- Email: [email protected]
Preconditioning with transcranial direct current stimulation sensitizes the motor cortex to rapid-rate transcranial magnetic stimulation and controls the direction of after-effects
Preconditioning with Transcranial Direct Current Stimulation Sensitizes the Motor Cortex to Rapid-Rate Transcranial Magnetic Stimulation and Controls the Direction of After-Effects Nicolas Lang, Hartwig R. Siebner, Diana Ernst, Michael A. Nitsche, Walter Paulus, Roger N. Lemon, and John C. Rothwell Background: Rapid-rate repetitive transcranial magnetic stimulation (rTMS) can produce a lasting increase in cortical excitability in healthy subjects or induce beneficial effects in patients with neuropsychiatric disorders; however, the conditioning effects of rTMS are often subtle and variable, limiting therapeutic applications. Here we show that magnitude and direction of after-effects induced by rapid-rate rTMS depend on the state of cortical excitability before stimulation and can be tuned by preconditioning with transcranial direct current stimulation (tDCS). Methods: Ten healthy volunteers received a 20-sec train of 5-Hz rTMS given at an intensity of individual active motor threshold to the left primary motor hand area. This interventional protocol was preconditioned by 10 min of anodal, cathodal, or sham tDCS. We used single-pulse TMS to assess corticospinal excitability at rest before, between, and after the two interventions. Results: The 5-Hz rTMS given after sham tDCS failed to produce any after-effect, whereas 5-Hz rTMS led to a marked shift in corticospinal excitability when given after effective tDCS. The direction of rTMS-induced plasticity critically depended on the polarity of tDCS conditioning. Conclusions: Preconditioning with tDCS enhances cortical plasticity induced by rapid-rate rTMS and can shape the direction of rTMS-induced after-effects. Key Words: Repetitive transcranial magnetic stimulation, transcranial direct current stimulation, homeostatic plasticity, human motor cortex, cortical excitability
R
epetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) are noninvasive methods to induce changes in cortical excitability that outlast the stimulation for at least several minutes (Hallett et al 1999; Lang et al 2004; Nitsche and Paulus 2000; Siebner and Rothwell 2003). The direction of after-effects depends on the frequency of rTMS and on the polarity of tDCS. For rTMS, lower frequencies, in the range of 1 Hz, can produce long-lasting inhibition (SICI) of motor cortical excitability (Chen et al 1997; Touge et al 2001; Wassermann et al 1996), whereas rapid-rate rTMS, at frequencies of ⱖ5 Hz, can induce long-lasting facilitation (Peinemann et al 2004). For tDCS, longer periods (e.g., 10 min or more) of stimulation can alter cortical excitability for up to 1 hour. Anodal tDCS provokes a lasting facilitation, whereas cathodal tDCS induces inhibition (Nitsche and Paulus 2001; Nitsche et al 2003). In recent years, rTMS has attracted considerable interest as a therapeutic tool in neuropsychiatry. Although its mechanism of action is largely unknown, rTMS has been used in numerous clinical trials to improve a variety of neuropsychiatric diseases, such as major depression, obsessive-compulsive disorder, schizophrenia, tic disorders, epilepsy, Parkinson’s disease, and dystonia (for reviews,
From the Sobell Department of Motor Neuroscience and Movement Disorders (NL, HRS, RNL, JCR), Institute of Neurology, London, United Kingdom; Department of Clinical Neurophysiology (NL, DE, MAN, WP), Georg-August University, Goettingen; and the Department of Neurology (HRS), Christian-Albrechts-University, Kiel, Germany. Address reprint requests to Dr. Nicolas Lang, Department of Clinical Neurophysiology, University of Goettingen, Robert-Koch-Strasse 40, 37075 Goettingen, Germany; E-mail: [email protected]. Received March 28, 2004; revised July 6, 2004; accepted July 26, 2004.
0006-3223/04/$30.00 doi:10.1016/j.biopsych.2004.07.017
see Sommer and Paulus 2003; Wassermann and Lisanby 2001); however, the clinical effects have often been subtle and variable, and the therapeutic potential of rTMS is still open to debate. A major problem when applying rTMS in patients is that pre-existing abnormalities in function and excitability might alter the responsiveness of the targeted area to rTMS (Siebner et al 1999a). Recently, it could be shown that a preconditioning session of tDCS can be used to shape the conditioning effect of subsequent low-frequency rTMS on the primary motor cortex (Siebner et al 2004). The preconditioning effects exhibited a “homeostatic” pattern: “facilitatory preconditioning” with anodal tDCS caused a subsequent period of 1-Hz rTMS to reduce corticospinal excitability to below baseline levels for more than 20 min. Conversely, “inhibitory preconditioning” with cathodal tDCS resulted in 1-Hz rTMS increasing corticospinal excitability for at least 20 min. This pattern was interpreted in the context of homeostatic plasticity in the human cerebral cortex. Although activity-dependent synaptic plasticity is required for modification of network properties, homeostatic mechanisms make sure that plastic changes only occur within a physiologically useful range and allow for network stability (Turrigiano and Nelson 2004). Siebner et al (2004) used longer periods of low-frequency rTMS (15 min of 1 Hz at 85% resting motor threshold [RMT]) to assess the preconditioning effects of tDCS. Although this rTMS protocol had no consistent effect on corticospinal excitability, it had sometimes produced a lasting suppression of corticospinal excitability in previous studies (Fitzgerald et al 2002; Romero et al 2002; Touge et al 2001). In the present study, we used a comparatively short period of low-intensity rapid-rate rTMS (20 sec of 5-Hz rTMS at active motor threshold [AMT]) to shape corticospinal excitability. In contrast to the previous study by Siebner et al (2004), this 5-Hz rTMS protocol is well below the threshold for provoking a change in corticospinal excitability when tested with single-pulse TMS (Di Lazzaro et al 2002; Quartarone et al, in press; Takano et al, in press). Indeed, a recent study demonstrated that more than 600 stimuli of 5-Hz rTMS need to be given to the primary motor hand area (M1) to induce a BIOL PSYCHIATRY 2004;56:634 – 639 © 2004 Society of Biological Psychiatry
N. Lang et al consistent increase in corticospinal excitability (Quartarone et al, in press). We reasoned that, if preconditioning could alter the effectiveness of subsequent 5-Hz rTMS to change corticospinal excitability, this would unequivocally indicate that it had sensitized M1 to the conditioning effects of rapid-rate rTMS.
Methods and Materials Subjects Ten healthy volunteers (five women; aged 24.3 ⫾ 1.9 years [mean ⫾ SD]) participated in the experiments. All were consistent right-handers according to the 10-item version of the Edinburgh Handedness Inventory (Oldfield 1971). Before participation, subjects gave their written informed consent for the study. Subjects were seated in a comfortable reclining chair with a mounted head rest. Experimental procedures were carried out with approval of the joint ethics committee of the National Hospital for Neurology and Neurosurgery and the Institute of Neurology, London and of the ethics committee of the University of Goettingen and were performed according to the ethical standards laid down in the Declaration of Helsinki. Experimental Procedures Experiment 1. The main experiment was designed to explore the preconditioning effect of tDCS on the after-effects of rapid-rate rTMS on corticospinal excitability. Experimental procedures are summarized in Figure 1. Preconditioning consisted of a 10-min session of tDCS. Anodal (⫹1 mA), cathodal (⫺1 mA), or sham (0 mA) tDCS were applied to the left M1 in three different sessions, which were performed at least 3 days apart. After tDCS, participants received a 100-stimuli train of 5-Hz rTMS to the left M1. The intensity of rTMS was individually adjusted to AMT. It has been shown that this rTMS protocol can induce a transient decrease in short-latency intracortical inhibition (SICI) but is subthreshold for provoking a lasting change in corticospinal excitability tested with single-pulse TMS (Di Lazzaro et al 2002). The order of interventions was pseudorandomized and balanced across participants. Participants were blinded to the types of tDCS and rTMS. Single-pulse TMS was used to probe the after-effects of tDCS and rTMS. Suprathreshold stimuli were given to the left M1, and the amplitude of the motor evoked potential (MEP) induced in the contralateral first dorsal interosseus (FDI) muscle was taken as a measure of corticospinal excitability. Measurements were performed in four blocks: before preconditioning (PRE), between the two interventions (INTER), and twice after rTMS (POST-1, POST-2). Each block consisted of 40 trials and lasted for
BIOL PSYCHIATRY 2004;56:634 – 639 635 approximately 8 min. The intertrial interval was approximately 12 sec. At the beginning of each experiment, the intensity for single-pulse TMS was adjusted to evoke MEPs of approximately 1-mV peak-to-peak amplitude in the relaxed contralateral FDI muscle. This intensity was used throughout the experiment. Experiment 2. Five subjects participated in a first control experiment, which investigated the changes in corticospinal excitability evoked by tDCS preconditioning alone. In two different sessions, participants received 10 min of anodal or cathodal tDCS followed by sham rTMS. Corticospinal excitability was assessed as in the main experiment. Experiment 3. In a second control experiment on five subjects, we tested the possibility that the results from the first experiment were due to changes in motor thresholds. In two separate sessions, participants received anodal or cathodal tDCS followed by real 5-Hz rTMS as in the main experiment. Resting motor threshold and AMT were determined before (PRE) as well as after tDCS (INTER) and rTMS (POST-1). Transcranial Stimulation For rapid-rate rTMS and single-pulse TMS, we used the same standard figure-of-eight coil connected to a Magstim Rapid stimulator (Magstim Company, Whitland, Dyfed, United Kingdom). Transcranial magnetic stimulation was always given over the optimal site to provoke a maximum motor response in the relaxed right FDI muscle (referred to as motor hotspot). The coil was placed tangentially to the subject’s scalp with the handle pointing posterolaterally at a 45° angle from the midline (Figure 1). Magnetic stimuli had a biphasic configuration, and the initial phase of the biphasic pulse induced a posterior-to-anterior current in M1. The procedures for rTMS and single-pulse TMS were identical across all experiments. For sham rTMS, the coil placed over the motor hot spot was disconnected from the stimulator. The TMS stimulator was then discharged through a second coil, which was fixed to a coil holder positioned approximately 50 cm behind the subject’s head. This provided a similar noise compared to real rTMS. Bipolar tDCS was applied by a battery-driven constant-current stimulator (Schneider Electronic, Gleichen, Germany) through conductive-rubber electrodes, placed in two saline-soaked sponges (5 ⫻ 7 cm). The first electrode was positioned over the motor hotspot, as revealed by TMS, and the second electrode was placed above the contralateral orbit. Transcranial DCS polarity refers to the electrode over the left M1. During DC stimulation, a constant current output was monitored with a built-in amperemeter. For sham tDCS, the DC stimulator was only
Figure 1. Experimental approach. Anodal, cathodal, or sham transcranial direct current stimulation (tDCS) was applied to the left primary motor hand area (M1) on separate days. Direct current (⫾1 mA, or sham) was given for 10 min through two large-sized electrodes placed over M1 and the contralateral frontal pole. Ten minutes after tDCS preconditioning, 100 biphasic pulses of 5-Hz repetitive transcranial magnetic stimulation (rTMS) were given to the left M1 at individual active motor threshold (AMT). We used suprathreshold single-pulse TMS to induce motor responses in the contralateral first dorsal interosseus muscle. The amplitude of motor evoked potentials (MEPs) was used to probe changes in corticospinal excitability induced by transcranial stimulation. Corticospinal excitability was assessed in blocks of 40 consecutive MEPs before (PRE), after tDCS (INTER), and twice after rTMS (POST-1, POST-2).
www.elsevier.com/locate/biopsych
636 BIOL PSYCHIATRY 2004;56:634 – 639
N. Lang et al
switched on for 5 sec at the beginning of the sham session and was then turned off. This produced a short-lasting skin sensation comparable to real tDCS. Electromyographic Measurements and Data Analysis Electromyographic responses were recorded through a pair of Ag-AgCl surface electrodes placed over the right FDI muscle, with a belly tendon montage. Raw signals were amplified (1K), band-pass filtered (10 Hz–1 kHz), and then digitized with a CED 1401 AD converter (Cambridge Electronic Design, Cambridge, United Kingdom) controlled by Signal Software (CED, version 2.13) and stored on a personal computer for off-line analysis. Complete relaxation was continuously controlled through auditory and visual feedback of EMG activity. Resting and active motor thresholds were determined as reported previously (Rothwell et al 1999). According to this, the TMS coil was placed over the motor hotspot, and the stimulus intensity was progressively reduced in 5% steps of maximum stimulator output until a level was reached below which reliable EMG responses disappeared. For RMT, a reliable response in relaxed muscle can be defined as an MEP of 50 –100 V occurring in 50% of 10 –20 consecutive trials. In active muscle, the minimal response size to determine AMT might be approximately 200 –300 V because of the difficulty in distinguishing it from the background activity (Rothwell et al 1999). With mean peak-to-peak MEP amplitudes (in millivolts) as dependent variable, the effects of tDCS and rTMS on corticospinal excitability were evaluated with two-factorial repeated-measurement analyses of variance (ANOVAs) with preconditioning and time of measurement as within-subject factors (experiments 1 and 2). To assess stimulation-induced changes in motor threshold, we computed a three-factorial ANOVA for repeated measurements with preconditioning (cathodal vs. anodal tDCS), block of measurement (PRE, INTER, and POST-1), and motor state (relaxation vs. tonic contraction) as within-subject factors (experiment 3) and motor threshold (in percent of maximum stimulator output) as dependent variable. The Greenhouse-Geisser method was used when necessary to correct for nonsphericity. Conditional on a significant F value, paired-samples two-tailed t tests were used for post hoc comparisons. Pearson’s correlation coefficient was used to examine the relationship between MEP changes induced by tDCS and subsequent rTMS. For all statistical analyses, p values of ⬍.05 were considered significant. Results are given as mean and SEM.
Results None of the subjects reported any adverse effects during or after the experiments. Experiment 1 Figure 2 illustrates the effects of tDCS preconditioning on rTMS-induced changes in corticospinal excitability as indexed by changes in mean MEP amplitude. Analysis of variance revealed a significant interaction between preconditioning and time of measurement [F(3.2,28.4) ⫽ 31.2, p ⬍ .001]. Post hoc t tests demonstrated that MEP amplitudes at INTER were facilitated after anodal tDCS [t (9) ⫽ ⫺7.7, p ⬍ .001], showing a mean increase of 49% ⫾ 7% compared with PRE. Conversely, cathodal tDCS induced a significant decrease in MEP amplitude of 34% ⫾ 4% [t (9) ⫽ 8.4, p ⬍ .001]. Sham tDCS had no effect on MEP size (Figure 2, upper panel). When given after cathodal tDCS, 5-Hz rTMS resulted in a marked facilitation of corticospinal excitability. Mean MEP amplitudes increased by 51% ⫾ 7% and 74% ⫾ 4% at POST-1 and POST-2, respectively, compared with INTER (Figure 2, lower panel). This was confirmed by pairwise t tests, which demonwww.elsevier.com/locate/biopsych
Figure 2. Time course of corticospinal excitability. (A) Mean amplitudes of motor evoked potentials (MEPs) evoked by single-pulse transcranial magnetic stimulation (TMS) before transcranial direct current stimulation (tDCS) (PRE), between tDCS and repetitive (r)TMS (INTER), and after rTMS (POST-1 and POST-2). (B) Relative changes of MEP amplitudes within the first and second block of post-rTMS measurements (POST-1 and POST-2) compared with mean amplitudes immediately before rTMS (INTER). The type of preconditioning tDCS had a strong effect on the magnitude and direction of after-effects produced by subsequent 5-Hz rTMS. No effects on corticospinal excitability occurred after sham preconditioning. Depending on the polarity, effective tDCS resulted in a bidirectional modulation of the after-effects induced by subsequent 5-Hz rTMS. The 5-Hz rTMS given after “inhibitory” preconditioning (cathodal tDCS) resulted in a significant increase of corticospinal excitability. Conversely, 5-Hz rTMS after “facilitatory” preconditioning (anodal tDCS) caused a decrease in corticospinal excitability. The conditioning effect of rTMS gradually built up during the 20 min after the end of rTMS. Error bars represent SEM.
strated significant differences in MEP amplitude between INTER and POST-1 [t (9) ⫽ ⫺8.1, p ⬍ .001] and between POST-1 and POST-2 [t (9) ⫽ ⫺2.8, p ⫽ .020]. When applied after anodal tDCS,
N. Lang et al
Figure 3. Correlation between motor evoked potential (MEP) changes induced by preconditioning transcranial direct current stimulation (tDCS) and subsequent 5-Hz repetitive transcranial magnetic stimulation (rTMS). The graph plots relative changes in MEP amplitude provoked by tDCS preconditioning (INTER–PRE) against relative changes in MEP amplitude induced by rTMS (POST-1–INTER). r equals Pearson’s correlation coefficient. There is an inverse linear relationship between after-effects provoked by tDCS and rTMS: the larger the change in MEP amplitude after preconditioning, the stronger the reversal in MEP amplitude provoked by rTMS.
5-Hz rTMS had the opposite effect on corticospinal excitability. In this instance, the same 5-Hz rTMS protocol induced relative decreases in mean MEP amplitude of 28% ⫾ 4% (POST-1) and 43% ⫾ 5% (POST-2) compared with INTER (Figure 2, lower panel). Pairwise t tests revealed significant differences in MEP amplitude between INTER and POST-1 [t (9) ⫽ 6.2, p ⬍ .001] and between POST-1 and POST-2 [t (9) ⫽ 2.8, p ⫽ .022]. In accordance with Di Lazzaro et al (2002), subthreshold 5-Hz rTMS given after sham tDCS had no effect on MEP size. Following up on the main ANOVA, we computed separate ANOVAs for each block of measurements (PRE, INTER, POST-1, and POST-2), which only considered the factor preconditioning. A main effect of preconditioning was present for the second block of measurements [INTER; F (1.4,12.2) ⫽ 44.5, p ⬍ .001] and for the fourth block of measurements [POST-2; F (1.9,17.0) ⫽ 15.7, p ⬍ .001], but not for the first and third block of measurements (PRE and POST-1). Accordingly, post hoc comparisons revealed significant differences in MEP amplitude for the second block [INTER: anodal–sham: t (9) ⫽ 5.3, p ⬍ .001; cathodal– sham: t (9) ⫽ ⫺6.3, p ⬍ .001; anodal– cathodal: t (9) ⫽ 7.6, p ⬍ .001] and for the fourth block (POST-2: anodal–sham: t (9) ⫽ ⫺3.2, p ⫽ .012; cathodal–sham: t (9) ⫽ 2.5, p ⫽ .032; anodal– cathodal: t (9) ⫽ ⫺5.3, p ⬍ .001]. For preconditioning with cathodal tDCS, there was an inverse correlation between the magnitude of MEP amplitude changes induced by tDCS and subsequent rTMS (Figure 3). Subjects who had the largest decrease in excitability after cathodal tDCS showed the greatest facilitation after rTMS (Pearson’s correlation: r ⫽ ⫺.86, p ⫽ .001). Likewise, subjects with the largest increase
BIOL PSYCHIATRY 2004;56:634 – 639 637
Figure 4. Effects of transcranial direct current stimulation (tDCS) followed by sham repetitive transcranial magnetic stimulation (rTMS). The figure plots mean amplitudes of MEPs evoked by single-pulse TMS before tDCS (PRE), between tDCS and rTMS (INTER), and after rTMS (POST-1 and POST-2). Preconditioning with anodal and cathodal tDCS followed by sham rTMS induced a stable, polarity-dependent shift in the level of corticospinal excitability. Error bars represent SEM.
in excitability after anodal tDCS tended to express the strongest inhibition after rTMS (Pearson’s correlation: r ⫽ ⫺.56, p ⫽ .094). Experiment 2 Consistent with previous reports (Lang et al 2004; Nitsche and Paulus 2001; Nitsche et al 2003; Siebner et al 2004), the second experiment demonstrated that 10 min of tDCS can produce a polarity-specific change in corticospinal excitability (Figure 4), which remained stable throughout the course of the experiment and was not affected by sham rTMS. This was also reflected in the ANOVA, which revealed an interaction between preconditioning and time of measurement [F (1.5,6.1) ⫽ 33.3, p ⫽ .001]. Pairwise comparisons between measurements at baseline (PRE) and after tDCS confirmed that MEP amplitudes were consistently facilitated after anodal tDCS and inhibited after cathodal tDCS (p ⱕ .003). Experiment 3 In the third experiment, we found no evidence for a change in motor threshold in response to tDCS or rTMS. The mean values are given in Table 1. The three-factorial ANOVA only showed a main effect of motor state [F (1.0,4.0) ⫽ 29.3, p ⫽ .006]. There was no main effect of block of measurement and no interaction between the three factors. Table 1. Active and Resting Motor Thresholds During the Course of the Experiments
Active motor thresholds Anodal tDCS and 5-Hz rTMS Cathodal tDCS and 5-Hz rTMS Resting motor thresholds Anodal tDCS and 5-Hz rTMS Cathodal tDCS and 5-Hz rTMS
PRE
INTER
POST-1
47 ⫾ 3 47 ⫾ 3
47 ⫾ 3 46 ⫾ 3
47 ⫾ 3 46 ⫾ 3
62 ⫾ 3 61 ⫾ 3
62 ⫾ 3 62 ⫾ 3
62 ⫾ 3 61 ⫾ 3
Values are percent of maximum stimulator output (⫾ SEM). PRE, before preconditioning; INTER, between the two interventions; POST-1, after rTMS.
www.elsevier.com/locate/biopsych
638 BIOL PSYCHIATRY 2004;56:634 – 639 Discussion Using the MEP amplitude as an index of corticospinal excitability, we found that a preconditioning session of tDCS can markedly reduce the threshold for subsequent rapid-rate rTMS to provoke a lasting change in corticospinal excitability. In accordance with previous work (Di Lazzaro et al 2002; Quartarone et al, in press; Takano et al, in press), a short period of subthreshold 5-Hz rTMS given alone (i.e., after sham tDCS) failed to produce any change in corticospinal excitability when tested with single-pulse TMS; however, this “ineffective” 5-Hz rTMS became capable of provoking prominent changes in corticospinal excitability if rTMS was preceded by a 10-min period of tDCS. The implication is that tDCS preconditioning can provide an effective means for sensitizing the corticospinal projection neurons to the conditioning effects of 5-Hz rTMS. “Inhibitory preconditioning” with cathodal tDCS resulted in 5-Hz rTMS increasing corticospinal excitability above baseline levels, whereas “facilitatory preconditioning” with anodal tDCS caused subsequent 5-Hz rTMS to reduce corticospinal excitability to below baseline levels. Those subjects who had the largest change in MEP amplitude from tDCS also had the greatest change in the response to rTMS. Interestingly, the correlation was more consistent when rTMS developed a facilitatory effect, which might reflect the “natural” response of the cortex to this rapidrate paradigm. To summarize, it seems that a previous change in corticospinal excitability induced by tDCS caused 5-Hz rTMS to provoke a strong shift in excitability that was opposite in sign. This pattern of preconditioning effects is compatible with the idea that corticospinal excitability is controlled by a homeostatic mechanism in human motor cortex that keeps corticospinal excitability within a physiologically useful range. In a recent study, an analogous homeostatic pattern could be observed, when low-frequency (1-Hz) rTMS was preconditioned with tDCS (Siebner et al 2004). When the excitability level of the corticospinal projection had been raised by 10 min of “facilitatory” anodal tDCS, subsequent 1-Hz rTMS led to a lasting reduction in corticospinal excitability. Conversely, when “inhibitory” cathodal tDCS was used to reduce corticospinal excitability, the same 1-Hz rTMS caused a sustained increase in corticospinal excitability. Taken together, it can be concluded that the conditioning effects of low-frequency (1-Hz) and high-frequency (5-Hz) rTMS follow the same homeostatic mechanisms. These data suggest that a homeostatic role of rTMS-induced plasticity needs to be taken into account when rTMS is used to promote plastic changes in healthy subjects and patients with neuropsychiatric disorders. Previous rTMS studies in healthy volunteers have shown that the frequency of rTMS mainly defines the direction of the change in corticospinal excitability when rTMS is given to the M1 without preconditioning. Low-frequency (1-Hz) rTMS induces a lasting attenuation of corticospinal excitability (Chen et al 1997; Touge et al 2001; Wassermann et al 1996), whereas high-frequency (ⱖ5-Hz) provokes a lasting facilitation of corticospinal excitability (Peinemann et al 2004). The tDCS preconditioning approach puts this rule into perspective, showing that “inhibitory” 1-Hz rTMS and “facilitatory” 5-Hz rTMS can induce either an increase or a decrease in corticospinal excitability, depending on the functional state of the M1 before or at the time of rTMS conditioning. Our findings have several important implications for the current use of rTMS in healthy subjects and patients. First, owing to differences in the functional state (i.e., the recent “history” of neuronal activity), it is plausible that the magnitude and direction of MEP modulation induced by rTMS differs considerably among www.elsevier.com/locate/biopsych
N. Lang et al healthy subjects (Maeda et al 2000; Romero et al 2002). Here, preconditioning provides an attractive approach to standardize and maximize the conditioning effects of rTMS (Iyer et al 2003; Siebner et al 2004). For instance, Iyer et al (2003) have recently shown that the inhibitory after-effect of low-frequency 1-Hz rTMS can be enhanced by preconditioning with 6-Hz rTMS in healthy subjects. Second, pathophysiology is likely to alter the functional state of the cortex and alter the response pattern provoked by rTMS compared with healthy control subjects (Siebner et al 1999a, 1999b). Therefore, it is problematic to predict the effects of rTMS in neuropsychiatric disorders (e.g., depression) by analogy with the conditioning effects in healthy subjects. Third, the observation that rTMS-induced after-effects can be flipped by preconditioning can also help to explain why an identical rTMS protocol produced opposite after-effects on the regional cerebral blood flow (as an index of regional neuronal activity) in the stimulated cortex when given to different cortical areas: 1800 pulses of subthreshold 1-Hz rTMS to left M1 increased rCBF in the stimulated M1 and connected areas (Lee et al 2003). Conversely, when given over the rostral portion of the dorsal premotor cortex, the same rTMS protocol provoked a decrease in rCBF in the stimulated dorsal premotor cortex and connected areas (Siebner et al 2003). These opposite after-effects might be due to different excitability levels at the time of rTMS conditioning. Fourth, when rTMS is used to treat depression, rTMS is usually given in multiple sessions over several days (Lisanby et al 2002). The rationale behind this procedure is to produce a cumulative effect of rTMS that augments and prolongs the beneficial effects. The present work raises the possibility that homeostatic mechanisms might counteract cumulative after-effects, limiting clinical efficacy of rTMS. Last, these homeostatic mechanisms render rTMS safe as excitability levels are kept within a physiologic range; however, underlying pathophysiology (e.g., epilepsy) or medication that might impair the effectiveness of homeostatic mechanisms might increase the risk of rTMS to induce seizures. The mechanism by which tDCS causes a sensitization and a directional tuning of corticospinal plasticity remains to be clarified. Siebner et al (2004) hypothesized that the homeostatic mechanism involved an altered integration of synaptic inputs in dendrites and at the cell body of corticospinal neurons or a shift in membrane excitability. The present study provides two pieces of information that favor an altered integration of synaptic inputs rather than a shift in membrane excitability. First, motor thresholds, both active and at rest, were unchanged by preconditioning (tDCS) and conditioning (5-Hz rTMS). Motor thresholds can be changed by alterations in membrane excitability (Ziemann et al 1996), so that the lack of any change in AMT and RMT is consistent with the idea that the homeostatic pattern was not predominantly due to a membrane effect. Second, the intensity of rTMS conditioning was individually adjusted to AMT. Intradural recordings of the epidural volleys evoked by TMS have shown that, at this intensity, rTMS preferentially activates corticospinal projection neurons transsynaptically (Di Lazzaro et al 2002). Therefore, it seems more likely that an altered integration of synaptic inputs than a shift in membrane excitability underlies the form of homeostatic plasticity revealed in this experiment. Because synaptic changes can build up over time, this might also explain why rTMS effects after conditioning only devolved with some delay. It should be noted that the time taken for conditioning effects of rTMS to become apparent is sometimes shorter and sometimes longer than that reported in the present study. Depending on the parameters of stimulation, 5-Hz rTMS can cause immediate and short-lasting effects on MEPs (Peinemann et al 2000; Wu
N. Lang et al et al 2000) or SICI (Di Lazzaro et al 2002), whereas on other occasions, the effects on MEPs can take up to 60 min to reach maximum (Gow et al 2004). Presumably, these different time courses indicate the existence of different types of after effect, which may well have different mechanisms. The intensity of rTMS (i.e., 100% AMT) was the same after either anodal or cathodal stimulation. Because the MEP differed in amplitude after tDCS of the two types, one might argue that the rTMS was stimulating a different set or number of neurons and hence that this might contribute to the effects we observed; however, the effects we observed were in fact the opposite to what might be expected if the stimulus intensity had been effectively changed. For example, anodal conditioning usually increases cortical excitability, and hence when it was applied before rTMS it might be argued to be equivalent to conditioning with a higher intensity of rTMS. Peinemann et al (2000) did use a higher intensity of 5-Hz rTMS (90% RMT) and actually found a short-lived increase in MEP amplitudes, which is the opposite of what we saw after anodal conditioning in the present experiments. In conclusion, the present study highlights an important biological feature of rTMS-induced plasticity. Though the experiments were not designed to explore the therapeutic effects of rTMS, the results have important implications for studies that aim to improve cerebral function in patients. The present data suggest that a preconditioning approach might be effective in enhancing the therapeutic efficacy of rTMS in patients and to produce more consistent after-effects. This study was supported by the Deutsche Forschungsgemeinschaft (Germany) within the European Graduiertenkolleg 632 “Neuroplasticity: from Molecules to Systems” and by the Medical Research Council (United Kingdom).
Chen R, Classen J, Gerloff C, Celnik P, Wassermann E, Hallet M, Cohen L (1997): Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology 48:1398 –1403. Di Lazzaro V, Oliviero A, Mazzone P, Pilato F, Saturno E, Dileone M, et al (2002): Short-term reduction of intracortical inhibition in the human motor cortex induced by repetitive transcranial magnetic stimulation. Exp Brain Res 147:108 –113. Fitzgerald PB, Brown TL, Daskalakis ZJ, Chen R, Kulkarni J (2001): Intensitydependent effects of 1 Hz rTMS on human corticospinal excitability. Clin Neurophysiol 113:1136 –1141. Gow D, Rothwell J, Hobson A, Thompson D, Hamdy S (2004): Induction of long-term plasticity in human swallowing motor cortex following repetitive cortical stimulation. Clin Neurophysiol 115:1044 –1051. Hallett M, Wassermann EM, Pascual-Leone A, Valls-Sole J (1999): Repetitive transcranial magnetic stimulation. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:105–113. Iyer MB, Schleper N, Wassermann EM (2003): Priming stimulation enhances the depressant effect of low-frequency repetitive transcranial magnetic stimulation. J Neurosci 23:10867–10872. Lang N, Nitsche MA, Paulus W, Rothwell JC, Lemon RN (2004): Effects of transcranial direct current stimulation over the human motor cortex on corticospinal and transcallosal excitability. Exp Brain Res 156:439 – 443. Lee L, Siebner HR, Rowe JB, Rizzo V, Rothwell JC, Frackowiak RS, Friston KJ (2003): Acute remapping within the motor system induced by lowfrequency repetitive transcranial magnetic stimulation. J Neurosci 23: 5308 –5318. Lisanby SH, Kinnunen LH, Crupain MJ (2002): Applications of TMS to therapy in psychiatry. J Clin Neurophysiol 19:344 –360. Maeda F, Keenan JP, Tormos JM, Topka H, Pascual-Leone A (2000): Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability. Exp Brain Res 133:425– 430.
BIOL PSYCHIATRY 2004;56:634 – 639 639 Nitsche M, Nitsche M, Klein C, Tergau F, Rothwell J, Paulus W (2003): Level of action of cathodal DC polarisation induced inhibition of the human motor cortex. Clin Neurophysiol 114:600 – 604. Nitsche MA, Paulus W (2000): Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol 527:633– 639. Nitsche MA, Paulus W (2001): Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology 57:1899 –1901. Oldfield RC (1971): The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia 9:97–113. Peinemann A, Lehner C, Mentschel C, Munchau A, Conrad B, Siebner HR (2000): Subthreshold 5-Hz repetitive transcranial magnetic stimulation of the human primary motor cortex reduces intracortical paired-pulse inhibition. Neurosci Lett 296:21–24. Peinemann A, Reimer B, Löer C, Quartarone A, Münchau A, Conrad B, Siebner HR (2004): Long-lasting increase in corticospinal excitability after 1800 pulses of subthreshold 5 Hz repetitive TMS to the primary motor cortex. Clin Neurophysiol 115:1519 –1526. Quartarone A, Bagnato S, Rizzo V, Morgante F, Sant’Angelo A, Battaglia A, et al (in press): Distinct changes in cortical and spinal excitability following high-frequency repetitive TMS to the human motor cortex. Exp Brain Res. Romero JR, Anschel D, Sparing R, Gangitano M, Pascual-Leone A (2002): Subthreshold low frequency repetitive transcranial magnetic stimulation selectively decreases facilitation in the motor cortex. Clin Neurophysiol 113:101–107. Rothwell JC, Hallett M, Berardelli A, Eisen A, Rossini P, Paulus W (1999): Magnetic stimulation: Motor evoked potentials. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:97–103. Siebner HR, Auer C, Conrad B (1999a): Abnormal increase in the corticomotor output to the affected hand during repetitive transcranial magnetic stimulation of the primary motor cortex in patients with writer’s cramp. Neurosci Lett 262:133–136. Siebner HR, Filipovic SR, Rowe JB, Cordivari C, Gerschlager W, Rothwell JC, et al (2003): Patients with focal arm dystonia have increased sensitivity to slow-frequency repetitive TMS of the dorsal premotor cortex. Brain 126: 2710 –2725. Siebner HR, Lang N, Rizzo V, Nitsche MA, Paulus W, Lemon RN, Rothwell JC (2004): Preconditioning of low-frequency repetitive transcranial magnetic stimulation with transcranial direct current stimulation: Evidence for homeostatic plasticity in the human motor cortex. J Neurosci 24: 3379 –3385. Siebner HR, Rothwell J (2003): Transcranial magnetic stimulation: New insights into representational cortical plasticity. Exp Brain Res 148:1–16. Siebner HR, Tormos JM, Ceballos-Baumann AO, Auer C, Catala MD, Conrad B, Pascual-Leone A (1999b): Low-frequency repetitive transcranial magnetic stimulation of the motor cortex in writer’s cramp. Neurology 52: 529 –537. Sommer M, Paulus W (2003): Pulse configuration and rTMS efficacy: A review of clinical studies. Suppl Clin Neurophysiol 56:33– 41. Takano B, Drzezga A, Peller M, Sax I, Schwaiger M, Lee L, Siebner HR (in press): Short-term modulation of regional excitability and blood flow in human motor cortex following rapid-rate transcranial magnetic stimulation: A TMS-PET study. Neuroimage. Touge T, Gerschlager W, Brown P, Rothwell JC (2001): Are the after-effects of low-frequency rTMS on motor cortex excitability due to changes in the efficacy of cortical synapses? Clin Neurophysiol 112:2138 –2145. Turrigiano GG, Nelson SB (2004): Homeostatic plasticity in the developing nervous system. Nat Rev Neurosci 5:97–107. Wassermann EM, Grafman J, Berry C, Hollnagel C, Wild K, Clark K, Hallet M (1996): Use and safety of a new repetitive transcranial magnetic stimulator. Electroencephalogr Clin Neurophysiol 101:412– 417. Wassermann EM, Lisanby SH (2001): Therapeutic application of repetitive transcranial magnetic stimulation: A review. Clin Neurophysiol 112:1367– 1377. Wu T, Sommer M, Tergau F, Paulus W (2000): Lasting influence of repetitive transcranial magnetic stimulation on intracortical excitability in human subjects. Neurosci Lett 16:37– 40. Ziemann U, Lonnecker S, Steinhoff BJ, Paulus W (1996): Effects of antiepileptic drugs on motor cortex excitability in humans: A transcranial magnetic stimulation study. Ann Neurol 40:367–378.
www.elsevier.com/locate/biopsych
R
epetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) are noninvasive methods to induce changes in cortical excitability that outlast the stimulation for at least several minutes (Hallett et al 1999; Lang et al 2004; Nitsche and Paulus 2000; Siebner and Rothwell 2003). The direction of after-effects depends on the frequency of rTMS and on the polarity of tDCS. For rTMS, lower frequencies, in the range of 1 Hz, can produce long-lasting inhibition (SICI) of motor cortical excitability (Chen et al 1997; Touge et al 2001; Wassermann et al 1996), whereas rapid-rate rTMS, at frequencies of ⱖ5 Hz, can induce long-lasting facilitation (Peinemann et al 2004). For tDCS, longer periods (e.g., 10 min or more) of stimulation can alter cortical excitability for up to 1 hour. Anodal tDCS provokes a lasting facilitation, whereas cathodal tDCS induces inhibition (Nitsche and Paulus 2001; Nitsche et al 2003). In recent years, rTMS has attracted considerable interest as a therapeutic tool in neuropsychiatry. Although its mechanism of action is largely unknown, rTMS has been used in numerous clinical trials to improve a variety of neuropsychiatric diseases, such as major depression, obsessive-compulsive disorder, schizophrenia, tic disorders, epilepsy, Parkinson’s disease, and dystonia (for reviews,
From the Sobell Department of Motor Neuroscience and Movement Disorders (NL, HRS, RNL, JCR), Institute of Neurology, London, United Kingdom; Department of Clinical Neurophysiology (NL, DE, MAN, WP), Georg-August University, Goettingen; and the Department of Neurology (HRS), Christian-Albrechts-University, Kiel, Germany. Address reprint requests to Dr. Nicolas Lang, Department of Clinical Neurophysiology, University of Goettingen, Robert-Koch-Strasse 40, 37075 Goettingen, Germany; E-mail: [email protected]. Received March 28, 2004; revised July 6, 2004; accepted July 26, 2004.
0006-3223/04/$30.00 doi:10.1016/j.biopsych.2004.07.017
see Sommer and Paulus 2003; Wassermann and Lisanby 2001); however, the clinical effects have often been subtle and variable, and the therapeutic potential of rTMS is still open to debate. A major problem when applying rTMS in patients is that pre-existing abnormalities in function and excitability might alter the responsiveness of the targeted area to rTMS (Siebner et al 1999a). Recently, it could be shown that a preconditioning session of tDCS can be used to shape the conditioning effect of subsequent low-frequency rTMS on the primary motor cortex (Siebner et al 2004). The preconditioning effects exhibited a “homeostatic” pattern: “facilitatory preconditioning” with anodal tDCS caused a subsequent period of 1-Hz rTMS to reduce corticospinal excitability to below baseline levels for more than 20 min. Conversely, “inhibitory preconditioning” with cathodal tDCS resulted in 1-Hz rTMS increasing corticospinal excitability for at least 20 min. This pattern was interpreted in the context of homeostatic plasticity in the human cerebral cortex. Although activity-dependent synaptic plasticity is required for modification of network properties, homeostatic mechanisms make sure that plastic changes only occur within a physiologically useful range and allow for network stability (Turrigiano and Nelson 2004). Siebner et al (2004) used longer periods of low-frequency rTMS (15 min of 1 Hz at 85% resting motor threshold [RMT]) to assess the preconditioning effects of tDCS. Although this rTMS protocol had no consistent effect on corticospinal excitability, it had sometimes produced a lasting suppression of corticospinal excitability in previous studies (Fitzgerald et al 2002; Romero et al 2002; Touge et al 2001). In the present study, we used a comparatively short period of low-intensity rapid-rate rTMS (20 sec of 5-Hz rTMS at active motor threshold [AMT]) to shape corticospinal excitability. In contrast to the previous study by Siebner et al (2004), this 5-Hz rTMS protocol is well below the threshold for provoking a change in corticospinal excitability when tested with single-pulse TMS (Di Lazzaro et al 2002; Quartarone et al, in press; Takano et al, in press). Indeed, a recent study demonstrated that more than 600 stimuli of 5-Hz rTMS need to be given to the primary motor hand area (M1) to induce a BIOL PSYCHIATRY 2004;56:634 – 639 © 2004 Society of Biological Psychiatry
N. Lang et al consistent increase in corticospinal excitability (Quartarone et al, in press). We reasoned that, if preconditioning could alter the effectiveness of subsequent 5-Hz rTMS to change corticospinal excitability, this would unequivocally indicate that it had sensitized M1 to the conditioning effects of rapid-rate rTMS.
Methods and Materials Subjects Ten healthy volunteers (five women; aged 24.3 ⫾ 1.9 years [mean ⫾ SD]) participated in the experiments. All were consistent right-handers according to the 10-item version of the Edinburgh Handedness Inventory (Oldfield 1971). Before participation, subjects gave their written informed consent for the study. Subjects were seated in a comfortable reclining chair with a mounted head rest. Experimental procedures were carried out with approval of the joint ethics committee of the National Hospital for Neurology and Neurosurgery and the Institute of Neurology, London and of the ethics committee of the University of Goettingen and were performed according to the ethical standards laid down in the Declaration of Helsinki. Experimental Procedures Experiment 1. The main experiment was designed to explore the preconditioning effect of tDCS on the after-effects of rapid-rate rTMS on corticospinal excitability. Experimental procedures are summarized in Figure 1. Preconditioning consisted of a 10-min session of tDCS. Anodal (⫹1 mA), cathodal (⫺1 mA), or sham (0 mA) tDCS were applied to the left M1 in three different sessions, which were performed at least 3 days apart. After tDCS, participants received a 100-stimuli train of 5-Hz rTMS to the left M1. The intensity of rTMS was individually adjusted to AMT. It has been shown that this rTMS protocol can induce a transient decrease in short-latency intracortical inhibition (SICI) but is subthreshold for provoking a lasting change in corticospinal excitability tested with single-pulse TMS (Di Lazzaro et al 2002). The order of interventions was pseudorandomized and balanced across participants. Participants were blinded to the types of tDCS and rTMS. Single-pulse TMS was used to probe the after-effects of tDCS and rTMS. Suprathreshold stimuli were given to the left M1, and the amplitude of the motor evoked potential (MEP) induced in the contralateral first dorsal interosseus (FDI) muscle was taken as a measure of corticospinal excitability. Measurements were performed in four blocks: before preconditioning (PRE), between the two interventions (INTER), and twice after rTMS (POST-1, POST-2). Each block consisted of 40 trials and lasted for
BIOL PSYCHIATRY 2004;56:634 – 639 635 approximately 8 min. The intertrial interval was approximately 12 sec. At the beginning of each experiment, the intensity for single-pulse TMS was adjusted to evoke MEPs of approximately 1-mV peak-to-peak amplitude in the relaxed contralateral FDI muscle. This intensity was used throughout the experiment. Experiment 2. Five subjects participated in a first control experiment, which investigated the changes in corticospinal excitability evoked by tDCS preconditioning alone. In two different sessions, participants received 10 min of anodal or cathodal tDCS followed by sham rTMS. Corticospinal excitability was assessed as in the main experiment. Experiment 3. In a second control experiment on five subjects, we tested the possibility that the results from the first experiment were due to changes in motor thresholds. In two separate sessions, participants received anodal or cathodal tDCS followed by real 5-Hz rTMS as in the main experiment. Resting motor threshold and AMT were determined before (PRE) as well as after tDCS (INTER) and rTMS (POST-1). Transcranial Stimulation For rapid-rate rTMS and single-pulse TMS, we used the same standard figure-of-eight coil connected to a Magstim Rapid stimulator (Magstim Company, Whitland, Dyfed, United Kingdom). Transcranial magnetic stimulation was always given over the optimal site to provoke a maximum motor response in the relaxed right FDI muscle (referred to as motor hotspot). The coil was placed tangentially to the subject’s scalp with the handle pointing posterolaterally at a 45° angle from the midline (Figure 1). Magnetic stimuli had a biphasic configuration, and the initial phase of the biphasic pulse induced a posterior-to-anterior current in M1. The procedures for rTMS and single-pulse TMS were identical across all experiments. For sham rTMS, the coil placed over the motor hot spot was disconnected from the stimulator. The TMS stimulator was then discharged through a second coil, which was fixed to a coil holder positioned approximately 50 cm behind the subject’s head. This provided a similar noise compared to real rTMS. Bipolar tDCS was applied by a battery-driven constant-current stimulator (Schneider Electronic, Gleichen, Germany) through conductive-rubber electrodes, placed in two saline-soaked sponges (5 ⫻ 7 cm). The first electrode was positioned over the motor hotspot, as revealed by TMS, and the second electrode was placed above the contralateral orbit. Transcranial DCS polarity refers to the electrode over the left M1. During DC stimulation, a constant current output was monitored with a built-in amperemeter. For sham tDCS, the DC stimulator was only
Figure 1. Experimental approach. Anodal, cathodal, or sham transcranial direct current stimulation (tDCS) was applied to the left primary motor hand area (M1) on separate days. Direct current (⫾1 mA, or sham) was given for 10 min through two large-sized electrodes placed over M1 and the contralateral frontal pole. Ten minutes after tDCS preconditioning, 100 biphasic pulses of 5-Hz repetitive transcranial magnetic stimulation (rTMS) were given to the left M1 at individual active motor threshold (AMT). We used suprathreshold single-pulse TMS to induce motor responses in the contralateral first dorsal interosseus muscle. The amplitude of motor evoked potentials (MEPs) was used to probe changes in corticospinal excitability induced by transcranial stimulation. Corticospinal excitability was assessed in blocks of 40 consecutive MEPs before (PRE), after tDCS (INTER), and twice after rTMS (POST-1, POST-2).
www.elsevier.com/locate/biopsych
636 BIOL PSYCHIATRY 2004;56:634 – 639
N. Lang et al
switched on for 5 sec at the beginning of the sham session and was then turned off. This produced a short-lasting skin sensation comparable to real tDCS. Electromyographic Measurements and Data Analysis Electromyographic responses were recorded through a pair of Ag-AgCl surface electrodes placed over the right FDI muscle, with a belly tendon montage. Raw signals were amplified (1K), band-pass filtered (10 Hz–1 kHz), and then digitized with a CED 1401 AD converter (Cambridge Electronic Design, Cambridge, United Kingdom) controlled by Signal Software (CED, version 2.13) and stored on a personal computer for off-line analysis. Complete relaxation was continuously controlled through auditory and visual feedback of EMG activity. Resting and active motor thresholds were determined as reported previously (Rothwell et al 1999). According to this, the TMS coil was placed over the motor hotspot, and the stimulus intensity was progressively reduced in 5% steps of maximum stimulator output until a level was reached below which reliable EMG responses disappeared. For RMT, a reliable response in relaxed muscle can be defined as an MEP of 50 –100 V occurring in 50% of 10 –20 consecutive trials. In active muscle, the minimal response size to determine AMT might be approximately 200 –300 V because of the difficulty in distinguishing it from the background activity (Rothwell et al 1999). With mean peak-to-peak MEP amplitudes (in millivolts) as dependent variable, the effects of tDCS and rTMS on corticospinal excitability were evaluated with two-factorial repeated-measurement analyses of variance (ANOVAs) with preconditioning and time of measurement as within-subject factors (experiments 1 and 2). To assess stimulation-induced changes in motor threshold, we computed a three-factorial ANOVA for repeated measurements with preconditioning (cathodal vs. anodal tDCS), block of measurement (PRE, INTER, and POST-1), and motor state (relaxation vs. tonic contraction) as within-subject factors (experiment 3) and motor threshold (in percent of maximum stimulator output) as dependent variable. The Greenhouse-Geisser method was used when necessary to correct for nonsphericity. Conditional on a significant F value, paired-samples two-tailed t tests were used for post hoc comparisons. Pearson’s correlation coefficient was used to examine the relationship between MEP changes induced by tDCS and subsequent rTMS. For all statistical analyses, p values of ⬍.05 were considered significant. Results are given as mean and SEM.
Results None of the subjects reported any adverse effects during or after the experiments. Experiment 1 Figure 2 illustrates the effects of tDCS preconditioning on rTMS-induced changes in corticospinal excitability as indexed by changes in mean MEP amplitude. Analysis of variance revealed a significant interaction between preconditioning and time of measurement [F(3.2,28.4) ⫽ 31.2, p ⬍ .001]. Post hoc t tests demonstrated that MEP amplitudes at INTER were facilitated after anodal tDCS [t (9) ⫽ ⫺7.7, p ⬍ .001], showing a mean increase of 49% ⫾ 7% compared with PRE. Conversely, cathodal tDCS induced a significant decrease in MEP amplitude of 34% ⫾ 4% [t (9) ⫽ 8.4, p ⬍ .001]. Sham tDCS had no effect on MEP size (Figure 2, upper panel). When given after cathodal tDCS, 5-Hz rTMS resulted in a marked facilitation of corticospinal excitability. Mean MEP amplitudes increased by 51% ⫾ 7% and 74% ⫾ 4% at POST-1 and POST-2, respectively, compared with INTER (Figure 2, lower panel). This was confirmed by pairwise t tests, which demonwww.elsevier.com/locate/biopsych
Figure 2. Time course of corticospinal excitability. (A) Mean amplitudes of motor evoked potentials (MEPs) evoked by single-pulse transcranial magnetic stimulation (TMS) before transcranial direct current stimulation (tDCS) (PRE), between tDCS and repetitive (r)TMS (INTER), and after rTMS (POST-1 and POST-2). (B) Relative changes of MEP amplitudes within the first and second block of post-rTMS measurements (POST-1 and POST-2) compared with mean amplitudes immediately before rTMS (INTER). The type of preconditioning tDCS had a strong effect on the magnitude and direction of after-effects produced by subsequent 5-Hz rTMS. No effects on corticospinal excitability occurred after sham preconditioning. Depending on the polarity, effective tDCS resulted in a bidirectional modulation of the after-effects induced by subsequent 5-Hz rTMS. The 5-Hz rTMS given after “inhibitory” preconditioning (cathodal tDCS) resulted in a significant increase of corticospinal excitability. Conversely, 5-Hz rTMS after “facilitatory” preconditioning (anodal tDCS) caused a decrease in corticospinal excitability. The conditioning effect of rTMS gradually built up during the 20 min after the end of rTMS. Error bars represent SEM.
strated significant differences in MEP amplitude between INTER and POST-1 [t (9) ⫽ ⫺8.1, p ⬍ .001] and between POST-1 and POST-2 [t (9) ⫽ ⫺2.8, p ⫽ .020]. When applied after anodal tDCS,
N. Lang et al
Figure 3. Correlation between motor evoked potential (MEP) changes induced by preconditioning transcranial direct current stimulation (tDCS) and subsequent 5-Hz repetitive transcranial magnetic stimulation (rTMS). The graph plots relative changes in MEP amplitude provoked by tDCS preconditioning (INTER–PRE) against relative changes in MEP amplitude induced by rTMS (POST-1–INTER). r equals Pearson’s correlation coefficient. There is an inverse linear relationship between after-effects provoked by tDCS and rTMS: the larger the change in MEP amplitude after preconditioning, the stronger the reversal in MEP amplitude provoked by rTMS.
5-Hz rTMS had the opposite effect on corticospinal excitability. In this instance, the same 5-Hz rTMS protocol induced relative decreases in mean MEP amplitude of 28% ⫾ 4% (POST-1) and 43% ⫾ 5% (POST-2) compared with INTER (Figure 2, lower panel). Pairwise t tests revealed significant differences in MEP amplitude between INTER and POST-1 [t (9) ⫽ 6.2, p ⬍ .001] and between POST-1 and POST-2 [t (9) ⫽ 2.8, p ⫽ .022]. In accordance with Di Lazzaro et al (2002), subthreshold 5-Hz rTMS given after sham tDCS had no effect on MEP size. Following up on the main ANOVA, we computed separate ANOVAs for each block of measurements (PRE, INTER, POST-1, and POST-2), which only considered the factor preconditioning. A main effect of preconditioning was present for the second block of measurements [INTER; F (1.4,12.2) ⫽ 44.5, p ⬍ .001] and for the fourth block of measurements [POST-2; F (1.9,17.0) ⫽ 15.7, p ⬍ .001], but not for the first and third block of measurements (PRE and POST-1). Accordingly, post hoc comparisons revealed significant differences in MEP amplitude for the second block [INTER: anodal–sham: t (9) ⫽ 5.3, p ⬍ .001; cathodal– sham: t (9) ⫽ ⫺6.3, p ⬍ .001; anodal– cathodal: t (9) ⫽ 7.6, p ⬍ .001] and for the fourth block (POST-2: anodal–sham: t (9) ⫽ ⫺3.2, p ⫽ .012; cathodal–sham: t (9) ⫽ 2.5, p ⫽ .032; anodal– cathodal: t (9) ⫽ ⫺5.3, p ⬍ .001]. For preconditioning with cathodal tDCS, there was an inverse correlation between the magnitude of MEP amplitude changes induced by tDCS and subsequent rTMS (Figure 3). Subjects who had the largest decrease in excitability after cathodal tDCS showed the greatest facilitation after rTMS (Pearson’s correlation: r ⫽ ⫺.86, p ⫽ .001). Likewise, subjects with the largest increase
BIOL PSYCHIATRY 2004;56:634 – 639 637
Figure 4. Effects of transcranial direct current stimulation (tDCS) followed by sham repetitive transcranial magnetic stimulation (rTMS). The figure plots mean amplitudes of MEPs evoked by single-pulse TMS before tDCS (PRE), between tDCS and rTMS (INTER), and after rTMS (POST-1 and POST-2). Preconditioning with anodal and cathodal tDCS followed by sham rTMS induced a stable, polarity-dependent shift in the level of corticospinal excitability. Error bars represent SEM.
in excitability after anodal tDCS tended to express the strongest inhibition after rTMS (Pearson’s correlation: r ⫽ ⫺.56, p ⫽ .094). Experiment 2 Consistent with previous reports (Lang et al 2004; Nitsche and Paulus 2001; Nitsche et al 2003; Siebner et al 2004), the second experiment demonstrated that 10 min of tDCS can produce a polarity-specific change in corticospinal excitability (Figure 4), which remained stable throughout the course of the experiment and was not affected by sham rTMS. This was also reflected in the ANOVA, which revealed an interaction between preconditioning and time of measurement [F (1.5,6.1) ⫽ 33.3, p ⫽ .001]. Pairwise comparisons between measurements at baseline (PRE) and after tDCS confirmed that MEP amplitudes were consistently facilitated after anodal tDCS and inhibited after cathodal tDCS (p ⱕ .003). Experiment 3 In the third experiment, we found no evidence for a change in motor threshold in response to tDCS or rTMS. The mean values are given in Table 1. The three-factorial ANOVA only showed a main effect of motor state [F (1.0,4.0) ⫽ 29.3, p ⫽ .006]. There was no main effect of block of measurement and no interaction between the three factors. Table 1. Active and Resting Motor Thresholds During the Course of the Experiments
Active motor thresholds Anodal tDCS and 5-Hz rTMS Cathodal tDCS and 5-Hz rTMS Resting motor thresholds Anodal tDCS and 5-Hz rTMS Cathodal tDCS and 5-Hz rTMS
PRE
INTER
POST-1
47 ⫾ 3 47 ⫾ 3
47 ⫾ 3 46 ⫾ 3
47 ⫾ 3 46 ⫾ 3
62 ⫾ 3 61 ⫾ 3
62 ⫾ 3 62 ⫾ 3
62 ⫾ 3 61 ⫾ 3
Values are percent of maximum stimulator output (⫾ SEM). PRE, before preconditioning; INTER, between the two interventions; POST-1, after rTMS.
www.elsevier.com/locate/biopsych
638 BIOL PSYCHIATRY 2004;56:634 – 639 Discussion Using the MEP amplitude as an index of corticospinal excitability, we found that a preconditioning session of tDCS can markedly reduce the threshold for subsequent rapid-rate rTMS to provoke a lasting change in corticospinal excitability. In accordance with previous work (Di Lazzaro et al 2002; Quartarone et al, in press; Takano et al, in press), a short period of subthreshold 5-Hz rTMS given alone (i.e., after sham tDCS) failed to produce any change in corticospinal excitability when tested with single-pulse TMS; however, this “ineffective” 5-Hz rTMS became capable of provoking prominent changes in corticospinal excitability if rTMS was preceded by a 10-min period of tDCS. The implication is that tDCS preconditioning can provide an effective means for sensitizing the corticospinal projection neurons to the conditioning effects of 5-Hz rTMS. “Inhibitory preconditioning” with cathodal tDCS resulted in 5-Hz rTMS increasing corticospinal excitability above baseline levels, whereas “facilitatory preconditioning” with anodal tDCS caused subsequent 5-Hz rTMS to reduce corticospinal excitability to below baseline levels. Those subjects who had the largest change in MEP amplitude from tDCS also had the greatest change in the response to rTMS. Interestingly, the correlation was more consistent when rTMS developed a facilitatory effect, which might reflect the “natural” response of the cortex to this rapidrate paradigm. To summarize, it seems that a previous change in corticospinal excitability induced by tDCS caused 5-Hz rTMS to provoke a strong shift in excitability that was opposite in sign. This pattern of preconditioning effects is compatible with the idea that corticospinal excitability is controlled by a homeostatic mechanism in human motor cortex that keeps corticospinal excitability within a physiologically useful range. In a recent study, an analogous homeostatic pattern could be observed, when low-frequency (1-Hz) rTMS was preconditioned with tDCS (Siebner et al 2004). When the excitability level of the corticospinal projection had been raised by 10 min of “facilitatory” anodal tDCS, subsequent 1-Hz rTMS led to a lasting reduction in corticospinal excitability. Conversely, when “inhibitory” cathodal tDCS was used to reduce corticospinal excitability, the same 1-Hz rTMS caused a sustained increase in corticospinal excitability. Taken together, it can be concluded that the conditioning effects of low-frequency (1-Hz) and high-frequency (5-Hz) rTMS follow the same homeostatic mechanisms. These data suggest that a homeostatic role of rTMS-induced plasticity needs to be taken into account when rTMS is used to promote plastic changes in healthy subjects and patients with neuropsychiatric disorders. Previous rTMS studies in healthy volunteers have shown that the frequency of rTMS mainly defines the direction of the change in corticospinal excitability when rTMS is given to the M1 without preconditioning. Low-frequency (1-Hz) rTMS induces a lasting attenuation of corticospinal excitability (Chen et al 1997; Touge et al 2001; Wassermann et al 1996), whereas high-frequency (ⱖ5-Hz) provokes a lasting facilitation of corticospinal excitability (Peinemann et al 2004). The tDCS preconditioning approach puts this rule into perspective, showing that “inhibitory” 1-Hz rTMS and “facilitatory” 5-Hz rTMS can induce either an increase or a decrease in corticospinal excitability, depending on the functional state of the M1 before or at the time of rTMS conditioning. Our findings have several important implications for the current use of rTMS in healthy subjects and patients. First, owing to differences in the functional state (i.e., the recent “history” of neuronal activity), it is plausible that the magnitude and direction of MEP modulation induced by rTMS differs considerably among www.elsevier.com/locate/biopsych
N. Lang et al healthy subjects (Maeda et al 2000; Romero et al 2002). Here, preconditioning provides an attractive approach to standardize and maximize the conditioning effects of rTMS (Iyer et al 2003; Siebner et al 2004). For instance, Iyer et al (2003) have recently shown that the inhibitory after-effect of low-frequency 1-Hz rTMS can be enhanced by preconditioning with 6-Hz rTMS in healthy subjects. Second, pathophysiology is likely to alter the functional state of the cortex and alter the response pattern provoked by rTMS compared with healthy control subjects (Siebner et al 1999a, 1999b). Therefore, it is problematic to predict the effects of rTMS in neuropsychiatric disorders (e.g., depression) by analogy with the conditioning effects in healthy subjects. Third, the observation that rTMS-induced after-effects can be flipped by preconditioning can also help to explain why an identical rTMS protocol produced opposite after-effects on the regional cerebral blood flow (as an index of regional neuronal activity) in the stimulated cortex when given to different cortical areas: 1800 pulses of subthreshold 1-Hz rTMS to left M1 increased rCBF in the stimulated M1 and connected areas (Lee et al 2003). Conversely, when given over the rostral portion of the dorsal premotor cortex, the same rTMS protocol provoked a decrease in rCBF in the stimulated dorsal premotor cortex and connected areas (Siebner et al 2003). These opposite after-effects might be due to different excitability levels at the time of rTMS conditioning. Fourth, when rTMS is used to treat depression, rTMS is usually given in multiple sessions over several days (Lisanby et al 2002). The rationale behind this procedure is to produce a cumulative effect of rTMS that augments and prolongs the beneficial effects. The present work raises the possibility that homeostatic mechanisms might counteract cumulative after-effects, limiting clinical efficacy of rTMS. Last, these homeostatic mechanisms render rTMS safe as excitability levels are kept within a physiologic range; however, underlying pathophysiology (e.g., epilepsy) or medication that might impair the effectiveness of homeostatic mechanisms might increase the risk of rTMS to induce seizures. The mechanism by which tDCS causes a sensitization and a directional tuning of corticospinal plasticity remains to be clarified. Siebner et al (2004) hypothesized that the homeostatic mechanism involved an altered integration of synaptic inputs in dendrites and at the cell body of corticospinal neurons or a shift in membrane excitability. The present study provides two pieces of information that favor an altered integration of synaptic inputs rather than a shift in membrane excitability. First, motor thresholds, both active and at rest, were unchanged by preconditioning (tDCS) and conditioning (5-Hz rTMS). Motor thresholds can be changed by alterations in membrane excitability (Ziemann et al 1996), so that the lack of any change in AMT and RMT is consistent with the idea that the homeostatic pattern was not predominantly due to a membrane effect. Second, the intensity of rTMS conditioning was individually adjusted to AMT. Intradural recordings of the epidural volleys evoked by TMS have shown that, at this intensity, rTMS preferentially activates corticospinal projection neurons transsynaptically (Di Lazzaro et al 2002). Therefore, it seems more likely that an altered integration of synaptic inputs than a shift in membrane excitability underlies the form of homeostatic plasticity revealed in this experiment. Because synaptic changes can build up over time, this might also explain why rTMS effects after conditioning only devolved with some delay. It should be noted that the time taken for conditioning effects of rTMS to become apparent is sometimes shorter and sometimes longer than that reported in the present study. Depending on the parameters of stimulation, 5-Hz rTMS can cause immediate and short-lasting effects on MEPs (Peinemann et al 2000; Wu
N. Lang et al et al 2000) or SICI (Di Lazzaro et al 2002), whereas on other occasions, the effects on MEPs can take up to 60 min to reach maximum (Gow et al 2004). Presumably, these different time courses indicate the existence of different types of after effect, which may well have different mechanisms. The intensity of rTMS (i.e., 100% AMT) was the same after either anodal or cathodal stimulation. Because the MEP differed in amplitude after tDCS of the two types, one might argue that the rTMS was stimulating a different set or number of neurons and hence that this might contribute to the effects we observed; however, the effects we observed were in fact the opposite to what might be expected if the stimulus intensity had been effectively changed. For example, anodal conditioning usually increases cortical excitability, and hence when it was applied before rTMS it might be argued to be equivalent to conditioning with a higher intensity of rTMS. Peinemann et al (2000) did use a higher intensity of 5-Hz rTMS (90% RMT) and actually found a short-lived increase in MEP amplitudes, which is the opposite of what we saw after anodal conditioning in the present experiments. In conclusion, the present study highlights an important biological feature of rTMS-induced plasticity. Though the experiments were not designed to explore the therapeutic effects of rTMS, the results have important implications for studies that aim to improve cerebral function in patients. The present data suggest that a preconditioning approach might be effective in enhancing the therapeutic efficacy of rTMS in patients and to produce more consistent after-effects. This study was supported by the Deutsche Forschungsgemeinschaft (Germany) within the European Graduiertenkolleg 632 “Neuroplasticity: from Molecules to Systems” and by the Medical Research Council (United Kingdom).
Chen R, Classen J, Gerloff C, Celnik P, Wassermann E, Hallet M, Cohen L (1997): Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology 48:1398 –1403. Di Lazzaro V, Oliviero A, Mazzone P, Pilato F, Saturno E, Dileone M, et al (2002): Short-term reduction of intracortical inhibition in the human motor cortex induced by repetitive transcranial magnetic stimulation. Exp Brain Res 147:108 –113. Fitzgerald PB, Brown TL, Daskalakis ZJ, Chen R, Kulkarni J (2001): Intensitydependent effects of 1 Hz rTMS on human corticospinal excitability. Clin Neurophysiol 113:1136 –1141. Gow D, Rothwell J, Hobson A, Thompson D, Hamdy S (2004): Induction of long-term plasticity in human swallowing motor cortex following repetitive cortical stimulation. Clin Neurophysiol 115:1044 –1051. Hallett M, Wassermann EM, Pascual-Leone A, Valls-Sole J (1999): Repetitive transcranial magnetic stimulation. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:105–113. Iyer MB, Schleper N, Wassermann EM (2003): Priming stimulation enhances the depressant effect of low-frequency repetitive transcranial magnetic stimulation. J Neurosci 23:10867–10872. Lang N, Nitsche MA, Paulus W, Rothwell JC, Lemon RN (2004): Effects of transcranial direct current stimulation over the human motor cortex on corticospinal and transcallosal excitability. Exp Brain Res 156:439 – 443. Lee L, Siebner HR, Rowe JB, Rizzo V, Rothwell JC, Frackowiak RS, Friston KJ (2003): Acute remapping within the motor system induced by lowfrequency repetitive transcranial magnetic stimulation. J Neurosci 23: 5308 –5318. Lisanby SH, Kinnunen LH, Crupain MJ (2002): Applications of TMS to therapy in psychiatry. J Clin Neurophysiol 19:344 –360. Maeda F, Keenan JP, Tormos JM, Topka H, Pascual-Leone A (2000): Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability. Exp Brain Res 133:425– 430.
BIOL PSYCHIATRY 2004;56:634 – 639 639 Nitsche M, Nitsche M, Klein C, Tergau F, Rothwell J, Paulus W (2003): Level of action of cathodal DC polarisation induced inhibition of the human motor cortex. Clin Neurophysiol 114:600 – 604. Nitsche MA, Paulus W (2000): Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol 527:633– 639. Nitsche MA, Paulus W (2001): Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology 57:1899 –1901. Oldfield RC (1971): The assessment and analysis of handedness: The Edinburgh Inventory. Neuropsychologia 9:97–113. Peinemann A, Lehner C, Mentschel C, Munchau A, Conrad B, Siebner HR (2000): Subthreshold 5-Hz repetitive transcranial magnetic stimulation of the human primary motor cortex reduces intracortical paired-pulse inhibition. Neurosci Lett 296:21–24. Peinemann A, Reimer B, Löer C, Quartarone A, Münchau A, Conrad B, Siebner HR (2004): Long-lasting increase in corticospinal excitability after 1800 pulses of subthreshold 5 Hz repetitive TMS to the primary motor cortex. Clin Neurophysiol 115:1519 –1526. Quartarone A, Bagnato S, Rizzo V, Morgante F, Sant’Angelo A, Battaglia A, et al (in press): Distinct changes in cortical and spinal excitability following high-frequency repetitive TMS to the human motor cortex. Exp Brain Res. Romero JR, Anschel D, Sparing R, Gangitano M, Pascual-Leone A (2002): Subthreshold low frequency repetitive transcranial magnetic stimulation selectively decreases facilitation in the motor cortex. Clin Neurophysiol 113:101–107. Rothwell JC, Hallett M, Berardelli A, Eisen A, Rossini P, Paulus W (1999): Magnetic stimulation: Motor evoked potentials. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:97–103. Siebner HR, Auer C, Conrad B (1999a): Abnormal increase in the corticomotor output to the affected hand during repetitive transcranial magnetic stimulation of the primary motor cortex in patients with writer’s cramp. Neurosci Lett 262:133–136. Siebner HR, Filipovic SR, Rowe JB, Cordivari C, Gerschlager W, Rothwell JC, et al (2003): Patients with focal arm dystonia have increased sensitivity to slow-frequency repetitive TMS of the dorsal premotor cortex. Brain 126: 2710 –2725. Siebner HR, Lang N, Rizzo V, Nitsche MA, Paulus W, Lemon RN, Rothwell JC (2004): Preconditioning of low-frequency repetitive transcranial magnetic stimulation with transcranial direct current stimulation: Evidence for homeostatic plasticity in the human motor cortex. J Neurosci 24: 3379 –3385. Siebner HR, Rothwell J (2003): Transcranial magnetic stimulation: New insights into representational cortical plasticity. Exp Brain Res 148:1–16. Siebner HR, Tormos JM, Ceballos-Baumann AO, Auer C, Catala MD, Conrad B, Pascual-Leone A (1999b): Low-frequency repetitive transcranial magnetic stimulation of the motor cortex in writer’s cramp. Neurology 52: 529 –537. Sommer M, Paulus W (2003): Pulse configuration and rTMS efficacy: A review of clinical studies. Suppl Clin Neurophysiol 56:33– 41. Takano B, Drzezga A, Peller M, Sax I, Schwaiger M, Lee L, Siebner HR (in press): Short-term modulation of regional excitability and blood flow in human motor cortex following rapid-rate transcranial magnetic stimulation: A TMS-PET study. Neuroimage. Touge T, Gerschlager W, Brown P, Rothwell JC (2001): Are the after-effects of low-frequency rTMS on motor cortex excitability due to changes in the efficacy of cortical synapses? Clin Neurophysiol 112:2138 –2145. Turrigiano GG, Nelson SB (2004): Homeostatic plasticity in the developing nervous system. Nat Rev Neurosci 5:97–107. Wassermann EM, Grafman J, Berry C, Hollnagel C, Wild K, Clark K, Hallet M (1996): Use and safety of a new repetitive transcranial magnetic stimulator. Electroencephalogr Clin Neurophysiol 101:412– 417. Wassermann EM, Lisanby SH (2001): Therapeutic application of repetitive transcranial magnetic stimulation: A review. Clin Neurophysiol 112:1367– 1377. Wu T, Sommer M, Tergau F, Paulus W (2000): Lasting influence of repetitive transcranial magnetic stimulation on intracortical excitability in human subjects. Neurosci Lett 16:37– 40. Ziemann U, Lonnecker S, Steinhoff BJ, Paulus W (1996): Effects of antiepileptic drugs on motor cortex excitability in humans: A transcranial magnetic stimulation study. Ann Neurol 40:367–378.
www.elsevier.com/locate/biopsych