Journal of Neurophysiology

Effect of Low-Frequency Repetitive Transcranial Magnetic Stimulation on Interhemispheric Inhibition

Pramod Kr. Pal, Ritsuko Hanajima, Carolyn A. Gunraj, Jie-Yuan Li, Aparna Wagle-Shukla, Francesca Morgante, Robert Chen

Abstract

We studied the effects of 1-Hz repetitive transcranial magnetic stimulation (rTMS) on the excitability of interhemispheric connections in 13 right-handed healthy volunteers. TMS was performed using figure-eight coils, and surface electromyography (EMG) was recorded from both first dorsal interosseous (FDI) muscles. A paired-pulse method with a conditioning stimulus (CS) to the motor cortex (M1) followed by a test stimulus to the opposite M1 was used to study the interhemispheric inhibition (ppIHI). Both CS and TS were adjusted to produce motor-evoked potentials of ∼1 mV in the contralateral FDI muscles. After baseline measurement of right-to-left IHI (pre-RIHI) and left-to-right IHI (pre-LIHI), rTMS was applied over left M1 at 1 Hz with 900 stimuli at 115% of resting motor threshold. After rTMS, ppIHI was studied using both the pre-rTMS CS (post-RIHI and post-LIHI) and an adjusted post-rTMS CS set to produce 1-mV motor evoked potentials (MEPs; post-RIHIadj and post-LIHIadj). The TS was set to produce 1-mV MEPs. There was a significant reduction in post-LIHI (P = 0.0049) and post-LIHIadj (P = 0.0169) compared with pre-LIHI at both interstimulus intervals of 10 and 40 ms. Post-RIHI was significantly reduced compared with pre-RIHI (P = 0.0015) but pre-RIHI and post-RIHIadj were not significantly different. We conclude that 1-Hz rTMS reduces IHI in both directions but is predominantly from the stimulated to the unstimulated hemisphere. Low-frequency rTMS may be used to modulate the excitability of IHI circuits. Treatment protocols using low-frequency rTMS to reduce cortical excitability in neurological and psychiatric conditions need to take into account their effects on IHI.

INTRODUCTION

Repetitive transcranial magnetic stimulation (rTMS) involves the application of regularly repeated magnetic pulses, and a stimulation rate of ≤1 Hz is referred to as low-frequency rTMS (Wassermann 1998). rTMS can activate or inhibit cortical activity depending on stimulation parameters and the effects of low-frequency stimulation are often opposite to those of high-frequency rTMS. Suprathreshold (Chen et al. 1997) and subthreshold (Romero et al. 2002) low-frequency rTMS reduce corticospinal excitability at the stimulation site. Reduction in cortical excitability, measured by suppression of size of motor-evoked potentials (MEPs), may last for >15 min after 15–25 min of stimulation at ∼1 Hz (Chen et al. 1997; Gerschlager et al. 2001). The neurophysiological effects of low-frequency rTMS form the basis of potential therapeutic application of rTMS in several neurological and psychiatric disorders (Hoffman et al. 2003; Siebner et al. 1999).

Cognition and smooth performance of voluntary movements is dependent on the proper functioning of the cortico-cortical excitatory and inhibitory processes. Single- and paired-pulse TMS protocols can test several different intracortical inhibitory and facilitatory processes (Chen 2004; Hallett 2000). These include short interval intracortical inhibition (SICI) and intracortical facilitation (ICF) (Kujirai et al. 1993), long interval intracortical inhibition (LICI) (Valls-Sole et al. 1992; Wassermann et al. 1996), and the contralateral silent period (cSP) (Cantello et al. 1992). rTMS have been shown to affect several of these processes, depending on the frequency and intensity of stimulation. Low-frequency rTMS may have no effect (Romeo et al. 2000) or shorten the cSP (Fierro et al. 2001), possibly through decreased excitability of inhibitory cortical neurons. After subthreshold 1-Hz rTMS, there is decreased ICF without concomitant change in SICI (Pascual-Leone et al. 1998; Romero et al. 2002) although a study using suprathreshold rTMS found no change in SICI or ICF (Plewnia et al. 2003). These reports suggest rTMS has different effects on different cortical circuits.

In addition to the effects in the stimulated area, rTMS also change the excitability of other cortical areas. For example, positron emission tomography studies showed that rTMS causes blood flow changes in remote but interconnected cortical areas (Paus et al. 1997, 2001). Low-frequency rTMS of the premotor cortex decreases the excitability of the motor cortex (Gerschlager et al. 2001), and low-frequency rTMS of the motor cortex may change the excitability of the contralateral motor cortex (Gorsler et al. 2003; Plewnia et al. 2003; Schambra et al. 2003; Wassermann et al. 1998).

Interhemispheric inhibition (IHI) is a physiologic phenomenon that may help to maintain hemispheric dominance in cognitive and motor tasks by suppressing undesired activity of the opposite hemisphere. There are two methods to evaluate IHI in humans: a paired-pulse method (will be termed ppIHI) and the ipsilateral silent period (iSP). ppIHI involves applying a conditioning stimulus (CS) to the motor cortex (M1), which inhibits the size of the MEP produced by a test stimulus (TS) applied to the opposite M1 6–50 ms later (Chen et al. 2003; Ferbert et al. 1992; Gerloff et al. 1998). At interstimulus intervals (ISIs) of ∼10 ms, there is inhibition of the M1 suggesting that the effect is mediated through a transcallosal pathway (Di Lazzaro et al. 1999; Ferbert et al. 1992), but subcortical connections may also be involved (Gerloff et al. 1998). iSP is the interruption of ongoing voluntary electromyographic (EMG) activity by magnetic stimulation of the ipsilateral M1 (Ferbert et al. 1992; Meyer et al. 1995). Studies in patients with lesions of the corpus callosum suggest that it is mediated through a transcallosal pathway (Meyer et al. 1995, 1998). iSP and ppIHI evoked by the paired-pulse method at short ISIs (8–10 ms) are likely mediated by different mechanisms whereas iSP and IHI evoked by long ISIs (∼40 ms) may involve similar circuits (Chen et al. 2003). Reduced IHI has been demonstrated by either ppIHI or iSP in patients with several neurological and psychiatric disorders (Daskalakis et al. 2002a; Niehaus et al. 2001; Schmierer et al. 2000; Trompetto et al. 2003) as well in professional musicians who started training at an early age (Ridding et al. 2000). Reduced resting ppIHI has been reported in patients with cortical stroke, whereas patients with subcortical stroke had normal resting ppIHI (Boroojerdi et al. 1996; Shimizu et al. 2002). Just before movement onset, ppIHI was reduced in normal subjects, but in patients with chronic subcortical stroke, ppIHI from the intact to the lesioned hemisphere persisted abnormally (Murase et al. 2004). Because IHI interacts with intracortical circuits (Daskalakis et al. 2002b), an imbalance of the normally present tonic IHI may explain the reduced intracortical inhibition and increased excitability in the unaffected hemisphere in patients with ischemic strokes in middle cerebral artery territory (Butefisch et al. 2003; Kimiskidis et al. 2002; Liepert et al. 2000; Manganotti et al. 2002).

Knowledge of how rTMS modulates IHI is useful in planning low-frequency rTMS treatment protocols to achieve the desired effects and to avoid potential adverse effects. Only one previous study examined the effects of low-frequency rTMS on IHI. Gilio et al. (2003) reported that 15 min of 1-Hz rTMS over the left M1 reduced ppIHI at short ISIs of 7 and 10 ms from left to right hemisphere. ppIHI at longer ISIs (15–75 ms), cSP and iSP from both hemispheres, and SICI and ICF in the right hemisphere were unaffected by the rTMS. However, in the study of Gilio et al. (2003), there was no change in the excitability of the stimulated left hemisphere, which may be desirable in treating patients with conditions associated with increased cortical excitability. Moreover, IHI from the unstimulated to the stimulated hemisphere was not examined. In the present study, we used a low-frequency rTMS protocol that has been shown to reduce motor cortex excitability (Chen et al. 1997; Fitzgerald et al. 2002) and examined the effect on interhemispheric connections between both motor cortices.

METHODS

Subjects

We studied 13 healthy volunteers (4 women and 9 men), aged 28–59 yr [40.9 ± 12.9 (SD) yr] in the main experiment. Subjects were recruited through advertisements in the community and postings within the hospital. All subjects gave their written informed consent, and the protocol was approved by the University Health Network Research Ethics Board in accordance with the declaration of Helsinki on the use of human subjects in experiments.

EMG recording

Surface EMG was recorded from the right and left first dorsal interosseous (FDI) muscles, using disposable disc electrodes placed in a tendon-belly arrangement over the bulk of the FDI muscle and the first metacarpal-phalangeal joint. The subjects were instructed to maintain muscle relaxation throughout the study, and the EMG was monitored on a computer screen and via speakers at high gain. The EMG signal was amplified (Model 2024F, Intronix Technologies, Bolton, Ontario, Canada), filtered (band-pass: 2 Hz to 2.5 kHz), digitized at 5 kHz (Micro 1401, Cambridge Electronics Design, Cambridge, UK) and stored in a laboratory computer for off-line analysis.

TMS procedure

A conditioning stimulus (CS)-test stimulus (TS) paradigm similar to that described by Ferbert et al. (1992) was used to evaluate IHI. A suprathreshold CS was delivered to the M1 followed by a suprathreshold TS delivered to the contralateral M1 at ISIs of 4, 10, and 40 ms. ISIs of 10 and 40 ms were chosen because they test different aspects of ppIHI (Chen et al. 2003), and ISIs of 4 ms may induce interhemispheric facilitation (IHF) (Hanajima et al. 2001). Both the left-to-right ppIHI (LIHI) and right-to-left ppIHI (RIHI) were examined. ppIHIs before rTMS are indicated by pre-LIHI and pre-RIHI and those after rTMS by post-LIHI and post-RIHI.

For the IHI studies, four Magstim 200 stimulators (Magstim, Dyfed, UK) and two Bistim modules were used. Each motor cortex was stimulated with two Magstim stimulators connected via a Bistim module to a 7-cm figure-eight coil. This setup allowed us to use different stimulus intensities for the one hemisphere in the same experimental run for the post-rTMS IHI study. The stimulators produced monophasic current.

Pre-rTMS IHI

The optimal positions for the right and left M1 to produce MEPs in the contralateral FDI muscles were determined and marked on the scalp to ensure identical placement of the coil throughout the experiment. The coil was placed tangentially to the scalp with the handle pointing backward and laterally at an approximate angle of 45° to the midsagittal line. For each M1, both the CS and TS were the minimum stimulus intensity (determined to the nearest 1% of the maximum stimulator output) that produces a peak-to-peak MEP amplitude of ≥1 mV in ≥5 of 10 trials (indicated as 1mVLPre for left M1 and 1mVRPre for right M1). In some subjects, it was not possible to hold both coils at the optimal positions because of the size of the coil and minor changes of coil positions were required. In these subjects, the stimulus intensities required to obtain ∼1-mV MEPs were determined with the coils at the adjusted positions, and same positions were used before and after rTMS.

The test conditions are shown in Table 1. Conditions 1–3 tested pre-RIHI and conditions 4–6 tested pre-LIHI. For pre-RIHI, the MEP amplitude for the TS alone from left M1 stimulation was taken from the MEP produced by the CS in condition 6pre. Similarly, for pre-LIHI, the MEP amplitude for TS alone was taken from the MEP produced by the CS in condition 3pre. Each run consisted of 10 trials of each of the six conditions delivered in a random order (60 trials) with 6 s between trials.

View this table:
TABLE 1.

Stimulus conditions used in pre-rTMS IHI study

rTMS procedure

rTMS of left M1 was performed with a 7-cm figure-eight coil connected to a Magstim Super Rapid stimulator. This stimulator produces biphasic currents. The coil was placed tangentially to the scalp with the handle pointing backwards and laterally at an approximate angle of 45° to the midsagittal line, and was perpendicular to the presumed direction of the central sulcus. The optimal position (rTMS position) for activating the right FDI was determined, and the position was marked over the scalp to ensure accurate placement of the coil during rTMS. The resting motor threshold (RMT) was determined and was the lowest intensity that produced MEPs of >50 μV in ≥5 of 10 trials.

Before rTMS, the baseline MEP size was recorded from right FDI muscle at rest by delivering 20 stimuli (115% of RMT) at 0.1 Hz. The rTMS procedure consisted of 900 stimuli (15 min) at 1Hz at 115% of RMT. The subjects were asked to keep the muscle relaxed, with EMG auditory feedback. When increased coil temperature was detected, rTMS was stopped for ∼30 s to replace the coil. All subjects required one or two replacements of the coil.

Post-rTMS IHI

After rTMS, the new stimulus intensity for each M1 that produced a peak-to-peak MEP amplitude of ≥1 mV from the contralateral FDI was determined (indicated as 1mVRpost for right M1 and 1mVLPost for left M1). This process took 3–4 min. The test conditions for post-rTMS RIHI and LIHI are shown in Table 2. Because ppIHI may be influenced by the CS intensity, we used both the pre-rTMS (the resultant IHI will be termed post-RIHI and post-LIHI) and the adjusted post-rTMS intensity that produce 1-mV MEPs (post-RIHIadj and post-LIHIadj) as the CS, whereas the TS was always the post-rTMS intensity that produced 1-mV MEPs (1mVPost). ISIs of 4, 10, and 40 ms were tested. Thus post-RIHIadj was studied in conditions 1Post to 3Post (Table 2), post-RIHI in conditions 4Post to 6Post, post-LIHIadj in conditions 7Post to 9Post, and post-LIHI in conditions 10Post to 12Post. For post-LIHI and post-LIHIadj, the MEP amplitude for the TS alone to the right M1 was taken from the MEP induced by the CS in condition 9Post. Similarly, for post-RIHI and post-RIHIadj, the MEP amplitude for the TS alone was taken from the MEP induced by the CS in condition 3Post. Each run consisted of 10 trials of each of the 12 conditions delivered in a random order (120 trials) with 6 s between trials.

View this table:
TABLE 2.

Stimulus conditions used in post-rTMS IHI study

Control experiment

We performed a control experiment to study the effects of rTMS current direction on IHI and cortical excitability. Four subjects (4 men, aged 20–61 yr) were recruited. rTMS was delivered over the left motor cortex with the handle of the coil pointing anteromedially, similar to a the study of Gilio et al. (2003), at 1 Hz for 15 min at 115% RMT. The rTMS paradigm as well as the pre-rTMS and the post-rTMS studies were otherwise identical to the main experiment.

Data analysis

The peak-to-peak MEP amplitude for each trial was measured off-line. Paired t-test were used to examine the differences between the stimulus intensities used for 1mVPre and 1mVPost, and between amplitudes of unconditioned (TS alone) MEPs for pre- and post-rTMS.

Effect of rTMS on MEP amplitude

To examine the changes in MEP amplitudes during the rTMS period, the mean MEP amplitudes of each block of 100 stimuli (9 blocks for 900 stimuli) were expressed as a percentage of the mean amplitude of the 20 MEPs recorded prior to rTMS. Some subjects were unable to relax the target muscle during rTMS. Therefore trials with voluntary EMG activity were excluded. To avoid subjective bias in determining which trials should be rejected, the background EMG area between the TMS stimulus artifact and MEP onset was calculated for each of the 20 pre-rTMS MEPs. A trial during the rTMS was excluded if the EMG area of the same period exceeded mean ± 2 SD of the baseline EMG area. In addition, a subject was excluded from this analysis if >50 trials were rejected in an epoch of 100 consecutive trials. The effects of duration of rTMS were evaluated by repeated-measures ANOVA. If the effect of stimulus duration was significant, Fisher's protected least-significant difference (PLSD) post hoc test was used to detect differences among different epochs of stimuli.

ppIHI

ppIHI was expressed as a ratio of the MEP amplitude for each conditioned trial to the mean unconditioned (TS alone) MEP amplitude for each experimental run. Ratios less than one indicate inhibition, and ratios greater than one indicate facilitation.

For ppIHI of each side, a repeated-measures ANOVA with test conditions (pre-IHI, post-IHI, and post-IHIadj) as the repeated measure and ISI (10 and 40 ms) as a dependent variable was performed. A separate repeated-measures ANOVA was performed to examine the effects of test conditions on ISI of 4 ms that may show interhemispheric facilitation. If the effects of test condition or ISI were significant, PLSD post hoc test was used to detect differences among the different test conditions or ISIs.

RESULTS

Three subjects reported mild headache after rTMS, and there were no other adverse effects. In the pre-rTMS studies, no IHI was detected in one subject at ISI of 10 ms and in two subjects at ISI of 40 ms. These subjects was excluded from the analysis of the effect of rTMS on IHI.

Changes in MEP amplitude during rTMS

The stimulation intensity for rTMS for the nine subjects was 64.0 ± 10.5% (mean ± SD) of stimulator output and the amplitude of the baseline MEP before rTMS was 1.29 ± 1.14 mV. Four subjects were unable to maintain muscle relaxation during rTMS and were excluded from the analysis of the effect of rTMS on MEP amplitude. Figure 1 shows the effect of duration of rTMS on the mean MEP amplitude in each epoch. Repeated-measures ANOVA showed a significant effect of time on MEP size (P = 0.032). Post hoc testing showed that the MEP amplitude of the last epoch of 100 stimuli was significantly lower than that of the baseline (P = 0.007).

FIG. 1.

Plot of motor-evoked potential (MEP) amplitude with number of stimuli delivered during 1-Hz stimulation. Data from 9 subjects. Four subjects who were not able to maintain muscle relaxation were excluded. For each subject, the mean MEP amplitudes from every 100 stimuli were normalized to the mean pre-repetitive transcranial magnetic stimulation (pre-rTMS) baseline MEP amplitude obtained from 20 stimuli delivered at 0.1 Hz using same stimulus intensity as for rTMS. Each error bar represents 1 SE. Repeated-measures ANOVA showed a significant effect of time on MEP size (P = 0.032). Post hoc testing showed that the MEP amplitude of the last epoch of 100 stimuli was significantly lower than that of the baseline (*P = 0.007).

Changes in MEP amplitude before and after rTMS

Before rTMS, the TS to produce 1-mV MEPs for left M1 was 57.4 ± 11.7% (1mVLpre, Tables 1 and 2) and for right M1 was 57.9 ± 12.5% (1mVRpre, Tables 1 and 2) of stimulator output, which increased after rTMS to 61.6 ± 14.7% (1mVLpost, Table 2) for left M1 and 60.1 ± 17.0% (1mVRpost, Tables 2) for right M1, the difference being significant only for left M1 (P = 0.023). There was no significant difference in the amplitudes of the unconditioned MEPs (TS alone) pre- and post-rTMS (right FDI: pre-rTMS = 1.44 ± 0.92 mV (generated by 1mVLpre pulse, Table 1), post-rTMS = 1.69 ± 0.84 mV (1mVLpost, Table 2); left FDI: pre-rTMS = 1.60 ± 1.11 mV (1mVRpre, Table 1), post-rTMS = 1.93 ± 1.35 mV (1mVRpost, Table 2).

Effect of left M1 rTMS on LIHI

The results are shown in Fig. 2. Repeated-measures ANOVA showed a significant effect of test conditions (pre-LIHI, post-LIHIadj, post-LIHI, P = 0.01) on LIHI. The effects of ISI or interaction between test conditions and ISI were not significant. Post hoc analysis showed a significantly reduced IHI for post-LIHI (P = 0.0049) and post-LIHIadj (P = 0.0169) compared with pre-LIHI, but no significant difference between post-LIHI and post-LIHIadj.

FIG. 2.

Effects of rTMS on interhemispheric inhibition of left motor cortex on right (LIHI) at different interstimulus intervals (ISIs). pre-LIHI, LIHI before rTMS; post-LIHI, LIHI after rTMS using post-rTMS adjusted test stimulus and pre-rTMS conditioning stimulus; post-LIHIadj, LIHI after rTMS using post-rTMS adjusted test as well as conditioning stimuli. Each error bar represents 1 SE. Repeated-measures ANOVA showed a significant effect of test conditions (P = 0.01) on LIHI but not of ISI or any significant interaction between test conditions and ISI. Post hoc analysis of the effect of ISI did not show any significant difference between 10 and 40 ms. Post hoc analysis of the effect of test condition on LIHI showed a significant difference between pre-LIHI and post-LIHI (*P = 0.0049) and between pre-LIHI and post-LIHIadj (**P = 0.0169), but not between post-LIHI and post-LIHIadj.

Effect of left M1 rTMS on RIHI

The results are shown in Fig. 3. Repeated-measures ANOVA showed a significant effect of test condition (pre-RIHI, post-RIHIadj, post-RIHI, P = 0.006) on RIHI. The effects of ISI or interaction between test conditions and ISI were not significant. Post hoc analysis showed a significant reduction in IHI for post-RIHI (P = 0.0015) compared with pre-RIHI, but no significant difference between pre-RIHI and post-RIHIadj and between post-RIHI and post-RIHIadj.

FIG. 3.

Effects of rTMS on interhemispheric inhibition from right to left motor cortex (RIHI) at different ISIs. pre-RIHI, RIHI before rTMS; post-RIHI, RIHI after rTMS using post-rTMS adjusted test stimulus and pre-rTMS conditioning stimulus; post-RIHIadj, RIHI after rTMS using post-rTMS adjusted test as well as conditioning stimuli. Each error bar represents 1 SE. Repeated-measures ANOVA showed a significant effect of test condition (P = 0.006) on RIHI but not of ISI or any significant interaction between time and ISI. Post hoc analysis of effect of ISI did not show any significant difference in RIHI between 10 and 40 ms. Post hoc analysis of the effect of test condition on RIHI showed a significant difference between pre-RIHI and post-RIHI (*P = 0.0015) but not between pre-RIHI and post-RIHIadj and between post-RIHI and post-RIHIadj.

Effect of rTMS on interhemispheric transmission at ISI of 4 ms

In the pre-rTMS study, IHF was absent at ISI of 4 ms on both sides. The mean ratio of the conditioned to unconditioned MEP amplitude of right FDI was 0.99 ± 0.24 and that of left FDI was 0.91 ± 0.25. In the post-rTMS period, the ratios did not change significantly on both sides. Using the pre-rTMS CS, the ratios were 1.06 ± 0.36 for the right FDI and 1.07 ± 0.40 the left FDI, and using the adjusted CS these ratios were 1.05 ± 0.31 for the right FDI and 1.04 ± 0.35 for the left FDI.

Control experiment

Changes in MEP amplitude during rTMS.

The stimulation intensity for rTMS was 57.7 ± 11.5% of stimulator output and the amplitude of the baseline MEP before rTMS was 0.66 ± 0.25 mV. Repeated-measures ANOVA showed no significant effect of time on MEP size (P = 0.931).

Changes in MEP amplitude before and after rTMS

Before rTMS, the TS to produce 1-mV MEPs for left M1 was 51.5 ± 7.8% and for the right M1 was 53.7 ± 5.9% of stimulator output. After rTMS the TS to produce 1mV MEPs for the left M1 was 53.7 ± 7.1% and for the right M1 was 53.7 ± 6.6%. The differences in TS intensity before and after rTMS were not significant. There was no significant difference in the amplitudes of the unconditioned MEPs before and after rTMS (right FDI: pre-rTMS = 1.38 ± 0.39 mV, post-rTMS = 1.54 ± 0.36; left FDI: pre-rTMS = 1.53 ± 0.44 mV, post-rTMS = 1.6 ± 0.62 mV).

Effect of left M1 rTMS on LIHI

The results are shown in Fig. 4. Repeated-measures ANOVA showed a significant effect of test conditions (pre-LIHI, post-LIHIadj, post-LIHI P = 0.0126) on LIHI. The effect of ISI or interaction between test conditions and ISI were not significant. Post hoc analysis showed a significantly reduced IHI for post-LIHI (P = 0.028) and post-LIHIadj (P = 0.0045) compared with pre-LIHI, but no significant difference between post-LIHI and post-LIHIadj. Because a previous study (Gilio et al. 2003) found changes only for ISI of 10 ms, we performed an additional analysis examining the effects of test conditions separately for ISIs of 10 and 40 ms. The effect of test condition was significant for ISI of 10ms (P = 0.05) but not for ISI of 40 ms (P = 0.187).

FIG. 4.

Effects of rTMS in the control experiment (with handle of coil pointing anteriomedially) on LIHI at different ISIs. Each error bar represents 1 SE. Repeated-measures ANOVA showed a significant effect of test conditions (P = 0.0126) on LIHI but not of ISI or significant interaction between test conditions and ISI. Post hoc analysis of the effect of ISI did not show any significant difference between 10 and 40 ms. Post hoc analysis of the effect of test condition on LIHI showed a significant difference between pre-LIHI and post-LIHI (*P = 0.028) and between pre-LIHI and post-LIHIadj (**P = 0.0045) but not between post-LIHI and post-LIHIadj.

Effect of left M1 rTMS on RIHI

Repeated-measures ANOVA showed no significant effect of test conditions (pre-RIHI, post-RIHIadj, post-RIHI P = 0.0949) on RIHI.

Effect of rTMS on interhemispheric transmission at ISI of 4 ms

Similar to the main experiment, there was no significant inhibition or facilitation pre-rTMS. The mean ratio of the conditioned to unconditioned MEP amplitude of right FDI was 0.98 ± 0.23 and that of left FDI was 0.97 ± 0.13, and there was no significant change post-rTMS.

DISCUSSION

Low frequency rTMS produced a significant reduction in the excitability of the ipsilateral M1, similar to earlier reports (Chen et al. 1997; Fitzgerald et al. 2002; Maeda et al. 2000; Muellbacher et al. 2000). This inhibitory effect persisted following rTMS because the stimulation intensity for left M1 for a 1-mV MEP from right FDI also increased significantly. Our results showed that 15 min of rTMS at 1 Hz over the left M1 significantly reduced the LIHI and to a less extent the RIHI. Because the post-rTMS assessment of IHI started 3–4 min after rTMS and lasted 12 min, this inhibitory effect on IHI lasted for ≥15 min after rTMS and affected ppIHI at both ISIs of 10 and 40 ms. Because only post-RIHI but not post-RIHIadj was significantly reduced compared with pre-RIHI, the reduction in RIHI can be reversed by slightly increasing the CS on the right M1 in the post-RIHIadj condition. There may be greater reduction of LIHI compared with RIHI as both the post-LIHI and post-LIHIadj were reduced compared with pre-rTMS values.

Interhemispheric facilitation

ISI of 4 ms was used to examine IHF. IHF is produced when the test MEPs are elicited using a posteriorly directed current at low stimulus intensities (Hanajima et al. 2001). Because we are primarily interested in ppIHI and due to the time constraints after rTMS, our testing paradigm was not optimal for eliciting IHF. This likely explains why IHF was not observed in our study. Our findings suggest that IHF was not grossly increased after rTMS and the reduction in ppIHI was probably not due to increased IHF. However, minor changes in IHF cannot be ruled out.

Previous studies on the effect of low-frequency rTMS on IHI

Our finding of decreased LIHI at ISI of 10 ms in the main experiment is similar to that of Gilio et al. (2003). However, Gilio et al. (2003) found no change in LIHI at longer intervals of 15–75 ms and no change in iSP. Moreover, Gilio et al. (2003) found that left rTMS led to increased MEP amplitude with right M1 stimulation while we found no significant change in the stimulus intensity needed to produce the same MEP as the pre-rTMS condition. The different findings may be related to different study design. While the number of pulses, stimulation frequency and the intensity of rTMS were similar (mean of 117% rest MT in Gilio et al. (2003) and 115% rest MT in the present study) in the two studies, in Gilio et al. (2003), the current direction for rTMS (handle pointing forward) was opposite to that used in our main experiment, and rTMS did not reduce the MEP size from the stimulated hemisphere. In our control experiment using the same current direction as Gilio et al. (2003), we confirmed their findings of no change in MEP amplitude from the stimulated hemisphere and reduced LIHI at ISI of 10 ms. The failure to find a significant difference between ISIs of 10 and 40 ms (no significant effect of ISI) may be related to the low number of subjects tested. Separate ANOVAs for ISI of 10 and 40 ms showed a significant effect for test condition only for ISI of 10 ms, but this point needs to be examined further in future studies. Thus our findings suggest that the rTMS with the first phase of the biphasic current inducing posterior to anterior current in the brain used in the main experiment reduce corticospinal excitability and IHI at both short and long ISIs. By contrast, rTMS with the first phase inducing anterior to posterior current (Gilio et al. 2003) has little effect on corticospinal excitability. Therefore the effects of 1-Hz rTMS on neuronal circuits are dependent on the current direction, and this should be taken into account in the design of treatment studies for neurological and psychiatric disorders using low-frequency rTMS.

Mechanism of rTMS induced changes in IHI

Because there are no known long-range inhibitory neurons that cross the corpus callosum, IHI is probably mediated by transcallosal excitatory fibers originating from the ipsilateral motor cortex that synapse with local circuits of inhibitory interneurons in the contralateral cortex, which finally inhibit the corticospinal neurons (Berlucci 1990). Therefore reduction of IHI can result from inhibition of transcallosal excitatory neurons in the originating M1 or inhibition of the inhibitory interneurons of the contralateral M1. Our results do not allow us to distinguish between these possibilities. The reduction in LIHI after left M1 rTMS can be due to inhibition of transcallosal fibers originating from the left M1 if there is reduced excitability of the transcallosal projection, similar to the changes in the corticospinal projection from the left M1. However, IHI interacts with intracortical inhibitory circuits (Daskalakis et al. 2002b) and repeated activation of transcallosal fibers by rTMS may lead to changes in inhibitory circuits in the right M1. This possibility is supported by the observation that increasing the conditioning stimulus intensity of the left M1 to compensate for the reduction in corticospinal excitability did not reverse the decreased LIHI after rTMS. It has been hypothesized that IHI may share common mechanism with LICI (Chen 2004; Daskalakis et al. 2002b), and LICI is likely related to the cSP (Wassermann et al. 1996). If this is correct, the slight reduction in RIHI can be related to the observation that 1-Hz rTMS shortens the cSP duration (Fierro et al. 2001) as these can be explained by reduced excitability of the inhibitory neurons mediating cSP and IHI in the stimulated (left) M1. However, changes in the excitability of transcallosal projection from the right M1 are also possible.

Abnormal IHI in neuropsychiatric disorders and potential therapeutic role of low-frequency rTMS

Low-frequency rTMS is being investigated as treatment of neurological and psychiatric disorders such as Parkinson's disease (Ikeguchi et al. 2003; Okabe et al. 2003; Shimamoto et al. 2001), dystonia (Siebner et al. 1999), hemispheric neglect (Brighina et al. 2003), and schizophrenia (Hoffman et al. 1999, 2000; Rollnik et al. 2000) because of the ability to reduce cortical excitability. Our findings suggest that these treatment regimens may also affect the transcallosal pathways.

Few studies have systematically studied IHI in disease states. Reduction or imbalance of IHI has been reported in several neurological and psychiatric conditions. These include decreased ppIHI at short ISIs in schizophrenia (Daskalakis et al. 2002a), which has been described as a disorder of dysfunctional cerebral connectivity (Crow 1998), and reduced iSP in multiple sclerosis with callosal lesions (Schmierer et al. 2000), corticobasal degeneration (Trompetto et al. 2003), and writer's cramp (Niehaus et al. 2001).

In patients suffering from acute ischemic stroke in middle cerebral artery territory, reduced intracortical inhibition and increased excitability of the unaffected hemisphere have been reported (Butefisch et al. 2003; Kimiskidis et al. 2002; Liepert et al. 2000; Manganotti et al. 2002). ppIHI from the unaffected to the affected hemisphere was found to be increased in chornic subcortical stroke just prior to movement onset (Murase et al. 2004). Such pathophysiological changes may explain the phenomena of neglect or extinction resulting from breakdown of the balance of hemispheric rivalry from unilateral hemispheric lesions (Kinsbourne 1977). There is a relative disinhibition of the unaffected hemisphere and generation of an unopposed orienting response toward ipsilesional space.

Low-frequency rTMS of the unaffected left parietal cortex has been reported to improve contralesional visuospatial hemineglect in right brain damaged patients for ≤15 days (Brighina et al. 2003). While it was suggested that this effect is due to rTMS-induced long-lasting depression of the left parietal cortex (Brighina et al. 2003), reduction in left to right IHI may also be involved and may help to balance transcallosal inhibitory activity. In the other conditions with asymmetrically reduced interhemispheric inhibitions, low-frequency rTMS over the side with better preserved IHI may be tested as a therapeutic option.

Reduction of IHI has been observed in professional musicians who began musical training at an early age (Ridding et al. 2000). Low-frequency (1 Hz) rTMS to the motor cortex was reported to improve ipsilateral finger movements (Kobayashi et al. 2004). These findings raise the possibility that reduction in interhemispheric inhibition induced by rTMS may improve hand dexterity.

In summary, 1-Hz suprathreshold rTMS over M1 reduces IHI. The effect is bi-directional, but it is more prominent from the stimulated to unstimulated hemisphere. Further studies on how different rTMS parameters such as frequency, stimulus intensity, and current direction affect IHI will give valuable information in designing rational treatment strategies using rTMS.

GRANTS

This work was supported by the Canadian Institutes of Health Research, Premier's Research Excellence Award, Canada Foundation for Innovation, Ontario Innovation Trust, and University Health Network Krembil Family in Neurology.

Acknowledgments

Present address of P. K. Pal: Dept. of Neurology, National Institute of Mental Health and Neurosciences, Bangalore-560029, Karnataka, India.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract