Dose-dependent effects of theta burst rTMS on cortical excitability and resting-state connectivity of the human motor system

Abstract

Theta burst stimulation (TBS), a specific protocol of repetitive transcranial magnetic stimulation (rTMS), induces changes in cortical excitability that last beyond stimulation. TBS-induced aftereffects, however, vary between subjects, and the mechanisms underlying these aftereffects to date remain poorly understood. Therefore, the purpose of this study was to investigate whether increasing the number of pulses of intermittent TBS (iTBS) (1) increases cortical excitability as measured by motor-evoked potentials (MEPs) and (2) alters functional connectivity measured using resting-state fMRI, in a dose-dependent manner. Sixteen healthy, human subjects received three serially applied iTBS blocks of 600 pulses over the primary motor cortex (M1 stimulation) and the parieto-occipital vertex (sham stimulation) to test for dose-dependent iTBS effects on cortical excitability and functional connectivity (four sessions in total). iTBS over M1 increased MEP amplitudes compared with sham stimulation after each stimulation block. Although the increase in MEP amplitudes did not differ between the first and second block of M1 stimulation, we observed a significant increase after three blocks (1800 pulses). Furthermore, iTBS enhanced resting-state functional connectivity between the stimulated M1 and premotor regions in both hemispheres. Functional connectivity between M1 and ipsilateral dorsal premotor cortex further increased dose-dependently after 1800 pulses of iTBS over M1. However, no correlation between changes in MEP amplitudes and functional connectivity was detected. In summary, our data show that increasing the number of iTBS stimulation blocks results in dose-dependent effects at the local level (cortical excitability) as well as at a systems level (functional connectivity) with a dose-dependent enhancement of dorsal premotor cortex-M1 connectivity.

Keywords: functional connectivity; iTBS; neural plasticity; premotor cortex; resting-state fMRI; supplementary motor area.

Figure 1.

Experimental design. A, Main experiment. Subjects took part in two MEP sessions (M1-iTBS_MEPs, sham-iTBS_MEPs) and two resting-state fMRI sessions (M1-iTBS_rs-fMRI, sham-iTBS_rs-fMRI) on four separate days. Using a within-subject design, each subject received three serially applied iTBS blocks over M1 (M1 stimulation) and over the parieto-occipital vertex (sham stimulation), each followed by the assessment of MEPs or resting-state fMRI. B, Supplemental control experiment. In a second experiment, a subgroup of 6 subjects additionally received one stimulation over M1 followed by two stimulations over the parieto-occipital vertex (supplemental control stimulation) to test for the specificity of additive aftereffects after serial iTBS over M1.

Figure 2.

Main experiment: MEP amplitudes normalized to baseline (gray) at different stimulation intensities relative to the RMT. A, sham stimulation. B, M1 stimulation. Dose-dependent iTBS aftereffects seem to be more pronounced at near-threshold stimulation intensities (90%–110% of the RMT) compared with higher stimulation intensities (120%–150% of the RMT).

Figure 3.

Main experiment: M1 versus sham stimulation. Changes in MEP amplitudes after M1 (squares) and sham stimulation (diamonds), normalized to baseline MEP amplitudes. Significant aftereffects after M1-iTBS compared with sham-iTBS or within stimulation conditions: *p ≤ 0.05 (Student's t test); **p ≤ 0.001 (Student's t test). M1-iTBS led to a significant increase in MEP amplitudes after iTBS600, iTBS1200, and iTBS1800 compared with sham stimulation and baseline. The increase after M1-iTBS1800 was significantly higher than that after M1-iTBS600 and M1-iTBS1200, whereas after sham-iTBS MEP amplitudes significantly decreased between iTBS1200 and iTBS1800.

Figure 4.

Supplemental control experiment: M1 versus supplemental control stimulation. Changes in MEP amplitudes after M1 (squares) and supplemental control stimulation (diamonds), normalized to baseline MEP amplitudes. *Significant aftereffects after M1-iTBS compared with supplemental control stimulation or baseline (p ≤ 0.05, Student's t test). One stimulation over M1 in the supplemental control experiment led to comparable results as obtained after M1-iTBS600 in the main experiment. After three blocks of iTBS over M1 (M1-iTBS1800), MEP amplitudes were significantly higher compared with one M1 stimulation followed by two stimulations over the parieto-occipital vertex.

Figure 5.

Changes in rsFC. M1 compared with sham stimulation, normalized to baseline values. Color bar represents t values. Only clusters surviving a cluster level FWE correction (p ≤ 0.05) are shown. A, Main experiment. M1-iTBS led to significantly higher changes in rsFC of M1 with bilateral premotor areas (dPMC, SMA) after all doses as well as with somatosensory and superior parietal cortex. B, Supplemental control experiment. iTBS1800 over M1 led to significantly higher correlations in the time courses between M1 and premotor areas (dPMC, SMA) as well as somatosensory/superior parietal cortex compared with a single M1-iTBS application followed by two sham stimulations over the vertex (supplemental control stimulation).

Figure 6.

ROI analysis. Dose-dependent changes in rsFC. Contrasts between the increase in rsFC compared with baseline between iTBS1800 and (A) iTBS600 or (B) iTBS1200. Color bar represents t values. The cross indicates the coordinate where dose-dependent increases were found for ipsilateral dPMC-M1 rsFC. p ≤ 0.05, small-volume FWE-corrected at the voxel level.