Validation of a new double-coil TMS method to assess corticospinal excitability bilaterally

• ¹ Institut de Neuroscience, Université Catholique de Louvain, Belgium

During the two last decades, many studies have applied single-coil transcranial magnetic stimulation (TMS) over one primary motor cortex (M1), eliciting motor-evoked potentials (MEPs) in a contralateral hand or limb muscle to track corticospinal excitability (CSE) changes in various contexts. MEPs represent a valuable indicator of the CSE at the time of stimulation (Bestmann & Duque, 2016; Grandjean et al., 2018)⁠ and their amplitude can be compared at different moments within a given condition and between different states (Bestmann & Krakauer, 2015; Vassiliadis et al., 2018)⁠. For example, in studies of action preparation, MEPs have been elicited at different times during the reaction time period preceding left and right hand responses in various tasks (Cos, Duque, & Cisek, 2014; Duque, Petitjean, & Swinnen, 2016; Quoilin, Lambert, Jacob, Klein, & Duque, 2016; Wilhelm, Grandjean, & Duque, 2017)⁠⁠. This work has led to considerable advances in the understanding of neural processes underlying action preparation, including the identification of various excitatory and inhibitory influences modulating CSE distinctively depending on the role of the probed muscle in the forthcoming response (Hadar, Rowe, Di Costa, Jones, & Yarrow, 2016)⁠ and on the task-setting (reviewed in Duque, Greenhouse, Labruna, & Ivry, 2017)⁠. However, a substantial caveat in the previously mentioned literature is that MEPs are usually only recorded from muscles of a single hand following the application of TMS over one M1 only. This occurs because applying TMS over both M1 in separate blocks doubles the duration of the experiment, making it impossible to fit all the conditions in a single session. Hence, in most experiments, the MEP data have only provided researchers with half of the story, increasing the risk of shortcuts in data interpretations. For example, in action preparation studies, it has been typically assumed that differences in MEP amplitudes between conditions in which a muscle (e.g. right index agonist) is either selected (e.g. right index response) or non-selected (e.g. left index response) for the forthcoming movement are due to the distinct function of the probed muscle in these two situations (Labruna et al., 2014)⁠. In other words, the stronger left MEP suppression in conditions where the target muscle is selected for the forthcoming movement compared to when it is non-selected has been commonly understood from the distinct function (selected vs. non- selected) of the left hand muscle in these two situations (Duque, Lew, Mazzocchio, Olivier, & Ivry, 2010; Labruna et al., 2014)⁠: preparatory inhibition is thought to be more prominent for selected than non-selected effector representations. Yet, there is a substantial confound here because besides the function, conditions also differ in regard to the hand being cued for the response (right or left). Hence, MEP differences may be due to the use of the dominant vs non-dominant hand, a possibility that has rarely been considered in the past (Wilhelm, Quoilin, Petitjean, & Duque, 2016)⁠. For the reasons outlined above, it would be useful to obtain MEPs in both hands at once, to track CSE bilaterally. For instance, in the hand choice example above, this would allow researchers to obtain markers of CSE changes associated with selected and non-selected hands within each trial, whether the dominant or non-dominant hand is chosen for the response. Obtaining MEPs in both hands requires using a double-coil TMS approach where the two M1 are stimulated concurrently (Grandjean et al., 2017b, 2017a)⁠. In this context, a critical issue is to avoid electromagnetic interference between the two stimulating coils. That is, if triggered exactly at the same time, the two magnetic fields (duration ∼200 μs) interact, exerting a force on the TMS coils; the resulting MEPs are smaller compared to when a single-coil TMS method is used (unpublished observations). Fortunately, this problem can be overcome by adding a short interval between the two M1 pulses. Another critical point is to make sure that the two descending volleys do not influence each other. Notably, cortical interactions through the corpus callosum can occur with intervals as short as 4 ms (Ferbert et al., 1992; Hanajima et al., 2001; reviewed in Reis et al., 2008)⁠. Hence, the inter-pulse interval should be kept shorter than that to avoid transcallosal interactions; yet, neural connections may still occur subcortically. In the present study, we tested a double-coil TMS method where the two M1 are stimulated with a 1 ms inter-pulse interval (double-coil1ms). MEPs elicited using this new approach (MEPdouble) were compared to those elicited using a standard single-coil TMS method (MEPsingle). These MEPs were recorded for five different intensities of stimulation (at 100, 115, 130, 145 or 160 % of the resting motor threshold, rMT) while participants were completely relaxed, at rest (n = 32). The goal of the present study was to assess whether double-coil1ms TMS yields comparable MEPs as single-coil TMS. LAY SUMMARY: Only a handful of non-invasive and non-painful techniques allow one to record neural activity in humans. Transcranial magnetic stimulation (TMS) is one of them. When applied over primary motor cortex (M1), TMS elicits motor-evoked potentials (MEPs) which provide us with a precise and muscle-specific readout of the excitatory status of the motor output system at the stimulation time. So far, studies have probed motor excitability unilaterally, by eliciting MEPs on one side of the body following single-pulse TMS over contralateral M1. Here, we aimed at evaluating a “Double-Coil” TMS method by which MEPs are elicited in both hands at once.