- Email: [email protected]
Interference of short-interval intracortical inhibition (SICI) and short-interval intracortical facilitation (SICF)
Clinical Neurophysiology 119 (2008) 2291–2297
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Interference of short-interval intracortical inhibition (SICI) and short-interval intracortical facilitation (SICF) Sinikka H. Peurala a,b,1, J. Florian M. Müller-Dahlhaus a,1, Noritoshi Arai a,c, Ulf Ziemann a,* a
Motor Cortex Group, Department of Neurology, Goethe University of Frankfurt, Schleusenweg 2-16, D-60528 Frankfurt am Main, Germany Department of Health Sciences, Finnish Centre for Interdisciplinary Gerontology, University of Jyväskylä, Finland c Department of Neurology, Graduate School of Medicine, University of Tokyo, Japan b
See Editorial, pages 2175–2176
a r t i c l e
i n f o
Article history: Accepted 24 May 2008 Available online 23 August 2008 Keywords: Short-interval intracortical inhibition Short-interval intracortical facilitation Human primary motor cortex Paired-pulse transcranial magnetic stimulation
a b s t r a c t Objective: Short-interval intracortical inhibition (SICI) is a widely used paired-pulse transcranial magnetic stimulation (TMS) measure to assess inhibition in human motor cortex. However, facilitatory processes may contaminate SICI under certain conditions. Here, we specifically address the contribution of short-interval intracortical facilitation (SICF). Methods: A SICF interstimulus interval (ISI) curve was obtained in nine healthy subjects according to an established paired-pulse TMS protocol [Ziemann U, Tergau F, Wassermann EM, Wischer S, Hildebrandt J, Paulus W. Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J Physiol (Lond) 1998a;511:181–190]. The individual ISI leading to SICF peak1, trough1, peak2, trough2 and peak3 was selected for the subsequent measurement of SICI intensity curves (SICIpeak1, SICItrough1, SICIpeak2, SICItrough2, SICIpeak3) using intensity variation of the first stimulus (S1) from 50% to 120% of active motor threshold (AMT) in the first dorsal interosseous muscle. Results: SICIpeak1 (mean ISI, 1.54 ms) and SICItrough1 (mean ISI, 1.97 ms) showed a sigmoid SICI increase with S1 intensity. SICItrough1 reached the strongest SICI and was therefore chosen for comparison with the other SICI curves. SICIpeak2 (mean ISI, 2.61 ms) was U-shaped with a similar increase at low S1 intensities, but a decrease when S1 intensity exceeded 90% AMT. Correlation analyses suggested that this decrease was caused by SICF. SICItrough2 (mean ISI, 3.50 ms) and SICIpeak3 (mean ISI, 4.26 ms) showed considerably less inhibition than SICItrough1 over the whole range of S1 intensities. Conclusions: Findings show that commonly accepted protocols of testing SICI (ISI of 2–3 ms, S1 intensity 95% AMT) bear the risk of measuring net inhibition contaminated by SICF. Significance: SICF may contribute to apparently reduced SICI in patients with neurological or psychiatric disorders. Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Short-interval intracortical inhibition (SICI) is a well established paired-pulse transcranial magnetic stimulation (TMS) measure to explore non-invasively inhibition in human motor cortex mediated by the gamma-aminobutyric acid A (GABAA) receptor (Kujirai et al., 1993; Ziemann et al., 1996b; Ziemann et al., 1996c; Di Lazzaro et al., 1998b, 2000, 2006a; Ilic et al., 2002; Müller-Dahlhaus et al., 2008). It has to be acknowledged that SICI is a relatively complex measure which consists of at least two phases of inhibition that occur at distinct interstimulus intervals (ISI) between the sub* Corresponding author. Tel.: +49 69 6301 5739; fax: +49 69 6301 4498. E-mail address: [email protected] (U. Ziemann). 1 These authors contributed equally to this work.
threshold first (S1) and suprathreshold second (S2) pulse. It is thought that the first phase of inhibition at very short ISI of 1 ms is, at least to some extent, accounted for by refractoriness of the neural elements that are responsible for the activation of corticospinal neurons, while a second phase of inhibition at longer intervals (2.0 to 4.5 ms) is a true synaptic inhibition mediated by the GABAA receptor (Fisher et al., 2002; Hanajima et al., 2003; Roshan et al., 2003). The magnitude of this second phase of SICI depends critically on the intensities of S1 and S2. Variation of S2 intensity at a given subthreshold intensity of S1 typically leads to a U-shaped variation of SICI magnitude. SICI peaks at S2 intensities that result in motor evoked potential (MEP) amplitudes of 1 mV (Sanger et al., 2001; Daskalakis et al., 2002; Ilic et al., 2002; Stefan et al., 2002; Müller-Dahlhaus et al., 2008). The low end of this curve is
1388-2457/$34.00 Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2008.05.031
2292
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
explained by the observation that only late indirect waves (I-waves) but not the first I-wave (I1-wave) of the TMS induced corticospinal volley are inhibited by S1 (Di Lazzaro et al., 1998b), and that low amplitude MEP is produced predominantly by the early recruited I1-wave (Di Lazzaro et al., 1998a). The high end of the U-shaped curve is interpreted as indicating that high-threshold corticospinal neurons are less susceptible to SICI compared to those that are already recruited in MEP of 1 mV in amplitude (Müller-Dahlhaus et al., 2008). In addition, high intensities of S2 are capable of exciting the corticospinal neurons directly at their descending axons (Di Lazzaro et al., 1998a), thereby circumventing synaptic inhibition of these neurons. For this reason, in most studies the intensity of S2 was set to elicit control MEP of on average 1 mV in amplitude. Typically, variation of S1 intensity at a given suprathreshold S2 intensity also results in a U-shaped SICI curve with maximum SICI occurring at S1 intensities 90% of the active motor threshold, or 70% of the resting motor threshold (Kujirai et al., 1993; Ziemann et al., 1996c; Schäfer et al., 1997; Ilic et al., 2002; Kossev et al., 2003; Orth et al., 2003). While the low end of the SICI intensity curve is explained by SICI threshold and increasing recruitment of inhibitory interneurons that contribute to SICI, the mechanisms of the high end of this curve are less clear. It was speculated that the decrease of SICI with S1 intensities above those resulting in maximum SICI indicates recruitment of facilitatory processes that superimpose with inhibition and, therefore, that SICI has to be considered a net inhibition at this range of S1 intensities (Ziemann, 2002). One candidate for such a facilitatory process is short-interval intracortical facilitation (SICF) (Tokimura et al., 1996; Ziemann et al., 1998a; Di Lazzaro et al., 1999; Hanajima et al., 2002; Ilic et al., 2002), but whether it contributes to the high end of the SICI intensity curve has not been formally addressed yet. Apparent deficiency of SICI was described in a multitude of neurological and neuropsychiatric disorders (for review, (Ziemann, 1999; Curra et al., 2002; Chen et al., 2008)), but recent studies shaded some doubt on as to whether these findings truly indicated abnormal SICI or exaggerated facilitation, or both (Bütefisch et al., 2003; MacKinnon et al., 2005). This puts to question by which protocol SICI is determined most appropriately. Here, we sought to investigate specifically the contribution of SICF to the high end of the SICI intensity curve. We demonstrate that SICF explains the decrease of SICI at high intensities of S1, in particular at discrete ISI that lead to strong SICF. In addition, we demonstrate that SICI is most strongly expressed at the individual ISI that results in the first trough of the SICF interstimulus interval curve. From these data, we develop a recommendation how to determine SICI specifically and appropriately.
2. Materials and methods 2.1. Subjects Nine subjects (two female) aged 30–43 years (mean ± SEM, 35 ± 1 years) participated in the study. None of the subjects had a history of neurological disease or was on CNS-active drugs at the time of the experiment. All the subjects were right-handed according to the Edinburgh Handedness Inventory (Oldfield, 1971). Written informed consent was obtained prior to participation. The experiments conformed to the Declaration of Helsinki and were approved by the ethics committee of the Johann Wolfgang Goethe-University of Frankfurt am Main, Germany. 2.2. Recording and stimulation procedures Subjects were seated comfortably in a reclining chair. The right forearm was placed in a pronated position on the arm rest. MEP
was recorded from the right first dorsal interosseous (FDI) muscle by surface EMG, using Ag–AgCl cup electrodes in a belly tendon montage. The EMG raw signal was amplified and band-pass filtered (20 Hz to 2 kHz; Counterpoint Mk2 electromyograph; Dantec, Skovlunde, Denmark), digitized at an A/D rate of 5 kHz (CED Micro 1401; Cambridge Electronic Design, Cambrigde, UK), and stored in a laboratory computer for online visual display and later offline analysis using customized data collection and conditional averaging software (Spike2Ò for Windows, version 3.05, CED). Focal TMS was applied over the hand area of the dominant (left) primary motor cortex (M1) through a figure-of-eight coil (diameter of each wing, 70 mm) using two Magstim 200 magnetic stimulators (Magstim Company, Carmarthenshire, Wales, UK) with a monophasic current waveform connected to a BiStim Module (Magstim). The coil was held tangentially to the scalp with the handle pointing backwards and rotated away from the mid-line by up to 45°. This way, the current induced in the brain is directed from lateral-posterior to medial-anterior, and the corticospinal system is being activated predominantly transsynaptically via horizontal corticocortical connections (Di Lazzaro et al., 2004). The optimal coil position for eliciting MEP in the right FDI was determined as the site, where stimulation at a slightly suprathreshold stimulus intensity consistently produced the largest MEP. This site was marked with a pen in order to ensure the consistent placement of the coil throughout the experiment. Resting motor threshold (RMT) was determined to the nearest 1% of maximum stimulator output (MSO) as the lowest stimulus intensity which elicited small MEP (>50 lV peak-to-peak amplitude) in at least five of ten consecutive trials. Active motor threshold (AMT) was determined during a slight isometric FDI contraction (10% of maximum voluntary contraction, monitored by audio-visual feedback of the EMG signal) and measured to the nearest 1% of MSO as the lowest stimulus intensity which produced an MEP of >100 lV in peak-to-peak amplitude as measured from the average of five consecutive sweeps. Finally, MEP1mV was determined as the stimulus intensity which elicited MEP of, on average, 1 mV in peakto-peak amplitude in the resting FDI. 2.2.1. Short-interval intracortical facilitation (SICF) The paired-pulse measurements were started by testing SICF as a function of the ISI between S1 and S2. According to an established protocol, the intensity of S1 was set to MEP1mV when given alone and the intensity of S2 was set to 90% RMT (Ziemann et al., 1998a; Hanajima et al., 2002). Sixteen ISI ranging from 1.5 to 4.5 ms were tested in 0.2 ms steps in each subject. SICF testing consisted of four blocks of 40 trials each. Each block was composed of five conditions presented eight times each in pseudo-random order: control (S1 given alone) and four paired-pulse conditions (S1 followed by S2) at one of four different ISI. SICF was expressed by the conditioned mean MEP at a given ISI as a percentage of the mean control MEP in the same block of trials. From these data, individual SICF-ISI curves were generated. 2.2.2. Short-interval intracortical inhibition (SICI) Previous studies revealed that three peaks of SICF occur at discrete ISI (peak1: 1.1–1.5 ms, peak2: 2.3–2.9 ms, peak3: 4.1–4.5 ms) and that these are separated by troughs (trough1, trough2) without significant facilitation (Ziemann et al., 1998a). In order to test the possibility of a contribution of SICF to SICI, those five ISI resulting in the three peaks and two troughs were selected from each individual SICF-ISI curve. At those five ISI, SICI was recorded as a function of S1 intensity in five different blocks of trials. The SICI intensity curves will be referred to as SICIpeak1, SICItrough1, SICIpeak2, SICItrough2 and SICIpeak3. For each curve, eight different S1 intensities ranging from 50% to 120% AMT in 10% steps of AMT were ap-
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
plied, while the intensity of S2 was set to result in MEP1mV when given alone. Each block consisted of 72 trials, i.e. eight paired-pulse conditions at the different S1 intensities and the control condition (S2 alone) repeated eight times each and presented in pseudo-random order. The order of blocks (i.e. SICIpeak1, SICItrough1, SICIpeak2, SICItrough2 and SICIpeak3) was pseudo-randomised and balanced across subjects. SICI was expressed by the conditioned mean MEP at a given S1 intensity as a percentage of the mean control MEP in the same block of trials. SICItrough1 was chosen as a reference SICI curve for comparison with all other SICI curves for the following reasons: (1) contamination by SICF should be minimal because SICF does not occur at this ISI; (2) SICI at this ISI (2 ms, see Section 3) reflects synaptic inhibition through the GABAA receptor rather than refractoriness of neural elements contributing to SICI (Ziemann et al., 1996b; Di Lazzaro et al., 2000; Fisher et al., 2002; Ilic et al., 2002; Hanajima et al., 2003; Roshan et al., 2003); (3) SICI at this ISI is maximally expressed and not contaminated by intracortical facilitation (ICF) that may start to intrude at slightly longer ISI > 3.0 ms (Ziemann et al., 1996c). For all experiments, we used an intertrial interval of 5 s with a random intertrial interval variation of 25% in order to minimize anticipation of the next trial. All paired-pulse TMS experiments were conducted in the resting FDI. Complete voluntary muscle relaxation was monitored audio-visually by high-gain EMG (50 lV/division). Trials contaminated with voluntary activity were discarded from the analysis.
2.3. Statistics To test for the effect of ISI on SICF, a one-way repeated measures analysis of variance (ANOVARM) was conducted with the withinsubject factor ISI (16 levels: 1.5–4.5 ms in 0.2 ms steps). Conditional on a significant main effect of ISI, post hoc two-tailed onesample t tests were conducted to identify ISI with significant MEP facilitation (SICF > 100%). To test for the effects of ISI and S1 intensity on SICI, a two-way ANOVARM was performed with the within-subject factors ISI (five levels: ISI of SICF peak1, trough1, peak2, trough2, and peak3) and S1 intensity (eight levels: 50%, 60%, 70%, 80%, 90%, 100%, 110%, and 120% AMT). To assess the relation between individual SICF and SICI, linear regression analyses were computed. These analyses were restricted to those SICI curves (i.e. SICIpeak2, SICItrough2 and SICIpeak3, Section 3) and S1 intensities which resulted in significant differences from SICItrough1 (i.e. the reference SICI curve). The ratio of SICF (r-SICF) obtained at the individual peak2, trough2 and peak3 over SICF at trough1 served as the independent variable of the regression analysis. The ratio of SICI (r-SICI) obtained for SICIpeak2, SICItrough2 and SICIpeak3 over SICItrough1 at a given intensity of S1 served as the dependent variable. The rationale for using ratios rather than differences is based on knowledge from cell physiological experiments on synaptic convergence of excitatory and inhibitory postsynaptic potentials, which often have shown non-linear integration of these inputs, resulting in strong inhibition shunting excitatory input (for review, (Shepherd, 1998)). In line with this, previous TMS experiments demonstrated that the interaction between strong SICI and intracortical facilitation is non-linear and most correctly estimated by multiplication rather than summation of the inhibitory and facilitatory effects (Ziemann et al., 1996c). All data are expressed as mean ±1 standard error of the mean (SEM). For all experiments, statistical significance was assumed if P < 0.05. All statistical analyses were conducted using StatView for Windows 5.0.1. software (SAS Institute Inc., Cary, NC, USA).
2293
3. Results None of the subjects reported any adverse effects. RMT was 43.2 ± 2.4% MSO, AMT was 34.7 ± 2.4% MSO, and the stimulus intensity to elicit MEP1mV was 55.0 ± 4.3% MSO. 3.1. Short-interval intracortical facilitation (SICF) There was a significant main effect of ISI on SICF (F15,120 = 2.59, P = 0.002). Post hoc analysis revealed that MEP elicited by S1 plus S2 was significantly larger than control MEP elicited by S1 alone at discrete ISI of 1.5 (peak1), 2.5–3.1 (peak2), and 4.5 ms (peak3). These peaks were separated by troughs (trough1, trough2) where S2 had no significant effect on MEP size (Fig. 1). The average ISI which induced individual maximum peaks and troughs were quite consistent across subjects: 1.54 ± 0.03 ms (peak1), 1.97 ± 0.03 ms (trough1), 2.61 ± 0.07 ms (peak2), 3.50 ± 0.07 ms (trough2), and 4.26 ± 0.06 ms (peak3). 3.2. Short-interval intracortical inhibition (SICI) There were significant main effects of ISI (F4,32 = 6.18, P < 0.001), S1 intensity (F7,56 = 12.15, P < 0.001) and a significant interaction effect of ISI with S1 intensity (F28,224 = 1.68, P = 0.022) on SICI. The interaction effect indicates that different ISI result in different SICI-S1 intensity curves. It was found that the ISI equalling trough1 led to the most pronounced SICI (Fig. 2A–D). Therefore, this curve (SICItrough1) was taken as a reference curve and compared by pairwise repeated measures ANOVAs to the four other SICI curves. There was no significant difference with the SICIpeak1 curve (Fig. 2A). The comparison with the SICIpeak2 curve revealed a trend for an effect of ISI (F1,8 = 5.15, P = 0.053) and a significant interaction of ISI with S1 intensity (F7,56 = 6.18, P < 0.0001) (Fig. 2B). This interaction effect was explained by a dissociation of both the curves at S1 intensities >90% AMT: While SICItrough1 reached a stable maximum inhibition at S1 intensities of 100–120% AMT, SICIpeak2 showed maximum inhibition at the S1 intensity of 90% AMT and decreased again with higher S1 intensities (Fig. 2B). In other words, the intensity curves of SICItrough1 and SICIpeak2 were sigmoid and U-shaped, respectively. The comparison of SICItrough1 with
Fig. 1. Short-interval intracortical facilitation (SICF) in the relaxed first dorsal interosseous muscle. The intensity of the first stimulus (S1) was set to produce, on average, 1 mV peak-to-peak motor evoked potential (MEP) amplitudes, when given alone; the intensity of the second stimulus (S2) was set at 90% of the resting motor threshold for eliciting MEP in the relaxed FDI. Data are means from nine subjects. The x-axis indicates the interstimulus interval between S1 and S2 (in ms); the y-axis shows the MEP size elicited by paired-pulse stimulation (S1 plus S2) expressed as a percentage of the control MEP size (MEP evoked by S1 alone). The dotted horizontal line indicates the 100% level. Error bars are SEM. Statistically significant facilitation (P < 0.05; two-tailed one-sample t tests).
2294
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
Fig. 2. Short-interval intracortical inhibition (SICI) in the relaxed first dorsal interosseous muscle as a function of the S1 intensity and interstimulus interval. A–D: In all diagrams, intensity of S1 is given on the x-axis (in % active motor threshold, AMT), and SICI is indicated on the y-axis by conditioned (S1 plus S2) motor evoked potential (MEP) size as a percentage of the unconditioned (S2 alone) control MEP. Values <100% (dashed horizontal line) indicate inhibition. The SICI curves with black circles and thick lines refer to the reference condition obtained at an interstimulus interval equalling the optimal individual interval (1.97 ± 0.03 ms) to elicit trough1 in the short-interval intracortical facilitation curve (SICItrough1). This curve is compared with SICI curves obtained at the interstimulus intervals leading to optimal peak1 (SICIpeak1, interval: 1.54 ± 0.03 ms) (A), peak 2 (SICIpeak2, interval: 2.61 ± 0.07 ms) (B), trough2 (SICItrough2, interval: 3.50 ± 0.07 ms) (C) and peak3 (SICIpeak3, interval: 4.26 ± 0.06 ms) (D) of the individual short-interval intracortical facilitation curve. All data are means ± SEM of nine subjects. Asterisks indicate significant differences (P < 0.05, two-tailed paired t tests) between curves at the given S1 intensities.
SICItrough2 revealed a significant effect of ISI (F1,8 = 18.3, P = 0.0027), but no interaction effect of ISI with S1 intensity (F7,56 = 1.1, P = 0.40) (Fig. 2C). Both the curves showed a sigmoid characteristic but SICItrough2 ran flatter compared to SICItrough1. Finally, the comparison of SICItrough1 with SICIpeak3 revealed significant effects of ISI (F1,8 = 20.1, P = 0.002) and the interaction of ISI with S1 intensity (F7,56 = 4.0, P = 0.0013) (Fig. 2D). These effects were explained by a flatter SICIpeak3 curve compared to the SICItrough1 curve, and by different curve characteristics (SICIpeak3: U-shaped, SICItrough1: sigmoid). Post hoc two-tailed paired t tests revealed that the differences between the various SICI curves with the SICItrough1 curve (reference curve) occurred in the range of high S1 intensities: 100–120% AMT for SICIpeak2, 90–120% AMT for SICItrough2 and 80–120% AMT for SICI peak3 (cf. asterisks in Fig. 2B–D). 3.3. Contribution of SICF to SICI In order to evaluate the contribution of SICF to any reduction in SICI, the individual magnitude of SICF was calculated by the ratio of SICF at a peak2, trough2 or peak3 over SICF at trough1 (r-SICF) and related by linear regression (independent variable) to the reduction of SICI which was calculated by the ratio of SICIpeak2, SICItrough2 or SICIpeak3 over SICItrough1 (r-SICI, dependent variable) (see Section 2). SICIpeak1 was not included into this analysis because SICIpeak1 was not different from SICItrough1 (cf. Fig. 2A). In addition, regression analysis was restricted to those intensities of S1, which resulted in a significant difference between any of the other SICI curves with SICItrough1 (cf. asterisks in Fig. 2B–D). A significant positive linear correlation was found for the comparison of peak2 with trough1 at an S1 intensity of 110% AMT (r = 0.76, P = 0.018; Fig. 3B) and 120% AMT (r = 0.85, P = 0.004; Fig. 3C). This indicates
that strong SICF leads to a large reduction of SICI. No significant correlation between r-SICF and r-SICI was found at an S1 intensity of 100% AMT (r = 0.26, P = 0.48; Fig. 3A) or at any of the comparisons of trough2 or peak3 with trough1 (data not shown). Regression analyses using differences (SICIpeak2–SICItrough1 vs. SICFpeak2–SICFtrough1) rather than ratios did not result in significant correlations (data not shown, all p > 0.2), supporting the notion that the interaction between SICF and SICI is non-linear, similar to a previously found non-linear interaction between intracortical facilitation (ICF) and SICI (Ziemann et al., 1996c). 4. Discussion 4.1. Short-interval intracortical facilitation (SICF) Our data replicate earlier findings that MEP facilitation occurs at distinct ISI of 1.5 ms, 2.5 ms and 4.5 ms (Tokimura et al., 1996; Ziemann et al., 1998a; Ziemann et al., 1998b; Chen and Garg, 2000). It is thought that these peaks reflect a facilitatory interaction between the effects of S1 and S2 at the axon initial segments of those excitatory interneurons in motor cortex which contribute to the generation of I-waves (Hanajima et al., 2001; Ilic et al., 2002). The peaks are separated by troughs where no facilitation occurs (cf. Fig. 1). If the hypothesis was correct that SICF contributed to a significant extent to SICI then it should be expected that this can be detected only or predominantly at those ISI where SICF occurs. In addition to the ISI, expression of SICF depends on the intensities of S1 and S2 (Ziemann et al., 1998a; Chen and Garg, 2000; Ilic et al., 2002). Of particular interest for the present experiments is the intensity of S1 because this one is varied in the SICI intensity curve measurements. The second SICF peak (which will be the focus of discussion, see below) has a S1 threshold of 90% RMT but oc-
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
Fig. 3. Correlation between r-SICF and r-SICI The ratio of SICF (r-SICF) obtained at the individual peak2 and trough1 of the SICF-ISI curve (S1 intensity MEP1mV; S2 intensity 90% RMT) served as the independent variable. The ratio of SICI (r-SICI) obtained at the ISI that resulted in the individual SICF peak2 and trough1 served as the dependent variable. Regression analyses were separately computed for SICI tested at an S1 intensity of 100% AMT (r = 0.26, P = 0.48) (A), 110% AMT (r = 0.76, P = 0.018) (B) and 120% AMT (r = 0.85, P = 0.004) (C). Note that large magnitudes of r-SICF resulted in strong reductions of SICI (i.e. large r-SICI) at the two higher intensities of S1.
curs at a delayed interval of 3.3–3.5 ms at this threshold S1 intensity (cf. Fig. 2 in (Ziemann et al., 1998a)). This peak becomes more clearly visible with higher S1 intensities of 100–130% RMT and the effective ISI decrease to 2.2–2.8 ms (Fig. 2 in (Ziemann et al., 1998a)). This increasing magnitude of SICF with the intensity of S1 is explained by the fact that the facilitatory interaction between S1 and S2 requires excitation by S1 of those interneurons which are responsible for the late I-waves (Hanajima et al., 2002; Ilic et al., 2002), which occurs to a significant extent only if S1 exceeds RMT (Di Lazzaro et al., 1998a). From this it follows that SICF should contribute to the SICIpeak2 intensity curve only at the high end part of the curve, i.e., where S1 is at least 90% RMT. 4.2. Short-interval intracortical inhibition (SICI) We show here that the shape of the SICI intensity curve and the magnitude of the maximum SICI differ with the interstimulus interval if adjusted to individual peaks and troughs of the SICF interstimulus interval curve. SICI trough1 exhibited a sigmoid shape (no decrease of SICI in the high intensity range of S1) and reached the strongest maximum inhibition when compared to all other curves (cf. Fig. 2A–D). Therefore, and because we expected no con-
2295
tamination by SICF, SICItrough1 was selected as reference curve for comparison with all other curves. To our knowledge, SICI intensity curves as a function of ISI have never been systematically studied, although previous studies tested SICI intensity curves at a single ISI and showed, similar to the present results, that the SICI curve could be sigmoid at an ISI of 2.0 ms (Chen et al., 1998; MacKinnon et al., 2005) but U-shaped at an ISI of 3 ms (Kujirai et al., 1993; Ziemann et al., 1996c; MacKinnon et al., 2005). This study is the first one to measure SICI intensity curves at those ISI that resulted in the individual peaks and troughs of the SICF ISI curve. The fact that we found stronger SICItrough1 in this study compared to SICI tested under similar conditions in one of our previous studies (cf. Fig. 1B in (Ilic et al., 2002) is first evidence in support of the notion that SICF may act to reduce SICI if not carefully avoided by choosing the individually determined ISI where no or almost no SICF occurs. The comparison of SICIpeak1 (ISI = 1.54 ± 0.03 ms) with SICItrough1 (ISI = 1.97 ± 0.03 ms) revealed that the two curves did not differ significantly as a whole although the maximum SICItrough1 measured at S1 = 100% AMT was stronger than maximum SICIpeak1 (14.3 ± 2.9% vs. 38.0 ± 8.8%, P = 0.017, Fig. 2A). The data are compatible with the earlier findings that indicated slightly weaker SICI at an ISI of 1.5 ms vs. 2.0 ms when using a threshold tracking protocol (Fisher et al., 2002). SICF did probably not contribute to this weaker expression of SICI at peak1 because regression analysis revealed no correlation between the ratios of SICFpeak1/SICFtrough1 vs. SICIpeak1/SICItrough1 at any of the S1 intensities from 90% to 120% AMT (all P > 0.1). The slight difference between maximum SICIpeak1 and SICItrough1 may be attributed to differences in the physiology of inhibition at these two ISI. Several findings suggested that refractoriness of axons excited by S1 contribute to inhibition at very short ISI of 1 ms, while true GABAAergic synaptic inhibition comes into play at longer ISI: voluntary contraction and drugs enhancing neurotransmission through the GABAA receptor had no effect on SICI at an ISI of 1.0 ms but modified SICI at ISI of 2–3 ms (Ziemann et al., 1996b; Ziemann et al., 1996a; Fisher et al., 2002). However, diazepam, a positive allosteric modulator at the GABAA receptor, enhanced SICI at an ISI of 1.5 ms, suggesting that synaptic inhibition is already operating at this interval (Ilic et al., 2002). Therefore, it appears that SICIpeak1 and SICItrough1 share more similarities than they are divided by differences, and this is further supported by the very similar course of these curves at low intensities of S1 (Fig. 2A), indicating no difference in SICI threshold, whereas a lower threshold would be expected for SICIpeak1 if inhibition at this interval was caused predominantly by refractoriness (Fisher et al., 2002). Finally, it cannot be entirely excluded that the differences between SICItrough1 and SICIpeak1 were to some extent underestimated in this study because individual SICF peak1 occurs at a range of ISI (1.1– 1.5 ms) (Ziemann et al., 1998a) while we have tested only ISI P 1.5 ms. The comparison of SICIpeak2 (ISI = 2.61 ± 0.07 ms) with SICItrough1 revealed a highly significant interaction of ISI with S1 intensity (Fig. 2B). This was explained by a weaker SICIpeak2 than SICItrough1 at S1 intensities P100% AMT, whereas the two curves showed an almost identical course at low S1 intensities of 50–90% AMT. The similarities in SICI threshold and recruitment of SICI with increasing S1 intensity strongly suggest that the physiology of SICI at SICF trough1 and SICF peak2 is similar if not identical. The divergence along the higher S1 intensities suggests recruitment of a highthreshold facilitatory process with SICIpeak2 but not SICItrough1. The significant correlation between the ratios SICFpeak2/SICFtrough1 vs. SICIpeak2/SICItrough1 strongly suggests that SICF contributes to this facilitatory process (cf. Fig. 3B and C). A similarly increasing contribution of facilitation with increasing S1 intensity was previously shown by the threshold tracking technique at an ISI of 2.5 ms (cf. Fig. 5 in (Fisher et al., 2002)). Therefore, SICI has to be considered as a net inhibition if tested at an ISI where SICF occurs (i.e.
2296
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
2.5 ms) and if S1 intensity exceeds the SICF threshold which is around 90% RMT (i.e. 100–110% AMT) (Ziemann et al., 1998a). In addition, it is possible that SICF is a net facilitation with contributions coming from SICI. This notion is supported by previous pharmacological evidence that showed decreases of SICF when neurotransmission through the GABAA receptor is enhanced by benzodiazepines (Ziemann et al., 1998b; Ilic et al., 2002). This interaction was not the focus of the present experiments but may have contributed to the observed significant correlations between the ratios SICFpeak2/SICFtrough1 vs. SICIpeak2/SICItrough1 (Fig. 3B and C). The SICItrough2 (ISI = 3.50 ± 0.07 ms) and SICIpeak3 (ISI = 4.26 ± 0.06 ms) curves were conspicuously different from SICItrough1 by running on a clearly flatter course (Fig. 2C and D). This is in agreement with earlier threshold tracking data, which have also indicated less inhibition at ISI in the range from 3.0 to 4.5 ms if compared to SICI at 2.5 ms (cf. Fig. 3B in (Fisher et al., 2002)). Our data do not allow firm conclusion to attribute this decrease in SICI to a certain mechanism. However, it is unlikely that it is indicating a true decline in the strength of GABAA receptor mediated inhibition as this inhibition is associated by inhibitory postsynaptic potentials with a clearly longer time course of some 20 ms (Connors et al., 1988) that can also be demonstrated by paired-pulse TMS in humans (Hanajima et al., 1998). A more likely explanation is that another facilitatory process, such as intracortical facilitation (ICF) comes into play at these longer interstimulus intervals (Kujirai et al., 1993; Ziemann et al., 1996c; Di Lazzaro et al., 2006b). This notion is consistent with the observations that (i) the threshold for ICF is around 80% AMT (Ziemann et al., 1996c), the S1 intensity at which the SICItrough2 and SICIpeak3 curves started to diverge clearly from the SICItrough1 curve in the present experiments (Fig. 2C and D); (ii) SICI interstimulus interval curves with S1 intensity set to 90% AMT often start to turn from inhibition to facilitation at ISI P 3.0 ms (Kujirai et al., 1993; Ziemann et al., 1996c). SICF did not seem to contribute to this SICI decrease as the regression analyses revealed no significant correlation between the ratios of SICF and SICI (see Section 3). 4.3. Practical significance Several studies demonstrated apparently reduced SICI at an ISI of 2.0 ms in the motor cortex of the unaffected hemisphere of patients after cerebral stroke (Liepert et al., 2000; Manganotti et al., 2002; Shimizu et al., 2002), or even a turn from SICI to facilitation (Liepert et al., 2005). However, all those studies used a fixed S1 intensity of 75–80% RMT or 95% AMT. A subsequent study measured the SICI intensity curve at an ISI of 2.0 ms and suggested that this apparent reduction in SICI or even turn to facilitation was caused by an exaggerated high-threshold facilitatory process, while SICI threshold and the magnitude of SICI along the low intensity part of the SICI curve were normal in the unaffected hemisphere of stroke patients as compared to the healthy controls (Bütefisch et al., 2003). Similarly, many studies have claimed that SICI is reduced in Parkinson’s disease (for review, (Cantello et al., 2002)). A more recent study, however, measured full SICI intensity curves and thereby revealed that SICI threshold and the low intensity part of the SICI curve are normal in Parkinson’s disease, whereas curves differ along the high intensity part (S1 > 100% AMT) suggesting excess facilitation in Parkinson’s disease (MacKinnon et al., 2005). These studies left the question unresolved if this abnormality in the high-intensity part of the SICI curve is caused by deficient SICI or exaggerated facilitation, or both. Our data signify that exaggerated SICF could have contributed, although it was not present in the healthy subjects of this study at an ISI of 2.0, but this might
be different in disease conditions. As a consequence, a reduction in SICI cannot be meaningfully interpreted as decreased inhibition or increased facilitation, if tested at only one S1 intensity, which is above the SICF threshold because, under this condition, SICI likely reflects net inhibition resulting from the summation of SICI and SICF. In order to estimate the contribution of excess facilitation, SICI intensity curves rather than SICI at a single stimulus intensity, and in addition, SICF should be measured. Acknowledgement This study was supported by a grant from the Academy of Finland to S.H.P. References Bütefisch CM, Netz J, Wessling M, Seitz RJ, Hömberg V. Remote changes in cortical excitability after stroke. Brain 2003;126:470–81. Cantello R, Tarletti R, Civardi C. Transcranial magnetic stimulation and Parkinson’s disease. Brain Res Brain Res Rev 2002;38:309–27. Chen R, Cros D, Curra A, Di Lazzaro V, Lefaucheur JP, Magistris MR, et al. The clinical diagnostic utility of transcranial magnetic stimulation: report of an IFCN committee. Clin Neurophysiol 2008;119:504–32. Chen R, Garg R. Facilitatory I wave interaction in proximal arm and lower limb muscle representations of the human motor cortex. J Neurophysiol 2000;83:1426–34. Chen R, Tam A, Butefisch C, Ziemann U, Rothwell JC, Cohen LG. Intracortical inhibition and facilitation in different representations of the human motor cortex. J Neurophysiology 1998;80:2870–81. Connors BW, Malenka RC, Silva LR. Two inhibitory postsynaptic potentials, and GABAA and GABAB receptor-mediated responses in neocortex of rat and cat. J Physiol (Lond) 1988;406:443–68. Curra A, Modugno N, Inghilleri M, Manfredi M, Hallett M, Berardelli A. Transcranial magnetic stimulation techniques in clinical investigation. Neurology 2002;59:1851–9. Daskalakis ZJ, Christensen BK, Fitzgerald PB, Roshan L, Chen R. The mechanisms of interhemispheric inhibition in the human motor cortex. J Physiol 2002;543:317–26. Di Lazzaro V, Oliviero A, Meglio M, Cioni B, Tamburrini G, Tonali P, et al. Direct demonstration of the effect of lorazepam on the excitability of the human motor cortex. Clin Neurophysiol 2000;111:794–9. Di Lazzaro V, Oliviero A, Pilato F, Saturno E, Dileone M, Mazzone P, et al. The physiological basis of transcranial motor cortex stimulation in conscious humans. Clin Neurophysiol 2004;115:255–66. Di Lazzaro V, Oliviero A, Profice P, Saturno E, Pilato F, Insola A, et al. Comparison of descending volleys evoked by transcranial magnetic and electric stimulation in conscious humans. Electroencephalogr Clin Neurophysiol 1998a;109:397–401. Di Lazzaro V, Pilato F, Dileone M, Ranieri F, Ricci V, Profice P, et al. GABAA receptor subtype specific enhancement of inhibition in human motor cortex. J Physiol 2006a;575:721–6. Di Lazzaro V, Pilato F, Oliviero A, Dileone M, Saturno E, Mazzone P, et al. Origin of facilitation of motor-evoked potentials after paired magnetic stimulation: direct recording of epidural activity in conscious humans. J Neurophysiol 2006b;96:1765–71. Di Lazzaro V, Restuccia D, Oliviero A, Profice P, Ferrara L, Insola A, et al. Magnetic transcranial stimulation at intensities below active motor threshold activates intracortical inhibitory circuits. Exp Brain Res 1998b;119:265–8. Di Lazzaro V, Rothwell JC, Oliviero A, Profice P, Insola A, Mazzone P, et al. Intracortical origin of the short latency facilitation produced by pairs of threshold magnetic stimuli applied to human motor cortex. Exp Brain Res 1999;129:494–9. Fisher RJ, Nakamura Y, Bestmann S, Rothwell JC, Bostock H. Two phases of intracortical inhibition revealed by transcranial magnetic threshold tracking. Exp Brain Res 2002;143:240–8. Hanajima R, Furubayashi T, Iwata NK, Shiio Y, Okabe S, Kanazawa I, et al. Further evidence to support different mechanisms underlying intracortical inhibition of the motor cortex. Exp Brain Res 2003;151:427–34. Hanajima R, Ugawa Y, Machii K, Mochizuki H, Terao Y, Enomoto H, et al. Interhemispheric facilitation of the hand motor area in humans. J Physiol 2001;531:849–59. Hanajima R, Ugawa Y, Terao Y, Enomoto H, Shiio Y, Mochizuki H, et al. Mechanisms of intracortical I-wave facilitation elicited with paired-pulse magnetic stimulation in humans. J Physiol 2002;538:253–61. Hanajima R, Ugawa Y, Terao Y, Sakai K, Furubayashi T, Machii K, et al. Paired-pulse magnetic stimulation of the human motor cortex: differences among I waves. J Physiol 1998;509:607–18. Ilic TV, Meintzschel F, Cleff U, Ruge D, Kessler KR, Ziemann U. Short-interval pairedpulse inhibition and facilitation of human motor cortex: the dimension of stimulus intensity. J Physiol 2002;545.1:153–67. Kossev AR, Siggelkow S, Dengler R, Rollnik JD. Intracortical inhibition and facilitation in paired-pulse transcranial magnetic stimulation: effect of
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297 conditioning stimulus intensity on sizes and latencies of motor evoked potentials. J Clin Neurophysiol 2003;20:54–8. Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A, et al. Corticocortical inhibition in human motor cortex. J Physiol (Lond) 1993;471:501–19. Liepert J, Hamzei F, Weiller C. Motor cortex disinhibition of the unaffected hemisphere after acute stroke. Muscle Nerve 2000;23:1761–3. Liepert J, Restemeyer C, Kucinski T, Zittel S, Weiller C. Motor strokes: the lesion location determines motor excitability changes. Stroke 2005;36:2648–53. MacKinnon CD, Gilley EA, Weis-McNulty A, Simuni T. Pathways mediating abnormal intracortical inhibition in Parkinson’s disease. Ann Neurol 2005;58:516–24. Manganotti P, Patuzzo S, Cortese F, Palermo A, Smania N, Fiaschi A. Motor disinhibition in affected and unaffected hemisphere in the early period of recovery after stroke. Clin Neurophysiol 2002;113:936–43. Müller-Dahlhaus JF, Liu Y, Ziemann U. Inhibitory circuits and the nature of their interactions in the human motor cortex: a pharmacological TMS study. J Physiol 2008;586:495–514. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. Orth M, Snijders AH, Rothwell J. The variability of intracortical inhibition and facilitation. Clin Neurophysiol 2003;114:2362–9. Roshan L, Paradiso GO, Chen R. Two phases of short-interval intracortical inhibition. Exp Brain Res 2003;151:330–7. Sanger TD, Garg RR, Chen R. Interactions between two different inhibitory systems in the human motor cortex. J Physiol 2001;530.2:307–17. Schäfer M, Biesecker JC, Schulze-Bonhage A, Ferbert A. Transcranial magnetic double stimulation: influence of the intensity of the conditioning stimulus. Electroencephalogr Clin Neurophysiol 1997;105:462–9. Shepherd GM. The synaptic organization of the brain. Oxford: Oxford University Press; 1998.
2297
Shimizu T, Hosaki A, Hino T, Sato M, Komori T, Hirai S, et al. Motor cortical disinhibition in the unaffected hemisphere after unilateral cortical stroke. Brain 2002;125:1896–907. Stefan K, Kunesch E, Benecke R, Cohen LG, Classen J. Mechanisms of enhancement of human motor cortex excitability induced by interventional paired associative stimulation. J Physiol 2002;543:699–708. Tokimura H, Ridding MC, Tokimura Y, Amassian VE, Rothwell JC. Short latency facilitation between pairs of threshold magnetic stimuli applied to human motor cortex. Electroencephalogr Clin Neurophysiol 1996;101:263–72. Ziemann U. Intracortical inhibition and facilitation in the conventional paired TMS paradigm. Electroencephalogr Clin Neurophysiol Suppl 1999;51:127–36. Ziemann U. Paired pulse techniques. In: Pascual-Leone A, Davey KJ, Rothwell J, Wassermann EM, Puri BK, editors. Handbook of transcranial magnetic stimulation. London: Arnold; 2002. p. 141–63. Ziemann U, Lönnecker S, Steinhoff BJ, Paulus W. The effect of lorazepam on the motor cortical excitability in man. Exp Brain Res 1996a;109:127–35. Ziemann U, Lönnecker S, Steinhoff BJ, Paulus W. Effects of antiepileptic drugs on motor cortex excitability in humans: a transcranial magnetic stimulation study. Ann Neurol 1996b;40:367–78. Ziemann U, Rothwell JC, Ridding MC. Interaction between intracortical inhibition and facilitation in human motor cortex. J Physiol (Lond) 1996c;496:873–81. Ziemann U, Tergau F, Wassermann EM, Wischer S, Hildebrandt J, Paulus W. Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J Physiol (Lond) 1998a;511:181–90. Ziemann U, Tergau F, Wischer S, Hildebrandt J, Paulus W. Pharmacological control of facilitatory I-wave interaction in the human motor cortex. A paired transcranial magnetic stimulation study. Electroencephalogr Clin Neurophysiol 1998b;109:321–30.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Interference of short-interval intracortical inhibition (SICI) and short-interval intracortical facilitation (SICF) Sinikka H. Peurala a,b,1, J. Florian M. Müller-Dahlhaus a,1, Noritoshi Arai a,c, Ulf Ziemann a,* a
Motor Cortex Group, Department of Neurology, Goethe University of Frankfurt, Schleusenweg 2-16, D-60528 Frankfurt am Main, Germany Department of Health Sciences, Finnish Centre for Interdisciplinary Gerontology, University of Jyväskylä, Finland c Department of Neurology, Graduate School of Medicine, University of Tokyo, Japan b
See Editorial, pages 2175–2176
a r t i c l e
i n f o
Article history: Accepted 24 May 2008 Available online 23 August 2008 Keywords: Short-interval intracortical inhibition Short-interval intracortical facilitation Human primary motor cortex Paired-pulse transcranial magnetic stimulation
a b s t r a c t Objective: Short-interval intracortical inhibition (SICI) is a widely used paired-pulse transcranial magnetic stimulation (TMS) measure to assess inhibition in human motor cortex. However, facilitatory processes may contaminate SICI under certain conditions. Here, we specifically address the contribution of short-interval intracortical facilitation (SICF). Methods: A SICF interstimulus interval (ISI) curve was obtained in nine healthy subjects according to an established paired-pulse TMS protocol [Ziemann U, Tergau F, Wassermann EM, Wischer S, Hildebrandt J, Paulus W. Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J Physiol (Lond) 1998a;511:181–190]. The individual ISI leading to SICF peak1, trough1, peak2, trough2 and peak3 was selected for the subsequent measurement of SICI intensity curves (SICIpeak1, SICItrough1, SICIpeak2, SICItrough2, SICIpeak3) using intensity variation of the first stimulus (S1) from 50% to 120% of active motor threshold (AMT) in the first dorsal interosseous muscle. Results: SICIpeak1 (mean ISI, 1.54 ms) and SICItrough1 (mean ISI, 1.97 ms) showed a sigmoid SICI increase with S1 intensity. SICItrough1 reached the strongest SICI and was therefore chosen for comparison with the other SICI curves. SICIpeak2 (mean ISI, 2.61 ms) was U-shaped with a similar increase at low S1 intensities, but a decrease when S1 intensity exceeded 90% AMT. Correlation analyses suggested that this decrease was caused by SICF. SICItrough2 (mean ISI, 3.50 ms) and SICIpeak3 (mean ISI, 4.26 ms) showed considerably less inhibition than SICItrough1 over the whole range of S1 intensities. Conclusions: Findings show that commonly accepted protocols of testing SICI (ISI of 2–3 ms, S1 intensity 95% AMT) bear the risk of measuring net inhibition contaminated by SICF. Significance: SICF may contribute to apparently reduced SICI in patients with neurological or psychiatric disorders. Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Short-interval intracortical inhibition (SICI) is a well established paired-pulse transcranial magnetic stimulation (TMS) measure to explore non-invasively inhibition in human motor cortex mediated by the gamma-aminobutyric acid A (GABAA) receptor (Kujirai et al., 1993; Ziemann et al., 1996b; Ziemann et al., 1996c; Di Lazzaro et al., 1998b, 2000, 2006a; Ilic et al., 2002; Müller-Dahlhaus et al., 2008). It has to be acknowledged that SICI is a relatively complex measure which consists of at least two phases of inhibition that occur at distinct interstimulus intervals (ISI) between the sub* Corresponding author. Tel.: +49 69 6301 5739; fax: +49 69 6301 4498. E-mail address: [email protected] (U. Ziemann). 1 These authors contributed equally to this work.
threshold first (S1) and suprathreshold second (S2) pulse. It is thought that the first phase of inhibition at very short ISI of 1 ms is, at least to some extent, accounted for by refractoriness of the neural elements that are responsible for the activation of corticospinal neurons, while a second phase of inhibition at longer intervals (2.0 to 4.5 ms) is a true synaptic inhibition mediated by the GABAA receptor (Fisher et al., 2002; Hanajima et al., 2003; Roshan et al., 2003). The magnitude of this second phase of SICI depends critically on the intensities of S1 and S2. Variation of S2 intensity at a given subthreshold intensity of S1 typically leads to a U-shaped variation of SICI magnitude. SICI peaks at S2 intensities that result in motor evoked potential (MEP) amplitudes of 1 mV (Sanger et al., 2001; Daskalakis et al., 2002; Ilic et al., 2002; Stefan et al., 2002; Müller-Dahlhaus et al., 2008). The low end of this curve is
1388-2457/$34.00 Ó 2008 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2008.05.031
2292
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
explained by the observation that only late indirect waves (I-waves) but not the first I-wave (I1-wave) of the TMS induced corticospinal volley are inhibited by S1 (Di Lazzaro et al., 1998b), and that low amplitude MEP is produced predominantly by the early recruited I1-wave (Di Lazzaro et al., 1998a). The high end of the U-shaped curve is interpreted as indicating that high-threshold corticospinal neurons are less susceptible to SICI compared to those that are already recruited in MEP of 1 mV in amplitude (Müller-Dahlhaus et al., 2008). In addition, high intensities of S2 are capable of exciting the corticospinal neurons directly at their descending axons (Di Lazzaro et al., 1998a), thereby circumventing synaptic inhibition of these neurons. For this reason, in most studies the intensity of S2 was set to elicit control MEP of on average 1 mV in amplitude. Typically, variation of S1 intensity at a given suprathreshold S2 intensity also results in a U-shaped SICI curve with maximum SICI occurring at S1 intensities 90% of the active motor threshold, or 70% of the resting motor threshold (Kujirai et al., 1993; Ziemann et al., 1996c; Schäfer et al., 1997; Ilic et al., 2002; Kossev et al., 2003; Orth et al., 2003). While the low end of the SICI intensity curve is explained by SICI threshold and increasing recruitment of inhibitory interneurons that contribute to SICI, the mechanisms of the high end of this curve are less clear. It was speculated that the decrease of SICI with S1 intensities above those resulting in maximum SICI indicates recruitment of facilitatory processes that superimpose with inhibition and, therefore, that SICI has to be considered a net inhibition at this range of S1 intensities (Ziemann, 2002). One candidate for such a facilitatory process is short-interval intracortical facilitation (SICF) (Tokimura et al., 1996; Ziemann et al., 1998a; Di Lazzaro et al., 1999; Hanajima et al., 2002; Ilic et al., 2002), but whether it contributes to the high end of the SICI intensity curve has not been formally addressed yet. Apparent deficiency of SICI was described in a multitude of neurological and neuropsychiatric disorders (for review, (Ziemann, 1999; Curra et al., 2002; Chen et al., 2008)), but recent studies shaded some doubt on as to whether these findings truly indicated abnormal SICI or exaggerated facilitation, or both (Bütefisch et al., 2003; MacKinnon et al., 2005). This puts to question by which protocol SICI is determined most appropriately. Here, we sought to investigate specifically the contribution of SICF to the high end of the SICI intensity curve. We demonstrate that SICF explains the decrease of SICI at high intensities of S1, in particular at discrete ISI that lead to strong SICF. In addition, we demonstrate that SICI is most strongly expressed at the individual ISI that results in the first trough of the SICF interstimulus interval curve. From these data, we develop a recommendation how to determine SICI specifically and appropriately.
2. Materials and methods 2.1. Subjects Nine subjects (two female) aged 30–43 years (mean ± SEM, 35 ± 1 years) participated in the study. None of the subjects had a history of neurological disease or was on CNS-active drugs at the time of the experiment. All the subjects were right-handed according to the Edinburgh Handedness Inventory (Oldfield, 1971). Written informed consent was obtained prior to participation. The experiments conformed to the Declaration of Helsinki and were approved by the ethics committee of the Johann Wolfgang Goethe-University of Frankfurt am Main, Germany. 2.2. Recording and stimulation procedures Subjects were seated comfortably in a reclining chair. The right forearm was placed in a pronated position on the arm rest. MEP
was recorded from the right first dorsal interosseous (FDI) muscle by surface EMG, using Ag–AgCl cup electrodes in a belly tendon montage. The EMG raw signal was amplified and band-pass filtered (20 Hz to 2 kHz; Counterpoint Mk2 electromyograph; Dantec, Skovlunde, Denmark), digitized at an A/D rate of 5 kHz (CED Micro 1401; Cambridge Electronic Design, Cambrigde, UK), and stored in a laboratory computer for online visual display and later offline analysis using customized data collection and conditional averaging software (Spike2Ò for Windows, version 3.05, CED). Focal TMS was applied over the hand area of the dominant (left) primary motor cortex (M1) through a figure-of-eight coil (diameter of each wing, 70 mm) using two Magstim 200 magnetic stimulators (Magstim Company, Carmarthenshire, Wales, UK) with a monophasic current waveform connected to a BiStim Module (Magstim). The coil was held tangentially to the scalp with the handle pointing backwards and rotated away from the mid-line by up to 45°. This way, the current induced in the brain is directed from lateral-posterior to medial-anterior, and the corticospinal system is being activated predominantly transsynaptically via horizontal corticocortical connections (Di Lazzaro et al., 2004). The optimal coil position for eliciting MEP in the right FDI was determined as the site, where stimulation at a slightly suprathreshold stimulus intensity consistently produced the largest MEP. This site was marked with a pen in order to ensure the consistent placement of the coil throughout the experiment. Resting motor threshold (RMT) was determined to the nearest 1% of maximum stimulator output (MSO) as the lowest stimulus intensity which elicited small MEP (>50 lV peak-to-peak amplitude) in at least five of ten consecutive trials. Active motor threshold (AMT) was determined during a slight isometric FDI contraction (10% of maximum voluntary contraction, monitored by audio-visual feedback of the EMG signal) and measured to the nearest 1% of MSO as the lowest stimulus intensity which produced an MEP of >100 lV in peak-to-peak amplitude as measured from the average of five consecutive sweeps. Finally, MEP1mV was determined as the stimulus intensity which elicited MEP of, on average, 1 mV in peakto-peak amplitude in the resting FDI. 2.2.1. Short-interval intracortical facilitation (SICF) The paired-pulse measurements were started by testing SICF as a function of the ISI between S1 and S2. According to an established protocol, the intensity of S1 was set to MEP1mV when given alone and the intensity of S2 was set to 90% RMT (Ziemann et al., 1998a; Hanajima et al., 2002). Sixteen ISI ranging from 1.5 to 4.5 ms were tested in 0.2 ms steps in each subject. SICF testing consisted of four blocks of 40 trials each. Each block was composed of five conditions presented eight times each in pseudo-random order: control (S1 given alone) and four paired-pulse conditions (S1 followed by S2) at one of four different ISI. SICF was expressed by the conditioned mean MEP at a given ISI as a percentage of the mean control MEP in the same block of trials. From these data, individual SICF-ISI curves were generated. 2.2.2. Short-interval intracortical inhibition (SICI) Previous studies revealed that three peaks of SICF occur at discrete ISI (peak1: 1.1–1.5 ms, peak2: 2.3–2.9 ms, peak3: 4.1–4.5 ms) and that these are separated by troughs (trough1, trough2) without significant facilitation (Ziemann et al., 1998a). In order to test the possibility of a contribution of SICF to SICI, those five ISI resulting in the three peaks and two troughs were selected from each individual SICF-ISI curve. At those five ISI, SICI was recorded as a function of S1 intensity in five different blocks of trials. The SICI intensity curves will be referred to as SICIpeak1, SICItrough1, SICIpeak2, SICItrough2 and SICIpeak3. For each curve, eight different S1 intensities ranging from 50% to 120% AMT in 10% steps of AMT were ap-
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
plied, while the intensity of S2 was set to result in MEP1mV when given alone. Each block consisted of 72 trials, i.e. eight paired-pulse conditions at the different S1 intensities and the control condition (S2 alone) repeated eight times each and presented in pseudo-random order. The order of blocks (i.e. SICIpeak1, SICItrough1, SICIpeak2, SICItrough2 and SICIpeak3) was pseudo-randomised and balanced across subjects. SICI was expressed by the conditioned mean MEP at a given S1 intensity as a percentage of the mean control MEP in the same block of trials. SICItrough1 was chosen as a reference SICI curve for comparison with all other SICI curves for the following reasons: (1) contamination by SICF should be minimal because SICF does not occur at this ISI; (2) SICI at this ISI (2 ms, see Section 3) reflects synaptic inhibition through the GABAA receptor rather than refractoriness of neural elements contributing to SICI (Ziemann et al., 1996b; Di Lazzaro et al., 2000; Fisher et al., 2002; Ilic et al., 2002; Hanajima et al., 2003; Roshan et al., 2003); (3) SICI at this ISI is maximally expressed and not contaminated by intracortical facilitation (ICF) that may start to intrude at slightly longer ISI > 3.0 ms (Ziemann et al., 1996c). For all experiments, we used an intertrial interval of 5 s with a random intertrial interval variation of 25% in order to minimize anticipation of the next trial. All paired-pulse TMS experiments were conducted in the resting FDI. Complete voluntary muscle relaxation was monitored audio-visually by high-gain EMG (50 lV/division). Trials contaminated with voluntary activity were discarded from the analysis.
2.3. Statistics To test for the effect of ISI on SICF, a one-way repeated measures analysis of variance (ANOVARM) was conducted with the withinsubject factor ISI (16 levels: 1.5–4.5 ms in 0.2 ms steps). Conditional on a significant main effect of ISI, post hoc two-tailed onesample t tests were conducted to identify ISI with significant MEP facilitation (SICF > 100%). To test for the effects of ISI and S1 intensity on SICI, a two-way ANOVARM was performed with the within-subject factors ISI (five levels: ISI of SICF peak1, trough1, peak2, trough2, and peak3) and S1 intensity (eight levels: 50%, 60%, 70%, 80%, 90%, 100%, 110%, and 120% AMT). To assess the relation between individual SICF and SICI, linear regression analyses were computed. These analyses were restricted to those SICI curves (i.e. SICIpeak2, SICItrough2 and SICIpeak3, Section 3) and S1 intensities which resulted in significant differences from SICItrough1 (i.e. the reference SICI curve). The ratio of SICF (r-SICF) obtained at the individual peak2, trough2 and peak3 over SICF at trough1 served as the independent variable of the regression analysis. The ratio of SICI (r-SICI) obtained for SICIpeak2, SICItrough2 and SICIpeak3 over SICItrough1 at a given intensity of S1 served as the dependent variable. The rationale for using ratios rather than differences is based on knowledge from cell physiological experiments on synaptic convergence of excitatory and inhibitory postsynaptic potentials, which often have shown non-linear integration of these inputs, resulting in strong inhibition shunting excitatory input (for review, (Shepherd, 1998)). In line with this, previous TMS experiments demonstrated that the interaction between strong SICI and intracortical facilitation is non-linear and most correctly estimated by multiplication rather than summation of the inhibitory and facilitatory effects (Ziemann et al., 1996c). All data are expressed as mean ±1 standard error of the mean (SEM). For all experiments, statistical significance was assumed if P < 0.05. All statistical analyses were conducted using StatView for Windows 5.0.1. software (SAS Institute Inc., Cary, NC, USA).
2293
3. Results None of the subjects reported any adverse effects. RMT was 43.2 ± 2.4% MSO, AMT was 34.7 ± 2.4% MSO, and the stimulus intensity to elicit MEP1mV was 55.0 ± 4.3% MSO. 3.1. Short-interval intracortical facilitation (SICF) There was a significant main effect of ISI on SICF (F15,120 = 2.59, P = 0.002). Post hoc analysis revealed that MEP elicited by S1 plus S2 was significantly larger than control MEP elicited by S1 alone at discrete ISI of 1.5 (peak1), 2.5–3.1 (peak2), and 4.5 ms (peak3). These peaks were separated by troughs (trough1, trough2) where S2 had no significant effect on MEP size (Fig. 1). The average ISI which induced individual maximum peaks and troughs were quite consistent across subjects: 1.54 ± 0.03 ms (peak1), 1.97 ± 0.03 ms (trough1), 2.61 ± 0.07 ms (peak2), 3.50 ± 0.07 ms (trough2), and 4.26 ± 0.06 ms (peak3). 3.2. Short-interval intracortical inhibition (SICI) There were significant main effects of ISI (F4,32 = 6.18, P < 0.001), S1 intensity (F7,56 = 12.15, P < 0.001) and a significant interaction effect of ISI with S1 intensity (F28,224 = 1.68, P = 0.022) on SICI. The interaction effect indicates that different ISI result in different SICI-S1 intensity curves. It was found that the ISI equalling trough1 led to the most pronounced SICI (Fig. 2A–D). Therefore, this curve (SICItrough1) was taken as a reference curve and compared by pairwise repeated measures ANOVAs to the four other SICI curves. There was no significant difference with the SICIpeak1 curve (Fig. 2A). The comparison with the SICIpeak2 curve revealed a trend for an effect of ISI (F1,8 = 5.15, P = 0.053) and a significant interaction of ISI with S1 intensity (F7,56 = 6.18, P < 0.0001) (Fig. 2B). This interaction effect was explained by a dissociation of both the curves at S1 intensities >90% AMT: While SICItrough1 reached a stable maximum inhibition at S1 intensities of 100–120% AMT, SICIpeak2 showed maximum inhibition at the S1 intensity of 90% AMT and decreased again with higher S1 intensities (Fig. 2B). In other words, the intensity curves of SICItrough1 and SICIpeak2 were sigmoid and U-shaped, respectively. The comparison of SICItrough1 with
Fig. 1. Short-interval intracortical facilitation (SICF) in the relaxed first dorsal interosseous muscle. The intensity of the first stimulus (S1) was set to produce, on average, 1 mV peak-to-peak motor evoked potential (MEP) amplitudes, when given alone; the intensity of the second stimulus (S2) was set at 90% of the resting motor threshold for eliciting MEP in the relaxed FDI. Data are means from nine subjects. The x-axis indicates the interstimulus interval between S1 and S2 (in ms); the y-axis shows the MEP size elicited by paired-pulse stimulation (S1 plus S2) expressed as a percentage of the control MEP size (MEP evoked by S1 alone). The dotted horizontal line indicates the 100% level. Error bars are SEM. Statistically significant facilitation (P < 0.05; two-tailed one-sample t tests).
2294
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
Fig. 2. Short-interval intracortical inhibition (SICI) in the relaxed first dorsal interosseous muscle as a function of the S1 intensity and interstimulus interval. A–D: In all diagrams, intensity of S1 is given on the x-axis (in % active motor threshold, AMT), and SICI is indicated on the y-axis by conditioned (S1 plus S2) motor evoked potential (MEP) size as a percentage of the unconditioned (S2 alone) control MEP. Values <100% (dashed horizontal line) indicate inhibition. The SICI curves with black circles and thick lines refer to the reference condition obtained at an interstimulus interval equalling the optimal individual interval (1.97 ± 0.03 ms) to elicit trough1 in the short-interval intracortical facilitation curve (SICItrough1). This curve is compared with SICI curves obtained at the interstimulus intervals leading to optimal peak1 (SICIpeak1, interval: 1.54 ± 0.03 ms) (A), peak 2 (SICIpeak2, interval: 2.61 ± 0.07 ms) (B), trough2 (SICItrough2, interval: 3.50 ± 0.07 ms) (C) and peak3 (SICIpeak3, interval: 4.26 ± 0.06 ms) (D) of the individual short-interval intracortical facilitation curve. All data are means ± SEM of nine subjects. Asterisks indicate significant differences (P < 0.05, two-tailed paired t tests) between curves at the given S1 intensities.
SICItrough2 revealed a significant effect of ISI (F1,8 = 18.3, P = 0.0027), but no interaction effect of ISI with S1 intensity (F7,56 = 1.1, P = 0.40) (Fig. 2C). Both the curves showed a sigmoid characteristic but SICItrough2 ran flatter compared to SICItrough1. Finally, the comparison of SICItrough1 with SICIpeak3 revealed significant effects of ISI (F1,8 = 20.1, P = 0.002) and the interaction of ISI with S1 intensity (F7,56 = 4.0, P = 0.0013) (Fig. 2D). These effects were explained by a flatter SICIpeak3 curve compared to the SICItrough1 curve, and by different curve characteristics (SICIpeak3: U-shaped, SICItrough1: sigmoid). Post hoc two-tailed paired t tests revealed that the differences between the various SICI curves with the SICItrough1 curve (reference curve) occurred in the range of high S1 intensities: 100–120% AMT for SICIpeak2, 90–120% AMT for SICItrough2 and 80–120% AMT for SICI peak3 (cf. asterisks in Fig. 2B–D). 3.3. Contribution of SICF to SICI In order to evaluate the contribution of SICF to any reduction in SICI, the individual magnitude of SICF was calculated by the ratio of SICF at a peak2, trough2 or peak3 over SICF at trough1 (r-SICF) and related by linear regression (independent variable) to the reduction of SICI which was calculated by the ratio of SICIpeak2, SICItrough2 or SICIpeak3 over SICItrough1 (r-SICI, dependent variable) (see Section 2). SICIpeak1 was not included into this analysis because SICIpeak1 was not different from SICItrough1 (cf. Fig. 2A). In addition, regression analysis was restricted to those intensities of S1, which resulted in a significant difference between any of the other SICI curves with SICItrough1 (cf. asterisks in Fig. 2B–D). A significant positive linear correlation was found for the comparison of peak2 with trough1 at an S1 intensity of 110% AMT (r = 0.76, P = 0.018; Fig. 3B) and 120% AMT (r = 0.85, P = 0.004; Fig. 3C). This indicates
that strong SICF leads to a large reduction of SICI. No significant correlation between r-SICF and r-SICI was found at an S1 intensity of 100% AMT (r = 0.26, P = 0.48; Fig. 3A) or at any of the comparisons of trough2 or peak3 with trough1 (data not shown). Regression analyses using differences (SICIpeak2–SICItrough1 vs. SICFpeak2–SICFtrough1) rather than ratios did not result in significant correlations (data not shown, all p > 0.2), supporting the notion that the interaction between SICF and SICI is non-linear, similar to a previously found non-linear interaction between intracortical facilitation (ICF) and SICI (Ziemann et al., 1996c). 4. Discussion 4.1. Short-interval intracortical facilitation (SICF) Our data replicate earlier findings that MEP facilitation occurs at distinct ISI of 1.5 ms, 2.5 ms and 4.5 ms (Tokimura et al., 1996; Ziemann et al., 1998a; Ziemann et al., 1998b; Chen and Garg, 2000). It is thought that these peaks reflect a facilitatory interaction between the effects of S1 and S2 at the axon initial segments of those excitatory interneurons in motor cortex which contribute to the generation of I-waves (Hanajima et al., 2001; Ilic et al., 2002). The peaks are separated by troughs where no facilitation occurs (cf. Fig. 1). If the hypothesis was correct that SICF contributed to a significant extent to SICI then it should be expected that this can be detected only or predominantly at those ISI where SICF occurs. In addition to the ISI, expression of SICF depends on the intensities of S1 and S2 (Ziemann et al., 1998a; Chen and Garg, 2000; Ilic et al., 2002). Of particular interest for the present experiments is the intensity of S1 because this one is varied in the SICI intensity curve measurements. The second SICF peak (which will be the focus of discussion, see below) has a S1 threshold of 90% RMT but oc-
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
Fig. 3. Correlation between r-SICF and r-SICI The ratio of SICF (r-SICF) obtained at the individual peak2 and trough1 of the SICF-ISI curve (S1 intensity MEP1mV; S2 intensity 90% RMT) served as the independent variable. The ratio of SICI (r-SICI) obtained at the ISI that resulted in the individual SICF peak2 and trough1 served as the dependent variable. Regression analyses were separately computed for SICI tested at an S1 intensity of 100% AMT (r = 0.26, P = 0.48) (A), 110% AMT (r = 0.76, P = 0.018) (B) and 120% AMT (r = 0.85, P = 0.004) (C). Note that large magnitudes of r-SICF resulted in strong reductions of SICI (i.e. large r-SICI) at the two higher intensities of S1.
curs at a delayed interval of 3.3–3.5 ms at this threshold S1 intensity (cf. Fig. 2 in (Ziemann et al., 1998a)). This peak becomes more clearly visible with higher S1 intensities of 100–130% RMT and the effective ISI decrease to 2.2–2.8 ms (Fig. 2 in (Ziemann et al., 1998a)). This increasing magnitude of SICF with the intensity of S1 is explained by the fact that the facilitatory interaction between S1 and S2 requires excitation by S1 of those interneurons which are responsible for the late I-waves (Hanajima et al., 2002; Ilic et al., 2002), which occurs to a significant extent only if S1 exceeds RMT (Di Lazzaro et al., 1998a). From this it follows that SICF should contribute to the SICIpeak2 intensity curve only at the high end part of the curve, i.e., where S1 is at least 90% RMT. 4.2. Short-interval intracortical inhibition (SICI) We show here that the shape of the SICI intensity curve and the magnitude of the maximum SICI differ with the interstimulus interval if adjusted to individual peaks and troughs of the SICF interstimulus interval curve. SICI trough1 exhibited a sigmoid shape (no decrease of SICI in the high intensity range of S1) and reached the strongest maximum inhibition when compared to all other curves (cf. Fig. 2A–D). Therefore, and because we expected no con-
2295
tamination by SICF, SICItrough1 was selected as reference curve for comparison with all other curves. To our knowledge, SICI intensity curves as a function of ISI have never been systematically studied, although previous studies tested SICI intensity curves at a single ISI and showed, similar to the present results, that the SICI curve could be sigmoid at an ISI of 2.0 ms (Chen et al., 1998; MacKinnon et al., 2005) but U-shaped at an ISI of 3 ms (Kujirai et al., 1993; Ziemann et al., 1996c; MacKinnon et al., 2005). This study is the first one to measure SICI intensity curves at those ISI that resulted in the individual peaks and troughs of the SICF ISI curve. The fact that we found stronger SICItrough1 in this study compared to SICI tested under similar conditions in one of our previous studies (cf. Fig. 1B in (Ilic et al., 2002) is first evidence in support of the notion that SICF may act to reduce SICI if not carefully avoided by choosing the individually determined ISI where no or almost no SICF occurs. The comparison of SICIpeak1 (ISI = 1.54 ± 0.03 ms) with SICItrough1 (ISI = 1.97 ± 0.03 ms) revealed that the two curves did not differ significantly as a whole although the maximum SICItrough1 measured at S1 = 100% AMT was stronger than maximum SICIpeak1 (14.3 ± 2.9% vs. 38.0 ± 8.8%, P = 0.017, Fig. 2A). The data are compatible with the earlier findings that indicated slightly weaker SICI at an ISI of 1.5 ms vs. 2.0 ms when using a threshold tracking protocol (Fisher et al., 2002). SICF did probably not contribute to this weaker expression of SICI at peak1 because regression analysis revealed no correlation between the ratios of SICFpeak1/SICFtrough1 vs. SICIpeak1/SICItrough1 at any of the S1 intensities from 90% to 120% AMT (all P > 0.1). The slight difference between maximum SICIpeak1 and SICItrough1 may be attributed to differences in the physiology of inhibition at these two ISI. Several findings suggested that refractoriness of axons excited by S1 contribute to inhibition at very short ISI of 1 ms, while true GABAAergic synaptic inhibition comes into play at longer ISI: voluntary contraction and drugs enhancing neurotransmission through the GABAA receptor had no effect on SICI at an ISI of 1.0 ms but modified SICI at ISI of 2–3 ms (Ziemann et al., 1996b; Ziemann et al., 1996a; Fisher et al., 2002). However, diazepam, a positive allosteric modulator at the GABAA receptor, enhanced SICI at an ISI of 1.5 ms, suggesting that synaptic inhibition is already operating at this interval (Ilic et al., 2002). Therefore, it appears that SICIpeak1 and SICItrough1 share more similarities than they are divided by differences, and this is further supported by the very similar course of these curves at low intensities of S1 (Fig. 2A), indicating no difference in SICI threshold, whereas a lower threshold would be expected for SICIpeak1 if inhibition at this interval was caused predominantly by refractoriness (Fisher et al., 2002). Finally, it cannot be entirely excluded that the differences between SICItrough1 and SICIpeak1 were to some extent underestimated in this study because individual SICF peak1 occurs at a range of ISI (1.1– 1.5 ms) (Ziemann et al., 1998a) while we have tested only ISI P 1.5 ms. The comparison of SICIpeak2 (ISI = 2.61 ± 0.07 ms) with SICItrough1 revealed a highly significant interaction of ISI with S1 intensity (Fig. 2B). This was explained by a weaker SICIpeak2 than SICItrough1 at S1 intensities P100% AMT, whereas the two curves showed an almost identical course at low S1 intensities of 50–90% AMT. The similarities in SICI threshold and recruitment of SICI with increasing S1 intensity strongly suggest that the physiology of SICI at SICF trough1 and SICF peak2 is similar if not identical. The divergence along the higher S1 intensities suggests recruitment of a highthreshold facilitatory process with SICIpeak2 but not SICItrough1. The significant correlation between the ratios SICFpeak2/SICFtrough1 vs. SICIpeak2/SICItrough1 strongly suggests that SICF contributes to this facilitatory process (cf. Fig. 3B and C). A similarly increasing contribution of facilitation with increasing S1 intensity was previously shown by the threshold tracking technique at an ISI of 2.5 ms (cf. Fig. 5 in (Fisher et al., 2002)). Therefore, SICI has to be considered as a net inhibition if tested at an ISI where SICF occurs (i.e.
2296
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297
2.5 ms) and if S1 intensity exceeds the SICF threshold which is around 90% RMT (i.e. 100–110% AMT) (Ziemann et al., 1998a). In addition, it is possible that SICF is a net facilitation with contributions coming from SICI. This notion is supported by previous pharmacological evidence that showed decreases of SICF when neurotransmission through the GABAA receptor is enhanced by benzodiazepines (Ziemann et al., 1998b; Ilic et al., 2002). This interaction was not the focus of the present experiments but may have contributed to the observed significant correlations between the ratios SICFpeak2/SICFtrough1 vs. SICIpeak2/SICItrough1 (Fig. 3B and C). The SICItrough2 (ISI = 3.50 ± 0.07 ms) and SICIpeak3 (ISI = 4.26 ± 0.06 ms) curves were conspicuously different from SICItrough1 by running on a clearly flatter course (Fig. 2C and D). This is in agreement with earlier threshold tracking data, which have also indicated less inhibition at ISI in the range from 3.0 to 4.5 ms if compared to SICI at 2.5 ms (cf. Fig. 3B in (Fisher et al., 2002)). Our data do not allow firm conclusion to attribute this decrease in SICI to a certain mechanism. However, it is unlikely that it is indicating a true decline in the strength of GABAA receptor mediated inhibition as this inhibition is associated by inhibitory postsynaptic potentials with a clearly longer time course of some 20 ms (Connors et al., 1988) that can also be demonstrated by paired-pulse TMS in humans (Hanajima et al., 1998). A more likely explanation is that another facilitatory process, such as intracortical facilitation (ICF) comes into play at these longer interstimulus intervals (Kujirai et al., 1993; Ziemann et al., 1996c; Di Lazzaro et al., 2006b). This notion is consistent with the observations that (i) the threshold for ICF is around 80% AMT (Ziemann et al., 1996c), the S1 intensity at which the SICItrough2 and SICIpeak3 curves started to diverge clearly from the SICItrough1 curve in the present experiments (Fig. 2C and D); (ii) SICI interstimulus interval curves with S1 intensity set to 90% AMT often start to turn from inhibition to facilitation at ISI P 3.0 ms (Kujirai et al., 1993; Ziemann et al., 1996c). SICF did not seem to contribute to this SICI decrease as the regression analyses revealed no significant correlation between the ratios of SICF and SICI (see Section 3). 4.3. Practical significance Several studies demonstrated apparently reduced SICI at an ISI of 2.0 ms in the motor cortex of the unaffected hemisphere of patients after cerebral stroke (Liepert et al., 2000; Manganotti et al., 2002; Shimizu et al., 2002), or even a turn from SICI to facilitation (Liepert et al., 2005). However, all those studies used a fixed S1 intensity of 75–80% RMT or 95% AMT. A subsequent study measured the SICI intensity curve at an ISI of 2.0 ms and suggested that this apparent reduction in SICI or even turn to facilitation was caused by an exaggerated high-threshold facilitatory process, while SICI threshold and the magnitude of SICI along the low intensity part of the SICI curve were normal in the unaffected hemisphere of stroke patients as compared to the healthy controls (Bütefisch et al., 2003). Similarly, many studies have claimed that SICI is reduced in Parkinson’s disease (for review, (Cantello et al., 2002)). A more recent study, however, measured full SICI intensity curves and thereby revealed that SICI threshold and the low intensity part of the SICI curve are normal in Parkinson’s disease, whereas curves differ along the high intensity part (S1 > 100% AMT) suggesting excess facilitation in Parkinson’s disease (MacKinnon et al., 2005). These studies left the question unresolved if this abnormality in the high-intensity part of the SICI curve is caused by deficient SICI or exaggerated facilitation, or both. Our data signify that exaggerated SICF could have contributed, although it was not present in the healthy subjects of this study at an ISI of 2.0, but this might
be different in disease conditions. As a consequence, a reduction in SICI cannot be meaningfully interpreted as decreased inhibition or increased facilitation, if tested at only one S1 intensity, which is above the SICF threshold because, under this condition, SICI likely reflects net inhibition resulting from the summation of SICI and SICF. In order to estimate the contribution of excess facilitation, SICI intensity curves rather than SICI at a single stimulus intensity, and in addition, SICF should be measured. Acknowledgement This study was supported by a grant from the Academy of Finland to S.H.P. References Bütefisch CM, Netz J, Wessling M, Seitz RJ, Hömberg V. Remote changes in cortical excitability after stroke. Brain 2003;126:470–81. Cantello R, Tarletti R, Civardi C. Transcranial magnetic stimulation and Parkinson’s disease. Brain Res Brain Res Rev 2002;38:309–27. Chen R, Cros D, Curra A, Di Lazzaro V, Lefaucheur JP, Magistris MR, et al. The clinical diagnostic utility of transcranial magnetic stimulation: report of an IFCN committee. Clin Neurophysiol 2008;119:504–32. Chen R, Garg R. Facilitatory I wave interaction in proximal arm and lower limb muscle representations of the human motor cortex. J Neurophysiol 2000;83:1426–34. Chen R, Tam A, Butefisch C, Ziemann U, Rothwell JC, Cohen LG. Intracortical inhibition and facilitation in different representations of the human motor cortex. J Neurophysiology 1998;80:2870–81. Connors BW, Malenka RC, Silva LR. Two inhibitory postsynaptic potentials, and GABAA and GABAB receptor-mediated responses in neocortex of rat and cat. J Physiol (Lond) 1988;406:443–68. Curra A, Modugno N, Inghilleri M, Manfredi M, Hallett M, Berardelli A. Transcranial magnetic stimulation techniques in clinical investigation. Neurology 2002;59:1851–9. Daskalakis ZJ, Christensen BK, Fitzgerald PB, Roshan L, Chen R. The mechanisms of interhemispheric inhibition in the human motor cortex. J Physiol 2002;543:317–26. Di Lazzaro V, Oliviero A, Meglio M, Cioni B, Tamburrini G, Tonali P, et al. Direct demonstration of the effect of lorazepam on the excitability of the human motor cortex. Clin Neurophysiol 2000;111:794–9. Di Lazzaro V, Oliviero A, Pilato F, Saturno E, Dileone M, Mazzone P, et al. The physiological basis of transcranial motor cortex stimulation in conscious humans. Clin Neurophysiol 2004;115:255–66. Di Lazzaro V, Oliviero A, Profice P, Saturno E, Pilato F, Insola A, et al. Comparison of descending volleys evoked by transcranial magnetic and electric stimulation in conscious humans. Electroencephalogr Clin Neurophysiol 1998a;109:397–401. Di Lazzaro V, Pilato F, Dileone M, Ranieri F, Ricci V, Profice P, et al. GABAA receptor subtype specific enhancement of inhibition in human motor cortex. J Physiol 2006a;575:721–6. Di Lazzaro V, Pilato F, Oliviero A, Dileone M, Saturno E, Mazzone P, et al. Origin of facilitation of motor-evoked potentials after paired magnetic stimulation: direct recording of epidural activity in conscious humans. J Neurophysiol 2006b;96:1765–71. Di Lazzaro V, Restuccia D, Oliviero A, Profice P, Ferrara L, Insola A, et al. Magnetic transcranial stimulation at intensities below active motor threshold activates intracortical inhibitory circuits. Exp Brain Res 1998b;119:265–8. Di Lazzaro V, Rothwell JC, Oliviero A, Profice P, Insola A, Mazzone P, et al. Intracortical origin of the short latency facilitation produced by pairs of threshold magnetic stimuli applied to human motor cortex. Exp Brain Res 1999;129:494–9. Fisher RJ, Nakamura Y, Bestmann S, Rothwell JC, Bostock H. Two phases of intracortical inhibition revealed by transcranial magnetic threshold tracking. Exp Brain Res 2002;143:240–8. Hanajima R, Furubayashi T, Iwata NK, Shiio Y, Okabe S, Kanazawa I, et al. Further evidence to support different mechanisms underlying intracortical inhibition of the motor cortex. Exp Brain Res 2003;151:427–34. Hanajima R, Ugawa Y, Machii K, Mochizuki H, Terao Y, Enomoto H, et al. Interhemispheric facilitation of the hand motor area in humans. J Physiol 2001;531:849–59. Hanajima R, Ugawa Y, Terao Y, Enomoto H, Shiio Y, Mochizuki H, et al. Mechanisms of intracortical I-wave facilitation elicited with paired-pulse magnetic stimulation in humans. J Physiol 2002;538:253–61. Hanajima R, Ugawa Y, Terao Y, Sakai K, Furubayashi T, Machii K, et al. Paired-pulse magnetic stimulation of the human motor cortex: differences among I waves. J Physiol 1998;509:607–18. Ilic TV, Meintzschel F, Cleff U, Ruge D, Kessler KR, Ziemann U. Short-interval pairedpulse inhibition and facilitation of human motor cortex: the dimension of stimulus intensity. J Physiol 2002;545.1:153–67. Kossev AR, Siggelkow S, Dengler R, Rollnik JD. Intracortical inhibition and facilitation in paired-pulse transcranial magnetic stimulation: effect of
S.H. Peurala et al. / Clinical Neurophysiology 119 (2008) 2291–2297 conditioning stimulus intensity on sizes and latencies of motor evoked potentials. J Clin Neurophysiol 2003;20:54–8. Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A, et al. Corticocortical inhibition in human motor cortex. J Physiol (Lond) 1993;471:501–19. Liepert J, Hamzei F, Weiller C. Motor cortex disinhibition of the unaffected hemisphere after acute stroke. Muscle Nerve 2000;23:1761–3. Liepert J, Restemeyer C, Kucinski T, Zittel S, Weiller C. Motor strokes: the lesion location determines motor excitability changes. Stroke 2005;36:2648–53. MacKinnon CD, Gilley EA, Weis-McNulty A, Simuni T. Pathways mediating abnormal intracortical inhibition in Parkinson’s disease. Ann Neurol 2005;58:516–24. Manganotti P, Patuzzo S, Cortese F, Palermo A, Smania N, Fiaschi A. Motor disinhibition in affected and unaffected hemisphere in the early period of recovery after stroke. Clin Neurophysiol 2002;113:936–43. Müller-Dahlhaus JF, Liu Y, Ziemann U. Inhibitory circuits and the nature of their interactions in the human motor cortex: a pharmacological TMS study. J Physiol 2008;586:495–514. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. Orth M, Snijders AH, Rothwell J. The variability of intracortical inhibition and facilitation. Clin Neurophysiol 2003;114:2362–9. Roshan L, Paradiso GO, Chen R. Two phases of short-interval intracortical inhibition. Exp Brain Res 2003;151:330–7. Sanger TD, Garg RR, Chen R. Interactions between two different inhibitory systems in the human motor cortex. J Physiol 2001;530.2:307–17. Schäfer M, Biesecker JC, Schulze-Bonhage A, Ferbert A. Transcranial magnetic double stimulation: influence of the intensity of the conditioning stimulus. Electroencephalogr Clin Neurophysiol 1997;105:462–9. Shepherd GM. The synaptic organization of the brain. Oxford: Oxford University Press; 1998.
2297
Shimizu T, Hosaki A, Hino T, Sato M, Komori T, Hirai S, et al. Motor cortical disinhibition in the unaffected hemisphere after unilateral cortical stroke. Brain 2002;125:1896–907. Stefan K, Kunesch E, Benecke R, Cohen LG, Classen J. Mechanisms of enhancement of human motor cortex excitability induced by interventional paired associative stimulation. J Physiol 2002;543:699–708. Tokimura H, Ridding MC, Tokimura Y, Amassian VE, Rothwell JC. Short latency facilitation between pairs of threshold magnetic stimuli applied to human motor cortex. Electroencephalogr Clin Neurophysiol 1996;101:263–72. Ziemann U. Intracortical inhibition and facilitation in the conventional paired TMS paradigm. Electroencephalogr Clin Neurophysiol Suppl 1999;51:127–36. Ziemann U. Paired pulse techniques. In: Pascual-Leone A, Davey KJ, Rothwell J, Wassermann EM, Puri BK, editors. Handbook of transcranial magnetic stimulation. London: Arnold; 2002. p. 141–63. Ziemann U, Lönnecker S, Steinhoff BJ, Paulus W. The effect of lorazepam on the motor cortical excitability in man. Exp Brain Res 1996a;109:127–35. Ziemann U, Lönnecker S, Steinhoff BJ, Paulus W. Effects of antiepileptic drugs on motor cortex excitability in humans: a transcranial magnetic stimulation study. Ann Neurol 1996b;40:367–78. Ziemann U, Rothwell JC, Ridding MC. Interaction between intracortical inhibition and facilitation in human motor cortex. J Physiol (Lond) 1996c;496:873–81. Ziemann U, Tergau F, Wassermann EM, Wischer S, Hildebrandt J, Paulus W. Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J Physiol (Lond) 1998a;511:181–90. Ziemann U, Tergau F, Wischer S, Hildebrandt J, Paulus W. Pharmacological control of facilitatory I-wave interaction in the human motor cortex. A paired transcranial magnetic stimulation study. Electroencephalogr Clin Neurophysiol 1998b;109:321–30.