Ventricular arrhythmia ablation lesions detectability and temporal changes on cardiac magnetic resonance

• Published in: Pacing and clinical electrophysiology Vol. 43; no. 3; pp. 314 - 321
• Main Authors: Vunnam, Rama, Maheshwari, Varun, Jeudy, Jean, Ghzally, Yousra, Imanli, Hasan, Abdulghani, Mohammed, Mahat, Jagat B., Timilsina, Saroj, Restrepo, Alejandro, See, Vincent, Shorofsky, Stephen, Dickfeld, Timm
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.03.2020
• Subjects: ablation; Arrhythmia; Cardiac arrhythmia; cardiac magnetic resonance; Catheter Ablation; Contrast Media; Electrophysiologic Techniques, Cardiac; Female; Gadolinium; Heart; Humans; Lesions; Magnetic Resonance Imaging, Cine - methods; Male; Meglumine - analogs & derivatives; Microvasculature; Middle Aged; Myocardium; Organometallic Compounds; Radiofrequency ablation; Retrospective Studies; Tachycardia, Ventricular - diagnostic imaging; Tachycardia, Ventricular - physiopathology; Tachycardia, Ventricular - surgery; Ventricle; ventricular arrhythmia; Ventricular Premature Complexes - diagnostic imaging; Ventricular Premature Complexes - physiopathology; Ventricular Premature Complexes - surgery; ventricular arrhythmia; cardiac magnetic resonance; ablation
• ISSN: 0147-8389; 1540-8159
• DOI: 10.1111/pace.13886

Background Cardiac magnetic resonance (CMR) characteristics of ventricular radiofrequency ablation (RFA) lesions have only been incompletely defined. Aim To determine the detectability and imaging characteristics of ventricular RFA lesions in an unselected patient cohort undergoing ventricular arrhythmia ablation. Methods and results A retrospective chart review (n = 249) identified 36 patients with either pre‐/postablation CMR (n = 14) or only postablation CMR (n = 22). Ablation lesions could be identified in 50% (n = 18) of patients. Nonvisualized lesions had more preexisting transmural late gadolinium enhancement (LGE) >75% at the ablation sites (21% vs 0.0%, P = .042), more prevalent ICD artifact (50% vs 0%, P = .001), and lower ejection fraction (35.8 ± 14.2% vs 45.3 ± 13.4%, P = .048). Early CMR imaging demonstrated a central “black” signal void (microvascular obstruction [MVO], n = 12, 67%) up to 32 days post‐RFA, whereas late imaging showed a homogenously “white” gadolinium enhancement pattern (n = 6, 33%). MVO was only observed in nonfibrotic myocardium without preexisting LGE (n = 12) but was not observed in the scar with preexisting LGE (n = 3, P = .002) suggesting different wash‐in/wash‐out kinetics in scar/nonscar myocardium. Signal intensity (1909 vs 2534, P = .009) and contrast‐to‐noise ratio (−7.8 vs 16.3, P = .009) were significantly different between MVO and LGE lesions, respectively. Conclusion Ventricular ablation lesions visualization is negatively affected by preexisting transmural scar, ICD artifact, and low ejection fraction. The transition of “black” MVO appearance to “white” LGE appearance on CMR occurs around 1 month following ablation, suggesting a change in histological characteristics of ablation lesions. This may affect the utility of CMR in the evaluation of the ventricular lesions, when undergoing real‐time or repeat VT ablations.