1350558 - Characterization of cardiac PRFS thermometry at 0.55T

CMR is a promising candidate for the guidance of catheter ablation of cardiac arrhythmias. CMR enables real-time monitoring of tissue temperature in the myocardium using the proton resonance frequency shift (PRFS) method, which can be used to characterize ablation lesions^1,2,3. Low field scanners (< 1 T) are promising for this application compared to current higher field scanners as more devices can safely be introduced whilst reducing the artifact level they have on images, and the associated costs are lower⁴. In this study, we present the initial characterization of cardiac PRFS thermometry at 0.55T.

Methods:

Five healthy subjects (2f, 3m) were scanned on a 0.55T scanner (MAGNETOM Free.Max, Siemens Healthcare, Erlangen, Germany). CMR-thermometry was performed using a prototype ECG-triggered (Invivo, Orlando, FL) single shot EPI sequence with the following parameters: TE = 40 ms, TR = 76 ms, BW = 1086 Hz/px, FOV = 350x350 mm², in-plane voxel size = 2.7x2.7 mm², slice thickness = 6 mm, slices = 3, GRAPPA factor = 2, flip angle = 75°, echo train length = 66 ms, number of dynamics = 300. EPI distortion correction was applied.

Motion-corrected temperature maps were generated offline using a post-processing pipeline previously reported⁵ including multi-baseline correction to correct for respiratory induced phase variations (look-up-table of length 30), non-rigid image registration and temporal filtering. Stability of thermometry was calculated as the standard deviation over time of thermometry values per pixel inside the myocardium.

Results:

CMR thermometry images were successfully acquired and examples are shown in Figure 1. Temperature stability was comparable over the entire myocardium and in all slices.

Figure 2 shows the stability of thermometry per subject, reported as mean±SD over all myocardial voxels in the 3 acquired slices. Over all subjects, the thermometry stability was 1.8±0.8 °C.

Conclusion:

CMR-thermometry in the left ventricle is feasible at 0.55T with a stability of 1.8±0.8 °C. Image distortion, slice coverage, and spatial resolution could be improved and will be the topic of further studies.