Background: Single-center data have demonstrated the feasibility of image-guidance for ventricular tachycardia (VT) ablation. The MUSIC network was created to collaborate on an international level.
Objective: To analyze the feasibility, safety and efficacy of image-guided VT ablation in a diversity of centers.
Methods: Across 14 centers in USA, Europe and Australia, the MUSIC software (IHU Liryc, Bordeaux, Inria Sophia Antipolis, France) was used to generate patient-specific image-based 3D models including detailed anatomy (cardiac chambers, papillary muscles, left phrenic nerve, coronary arteries and veins, aorta, ICD leads) and structural substrate (wall thinning / intramural fat / calcification from arterial CT, late enhancement from late-enhanced CT or MRI). These models were used in real-time to assist the VT ablation procedures. Procedural techniques and endpoints were left at the discretion of the operators, including how imaging was used to optimize mapping and ablation strategies. Imaging, patient characteristics, procedural and follow-up data were collected.
Results: In total, 840 MUSIC-guided VT procedures were analyzed (age 60±15 years, 79% male, 70% with ICD, 43% ICM, 44% NICM, 4% congenital, 9% no structural heart disease, 64% first procedure, 36% redo). Imaging consisted of arterial CT in 86%, late-enhanced CT in 8%, late-enhanced MRI in 61%. ICM patients were older than NICM (66±11yrs vs. 56±16yrs, p<0.0001) and had a lower LVEF (35±11% vs. 43±15%, p>0.0001). Procedure times and RF duration were highly variable across patients and sites (300 [226-420] min and 36 [21-60] min, respectively). Dedicated mapping catheters were used in 44%, and epicardial approaches in 23%. Complete LAVA elimination was obtained in 66% and VT non-inducibility in 85%. Acute complications occurred more in epicardial procedures [total: 40/840 (5%), epi: 8%, non-epi: 4%, p 0.01]. No coronary or phrenic nerve damage was observed. Over a median follow-up of 17 [6-34] months, the VT recurrence rate was 24% in ICM vs. 23% in NICM (p=0.77). The mortality rate was 11% in ICM vs. 13% in NICM (p=0.44).
Conclusion: Image-integrated VT ablation is reproducible, safe and effective in a diversity of centers, with low VT recurrence rates over a median 17-month follow-up.