Interventional CMR Course
SCMR 22nd Annual Scientific Sessions
Radiofrequency (RF) ablation for ventricular tachycardia (VT) in structural heart disease is limited by recurrence rates of up to 37%, even after acutely successful ablation. Significant gaps between RF lesions and the presence of transient edema after ablation are thought to play a role in recurrence. In this study, we investigated the relationship between electrophysiology (EP) measures of excitability and CMR visualization of RF lesions. We hypothesized that the extent of reduced excitability post-ablation is related to edema seen in T2 maps<./p>
CMR-guided ablation was performed in 8 healthy swine which remained in the 1.5T scanner (GE MR450w) throughout the interventions. CINE images (TR/TE=5/2ms, slice thickness=6mm, res.=1.5x1.25mm) served as anatomical roadmaps for guidance of an actively-tracked 9F Vision-MR catheter (Imricor Medical Systems) within the left ventricle (LV). Ablation (60-120s, 30-35W) was followed by imaging and electroanatomic voltage mapping (EAVM). RF lesions were visualized using native T1-weighted IR-SSFP (TI=860±180ms, VPS=16, FA= 45°, TR/TE=5.6/2.0ms, FOV=24cm, matrix=192x160, slice thickness=6mm) and T2 mapping with a spiral readout (4 TEs between 3-184ms, TR=2 R-R intervals, matrix=3072x10, FOV=24cm, res.=1.3x1.3mm, slice thickness=6mm). Bipolar electrograms (EGMs) were recorded while incrementally probing the endocardial surface.
The T1-derived lesion and T2-derived edema were segmented using methods described previously (Fig 1). T2 and tissue classification (myocardium, lesion core, or edema--assigned by the majority of pixels) in radial LV sectors were compared with EGMs from corresponding locations. Image-based properties within 6mm of the endocardium were used, per the catheter sensitive depth for EGMs measured previously. EGM peak-to-peak amplitudes associated with each tissue class were compared using the Kruskal-Wallis test. Results are given as mean±SD and differences considered significant when p<0.05.
10 RF lesions were created, followed by imaging and EAVM (Fig 2), with 895 EGMs in total. EGM amplitude was significantly reduced in the broader ablated tissue (union of T2-derived edema and T1-derived lesion regions, N=208, 2.8±1.4mV, skewness=1.2) compared to healthy tissue (N=687, 3.8±1.9mV, skewness=0.8; p<0.05; Fig 3). Healthy EGM amplitude was significantly higher than in T2-derived edema regions alone (N=153 EGMs; p<0.05) but not significantly higher than in T1-derived lesion alone (N=14 EGMs; p=0.3). EGM amplitudes in ablated tissue tended to be higher than those generally seen in dense scar, potentially due to incomplete destruction of tissue within the catheter sensitive volume.
Reduced EGM amplitude suggests suppressed excitability near RF lesions due to both the permanent lesion core and surrounding transient edema. These findings may help elucidate VT recurrence mechanisms and the role of CMR in ablation.