Contralateral silent period (cSP) assessment is a promising biomarker to index cortical excitability and treatment response. We demonstrate a protocol to assess cSP intended for studying M1 corticospinal inhibition of upper and lower limbs.
Contralateral silent period (cSP) is a period of suppression in the background electrical muscle activity captured by electromyography (EMG) after a motor evoked potential (MEP). To obtain this, an MEP is elicited by a suprathreshold transcranial magnetic stimulation (TMS) pulse delivered to the primary motor cortex (M1) of the target muscle selected, while the participant provides a standardized voluntary target muscle contraction. The cSP is a result of inhibitory mechanisms that occur after the MEP; it provides a broad temporal assessment of spinal inhibition in its initial ~50 ms, and cortical inhibition after. Researchers have tried to better understand the neurobiological mechanism behind the cSP to validate it as a potential diagnostic, surrogate, and predictive biomarker for different neuropsychiatric diseases. Therefore, this article describes a method to measure M1 cSP of lower and upper limbs, including a selection of target muscle, electrode placement, coil positioning, method of measuring voluntary contraction stimulation, intensity setup, and data analysis to obtain a representative result. It has the educational objective of giving a visual guideline in performing a feasible, reliable, and reproducible cSP protocol for lower and upper limbs and discussing practical challenges of this technique.
The silent period (SP) is a period of electromyographic (EMG) silence that follows a motor-evoked potential (MEP) induced by transcranial magnetic stimulation (TMS) applied during sustained muscle contraction. The suprathreshold TMS pulse can either be applied to the contralateral or ipsilateral primary motor cortex (M1) of the target muscle from which the EMG activity is being recorded yielding two phenomena: contralateral silent period (cSP) and ipsilateral silent period (iSP).
Even though iSP and cSP share similar features, they may reflect slightly different components. The first is thought to reflect transcallosal inhibition and thus be entirely of cortical origin1,2. Conversely, cSP is investigated as a possible surrogate of corticospinal inhibition, most likely mediated by gamma-aminobutyric acid (GABA) B receptors within M13,4,5.
Supporting the role of cSP in GABA-mediated pathways, previous works have found an increase in cSP duration after oral administration of GABA-enhancing components5,6,7,8. Still, spinal processes are also involved in altering its duration. The earlier phase (<50 ms) of the cSP is associated with decreased H-reflex values3-a reflex that is a product of peripheral neurocircuitry and that quantifies the excitability of spinal neurons9. Spinal processing is thought to be mediated through the activation of Renshaw cells, motoneuron after-hyperpolarization, and postsynaptic inhibition by spinal interneurons10,11,12,13,14.
Despite spinal contribution, cSP results mainly from the activation of cortical inhibitory neurons, which are responsible for generating the later part of the cSP (50-200 ms)3,10,13,15,16. In that respect, the early part of cSP duration has been associated with spinal inhibition mechanisms, whereas long cSPs request larger cortical inhibitory mechanisms3,13,17,18.
Therefore, cSP is a promising biomarker candidate for corticospinal maladaptation due to neurological disorders, whereas more significant cSP durations potentially reflect an increase in corticospinal inhibition and vice versa5,11. Accordingly, previous works have found an association between cSP duration and pathologies such as dystonia, Parkinson's Disease, chronic pain, stroke, and other neurodegenerative and psychiatric conditions19,20,21,22. To illustrate, in a knee osteoarthritis cohort, a higher intracortical inhibition (as indexed by cSP) was associated with younger age, greater cartilage degeneration, and less cognitive performance in the Montreal cognitive assessment scale23. Moreover, cSP changes could also longitudinally index treatment response and motor recovery24,25,26,27,28,29,30.
As promising as the role of cSP in the neuropsychiatry field is, a challenging aspect of its assessment is that it can be too sensitive to protocol variations. For instance, the cSP duration (~100-300 ms)11 is distinguishable between upper and lower limbs. Salerno et al. found an average cSP duration of 121.2 ms (± 32.5) for the first dorsal interosseous muscle (FDI) and 75.5 ms (± 21) for the tibialis anterior muscle (TA), in a sample of fibromyalgia patients31. Thus, the literature conveys a myriad of divergences in the parameters used to elicit cSPs, which in turn jeopardizes comparability across studies and delays the translation to clinical practice. Within a similar population, protocols have been heterogeneous regarding the suprathreshold TMS pulse setting used to stimulate M1 and the target muscle, for example. On top of that, researchers have failed to properly report the parameters used in their protocols.
Therefore, the goal is to provide a visual guideline on how to apply a feasible, reliable, and easily reproducible cSP protocol for evaluating M1 corticospinal excitability of upper and lower limbs and to discuss the practical methodological challenges of that procedure. Also, to help illustrate the reasoning for the choice of parameters, we conducted a non-exhaustive literature review on Pubmed/MEDLINE to identify published papers on cSP in chronic pain and rehabilitation populations, using the search term: Rehabilitation (Mesh) or rehabilitation or chronic pain or stroke and terms such as transcranial magnetic stimulation and single pulse or cortical silent period. No inclusion criteria were defined for the extraction, and pooled results are displayed in Table 1 for illustrative purposes only.
This protocol involves research on human subjects and is in alliance with institutional and ethical guidelines of local ethical committees and the Declaration of Helsinki. Informed consent was obtained from subjects for using their data in the study.
1. Pre-experimental procedures
2. Appropriate instructions to the patients
3. Experimental procedures (Figure 1)
After following the step-by-step procedure, the delivery of a suprathreshold TMS pulse (120% of the RMT) will elicit an observable MEP in the EMG recording of the target muscle, and a subsequent period of background EMG activity suppression of approximately 150 ms to 300 ms (Figure 2). From that EMG pattern, it's possible to calculate the cSP metrics. The most reported outcomes are the duration (in the range of ms) of the relative and absolute SP. The relative SP is measured from the MEP start to the reappearance of EMG activity. One alternative is using the amplified motor stimulated output (MSO = 120% of RMT, as per the protocol) to establish the onset of relative SP. Since the true onset at a network level cannot be known, select the MEP onset as the initial start point to increase experimental reliability13. On the other hand, the absolute SP can be measured from the end of the MEP to the onset of the reemergence of voluntary EMG activity. For instance, using a recording of the resting EMG activity of the subject as reference for qualitative comparison. Those temporal parameters can be identified manually or using automated software37.
A fundamental methodological question for accurate cSP calculation is the definition of the reemergence of EMG background activity. Two approaches can be explored here: The first is using individual trial calculation. In this case, the calculation is based on trial-by-trial measure, using each recording to calculate the cSP duration. Then, a mean (or median) of the individual trials can be calculated and reported. The second approach is using rectified multiple trials. For this approach, all trials will be rectified, and then have to be averaged and overlapped with each other. Then, using the rectified and averaged traces, calculate the cSP duration using average temporal marks. The main advantage of this method is its precision and easier identification of the reappearance of voluntary EMG activity relative to the tonic baseline EMG level36. Using rectified average is advantageous because it is more comparable and reduces the between-subject variability.
It is important to mention that the cSP duration can be prolonged as a sigmoid function of stimulus intensity38, but it is barely affected by the degree of intentional contraction of the target muscle39. Moreover, MEP amplitudes increase with increases in stimulus intensities. Kojima et al. demonstrated that these increases in MEP amplitude (secondary to increases in intensity) is also accompanied by increases in cSP duration40. This behavior is expected since the MEP and cSP duration are thought to be influenced by common factors38. These common factors seem to be present throughout the cortical spinal tract and not in the motor unit; since the increases in stimulus intensity increases both, but increases in muscles contraction do not affect the cSP duration.
With this discussion, it is possible to conclude that stimulus intensity and muscle contraction must be carefully considered during the analysis and interpretation of findings. The cSP is characterized by a linear increase due to the SI, but then a plateau is reached at high intensities; this pattern is highly variable across subjects39, since they could have unique slopes and different plateau intensities. One alternative analysis could include evaluating the cSP during progressively increased intensities to perform an input-output (I/O) curve, and then the cSP can be obtained using the intensity at which the I/O curve reaches the plateau41,42. Finally, since cSP is affected by any activity or exposure that can produce cortical excitability and inhibition changes, it is recommended to assess and record general confounders in the analysis. For instance, using a reporting checklist for TMS experiments43.
cSP interpretation
The TMS test in the current study was used to show the implementation of a feasible and versatile biomarker of M1 inhibition. In general, the longer the duration of the SP, a higher corticospinal M1 inhibition is observed44. However, several factors need to be considered for its interpretation. First, the cSP is defined by both spinal and cortico-subcortical processes45. The spinal components account for approximately the first 50 ms46. The remaining duration is highly influenced by cortical mechanisms such as M1 interneurons inhibition and other inhibitory afferences within M1 (from subcortical regions and other cortices), mediated mainly by GABAergic B neurons after an important cortical activation eliciting MEPs6. It has been suggested that the role of this inhibition is to prevent unwanted movements and sustain motor control47. Second, behavioral and cognitive factors can impact the CSP duration, as well as motor and non-motor neuropsychiatric disorders45,48. Due to this dual nature of cSP, its values need to be interpreted within the experiment context (target population and the use of concomitant motor control tasks).
Figure 1: Experimental steps. 1. Electrode placement on the belly of the FDI muscle 2. Positioning of the dynamometer between fingers. 3. Voluntary contraction of the target muscle to test the standardization of 20% MVC 4. Head measurements and TMS pulses for identification of the hotspot and the RMT (lowest stimuli that elicits an MEP of at least 50 mV in five out of ten trials) 5. CSP protocol, consistent of 10 pulses with 120% RMT spaced out by 10s, during sustained muscle contraction. In the bottom center figure, the small red rectangle represents a single TMS pulse and divides pre-TMS stimulus (sustained muscle contraction and background EMG activity) and the cSP recording. CSP is considered from the start of the MEP until the reemergence of EMG baseline activity, represented inside the blue rectangle. In the yellow rectangle, the MEP latency is shown. Please click here to view a larger version of this figure.
Figure 2: MEP in the EMG recording of the target muscle. On the X-axis, milliseconds (ms), and the Y axis, millivolts (mV) of the EMG signal. From left to right: the red line indicates the background electrical muscle activity before the MEP, subsequently, after the electric effect of the TMS pulse is observed it is followed by the motor-evoked potential. After the MEP, there is a suppression of the EMG signal known as the SP. It can be relative, counting the interval from the beginning of the MEP wave to the return of EMG background activity or absolute SP, counting the interval from the end of the MEP to the return of the background wave. Please click here to view a larger version of this figure.
The default SI to elicit MEP and SPs can vary according to the population. Intensities as low as 80% RMT have been shown to elicit cSP in healthy individuals39, still studies on both healthy and diseased populations have used intensities as high as 150% RMT49,50,51. Although this source of heterogeneity can be inherent to the nature of the target population, it should not be neglected as different SIs have shown to independently (regardless of muscle contraction force) dictate the period of silent EMG activity following the MEP39,49,52. RMTs from 110% to 120% have successfully elicited SPs in a broad range of populations, while still being tolerable for participants53,54. However, 110% RMT might be borderline, since SIs lower than 110% have failed to elicit SPs or elicited SPs with duration shorter than 50 ms39, which might represent solely spinal rather than cortical or corticospinal components of M1 inhibition. Moreover, higher SIs are associated with diminished stimulation focality and increased patient discomfort-especially in diseased populations with higher RMTs55, in which high percentages of suprathreshold stimuli could correspond to close-to-maximum stimulator output. This can jeopardize participants' adherence to the used protocol56. Even though using 120% RMT seems to be the safest and most suitable SI setup overall, researchers should standardize the SI by checking previous successful experiments done in the population of interest. Standardization across similar populations is important to facilitate further pooled statistics.
The experiment is typically done using a single SI, but some studies have investigated cSP responses at more than one stimulus setup53,57,58,59,60. In the context of conditions with no clear pathophysiology or previous literature, or in which the understanding of the SP behavior is the objective of the study, it is recommended to plot cSPs against subsequent increasing stimulus intensities (i.e., 10% increment in a sigmoid stimulus-response curve)42. In that case, the researcher should consider adding to the protocol resting breaks to avoid muscle fatigue. Even though still contradictory, there is a considerable amount of evidence that cSPs are not influenced by the level of muscle contraction39,61,62; however, it is impacted by muscle fatigue63,64,65,66. A value of 20% of maximum voluntary contraction (MVC) has consistently shown to successfully elicit SPs with being less likely to induce fatigue60,67,68,69.
Another significant parameter that may contribute to the heterogeneity of cSP results in the literature is the selected muscle to evaluate cSP after the TMS stimuli. Studies have found that different muscles may recruit distinctive motor neuron networks, which in turn will have diverging cSP effects. This is true not only for upper versus lower limb musculature, but also for proximal and distal muscles of the same limb. For instance, in two separate studies, Van Kuijk and colleagues have conveyed a more significant sensitivity to TMS parameters, such as cSP, in distal upper-extremity muscles as compared to proximal muscles70,71. And although this difference has not always been statistically significant71, it is still noteworthy and may contribute to heterogeneous results. Moreover, a significant difference in cSP responses in upper and lower limb muscles has also been denoted in studies about fatigue, with upper limbs conveying 30% longer SPs than lower limbs72. Thus, to reduce heterogeneity in cSP results, it is important to standardize the muscle in which the cSP assessment will be evaluated as some are more sensitive to TMS stimuli than others. Therefore, different muscles can drastically change the procedure specificities and interpretation. To illustrate, cSP is also used to evaluate cortical excitability in deeper muscles, such as the laryngeal motor units. Applying the cSP protocol to these structures comes with unique challenges. An example is that of the laryngeal motor cortex; the stimulation site of this protocol is near the EMG electrode, which can increase the number of artifacts requiring adjustments to the EMG amplifier73. Also, needle electrodes that penetrate the skin are needed to measure the EMG activity of these muscles, making the placement, and if needed, reallocation of the electrodes difficult, as well as changing the interpretation of results. Therefore, a limitation of this methodological paper is that its scope is restricted to illustrating a protocol for upper and lower limbs, and that does encompass, for example, the field that explores cSP as a marker of corticobulbar inhibition or psychiatry conditions.
In that matter, literature search supports that the FDI is the most commonly used target muscle for studying upper limb M1 corticospinal inhibition. Reasons include but are not limited to its superficial and large cortical representation in the motor cortex, the lowest motor threshold for stimulation, and the simplicity to perform its isolated and sustained contraction as well as the positioning of the electrodes73,74. For the lower limbs, the use of the TA muscle is most frequent-likely due to its larger cortical representation in comparison to other leg muscles75. Also, the easiness to be isolated from the activity of large muscle groups that compose the lower limbs' musculature plays a role. Despite the importance of lower limb (LL) rehabilitation in the field, fewer studies use the LL MEP given its particular challenges. The brain anatomical region of the LL is more medial and deeper in the inter-hemispheric fissure compared to the upper limbs. However, the use of Neuronavigation has improved the accuracy of the stimuli36, while the use of a double cone coil has successfully targeted LL regions, including the TA muscle, showing lower LL MT than other coil types76,77,78,79, and currently being the standard recommendation to target LL36,44. However, the usage of modern navigation technologies should be contemplated parallel to the feasibility of the protocol. Jung et al. (2010) revealed no significant difference in MEP variability and reproducibility between non-navigated TMS and TMS navigation, which reach a comparable performance level80. Using non-navigated TMS can be more cost-effective in specific circumstances (i.e., limited resources), and therefore was the preferred approach for this protocol that aims to demonstrate a feasible, easy, and reproducible cSP assessment.
Given the promising and versatile use of cSP as a corticospinal inhibition biomarker in different neurological disorders, it is essential to provide researchers with a feasible, reproducible, and still reliable cSP protocol for upper and lower limbs. We highlight that only a few muscles could be represented in the experiment, leading to the absence of investigation of cSP for corticobulbar inhibition. Moreover, the results of the non-exhaustive search provided in Table 1 are not an attempt to summarize existing data but rather to illustrate part of the rationale behind the choice of parameters and insights, therefore, conducted with no scientific rigor. Hopefully, this methodological paper will help researchers advance the investigation of cSP as a biomarker for M1 corticospinal inhibition.
Table 1: Different parameters used on cSP protocols. We extracted data of cSP experiments from 117 different articles. Results are reported if the paradigm was used in ≥2 experiments, otherwise they were gathered within others. †Includes articles that either did not report the method of standardization or that reported not applying standardization. Abbreviations: MVC = maximum voluntary contraction. Please click here to download this Table.
The authors have nothing to disclose.
No acknowledgments.
Alcohol pads | Medline | Preparation with 70% isopropyl alcohol | |
Conductive gel | Weaver and Company | Used on the electrode | |
Echo Pinch | JTECH medical | 0902A302 | Digital dynamometer. |
Mega-EMG | Soterix Medical | NS006201 | Digital multiple channel EMG with built in software. |
MEGA-TMS coil | Soterix Medical | NS063201 | 8 shaped TMS coil |
Mega-TMS stimulator | Soterix Medical | 6990061 | Single Pulse TMS |
Neuro-MEP.NET | Soterix Medical | EMG software used to analyse the muscles eletrical activity. | |
Swim cap | Kiefer |