Cerebellar transcranial magnetic stimulation: the role of coil geometry and tissue depth

Abstract

Background: While transcranial magnetic stimulation (TMS) coil geometry has important effects on the evoked magnetic field, no study has systematically examined how different coil designs affect the effectiveness of cerebellar stimulation.

Hypothesis: The depth of the cerebellar targets will limit efficiency. Angled coils designed to stimulate deeper tissue are more effective in eliciting cerebellar stimulation.

Methods: Experiment 1 examined basic input-output properties of the figure-of-eight, batwing and double-cone coils, assessed with stimulation of motor cortex. Experiment 2 assessed the ability of each coil to activate cerebellum, using cerebellar-brain inhibition (CBI). Experiment 3 mapped distances from the scalp to cerebellar and motor cortical targets in a sample of 100 subjects' structural magnetic resonance images.

Results: Experiment 1 showed batwing and double-cone coils have significantly lower resting motor thresholds, and recruitment curves with steeper slopes than the figure-of-eight coil. Experiment 2 showed the double-cone coil was the most efficient for eliciting CBI. The batwing coil induced CBI only at higher stimulus intensities. The figure-of-eight coil did not elicit reliable CBI. Experiment 3 confirmed that cerebellar tissue is significantly deeper than primary motor cortex tissue, and we provide a map of scalp-to-target distances.

Conclusions: The double-cone and batwing coils designed to stimulate deeper tissue can effectively stimulate cerebellar targets. The double-cone coil was found to be most effective. The depth map provides a guide to the accessible regions of the cerebellar volume. These results can guide coil selection and stimulation parameters when designing cerebellar TMS studies.

Keywords: Batwing coil; Cerebellar brain inhibition; Cerebello brain inhibition; Cerebellum; Deep TMS; Double cone coil; Figure-of-eight coil; TMS; TMS coil geometry.

Copyright © 2014 The Authors. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Primary motor cortex excitability as a function of coil geometry. A) Mean intensity (% of MSO) required to achieve motor threshold with each coil design. B) MEP recruitment curves for each coil, with regression lines fit to the approximately linear part of the curve, i.e. 90%–140% of RMT ([24], [25], [26]). Smaller circles indicate points that were not included in the slope estimation. C) Slope parameters for the recruitment curve fits shown in panel B. F8: Figure-of-eight coil, B: Batwing coil, DC: Double-cone coil. Error bars represent ± 1 standard error of the mean (SEM).

Figure 2

Cerebello-brain inhibition ratios for the three different coil designs. Circles present mean group data for each block of MEPs collected, with the data normalized by dividing the mean conditioned MEP amplitude by the mean control MEP amplitude. *indicates a significant difference between the conditioned and control MEP amplitude (all Bonferroni corrected one tailed t-test, P < 0.00417).

Figure 3

Perceived discomfort of cerebellar TMS. Ratings were on a scale ranging from 1 (“No discomfort”) to 7 (“Unbearable”). Error bars represent ± 1 SEM. The dashed line separates conditions where CBI was not (below) and was (above) induced.

Figure 4

Depth of different tissue types in relation to target scalp locations. CB: cerebellar gray matter, CB V: Cerebellar lobule V hand representation, CB VIII: Cerebellar lobule VIII hand representation, Light gray: scalp location 3L, Darker gray: scalp location 3L1I. Error bars represent ± 1 SEM.

Figure 5

Data obtained from structural MRI scans from a sample of 100 subjects. A) For each subject a grid of scalp positions at 1 cm intervals was calculated relative to the inion. B) Cartesian distances from the grid of scalp positions to the closest gray matter (mm). Lighter colors indicate shorter distances. C) Number of participants in which the shortest path to cerebellar tissue took a path through the occipital cortex. Lighter colors indicate positions where this path was less likely to pass through occipital tissue.