Accumulated thermal dose in MRI-guided focused ultrasound for essential tremor: repeated sonications with low focal temperatures

Abstract

Objective: The object of this study was to correlate lesion size with accumulated thermal dose (ATD) in transcranial MRI-guided focused ultrasound (MRgFUS) treatments of essential tremor with focal temperatures limited to 50°C-54°C.

Methods: Seventy-five patients with medically refractory essential tremor underwent MRgFUS thalamotomy at the authors' institution. Intraoperative MR thermometry was performed to measure the induced temperature and thermal dose distributions (proton resonance frequency shift coefficient = -0.00909 ppm/°C). In 19 patients, it was not possible to raise the focal temperature above 54°C because of unfavorable skull characteristics and/or the pain associated with cranial heating. In this patient subset, sonications with focal temperatures between 50°C and 54°C were repeated (5.1 ± 1.5, mean ± standard deviation) to accumulate a sufficient thermal dose for lesion formation. The ATD profile sizes (17, 40, 100, 200, and 240 cumulative equivalent minutes at 43°C [CEM43]) calculated by combining axial MR thermometry data from individual sonications were correlated with the corresponding lesion sizes measured on axial T1-weighted (T1w) and T2-weighted (T2w) MR images acquired 1 day posttreatment. Manual corrections were applied to the MR thermometry data prior to thermal dose accumulation to compensate for off-resonance-induced spatial-shifting artifacts.

Results: Mean lesion sizes measured on T2w MRI (5.0 ± 1.4 mm) were, on average, 28% larger than those measured on T1w MRI (3.9 ± 1.4 mm). The ATD thresholds found to provide the best correlation with lesion sizes measured on T2w and T1w MRI were 100 CEM43 (regression slope = 0.97, R2 = 0.66) and 200 CEM43 (regression slope = 0.98, R2 = 0.89), respectively, consistent with data from a previous study of MRgFUS thalamotomy via repeated sonications at higher focal temperatures (≥ 55°C). Two-way linear mixed-effects analysis revealed that dominant tremor subscores on the Fahn-Tolosa-Marin Clinical Rating Scale for Tremor (CRST) were statistically different from baseline at 3 months and 1 year posttreatment in both low-temperature (50°C-54°C) and high-temperature (≥ 55°C) patient cohorts. No significant fixed effect on the dominant tremor scores was found for the temperature cohort factor.

Conclusions: In transcranial MRgFUS thalamotomy for essential tremor, repeated sonications with focal temperatures between 50°C and 54°C can accumulate a sufficient thermal dose to generate lesions for clinically relevant tremor suppression up to 1 year posttreatment, and the ATD can be used to predict the size of the resulting ablation zones measured on MRI. These data will serve to guide future clinical MRgFUS brain procedures, particularly those in which focal temperatures are limited to below 55°C.

Keywords: MR thermometry; focused ultrasound; functional neurosurgery; thalamotomy; thermal ablation; thermal dose.

FIG. 1.

Axial T2w MR images acquired 1 day posttreatment showing a unilateral thalamotomy in the VIM nucleus (dashed square). Insets: Magnified views of the same patient’s T2w MRI, T1w MRI, and ATD profiles, respectively (17, 40, 100, 200, and 240 CEM₄₃) (FOV = 12 × 12 mm). In this patient, six sonications with focal temperatures between 50°C and 54°C (duration range = 15–45 seconds) were applied to generate a lesion.

FIG. 2.

Slopes of linear regressions between the ATD profile size (17, 40, 100, 200, and 240 CEM₄₃) and the lesion size measured on T1w and T2w MR images acquired 1 day posttreatment. Error bars represent 95% confidence intervals (± 1.96 SDs). Logarithmic fits were applied to the data and added to the plots.

FIG. 3.

Scatterplots (A and C) and Bland-Altman plots (B and D) [difference(x, y) = x − y, mean(x, y) = (x + y)/2] of the comparison of lesion diameters estimated via ATD profile size with those measured via MRI acquired 1 day posttreatment. Comparison of the 200 CEM₄₃ ATD profile size with the lesion size measured on T1w MRI (A and B). Comparison of the 100 CEM₄₃ ATD profile size with the lesion size measured on T2w MRI (C and D). Solid lines (A and C) represent linear regressions, and data from a previous clinical study with repeated sonications at higher focal temperatures (≥ 55°C) are plotted for comparison. Solid and dashed lines (B and D) represent the mean bias and limits of agreement (mean bias ± 1.96 standard deviations), respectively. T = temperature.

FIG. 4.

Scatterplots of dominant tremor CRST subscores at baseline, 3 months posttreatment, and 1 year posttreatment for both the low and high-temperature cohorts. Statistical analysis was performed on both cohorts: *p < 0.05 and **p < 0.0001 (ns = not significant, p > 0.05) on post hoc Tukey multiple comparison testing (adjusted p values). Horizontal lines and error bars represent mean and standard deviation values, respectively. The number of patients who completed each follow-up examination (n) is listed for each group.