T123. Amplitudes of motor evoked potentials are modulated by preceding a band oscillations: A closed-loop EEG-TMS study

Abstract

Introduction The amplitudes of MEPs in the resting state are considerably variable in each TMS. This variability of MEP amplitudes may reflect the fluctuation of the excitability of the primary motor cortex (M1). The neural substrates of determining MEP amplitudes for each TMS are still under the debate. There have been a few reports that deals with the relationship between MEP amplitudes and EEG oscillations in the offline analysis (e.g. Maki and Ilmoniemi, 2010). However, an attempt to explore the link between M1 and neural oscillations by a closed-loop EEG-MEP system is scarce. We have developed the closed-loop system and examined the relationship between EEG oscillations and MEP amplitudes. Methods Eighteen healthy adults (21–47 y.o, M:F = 15:3) participated in this study. EEGs were recorded at C3 with four surrounding electrodes, referred to the linked earlobe. Then, EEG signals at the C3 electrode were remontaged by source derivation and Fourier analyzed in the online to obtain EEG power of α or β band. The individual peak frequency of each band was determined in each subject before the MEP sessions using EEGs for about one minute in the resting state. When the online EEG power during MEP sessions reached the predefined threshold of high or low EEG power, a TMS pulse was delivered to record MEPs. About 50 MEPs were obtained for each of high/low power conditions. After recording MEPs, the MEP amplitudes of the high EEG power were compared with those of low power. The analyses were conducted for each of α and β band, respectively. Results MEPs were successfully recorded by the closed-loop system using EEG at C3. MEP amplitudes of the high EEG power condition were significantly higher than those of the low power condition in the α band. On the contrary, MEP amplitudes of the closed-loop by the β band were not significantly different between the conditions of high EEG power and that of low power. Conclusion We demonstrated that the high EEG power in α band resulted in higher MEP amplitudes compared with those of the low power by the customized closed-loop EEG-TMS system. Thus, the higher α power at C3 can be indicative of the higher M1 excitability. Event-related desynchronization is often reported that α or β power decreases before the hand movement (Stancak and Pfurtscheller, 1995). However, this experiment was conducted under the resting state. Therefore, the mechanism that modulates M1 excitability may be different from that of the premovement period. The α power around the central region is reported to originate from the primary sensory cortex (S1) while β power from M1 (Salmelin et al., 1995). It is also well known that M1 and S1 are tightly linked and modulate excitability between each other. Therefore, S1 oscillations may play a more important role to determine M1 excitability through M1-S1 connectivity.