Download PDF version
Link to Max Planck Institute for Biological Cybernetics
TMS is an attractive technique to study human brain function as it non-invasively induces neural activity in the targeted cortical region, with little or no discomfort for the investigated subject. In addition, by applying TMS to reversibly interfere with ongoing brain activity during task execution and monitoring the effects on behavior, it can be used to establish causal links between brain structures and function (the so-called “virtual-lesion approach”1). As TMS applied to the motor cortex causes clear-cut responses in form of motor-evoked potentials, much of the knowledge on TMS stems from studies targeting this area. However, the characterization of TMS effects on other cortical regions is less straight-forward. Additionally, TMS can induce remote effects in anatomically connected brain areas, and these effects might contribute to the observed behavioral changes2. Taken together, these challenges inspired attempts to combine TMS with brain imaging techniques such as electroencephalography 3, positron emission tomography4,5 and functional magnetic resonance imaging (fMRI)6,7 in order to get a more complete picture of both the local and remote effects of TMS on brain activity.
fMRI has become one of the standard tools in human cognitive neuroscience. It has a quite good spatial resolution in the order of millimeters. It assesses neuronal activity indirectly based on the detection of local changes of the blood oxygenation level (BOLD effect8), resulting in a temporal resolution in the order of seconds. Most commonly, so called gradient-echo echo-planar imaging (GE EPI) is used to acquire repeated images of the whole brain every few (typically 2 to 3) seconds in the course of an experiment. While this temporal resolution is inferior to that of TMS (a few milliseconds), it is still good enough to capture the transient brain responses to single TMS pulses or to short rTMS trains.
Bohning et al.6 were the first to demonstrate the feasibility of fMRI to directly measure the BOLD activations caused by TMS pulses applied to the motor cortex, despite the potentially destructive impact of TMS on the MR images. Much of the subsequent work on interleaved (or concurrent) TMS-fMRI was devoted to methological and technological improvements, so that this technique is now safe to use and the impact of TMS on the EPI image quality is minimized. For example, the static magnetic field of the MR scanner causes strong mechanical forces on the TMS coil when a pulse is fired9. This results in increased vibration and noise levels (the “coil click” is louder). Standard TMS coils are not designed to cope with the additional forces and can therefore be destroyed. As a consequence, a custom-designed TMS coil from MagVenture was already used in the initial studies of Bohning et al.6. On the other hand, the TMS stimulator induces RF noise that can decrease the MR image quality. High-current filters can be added to the TMS coil cable in order to prevent the noise from entering the MR cabin9. The TMS coil causes marked static signal dropouts and distortions in the EPI images. While it seems impossible to get completely rid of these effects, their extent can be markedly reduced by a proper orientation of the imaging slices parallel to the TMS coil plane7. Finally, applying the TMS pulses directly during image acquisition severely deteriorates the EPI images. As a consequence, short temporal pauses are introduced in the MR acquisition, and the TMS pulses are applied within these gaps10. In combination, these measures allow to acquire EPI images of excellent quality.
Figure 1: (a) rTMS-trains consisting of 10 stimuli applied every 200 ms were interleaved with the MR acquistion (b) Brain responses to rTMS stimulation. Shown is the group activation map for 5 subjects (threshold z=2.3 voxel level, p=0.05 cluster level; FSL FLAME mixed effects analysis; MNI space) (c) Brain activation caused by volitional movements acoustically triggered by rTMS trains at low intensity (same threshold level) (d) Overlap between rTMS- and movement-related activations. Adapted from Moisa et al.11.
As an example, Figure 1 shows the brain activation caused by the stimulation of the motor representation of a small hand muscle in five healthy subjects. A MagVenture MagPro X100 together with a MR-compatible figure-8 coil (MagVenture MRi-B88) with minimized acoustical noise and vibrations was used for stimulation. The TMS protocol consisted of short rTMS trains (5Hz for 2sec at 110% rMT; Fig. 1a) that were followed by rest periods of 17 sec. This was repeated 25 times. In addition, the overall experiment was repeated with the TMS intensity set to 50% rMT and the subject performed volitional thumb movements acoustically triggered by the TMS coil clicks. The group activation pattern in response to rTMS stimulation is shown in Fig. 1b. In concordance with previous findings12,13, significant BOLD responses are observed in the stimulated region (M1/S1) and in connected brain areas (SMA/CMA, thalamus ipsilateral to stimulated M1, bilateral putamen, bilateral cerebellum). The coil clicks induce additional activations in auditory regions (bilateral auditory cortices, ipsilateral inferior colliculus). The rTMS-induced activations exhibit a robust spatial overlap with those obtained for volitional movement (Fig. 1c&d). Except for the control region in white matter, they show the expected BOLD shape (Fig. 2). Taken together, the example demonstrates that fMRI can be used to reliably access the responses to TMS in the stimulated and in connected brain areas.
Figure 2: Average BOLD time courses for regions-of-interest corresponding to the motor system (top row), the auditory system (bottom row; first three plots) and a control region in white matter (bottom row; rightmost plot)
Up to now, most interleaved TMS/fMRI studies have targeted the motor system. However, recently more complex questions were tackled. Li et al.14 used it to assess the impact of lamotrigine on prefrontal-limbic circuits. Ruff et al.15,16 studied top-down effects of frontal and parietal regions on early visual areas. Sack et al.1 disturbed the parietal cortex in a visuo-spatial imagery task while simultaneously assessing the impact on the task-related activity in the stimulated area and in interconnected frontal regions. To conclude, interleaved TMS/fMRI it is becoming increasingly popular. It combines the advantages of the two single techniques: fMRI offers whole-brain coverage at a quite good spatial and temporal resolution while TMS allows us to establish causal structure-function links.
Axel Thielscher
MPI for Biological Cybernetics Tübingen, Germany
References
1. Sack AT, Kohler A, Bestmann S, et al. Imaging the brain activity changes underlying impaired visuospatial judgments: simultaneous FMRI, TMS, and behavioral studies. Cerebral Cortex 2007;17(12):2841-2852. 2. Walsh V, Pascual-Leone A. Transcranial Magnetic Stimulation: A Neurochronometrics of Mind. Cambridge, MA: MIT Press; 2003. 3. Ilmoniemi RJ, Virtanen J, Ruohonen J, et al. Neuronal responses to magnetic stimulation reveal cortical reactivity and connectivity. NeuroReport 1997;8:3537–3540. 4. Paus T, Jech R, Thompson CJ, Comeau R, Peters T, Evans AC. Transcranial magnetic stimulation during positron emission tomography: a new method for studying connectivity of the human cerebral cortex. J Neurosci 1997;17:3178-3184. 5. Siebner HR, Willoch F, Peller M, et al. Imaging brain activation induced by long trains of repetitive transcranial magnetic stimulation. NeuroReport 1998;9:943-948. 6. Bohning DE, Pecheny AP, Epstein CM, et al. Mapping transcranial magnetic stimulation (TMS) fields in vivo with MRI. Neuroreport 1997;8(11):2535-2538. 7. Baudewig J, Paulus W, Frahm J. Artifacts caused by transcranial magnetic stimulation coils and EEG electrodes in T(2)*-weighted echo-planar imaging. Magn Reson Imaging 2000;18(4):479-484. 8. Ogawa S, Tank DW, Menon R, et al. Intrinsic signal changes accompanying sensory stimulation: functional brain mapping with magnetic resonance imaging. PNAS 1992;89(13):5951-5955. 9. Bohning DE, Denslow S, Bohning PA, Lomarev MP, George MS. Interleaving fMRI and rTMS. Suppl Clin Neurophysiol 2003;56:42-54. 10. Bestmann S, Baudewig J, Frahm J. On the synchronization of transcranial magnetic stimulation and functional echo-planar imaging. J Magn Reson Imaging 2003;17(3):309-316. 11. Moisa M, Uludag K, Ugurbil K, Thielscher A. A new hardware-software coil positioning system for interleaved TMS/fMRI: A motor cortex stimulation study. 2008; Toronto. 12. Bestmann S, Baudewig J, Siebner HR, Rothwell JC, Frahm J. Functional MRI of the immediate impact of transcranial magnetic stimulation on cortical and subcortical motor circuits. E J Neurosc 2004;19(7):1950-1962. 13. Bohning DE, Shastri A, Lomarev MP, Lorberbaum JP, Nahas Z, George MS. BOLD-fMRI response vs. transcranial magnetic stimulation (TMS) pulse-train length: testing for linearity. J Magn Reson Imaging 2003;17(3):279-290. 14. Li X, Tenebäck CC, Nahas Z, et al. Interleaved transcranial magnetic stimulation/functional MRI confirms that lamotrigine inhibits cortical excitability in healthy young men. Neuropsychopharmacology 2004;29:1395-1407. 15. Ruff CC, Bestmann S, Blankenburg F, et al. Distinct causal influences of parietal versus frontal areas on human visual cortex: evidence from concurrent TMS-fMRI. Cerebral Cortex 2008;18:817-827. 16. Ruff CC, Blankenburg F, Bjoertomt O, et al. Concurrent TMS-fMRI and psychophysics reveal frontal influences on human retinotopic visual cortex. Curr Biol 2006;16:1479-1488.
|