zone, an arrhythmogenic substrate may arise. Areas of dense fibrosis intersperse with surviving myocardial bundles and form regions of conduction block that define reentry circuit borders. The presence of fibrosis between these surviving bundles prolongs the pathway for impulse propagation, creating slow conduction through the scar. The complex structure of these scars is likely to determine reentrant circuit characteristics and thereby VT characteristics. Late gadolinium-enhancement cardiac magnetic resonance (LGE-CMR) has increasingly been used to visualize the threedimensional (3D) geometry of myocardial fibrosis and has become the gold standard for detection of non-transmural scars because of its superior spatial resolution. Woie et al. have addressed an interesting question of whether myocardial scar characteristics as visualized and defined by LGE-CMR correlate with VT characteristics, specifically with the VT cycle length (VTCL). They hypothesized that the rate of spontaneous VTs is related to the size of the post-infarction scar or its border zone. They studied 24 patients with remote MI and indication for implantation of an internal cardioverter-defibrillator for primary or secondary prevention who underwent LGE-CMR for quantification of left ventricular (LV) volumes, function, size of core infarct, and size of the infarct border zone prior to device implantation. Patients were followed every 3 months during the first year after implantation. Twenty out of 24 patients had VTs with a median VTCL of 325 ms (range 239–438 ms). None of these patients received anti-arrhythmic drugs that might influence VTCL. Ventricular tachycardia cycle length was correlated with five proposed LGE-CMR scar characteristics: size of the scar, size of core infarct and the scar border zone, contour regularity of the scar, and number of core islands within the scar area. In univariate analysis, the scar characteristics that showed a strong correlation with VTCL were scar size, number of core islands, and size of the peri-infarction zone. Following the stepwise selection of different parameters (not fully described), the strongest positive correlation was found between VTCL and the core infarct size combined with the number of core islands. The authors concluded that large and dense scars assessed by LGE-CMR may prolong the time the excitation wavefront needs to propagate around the circuit. In animal post-infarction models, it has been demonstrated that areas of slow conduction, which determine the reentry circuit isthmus, coincided with areas where layers of viable myocardium are the thinnest and layers of fibrosis are the thickest, and these areas correlated with greater scar transmurality on CMR. Ventricular tachycardia cycle length is determined by circuit path length and conduction velocity; an increasing isthmus length contributes to the circuit path length resulting in a longer VTCL. Electroanatomical mapping studies in humans with VTs after MI have demonstrated isthmus lengths of 23+11 mm for a VTCL of 374+59 ms. Of interest, these isthmuses were found in very low-voltage areas (≤0.2 mV) assessed by bipolar voltage mapping. These low-voltage electrograms are usually found in areas with dense transmural infarction. These findings support an interesting correlation between slower VTs and larger scar areas, provided that these scars are transmural, and the authors should be encouraged to include scar transmurality in their analysis.