Assessment of vestibulocortical interactions during standing in healthy subjects

Abstract

The vestibular system is essential to produce adequate postural responses enabling voluntary movement. However, how the vestibular system influences corticospinal output during postural tasks is still unknown. Here, we examined the modulation exerted by the vestibular system on corticospinal output during standing. Healthy subjects (n = 25) maintained quiet standing, head facing forward with eyes closed. Galvanic vestibular stimulation (GVS) was applied bipolarly and binaurally at different delays prior to transcranial magnetic stimulation (TMS) which triggered motor evoked potentials (MEPs). With the cathode right/anode left configuration, MEPs in right Soleus (SOL) muscle were significantly suppressed when GVS was applied at ISI = 40 and 130ms before TMS. With the anode right/cathode left configuration, no significant changes were observed. Changes in the MEP amplitude were then compared to changes in the ongoing EMG when GVS was applied alone. Only the decrease in MEP amplitude at ISI = 40ms occurred without change in the ongoing EMG, suggesting that modulation occurred at a premotoneuronal level. We further investigated whether vestibular modulation could occur at the motor cortex level by assessing changes in the direct corticospinal pathways using the short-latency facilitation of the SOL Hoffmann reflex (H-reflex) by TMS. None of the observed modulation occurred at the level of motor cortex. Finally, using the long-latency facilitation of the SOL H-reflex, we were able to confirm that the suppression of MEP at ISI = 40ms occurred at a premotoneuronal level. The data indicate that vestibular signals modulate corticospinal output to SOL at both premotoneuronal and motoneuronal levels during standing.

Fig 1. Overview of the experimental procedures.

A. The schematic view of the experimental setup: Conditioning-test of the GVS on the MEP and on the H-reflex. The coil of the transcranial magnetic stimulation (TMS) was placed over the left primary motor cortex; The electrodes of the galvanic vestibular stimulation (GVS) were placed on the mastoid behind each ear; The H-reflex electrode was fastened to the right leg by a custom-made rectangular plate and straps. B. Experimental design. Time course of the experimental procedures during a session (2.5 hours). C. Data of one subject representing the effects of the bipolar galvanic vestibular stimulation (GVS) on the electromyographic (EMG) activity of the SOL muscles. The average rectified EMG responses of the right SOL with cathode right/anode left GVS (cathode behind the right ear) stimulation (30 trials at 3.5 mA) is shown in black, the control EMG, in grey. The two full vertical lines show the onset and offset of the 200 ms GVS pulse. The two dashed horizontal lines represent 1 SD above and below the mean background activity prior to the stimulation (solid horizontal line). D. Motor evoked potentials (MEP) induced by TMS in the right SOL (average of 10 stimuli). Grey trace: not stimulated; black trace: stimulated. E. Illustration of the stimulation paradigm where GVS and TMS are applied at different ISIs (30–130 ms) to MEP.

Fig 2. Short-Latency and long-latency facilitation paradigm.

A. Illustration of the short-latency facilitation where TMS pulse follows the application of the tibial nerve (ISI = -5 to -1 ms) that will trigger an H-Reflex. B. This paradigm induces a facilitation of the H-reflex amplitude and the earliest delay at which the facilitation occurred (arrow) was determined and used to assess influence of GVS on cortical neurons. C. To assess the effect of GVS on the facilitated H-reflex, GVS was applied 40 and 130 ms (cathode right/anode left) prior to the tibial nerve. D. Illustration of the long-latency facilitation where TMS pulse follows the application of the tibial nerve (only ISI -1, 2, 3 and 4 ms shown in this subject) that will trigger an H-Reflex. E. This paradigm induces a facilitation of the H-reflex amplitude and the delay at which the largest facilitation occurred (arrow) was determined and used to assess influence of GVS on corticospinal tract. F. To assess the effect of GVS on the facilitated H-reflex, GVS was applied 40 and 130 ms (cathode right/anode left) prior to the tibial nerve.

Fig 3. Modulation of motor evoked potential (MEP) in the right Soleus muscle (SOL) by galvanic vestibular stimulation (GVS).

A and E. Sample EMG traces of the average MEP (n = 10) of a representative subject illustrating the conditioning effects of GVS for four different interstimulus intervals (ISIs = 0, 40, 130 and 50 ms); control MEP (black) and conditioned MEP (grey). B and F. Timing of the effect of bipolar GVS on the MEP. The histogram bars correspond to the mean effect observed for all the subjects (n = 14). The dashed black line corresponds to the control MEP size. C and D. Significant effect of GVS on MEP in SOL compared to the effect of GVS on the H-reflex and background EMG activity in SOL for the same interstimulus interval (40 and 130 ms). The histogram bars correspond to the mean effect observed for all the subjects (n = 14). Circles represent data from each subject. Error bars represent the standard error of the mean. *p<0.05.

Fig 4. Modulation of the short-latency TMS-facilitated H-reflex in the Soleus muscle (SOL) by galvanic vestibular stimulation (GVS).

A. Short-latency facilitation of the H-reflex by TMS. Left trace (grey) is H-reflex in the Soleus of one standing subject (N = 10 sweeps). The dotted trace is the SOL EMG when only TMS is applied. The right trace (black) is SOL H-reflex when it was conditioned by the TMS at ISI = -3ms are superimposed, which leads to an increased H-reflex amplitude. B. Modulation of the facilitated H-reflex by GVS for the group. In the left graph, the 1^st and 2^nd histograms show a significant facilitation of the SOL H-reflex by TMS. The 3^rd and 4^th histograms show the effects of GVS at 2 different ISIs. Overall, there was no significant effect of GVS on the short-latency facilitated H-reflex. Mean value for each subject (circle) were also superimposed on the histogram bars. In the right graph, control M wave (bars; Y axis on the left) and background EMG level (line; Y axis on the right) are displayed for the group. C. Comparison between the mean background EMG level and the mean EMG level following application of subthreshold TMS alone. No significant effect is observed. * = p<0.05.

Fig 5. Modulation of the long-latency TMS-facilitated H-reflex in the Soleus muscle (SOL) by galvanic vestibular stimulation (GVS).

A. Late facilitation of the H-reflex by TMS. Left trace (grey) shows H-reflex in the Soleus of one standing subject (N = 10 sweeps). Middle trace (dotted) shows the SOL EMG when only TMS is applied. Right trace (black) shows the SOL H-reflex when it was conditioned by the TMS at ISI = 2ms, which leads to an increase in the H-reflex amplitude. B. Modulation of the facilitated H-reflex by GVS for the group. In the left graph, histograms representing effect of GVS on the facilitated H-reflex for the group. Control amplitude (dashed line) was that of the facilitated H-reflex (TMS on H). Mean value for each subject were also superimposed on the histogram bars (circles). In the right graph, control M wave (bars; Y axis on the left) and background EMG level (line; Y axis on the right) are displayed for the group. C. Comparison between the mean background EMG level and the mean EMG level following application of subthreshold TMS alone. No significant effect is observed. * = p<0.05.