By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
There is some limited evidence suggesting that the spaced application of repetitive transcranial magnetic stimulation (rTMS) protocols may extend the duration of induced neuroplastic changes. However, this has yet to be demonstrated in the human primary motor cortex (M1). We evaluated whether the paired application of an inhibitory rTMS protocol [continuous theta burst stimulation (cTBS)] at 10-min intervals prolonged the duration of induced M1 plasticity. Motor evoked potentials (MEPs) were recorded from the right first dorsal interosseous muscle before and following single and paired cTBS protocols applied with two intensities: 80% of active motor threshold (AMT80) and 70% of resting motor threshold (RMT70). Single cTBS protocols did not significantly influence MEP amplitudes. Whereas paired trains applied at AMT80 had no effect on MEP amplitudes, paired cTBS trains at RMT70 significantly reduced them. MEP amplitudes remained suppressed for at least 2 h following the second train. Control experiments suggested that the contraction used to establish active motor threshold prior to cTBS application may be responsible for blocking the effect of paired cTBS trains at AMT80. The results suggest that the spaced application of cTBS protocols may be an effective approach for establishing long-lasting M1 neuroplasticity only in the absence of prior voluntary motor activation. These findings may have important implications for the therapeutic application of rTMS.
Over recent years, there has been an explosion in the number of studies using non-invasive brain stimulation techniques to investigate the functioning of the human brain. Additionally, there is much interest in using these techniques to induce lasting neuroplastic changes that might be beneficial for the rehabilitation of patients with a wide range of neurological/psychiatric conditions, including depression (Fitzgerald et al., 2003; George et al., 2010), stroke (Khedr et al., 2005; Fregni et al., 2006), and Parkinson’s disease (Lomarev et al., 2006). Perhaps the most popular experimental protocol used to induce neuroplasticity within the human cortex involves the application of trains of transcranial magnetic stimuli [repetitive transcranial magnetic stimulation (rTMS)], a technique that induces lasting plasticity via mechanisms similar to the long-term potentiation (LTP) and long-term depression (LTD) observed in animal models (Cooke & Bliss, 2006; Huang et al., 2007; Ridding & Rothwell, 2007).
Although there is good evidence that the plastic changes induced by rTMS are caused by LTP/LTD-like mechanisms (Huang et al., 2007), the magnitude of such change is highly variable (Ridding & Ziemann, 2010) and commonly has a duration of < 1 h. This is considerably shorter in duration than the LTP/LTD induced in animal models, which may last anywhere from a few hours to several days (Malenka & Bear, 2004). Indeed, the comparatively short lifetime of rTMS-induced after-effects may limit the therapeutic potential of rTMS, with any positive effects being too short to have any long-term therapeutic benefit.
One potential approach to prolong the duration of experimentally induced human cortical plasticity involves the repeated application of rTMS protocols in a spaced manner. In animal models, the repeated application of stimulation protocols can enhance the lifetime of activity-dependent synaptic plasticity (Bliss & Gardner-Medwin, 1973; Abraham et al., 1993, 2002). Interestingly, a similar effect has been observed in human subjects, with multiple applications of an inhibitory rTMS paradigm to the frontal eye field producing behavioural effects that lasted for several hours, significantly longer than the effects observed following a single train (Nyffeler et al., 2006a). Whereas evidence for the prolongation of plastic effects has been observed in the human primary motor cortex (M1) with repeated applications of other non-invasive brain stimulation techniques (Monte-Silva et al., 2010), a similar extension of the duration of induced effects with the repeated application of rTMS protocols has yet to be demonstrated.
Therefore, in the present study we aimed to determine whether the paired application of rTMS protocols could prolong the duration of plasticity induced in the human M1. To do this, we employed the inhibitory rTMS paradigm of continuous theta burst stimulation (cTBS), a well-characterised technique for inducing neuroplastic changes in M1 (Huang et al., 2005). Given previous reports that prior voluntary motor activation can influence cTBS-induced M1 plasticity (Gentner et al., 2008), a secondary aim was to examine whether behavioural engagement of M1 prior to application of cTBS influenced the outcome. This was achieved by setting the cTBS intensity relative to either active motor threshold (AMT) or resting motor threshold (RMT).
Materials and methods
All participants were screened for any contraindications to transcranial magnetic stimulation (TMS) (Rossi et al., 2009), and gave informed written consent prior to their involvement in this study. A total of 22 healthy subjects (eight males) between the ages of 19 and 47 years (mean age, 24.2 ± 1.4 years) were investigated. Twelve subjects participated in Experiments 1–3 (five males; mean age, 26.3 ± 2.3 years), and six of the 12 returned for Experiment 4 (three males; mean age, 29.7 ± 4.0 years). Experiment 5 was performed on a total of nine subjects (four males; mean age 22.1 ± 3.7 years), two of whom had participated in Experiments 1–4, and Experiment 6 was performed on a total of seven subjects (four males; mean age, 28.9 ± 3.5 years), four of whom had participated in Experiments 1–3. All experiments conformed to the Declaration of Helsinki and were approved by the University of Adelaide Human Research Ethics Committee.
Stimulation and recording
Subjects were seated in a comfortable chair for each experimental session, and were directed to stay as relaxed as possible unless instructed otherwise. Changes in corticospinal excitability were assessed with single-pulse TMS to evoke motor evoked potentials (MEPs) from the right first dorsal interosseous (FDI) muscle. Surface electromyographic recordings were made from the right FDI muscle with Ag–AgCl surface electrodes and a sampling rate of 5 kHz. Signals were amplified with a gain of 1000 and filtered (20–1000 Hz) (Cambridge Electrical Design 1401, Cambridge, UK), before being stored on a computer for offline analysis.
Single-pulse TMS with a monophasic waveform was performed with a Magstim 200 magnetic stimulator (Magstim, Whitland, UK) connected to a figure-of-eight magnetic coil (external wing diameter, 90 mm) held tangentially to the skull over the left M1 with the handle pointing posteriorly and laterally at a 45° angle to the sagittal plane. A felt marker was used to mark the optimal scalp position for eliciting MEPs in the right FDI muscle at rest. Single-pulse TMS was applied at an intensity sufficient to evoke baseline MEPs of approximately 1-mV amplitude (measured peak-to-peak). Two blocks of 15 MEPs were recorded prior to the intervention in each experimental session, and averaged to yield a baseline measure of corticospinal excitability. All pre-intervention and post-intervention MEPs were recorded at a rate of 0.2 Hz (using 10% variance) unless indicated otherwise. Any change in MEP amplitude following intervention was used as a marker for M1 plasticity.
Theta burst stimulation (TBS) paradigm
In all sessions, cTBS was applied in trains of repetitive biphasic magnetic pulses to the left M1 with a Magstim Super Rapid stimulator (Magstim) connected to an air-cooled figure-of-eight coil. Each cTBS train consisted of 600 pulses applied in bursts of three pulses at 50 Hz, with bursts repeated at a frequency of 5 Hz, corresponding to a total train length of 40 s (Huang et al., 2005). When paired trains of cTBS were applied, an inter-train interval of 10 min was employed.
The stimulation intensity of cTBS was set relative to either AMT or RMT. AMT was defined as the minimum stimulus intensity necessary to evoke MEPs with peak-to-peak amplitudes > 200 μV in at least five of 10 consecutive trials while the subject performed a sub-maximal isometric contraction of their right FDI muscle. The force of contraction during AMT measurement was maintained at 10% of maximum voluntary contraction by the use of visual feedback displayed on an oscilloscope. RMT was defined as the minimum stimulus intensity necessary to evoke MEPs with peak-to-peak amplitudes > 50 μV in at least five of 10 consecutive trials while the subject was at rest (i.e. right FDI muscle relaxed).
Sham stimulation was delivered with a sham rTMS coil (placebo coil PN 3285-00; Magstim), which produced the same number and pattern of auditory stimuli as real cTBS without inducing an electric current in the brain.
A total of six experiments were included in this study (Fig. 1). For Experiment 1, subjects attended two sessions, receiving a single train of cTBS at 80% of AMT (AMT80) in one session, and paired trains of cTBS (also at an intensity of AMT80) in the other. For Experiment 2, subjects were again required to attend two sessions, receiving a single train of cTBS in one session and paired cTBS trains in the other; however, in this experiment, the intensity of cTBS trains was set at 70% of RMT (RMT70) instead of AMT80 (Gentner et al., 2008). Additionally, for this experiment, the single cTBS train was paired with a sham control (applied first). It should be noted that the sham stimulation condition was not used for Experiment 1. This was because we wanted the time interval between completion of AMT determination and the onset of cTBS application (the first application in the case of the paired cTBS intervention) to be the same for each of the relevant interventions.
Experiment 3 was a control study designed to investigate more closely the impact that the initial contraction used to establish AMT may have had on the response to paired cTBS trains. Subjects attended one session, in which they received paired trains of cTBS applied at RMT70 (as in Experiment 2); however, prior to application of the first cTBS train, subjects were instructed to sustain a sub-maximal isometric contraction of their right FDI muscle for 3 min (similar to the contraction performed during AMT measurement in Experiment 1). As with AMT measurement, subjects were provided visual feedback during the contraction to ensure that force production was maintained at 10% maximum voluntary contraction.
Although we expected that the absolute intensities employed for RMT70 and AMT80 would be very similar, we actually found that RMT70 stimulation resulted in a slightly higher intensity than that used for AMT80 (see Results). Therefore, in order to investigate whether this small difference in stimulation intensity might explain differences in the response to cTBS at RMT70 and AMT80, a second control experiment was conducted. For this experiment (i.e. Experiment 4), subjects attended one session in which they again received paired cTBS trains; however, the intensity of stimulation was set at 65% of RMT (RMT65).
For Experiments 1–4, blocks of 15 MEPs were recorded at 5, 10, 20, 30, 40, 50 and 60 min following completion of the intervention period. Each session was separated by at least 2 days.
To more comprehensively characterise the duration of after-effects produced by paired cTBS trains applied at RMT70, a fifth experiment (i.e. Experiment 5) was conducted. Subjects for this experiment attended one session in which they received paired trains of cTBS at RMT70. MEPs were then recorded in blocks of 15 trials at 5, 10, 20, 30, 40, 50, 60, 75, 90, 105 and 120 min following completion of the second cTBS train.
A final experiment (Experiment 6) was conducted to more fully explore the influence that a subject’s response to the first cTBS in a pair might have on the response to a paired cTBS paradigm. For this, subjects again received paired cTBS trains at RMT70, with MEPs recorded in blocks of 15 trials at 5, 10, 20, 30, 40, 50 and 60 min following completion of the second cTBS train (as in Experiments 1–4); however, following the first train and prior to application of the second train, a block of 42 MEPs were recorded at a frequency of 0.1 Hz to determine the response to the first cTBS train.
All experimental sessions were performed in the afternoon to prevent any time-of-day effects from influencing our results (Sale et al., 2007). Given the impact that physiological activity can have on TBS-induced neuroplasticity induction (Huang et al., 2008), surface electromyographic activity was monitored at all times in all subjects for each experiment to ensure that the FDI muscle was completely at rest during periods when a contraction was not required. Trials contaminated with background muscle activation prior to TMS application were excluded from analysis.
All statistical analyses were performed with pasw statistics version 17 (IBM SPSS, Armonk, NY, USA). MEP amplitudes were expressed as a percentage of baseline for comparisons between two or more interventions. Comparisons between single and paired cTBS for both AMT80 (Experiment 1) and RMT70 (Experiment 2) were performed with two-way repeated measures analysis of variance (anovaRM) with INTERVENTION (two levels – single cTBS and paired cTBS) and TIME (seven levels – 5, 10, 20, 30, 40, 50, and 60 min) as within-subject factors. A one-way anovaRM was then performed on raw data for each intervention separately (i.e. single cTBS at AMT80, paired cTBS at AMT80, single cTBS at RMT70, and paired cTBS at RMT70) with TIME (eight levels – baseline plus the seven post-intervention time points) as a within-subject factor. If a significant main effect of time was observed, paired t-tests were used to determine at which time points MEP amplitudes were significantly different from baseline. Multiple comparisons were corrected for by use of the false discovery rate procedure (FDRP) (Curran-Everett, 2000).
To investigate whether the response to paired cTBS trains might be influenced by the subject’s response to a single train, correlations between the responses to single and paired cTBS trains were performed with a two-tailed Pearson correlation coefficient test for both the AMT80 (Experiment 1) and RMT70 (Experiment 2) intensities. All data were expressed as percentage of baseline, and all seven post-intervention time points for the paired cTBS conditions were averaged for each subject prior to analysis. For the single cTBS data, two average response variables were calculated: the average of the entire 60-min follow-up period (to give a measure of the overall response to the single cTBS train); and the average for the first 10 min following intervention (as differences in the size and direction of effects observed during this time may be more likely to impact on the response to a second train applied 10 min later).
For Experiment 3, the response to paired cTBS at RMT70 primed with an initial voluntary contraction was compared with that produced by paired trains of cTBS in Experiments 1 and 2 (i.e. paired cTBS at AMT80 and RMT70, respectively) by use of anovaRM with within-subject factors INTERVENTION (three levels) and TIME (seven levels). Similarly, anovaRM (with factors INTERVENTION and TIME) was used to compare the response to paired cTBS at RMT65 in Experiment 4 with that of the paired cTBS conditions in Experiments 1 and 2; however, analysis was performed only on the subjects from Experiments 1 and 2 who also participated in Experiment 4 (giving a total of six subjects for this comparison). One-way anovaRM was again performed on raw data for Experiments 3 and 4, and, contingent on a significant main effect of TIME, post hoc analyses were performed with paired t-tests (corrected for multiple comparisons by the use of FDRP).
For Experiment 5, testing the duration of after-effects induced by paired cTBS at RMT70, a one-way anovaRM was performed on raw MEP data with TIME (12 levels – baseline, 5, 10, 20, 30, 40, 50, 60, 75, 90, 105, and 120 min) as the within-subject factor. Paired t-tests were again used for post hoc analyses (corrected for multiple comparisons by the use of FDRP).
Finally, data for Experiment 6 were analysed with two-tailed Pearson correlation coefficient tests to determine whether the response to the first train [average of 42 MEPs (expressed as a percentage of baseline MEP amplitude)] influenced the response to paired cTBS applied at RMT70. Three average response variables were calculated to characterise paired cTBS outcome: the average amplitude of MEPs (expressed as a percentage of baseline) recorded 5 min following the second cTBS train; the average amplitude of MEPs at the time of peak suppression; and the average amplitude of the MEPs across the entire 60-min follow-up period.
Data are presented as group means ± standard errors of the mean unless otherwise indicated. Statistical significance was accepted at P-values < 0.05 for all analyses.
Experiment 1 – single cTBS at AMT80 vs. paired cTBS at AMT80
Baseline MEP amplitudes for Experiment 1 were not significantly different between the single and paired cTBS interventions when applied at AMT80 (0.96 ± 0.06 mV and 0.93 ± 0.07 mV, respectively; P > 0.05). anovaRM also revealed no significant difference between the two interventions with respect to their effect on MEP amplitude (F1,11 = 3.47, P = 0.089). There appeared to be mild MEP suppression following a single train of cTBS at AMT80 (Fig. 2A); however, this change was not significant (TIME: F7,77 = 1.56, P = 0.161). There was no MEP suppression following paired cTBS trains at AMT80 (F3,29 = 1.34, P = 0.280) (Fig. 2B).
Experiment 2 – single cTBS at RMT70 vs. paired cTBS at RMT70
Baseline MEP amplitudes for Experiment 2 were not different when the single and paired cTBS interventions at RMT70 were compared (1.05 ± 0.06 mV and 1.00 ± 0.07 mV, respectively; P > 0.05). However, anovaRM revealed a highly significant main effect of intervention type on MEP amplitude (INTERVENTION: F1,11 = 22.82, P = 0.0006). No MEP suppression was observed following a single train of cTBS at RMT70 (F3,29 = 0.646, P = 0.572) (Fig. 3A); however, paired trains of cTBS at RMT70 induced strong MEP suppression (TIME: F7,77 = 5.37, P = 0.00005) (Fig. 3B). Post hoc analyses revealed that MEP amplitude was suppressed as compared with baseline when measured at 5, 10, 30, 40, 50 and 60 min after paired cTBS trains at RMT70 (for all, P ≤ 0.017; corrected with FDRP), with peak suppression occurring 50 min following the second cTBS train (48% of baseline MEP amplitude).
Relationship between the responses to single and paired cTBS trains
There was no significant correlation between the responses to a single cTBS train (average of all post-intervention time points) and those to paired cTBS trains for either the AMT80 paradigms (r = 0.20, P = 0.532) (Fig. 4A) or the RMT70 paradigms (r = 0.40, P = 0.198) (Fig. 4B). Similarly, when time points were averaged only for the first 10 min following intervention for the single cTBS condition, there was no significant correlation between single and paired cTBS responses, regardless of whether AMT80 paradigms (r = 0.26, P = 0.422) (Fig. 4C) or RMT70 paradigms (r = 0.46, P = 0.134) (Fig. 4D) were used.
Stimulation intensity of paired cTBS trains at AMT80 and RMT70
Analysis of the intensity at which paired cTBS trains were applied in the first two experiments revealed that a slightly (but significantly) lower absolute stimulation intensity was used for paired cTBS at AMT80 than at RMT70 (34.4% ± 2.2% of maximum stimulator output and 38.5% ± 1.9% of maximum stimulator output, respectively; P < 0.01).
Experiment 3 – paired cTBS trains at RMT70 primed with an initial voluntary contraction
Baseline MEP amplitudes for the paired cTBS intervention applied at RMT70 and primed with an initial voluntary contraction were not significantly different from those for the paired cTBS interventions of Experiments 1 and 2 (F2,22 = 1.37, P = 0.274). Comparison of the three conditions with respect to their effect on MEP amplitude revealed a main effect of INTERVENTION (F2,22 = 7.63, P = 0.003). Post hoc tests showed that this was attributable to greater MEP suppression with paired cTBS at RMT70 (with no initial contraction) (Fig. 3B) than with paired cTBS at AMT80 (Fig. 2B) (P = 0.002; corrected with FDRP) and paired cTBS at RMT70 primed with an initial contraction (Fig. 5A) (P = 0.013; corrected with FDRP). There was no difference in MEP amplitude change when paired cTBS at AMT80 (Fig. 2B) and paired cTBS at RMT70 primed with an initial contraction (Fig. 5A) were compared (P = 0.341). In addition, there was no main effect of TIME for paired cTBS at RMT70 primed with an initial contraction (F7,77 = 1.257, P = 0.283) (Fig. 5A).
Experiment 4 – paired cTBS trains at RMT65
In the subset of six subjects participating in this experiment, there was no difference between baseline MEP amplitudes for the paired cTBS intervention applied at RMT65 and the paired cTBS interventions of Experiments 1 and 2 (F2,10 = 0.135, P = 0.875). There was also no difference between the stimulation intensities used for paired cTBS at AMT80 and paired cTBS at RMT65 (34.2% ± 3.0% of maximum stimulator output and 33.3% ± 3.0% of maximum stimulator output, respectively; P > 0.05). There was a main effect of the different interventions on MEP amplitude (INTERVENTION: F2,10 = 8.32, P = 0.007). This was attributable to greater MEP suppression for paired cTBS at both RMT70 and RMT65 than for the paired cTBS intervention at AMT80 (P = 0.026 and P = 0.01, respectively; corrected with FDRP). There was no difference in the degree of MEP suppression following paired cTBS at RMT70 and paired cTBS at RMT65 (P = 0.601). One-way anovaRM revealed a main effect of TIME for paired cTBS at RMT65 (F3,17 = 4.77, P = 0.011) (Fig. 5B). This was attributable to suppression of MEP amplitudes measured at 10, 30, 40, 50 and 60 min after paired cTBS trains at RMT65 as compared with those measured at baseline (for all, P ≤ 0.029; corrected with FDRP), with peak suppression occurring 60 min following the second cTBS train (44% of baseline values).
Experiment 5 – duration of paired cTBS-induced after-effects when applied at RMT70
Strong MEP suppression was again observed for paired cTBS at RMT70 with the post-intervention follow-up period extended to 2 h (TIME: F5,41 = 4.472, P = 0.002) (Fig. 6A). MEP amplitudes measured at all time points up to and including 2 h after the second train of stimulation were depressed as compared with those at baseline (for all, P < 0.05; corrected with FDRP), with peak suppression occurring around 75–90 min following the second cTBS train (38% of baseline values).
Experiment 6 – relationship between the response to the first train and the outcome of paired cTBS applied at RMT70
A significant positive linear correlation was observed between the response to the first cTBS train and all three post-cTBS response variables (response recorded at 5 min, r = 0.856, P = 0.014; peak suppression, r = 0.913, P = 0.004; average response at all time points, r = 0.891, P = 0.007; Fig. 6B).
The two main findings of this study are that spaced pairs of cTBS trains induce a significantly greater neuroplastic response than single cTBS protocols, and that a voluntary contraction prior to paired cTBS trains at RMT70 abolishes the neuroplastic effect.
Single cTBS trains had no significant effect on corticospinal excitability
It should be noted that, although the change in corticospinal excitability induced in the present study by a single cTBS protocol at AMT80 was insufficient to reach a statistically significant level, there was a strong trend towards MEP suppression. Of the 12 subjects tested for this protocol, the majority responded to intervention with an overall suppression of MEP amplitudes; however, for two subjects, a facilitatory response to cTBS was observed (Fig. 4A). A similar variability between the profiles of subject responses to a single cTBS protocol has been reported previously (Martin et al., 2006; McAllister et al., 2011). Taken together, these findings reflect the considerable inter-individual variability that is common for rTMS protocols and that is probably attributable to a number of factors (Ridding & Ziemann, 2010), including genetic differences between subjects. For example, it has been demonstrated that a common single-nucleotide polymorphism of the gene encoding brain-derived neurotrophic factor can attenuate the response to a number of plasticity-inducing protocols (Kleim et al., 2006; Cheeran et al., 2008; Antal et al., 2010), including cTBS (Cheeran et al., 2008). Similarly, genetic variations in the gene encoding the NR2B subunit of the N-methyl-d-aspartate (NMDA) receptor have been linked to variations in the response profiles of subjects to the facilitatory intermittent TBS protocol (Mori et al., 2011). Although this has yet to be tested for cTBS, it is likely that at least some of the variability associated with cTBS protocols may also be mediated by differences in NMDA receptor genotype, given the dependency of the cTBS-induced neuroplastic response on NMDA receptor function (Huang et al., 2007).
Paired cTBS trains at RMT70 induced a long-lasting suppression of corticospinal excitability
Whereas there was no significant change in MEP amplitudes following a single cTBS train applied at either AMT80 or RMT70, paired cTBS trains applied at RMT70 significantly suppressed MEP amplitudes. Furthermore, this MEP suppression was present for all subjects – a response rate that is rare in conventional rTMS experiments (Fig. 4B and D) – and lasted for at least 2 h. All currently published data suggest that MEP suppression induced by a single cTBS train does not exceed 1 h (Huang et al., 2005; Gentner et al., 2008; Gamboa et al., 2010). The fact that paired cTBS at RMT70 induced strong MEP suppression exceeding 2 h in the present study suggests that the spaced application of multiple cTBS trains may be an effective approach for inducing long-lasting neuroplastic changes in corticospinal excitability when applied to the human M1.
Why are paired trains at RMT70 more effective than paired trains at AMT80?
Several factors may have contributed to the differing response to paired cTBS trains when applied at RMT70 or AMT80. First, small differences in the absolute stimulation intensity used in the RMT70 and AMT80 protocols may have played a role. The rationale for using RMT70 was that the resultant absolute intensity would be well matched to that used for the AMT80 protocols (Gentner et al., 2008). However, the absolute intensity for the RMT70 protocol was slightly, but significantly, higher than that used for the AMT80 protocol. Therefore, we performed a control experiment (Experiment 4) in which the response to paired cTBS at RMT65 was investigated. In this condition, the absolute stimulation intensity employed was well matched to that of the AMT80 condition. However, paired cTBS at RMT65 still suppressed MEP amplitudes to a similar degree as RMT70 stimulation. Therefore, it is unlikely that the small difference in absolute intensity employed in the AMT80 and RMT70 conditions was responsible for the different response profiles.
The second factor that may have been important for modifying the response to paired cTBS trains was the initial voluntary contraction used to establish AMT for the AMT80 protocols. In Experiment 3, we found that the MEP suppression seen following paired cTBS at RMT70 was abolished when a priming contraction (similar to that required during AMT assessment) was performed. Taken together, these findings suggest that the initial contraction performed during AMT assessment, and not the difference in stimulation intensity, was responsible for attenuating the response to paired cTBS trains.
Other factors that may have influenced the response to paired cTBS trains
To date, only one other study has examined the effect of repeated applications of cTBS to the human M1. Gamboa et al. (2011) applied paired cTBS trains at AMT80 with various inter-train intervals. Whereas a single cTBS train suppressed MEP amplitudes for up to 1 h, no suppression was observed when paired trains were applied at intervals of 2 and 5 min. Paired cTBS trains applied with a longer interval of 20 min reduced corticospinal excitability to a similar degree as a single cTBS train. It was proposed that some interaction, perhaps mediated by homeostatic mechanisms, was responsible for the lack of effects when two cTBS protocols were applied at short intervals, whereas at longer intervals the impact of the first train would have subsided such that no interaction with the second train occurred.
At first glance, it would appear that our results are in contrast to the findings of Gamboa et al. (2011). However, several factors may preclude direct comparison of the studies. First, a single cTBS train at both AMT80 and RMT70 failed to elicit significant MEP suppression in the present study, but Gamboa et al. observed a strong inhibitory effect with a single cTBS train. It is possible that the interaction between two cTBS trains differs depending on whether a single train modulated M1 excitability. We examined this possibility in two ways. First, using the data from Experiments 1 and 2, we performed a correlation analysis between the magnitude of a subject’s response to the single cTBS and paired cTBS protocols at both AMT80 and RMT70. Although these correlations were not significant, there was a suggestion of a positive correlation (especially for the RMT70 condition; see Fig. 4B and D). However, it should be noted that, for this analysis, subjects’ responses to single and paired cTBS trains were investigated on different days, and it is possible that an individual’s response to the single train protocol and the response to the first train in the paired protocol were different. Therefore, a second approach was employed to further investigate the influence of the response to the first cTBS train on the response to the paired trains. In Experiment 6, we used MEP amplitude measures between the cTBS trains to investigate the response to the first train. We found that the response to the first cTBS train was positively correlated with the response to paired cTBS (Fig. 6B), suggesting an interaction between the two cTBS trains that was non-homeostatic in nature. Thus, we consider it unlikely that the different response to paired cTBS trains in the present study and that reported by Gamboa et al. was caused by a stronger inhibitory response to the first cTBS train in the pair in the Gamboa et al. study.
The rationale for employing a 10-min interval in the present study was based on data from the rat hippocampus showing the repeated application of stimulation protocols at this interval to be highly effective in extending the duration of induced plasticity (Abraham et al., 2002). It is possible that the inter-train intervals used by Gamboa et al. (2011) were not optimal for prolonging the duration of cTBS-induced M1 suppression. Indeed, the degree to which repeated stimulation trains stabilise synaptic plasticity within the developing Xenopus visual system appears to be highly dependent on inter-train interval length, showing an inverted U-shaped relationship (Zhou et al., 2003). A similar relationship may influence the interaction of paired cTBS. Consequently, the intervals employed by Gamboa et al. may have been either too short or too long to interact in a manner that potentiates neuroplasticity.
Although the possibility exists that the use of sub-optimal intervals between cTBS protocols by Gamboa et al. may have prevented the progression of long-lasting M1 suppression, it should also be remembered that, in their study, AMT80 was used as the stimulation intensity of paired cTBS. Given our finding that an initial voluntary contraction may bias the response to paired cTBS trains, it is difficult to make direct comparisons between our results obtained with paired cTBS at RMT70 and those of Gamboa et al.
The induction of long-lasting neuroplastic change in regions other than M1 has been found with repeated applications of a slightly modified cTBS protocol (Nyffeler et al., 2006b). This modified cTBS protocol, when applied to the human frontal eye field as a single train, induced delays in the triggering of saccadic eye movements that lasted for < 1 h. However, when paired trains were applied at an interval of 15 min, the delay in saccade triggering lasted for well over 2 h (Nyffeler et al., 2006a). Despite differences in the stimulation parameters, the method used for assessing outcomes (i.e. behavioural as opposed to electrophysiological in the present study), and the cortical region investigated, the results of the present study obtained with paired cTBS trains at RMT70 are largely consistent with the effects reported by Nyffeler et al. (2006a).
Mechanisms by which paired cTBS increases neuroplasticity induction
It is not clear from the present study how paired cTBS protocols induce a stronger and more lasting neuroplastic effect than single applications. It is possible that this increased strength of stimulation could simply be a product of applying a greater number of pulses. However, we consider this to be unlikely, as a previous study showed that doubling the length of a cTBS train was not sufficient to increase MEP suppression; in fact, it was found that it reversed the response from suppression to facilitation (Gamboa et al., 2010).
Whereas an initial voluntary contraction had an attenuative effect on the long-lasting MEP suppression induced by paired cTBS protocols in the present study, it is interesting to note that the less persistent changes induced by single cTBS protocols in a number of previous studies have been largely resistant to this same disruption (Huang et al., 2005; Gentner et al., 2008; Gamboa et al., 2011). This might suggest that different mechanisms may be responsible for mediating the response to repeated cTBS protocols. Data from animal studies suggest that LTP and LTD each consist of at least two mechanistically distinct phases: a transient early phase that is dependent on post-translational modifications of pre-existing proteins, lasting for less than a few hours, and a more persistent late phase associated with new gene expression and increases in de novo protein synthesis, which can last many hours or even days (Krug et al., 1984; Huang & Kandel, 1994; Nguyen et al., 1994; Linden, 1996). The repeated application of stimulation protocols in a spaced manner has proved an effective method for inducing this protein synthesis-dependent late phase in animal models (Bliss & Gardner-Medwin, 1973; Abraham et al., 1993, 2002; Huang & Kandel, 1994). Furthermore, there is evidence that certain types of priming stimulation may selectively impair late-phase LTP while having no impact on the protein synthesis-independent early phase (Abraham et al., 2002; Woo & Nguyen, 2002; Young & Nguyen, 2005; Young et al., 2006). Although speculative at this stage, the long-lasting suppression of corticospinal excitability induced by paired cTBS protocols in the present study is consistent with the induction of late-phase LTD-like effects at excitatory synapses within the human M1. However, we cannot exclude the possibility that LTP-like effects at inhibitory synapses within M1 or changes in intrinsic neuronal excitability may have contributed to the results of the present study.
In conclusion, we have demonstrated that the repeated application of cTBS trains in a spaced manner is a highly effective approach for inducing long-lasting neuroplastic changes in corticospinal excitability when applied to the human M1. We have also shown that this long-lasting suppression only occurs in the absence of prior engagement of the motor cortical regions. Although the mechanisms responsible for the disruptive effect of an initial voluntary contraction are unclear, these results do suggest that having subjects perform a voluntary contraction of the hand muscles prior to cTBS application may be undesirable when spaced cTBS trains are used to induce lasting M1 plasticity. The findings of this study may have significant implications for the therapeutic use of rTMS.
This work is supported by a grant from the National Health and Medical research Council of Australia (ID 565302). M. R. Goldsworthy is an Australian Postgraduate Award (APA) and Robinson Institute Postgraduate Scholar. J. B. Pitcher is an M. S. McLeod Research Fellow. M. C. Ridding is supported by a National Health and Medical Research Council Senior Research Fellowship (ID 519313).