Abstract 15992: Ventricular Tachyarrhythmia Sensing Following Magnetic Resonance Imaging in Patients With an ICD System Designed for the MRI Environment

Objective: To evaluate the impact of MRI on ICD sensing of ventricular tachyarrhythmias. Methods: The Evera MRI study is the first randomized trial of an ICD designed specifically to be MRI safe. This was a world-wide trial of 263 pts implanted with an Evera MRI single or dual chamber ICD in 42 centers. Patients were randomized to either undergo an MRI scan (MRI group) or to not receive a study MRI scan. Per protocol, 156 MRI group pts underwent an MRI procedure at 9-12 weeks post-implant. Device detected spontaneous and induced VT/VF episodes in the MRI group occurring prior to and following MRI were evaluated by an independent Episode Review Committee blinded to the timing of events. Detection delay was computed as the sum of RR intervals of under sensed beats. A ≥5 second delay in detection due to under sensing was prospectively defined as clinically significant. P-values are adjusted for multiple episodes within a patient. Results: There were a total of 34 post-MRI VT/VF episodes in 24 pts, with 16 of those pts having 27 VT/VF episodes prior to MRI. The mean detection delay due to under sensing in the 23 pre-MRI treated episodes was 0.37 ± 0.57 seconds, compared to 0.25 ± 0.57 in 27 treated post-MRI episodes (p=0.29). The maximum detection delay was 2.19 seconds pre-MRI and 2.87 seconds post-MRI. Monomorphic VT and non-sustained VT/VF rarely had any under sensing with 1/11 (9%) pre-MRI episodes and 1/14 (7%) post-MRI episodes (p=0.81) having 2 and 1 under sensed beats, respectively. For polymorphic VT/VF episodes, 13/16 (81%) pre-MRI episodes had some detection delay vs. 10/20 (50%) in post-MRI episodes (p=0.052), but no detection delay was clinically significant. The duration of delay was 0.54 ± 0.61 vs 0.32 ± 0.64, respectively, p=0.28). Conclusion: Detection of VT/VF was excellent with no significant impact of MRI with this system.