P078 Modeling the effect of phase-triggered transcranial magnetic stimulation on cortical excitability

Abstract

Introduction Responses to transcranial magnetic stimulation show large variability, between subjects as well as within subject between trials. Can the nature of this variability be explained by the prominent oscillatory cortical rhythms, such as the mu-rhythm? Several studies show a dependency of TMS response to phase at stimulation onset ( Dugue et al., 2011 ; Kundu et al., 2014 ; Triesch et al., 2015 ). Objectives Our previous network model ( Rusu et al., 2014 ) displays several characteristics of D&I-waves, but was without background activity. Can we see a phase-dependency, when introducing oscillatory background activity and triggering a TMS pulse at specified phases (as possible in an online TMS-EEG setup)? Materials & methods The model consists of a network of a multicompartmental layer 5 cell and 300 excitatory and inhibitory layer 2/3 point neurons with feedforward projections onto the L5 cell. For the model output, L5 activity of 250 simulated subnetworks is pooled. The L2/3 cells receive input from a sine wave modulated poisson process. TMS is modelled as a current injection into the L2/3 soma and L5 axon. TMS pulse onset was varied with respect to phase of L2/3 background activity. Results The model shows a clear modulation of L5 firing rate depending on the phase at which the TMS pulse is applied (Fig. 1), as well as an systematic effect on amplitude and number of I-waves. The largest response can be elicited when applying the pulse in a phase when the L5 neurons are in a depolarized state, with the difference between peak and trough response regulated by stimulation intensity. Conclusion The network model extended with oscillatory spontaneous activity displays phase-dependency of TMS-response and can parsimoniously explain how the mu-rhythm contributes to the variability of TMS effects. Download high-res image (123KB) Download full-size image