Predicting tremor control and side effects during focused ultrasound thalamotomy using MR temperature imaging.

OBJECTIVE

Focused ultrasound (FUS) is a minimally invasive treatment option for essential tremor and tremor-dominant Parkinson's disease. Because thermal lesions are irreversible, it is critical to understand how lesion placement relates to patient outcomes. The aim of this study was to determine whether patient outcomes can be predicted using real-time estimation of heating and the accumulated thermal dose derived from MR temperature imaging (MRTI).

METHODS

MRTI, tremor control, and side effects were analyzed from 547 FUS thalamotomy treatments (481 patients, median age 74 years). Previously described methods were used to estimate the heating from 2D MRTI in different orientations and to generate phase maps from the MRTI data. After registration to an MRI atlas, we compared the thermal dose deposition of patients who reported side effects with that of patients who did not report side effects at time points ranging from 1 day to 1 year after FUS. Logistic regression and voxel-wise frequency mapping were used to compare the size and spatial extent of heating, thermal dose, and lesion segmentations to the frequency of different side effects and complete tremor control.

RESULTS

Relationships between the frequency of side effects and tremor control were similar for thermal dose contours and lesion segmentations. Weakness and imbalance were more frequent when heating, dose contours, and lesions extended inferolaterally into the internal capsule. Sensory deficits and dysgeusia were more frequent with increased posterior extent, and dysarthria was more frequent with superior/medial/anterior extent. While most areas associated with complete tremor response overlapped with those associated with side effects, a small anterior region was identified in which complete tremor response without side effects was more frequent. Increased lesion area resulted in more side effects and complete tremor response at early and late time points, but the corresponding areas for dose contours were only related to side effects and only at 1 day after FUS.

CONCLUSIONS

MRTI can provide similar results as lesion segmentations after estimating heating and the thermal dose in different orientations. Combining this approach with visualization of anatomy around the thalamus using filtered phase maps of MRTI is useful for real-time monitoring and optimization of FUS treatments.