Decrease in the excitability of spinal nerve function corresponding to the soleus muscle during terminal phase of heel-raise imagery Authors: Hirokazu Takasaki, Hitoshi Asai, Kenji Suehiro, Takashi Ishihama, Toshiaki Suzuki Published: 2024 ABSTRACT Abstract [Purpose] This study investigated the excitability of spinal nerve function in healthy participants by asking them to imagine raising and lowering their heels in a standing position while varying the timing of stimulation to induce an H-reflex from the soleus muscle during imagery. [Participants and Methods] Participants were instructed to begin their imagery at the fourth metronome sound and imagine that their heels would be maximally raised at the time of the optical signal, which occurred 1,000 ms after the metronome sound; moreover, they would be standing at the metronome sound 1,000 ms later. They were instructed to perform the imagery 21 consecutive times. The timing of the electrical stimulation was set for five conditions: 0, 200, 400, 600, and 800 ms after the metronome sound. [Results] Amplitude H/Mmax ratios were significantly higher at the 0, 200, 400, and 600 ms conditions than in the resting and 800 ms conditions. [Conclusion] The imagery task has the potential to increase the excitability of spinal nerve function from 0 ms to 600 ms after a metronome sound but then decrease it to the resting state at 800 ms.Keywords: Motor imagery, H-reflex, Spinal nerve function INTRODUCTION In recent years, stroke rehabilitation has centered on improving motor paralysis, and treatment methods for this condition have advanced dramatically. In particular, assistive robots1), repetitive transcranial magnetic stimulation2), and transcranial direct current electrical stimulation3) have all been effective modern treatment methods. However, even today, the number of patients with access to these therapies is limited, and they have not been widely adopted. Therefore, we focused on “motor imagery (MI)” a treatment for motor paralysis that does not require specialized equipment. It is effective for improving motor function after a stroke in both the upper and lower limbs4, 5), and it is also considered cost-effective due to its simplicity and low-cost6). MI is defined as a mental representation of movement without any visible movement7). Many studies have found that, while movement does not occur, it shares a neural network with motor execution and activates the premotor, supplementary motor, and primary motor cortices8,9,10,11). Furthermore, transcranial magnetic stimulation (TMS) studies have demonstrated that MI increases the excitability of the corticospinal tract12, 13). In short, the excitability of the spinal nerve function may increase, as the activation of the upper centers by MI is thought to affect the spinal cord via descending fibers. Because spinal nerve function is involved in motor control and serves as the final pathway connecting the upper centers to the peripheral effector muscles, the therapeutic effect of MI in rehabilitation must be reflected in spinal nerve function. However, there is no consensus on the excitability of spinal nerve function during MI, which has been classified as either increased14, 15) or unchanged16, 17) excitability. One factor may be influenced by differences in the timing of the electrical stimulation on examination of the peripheral nerves. Assessing the excitability of spinal nerve function with the H-reflex fails to detect all changes. Because the H-reflex evaluates only the reactivity of the spinal cord when electrical stimulation (ES) is applied to the peripheral nerves, the timing of the stimulation is essential. In reports of MI affecting the spinal cord14, 15), the timing of ES is unclear. However, because the imagery task involves isometric contraction of the plantar flexor ankle muscles, the excitability of spinal nerve function during imagery may have remained constant, capturing the increased excitability caused by the imagery effect. On the other hand, Stinear et al.16) devised an imagery task in which thumb flexion movements were timed to a metronome sound, with ES administered at 50 ms before the metronome sound began. Also, Hashimoto and Rothwell17) devised an imagery task in which wrist flexion and extension movements were timed to a metronome sound, with ES administered at 0 and 250 or 500 ms after the metronome sound began. In both reports, it is expected that the excitability of spinal nerve function during imagery would vary because joint movement is used as the imagery task. However, there is little evidence to determine the timing of ES. These studies may not accurately reflect the effects of MI on spinal nerve function. This study hypothesizes that the excitability of spinal nerve function during MI will be increased and sustained compared to rest, but only for a limited time. To test this hypothesis, the study aimed to examine the variability in the excitability of spinal nerve function by comparing the H-reflex amplitudes obtained from stimulation at various times during standing heel raise imagery. Another factor that may influence the excitability of spinal nerve function is the “vividness of the imagery”, which refers to the image’s quality. Moriuchi et al.18) used the visual analog scale (VAS) to assess MI vividness and found a positive correlation between the amplitude of motor-evoked potentials derived from the muscles involved in the imagery task via TMS and VAS values. In other words, the spinal cord is expected to be affected because vivid MI has been shown to significantly increase the excitability of the corticospinal tract. We chose to assess MI vividness using VAS following their method18). PARTICIPANTS AND METHODS The study design was a before-and-after of intervention study to investigate the effects of having the participants perform motor imagery. G*power software (version 3.1.9.5) was used for analysis of variance in repeated measures, within factors. The type of power analysis was set to “a priori: compute required sample size-given α, power, and effect size”. The effect size f was set to 0.25, α error probability was set to 0.05, power (1-β error probability) was set to 0.8, the number of groups was set to 1, the number of measurements was set to 6, correlation among repetitive measures was set to 0.5, and nonsphericity correction ε was set to 1. The sample size was calculated to be 19. Considering participant dropout, 25 healthy adults (17 males and 8 females with a mean age of 27.1 ± 4.4 years) were included to exceed the appropriate sample size. Participants were recruited through a bulletin board on the first floor of the main building of Kyoto Tanabe Central Hospital. The inclusion criteria for participants were healthy adults 20 years or older, and the exclusion criteria were those with a history of peripheral or central nerve disease. Also excluded were those with a history of musculoskeletal disorders such as sprains or fractures. All participants provided verbal and written informed consent before the study began. The study was approved by the Kanazawa University Ethics Review Committee (111095-1) and the Kyoto Tanabe Central Hospital Ethics Committee (2023-002) and followed the Declaration of Helsinki. The participants were positioned supine in half pants, with the right lower extremity on the examination side at 30° knee flexion and 20° ankle plantar flexion. The laboratory environment was a private room with a temperature set to 25 degrees Celsius and every effort was made to keep outside noise from entering. An evoked electromyogram measuring device (MEB-9400 Neuropack S1 (Nihon Kohden Corp., Tokyo, Japan)) was used to measure the H-reflex, which had a frequency range of 10 Hz to 3 kHz and a sampling rate of 10 kHz. As a pretreatment, the skin was shaved, abraded with sandpaper and cleaned with alcohol to reduce the skin impedance resistance to less than 5 kΩ at the electrode application site. Electromyography (EMG) of the right soleus muscle was recorded using Ag/AgCl viscous gel (NCS electrode NM-31, Nihon Kohden Corp.). The active electrode (2.6 × 1.9 cm) was the medial part of the tibia at the midpoint of the line connecting the popliteal fossa on the tibial nerve and the endocarp, the reference electrode (2.6 × 1.9 cm) was placed on the Achilles tendon, and the ground electrode (2.8 cm diameter) was located in the middle of the line connecting the active and reference electrodes. The stimulation site was determined by finding the location where H-reflex could be elicited without M-wave at low stimulus intensities, and where M-wave appeared as stimulus intensity was increased using a handy electrode called the Somato Control Box (RY-441B, Nihon Kohden Corp.). An anode (4.5 × 2.5 cm) was placed distally and a cathode (4.5 × 2.5 cm) was placed proximally (with inter-electrode distance of 2.0 cm) at that site using Ag/AgCl viscous gel (The Vitrode V disposable electrode, Nihon Kohden Corp.). The stimulation intensity was set at 120% of the M-wave appearance threshold according to Tanino et al.19), and percutaneous ES was applied to the posterior tibial nerve with square wave pulses of 1 ms in duration. The M-wave appearance threshold was determined by increasing the stimulus intensity in steps of 0.2 mA from below H-reflex threshold and confirming M-wave appearance at an amplitude sensitivity of 200 μV/D. In order to account for the effects of presynaptic inhibition caused by ES transmitted to the common peroneal nerve20), palpation revealed that the peroneal muscle group was not contracting21). In addition, to ensure consistent ES of the posterior tibial nerve, the variation in M-wave amplitude was monitored simultaneously during the measurement. The MI consisted of a heel raise task in which the heels were maximally raised for 1 s from a standing position before being lowered for 1 s. The participants were given a visual and verbal description of the task in the previous session. The participants were instructed to “imagine as if you were actually doing it”, to make the imagery first person. The participants were instructed to remain still and not move their extremities, trunk, or head and neck during the test. The Mmax wave was measured first, followed by the H-reflex. The Mmax wave was determined by increasing the stimulus intensity in steps of 1.0 mA from below the M-wave appearance threshold. The H-reflex was measured with an electrical stimulus timing controller (delay controller (East Medic CO.,Ltd, Ishikawa, Japan)), which was triggered by an analog signal produced at the start of the electronic metronome sound (EMS). EMS was generated 25 times at 0.5 Hz (2 s intervals) for each condition. The participants were instructed to begin the imagery at the fourth EMS and imagine their heels being maximally raised at the timing of the optical signal, which occurs 1,000 ms after EMS, and standing at EMS 1,000 ms after the optical signal. This MI was repeated 21 times in a row (one cool), and ES to derive the H-reflex per cool was applied five times at 0.1 Hz (10 s intervals). The H-reflex was measured at the 4th, 9th, 14th, 19th, and 24th EMS sessions. The timing conditions for ES to derive the H-reflex were 0, 200, 400, 600, and 800 ms after the onset of EMS. Furthermore, the resting condition of no imagery (the resting state during EMS onset) was included, resulting in six conditions performed in random order. Before the H-reflex measurement, the participants were given ample time to practice imagery of the heel raise task. After completing the imagery in each condition, the participants were instructed to mark a location on a line (100 mm), the two ends of which were labeled “0=not at all” and “10=very clear imagery” on the paper to assess MI vividness using the VAS. At the end of the study, in order to assess whether the participants were engaged in first-person imagery, they were asked to describe in detail what they imagined on the questionnaire. Because all data were determined to be normal by the Shapiro–Wilk test, parametric tests were used in this study. A one-way repeated measures analysis of variance was used to compare the M-wave derived in each condition (mean M-wave amplitude derived from five electrical stimuli per cool), the amplitude H/Mmax ratio (mean H-reflex amplitude derived from five electrical stimuli normalized by the maximum M-wave amplitude), and Shaffer’s modified t-test was used as a post hoc test. When the assumption of multisample sphericity failed (Mendoza’s multisample sphericity test, p<0.05), Greenhouse-Gerisser’s ε was used to calibrate degrees of freedom. The relationship between MI vividness and the amplitude H/Mmax ratio was also assessed using Pearson’s correlation coefficient (r). Modified R Commander (R4.2.1 Ver) was used for statistical analysis, with a significance level of 5%. RESULTS Because of the self-reflective survey, data from one individual who was unable to perform first-person imagery was excluded from this study, while data from the remaining 24 participants were used in the results. Table 1 shows the Mmax wave amplitude, H-reflex amplitude, and VAS obtained for each condition. The M-wave amplitudes obtained under each condition were not significantly different (p=0.24). In contrast, conditions had a significant effect on the amplitude H/Mmax ratio, with multiple comparison tests revealing significantly higher values for the 0, 200, 400, and 600 ms conditions compared to the resting condition (p<0.001). Also, the 0 ms (p<0.001), 200 ms (p=0.002), 400 ms (p<0.001), and 600 ms (p<0.001) conditions yielded significantly higher values than the 800 ms condition. Furthermore, the 400 ms condition significantly outperformed the 0 ms (p=0.005) and 200 ms (p=0.048) conditions (Table 2). There were no significant correlation coefficients between the amplitude H/Mmax ratio and MI vividness (Table 3). Table 1. M-wave maximum amplitude, H-reflex amplitude, and VAS score for each conditionResting condition0 ms condition200 ms condition400 ms condition600 ms condition800 ms conditionMmax-wave (mV)21.7 ± 6.9H-reflex (mV)7.9 ± 3.28.5 ± 3.38.5 ± 3.48.7 ± 3.58.5 ± 3.38.0 ± 3.1VAS score−72.3 ± 11.672.0 ± 11.371.3 ± 11.074.3 ± 10.074.3 ± 11.9VAS: visual analog scale. M-wave maximum amplitude, H-reflex amplitude, and VAS score for each condition VAS: visual analog scale. Table 2. M-wave amplitude and amplitude H/Mmax ratio compared across conditionsResting condition0 ms condition200 ms condition400 ms condition600 ms condition800 ms conditionM-wave amplitude (μV)112.1 ± 12.7110.5 ± 13.6110.6 ± 13.3111.2 ± 13.6111.2 ± 13.8109.9 ± 14.3Amplitude H/Mmax ratio (%)37.2 ± 11.939.8 ± 11.9*, §39.9 ± 12.9*, §41.1 ± 12.6*, †, ‡, §40.5 ± 12.7*, §37.9 ± 11.9*p<0.05, significantly different from resting condition; †p<0.05, significantly different from 0 ms condition; ‡p<0.05, significantly different from 200 ms condition; §p<0.05, significantly different from 800 ms condition. M-wave amplitude and amplitude H/Mmax ratio compared across conditions *p<0.05, significantly different from resting condition; †p<0.05, significantly different from 0 ms condition; ‡p<0.05, significantly different from 200 ms condition; §p<0.05, significantly different from 800 ms condition. Table 3. Correlation coefficient (r) between the amplitude H/Mmax ratio and VAS value for each condition0 ms condition200 ms condition400 ms condition600 ms condition800 ms conditionr0.14−0.06−0.04−0.09−0.22VAS: visual analog scale. Correlation coefficient (r) between the amplitude H/Mmax ratio and VAS value for each condition VAS: visual analog scale. DISCUSSION The spinal nerve function is the neural system’s final pathway for motor control, receiving commands from cortical and subcortical sources via descending fibers and transmitting motor information to the muscles. The function of the spinal nerves may influence movement performance. In fact, there have been reports of reduced spinal nerve function excitability following stroke due to the effects of motor paralysis, resulting in difficulty with voluntary movement22). In such cases, a method for increasing and facilitating muscle contraction is required. Therefore, it is significant that the effects of physiotherapy on stroke patients are mirrored in the excitability of spinal nerve function. In this study, the H-reflex of the evoked electromyogram was used to indirectly assess the excitability of spinal nerve function. Hoffmann described the H-reflex as an electrically evoked reflex23). In neurophysiology, the H-reflex is used to measure spinal motoneuron excitability23, 24). However, the H-reflex amplitude will be affected by the amount of neurotransmitter that is released from the Ia afferent terminals. In particular, presynaptic inhibition and post-activation depression have been proposed as mechanisms for suppressing their absolute amounts and lowering the H-reflex amplitude23, 24). Therefore, it is critical to minimize these consequences. Presynaptic inhibition is the activity of Ia inhibitory neurons in the spinal cord that act on the Ia afferent terminals to reduce the efficiency of synaptic transmission. The activation of Ia inhibitory interneurons involves ES of peripheral nerves innervating heteronymous muscles, passive and active movements of the limbs, and head displacement25,26,27). In this study, participants were placed in a comfortable supine position with their heads stable and instructed not to move their heads, necks, trunks, or limbs during testing. When ES was applied to the posterior tibial nerve, the stimulation site was determined by ensuring that the stimulation did not propagate to the common peroneal nerve, thereby reducing the effect of presynaptic inhibition. Second, post-activation depression is a phenomenon in which after Ia afferent are activated, some fibers become refractory and their number of mobilizations decreases. It has been suggested that the time interval between ES of Ia afferent plays a role in this. Pierrot-Deseilligny and Mazevet28) found that the insensitivity that occurs after activation of Ia afferent takes at least 10 s to recover. Therefore, the frequency of ES in this study was set at 0.1 Hz (10 s intervals), and it was also confirmed that the H-reflex amplitude did not decrease over time in the preliminary experiments of this study. The M-waves derived during the H-reflex testing were also evaluated. The M-wave amplitude is used to monitor whether the stimulation conditions remain constant28). In this study, the stimulus intensity was set at 1.2 times the threshold for the appearance of M-waves, so the M-waves were always derived, and the variability of the M-wave amplitude could be confirmed. The results show that there were no significant differences in M-wave amplitude between the six conditions, implying that they were all performed at similar stimulus intensities. Accordingly, the H-reflex used in this study is regarded as a reliable indicator for examining changes in the excitability of spinal nerve function under various conditions. The amplitude H/Mmax ratio was higher than the resting condition at 0, 200, 400, and 600 ms after the onset of EMS, but was equivalent to the resting condition at the 800 ms. This result indicates that the imagery which the heels were maximally raised for 1 s from a standing position with EMS increases the excitability of the spinal nerve function corresponding to the soleus muscle from 0 ms to 600 ms after the onset of EMS, but the excitability may decrease at the 800 ms. Previous research on MI and the excitability of spinal nerve function has resulted in reports concluding that it either increases14, 15) or remains unchanged16, 17). However, the findings of this study indicate that while the excitability of spinal nerve function increases and persists during MI, it may decrease at times. In other words, when assessing changes in the excitability of spinal nerve function with MI, the timing of ES for testing must be determined after capturing the phase that fully reflects the influence of the upper centers via descending fibers on spinal nerve function. Cowley et al.14) and Hale et al.15) have been able to capture the increased excitability of spinal nerve function caused by MI, although the timing of ES to derive the H-reflex during motor imagery is not clearly specified. This factor is thought to related to the content of the imagery. In both reports, the imagery task was set to isometric contraction of the plantar flexor ankle muscles, and the excitability of spinal nerve function during the imagery of isometric contraction sustained for a certain period was investigated. Thus, it is possible that the excitability of spinal nerve function was less variable. On the other hand, reports16, 17) indicating that the excitability of spinal nerve function remains unchanged with MI suggest that the increase in excitability is not accurately captured because the imagery task involved movement and the excitability of spinal nerve function varied over time. When researching neural excitability in MI, it is important to consider the possibility that excitability may fluctuate along with MI. Next, consider that the excitability of spinal nerve function decreased at 800 ms. Because the MI task is to maximally raise heels over 1 s from a standing position, the 800 ms is the period when plantar flexion movements of ankle joints are still occurring, and the excitability of spinal nerve function corresponding to the soleus muscle was expected to be increased. However, the results revealed that the excitability was lower than that at 0 ms, 200 ms, 400 ms, and 600 ms, indicating that the excitability of spinal nerve function corresponding to the soleus muscle was suppressed. It is possible that the 800 ms period was transitioning to a phase in which the raising of heels was stopped, and that the imagery of stopping the movement affected the spinal nerve function. Belvisi et al.29) used TMS to compare motor evoked potentials (MEP) amplitude derived from first dorsal interosseous (FDI) by applying magnetic stimulation to the unilateral motor cortex at 0 ms, 100 ms, 200 ms, and 500 ms after the onset of exercise in flexion of the index finger (10% maximal voluntary contraction of FDI) using TMS. The results showed that MEP amplitude increased significantly from resting to 0 ms, 100 ms, and 200 ms after the start of exercise, but there was no difference at 500 ms. Although the report does not specify the duration of the movement, the 500-ms period is considered to correspond to the phase when the index finger flexion movement is stopped. In short, it is demonstrated that when voluntary movements are stopped, the excitability of the corticospinal tracts corresponding to the muscles performing the primary actions is suppressed. However, the findings of this study alone do not adequately address the factors that prevented an increase in the excitability of spinal nerve function at the 800 ms time point. In the future, it is necessary to measure the muscle activity of the soleus using surface electromyography while the participants are actually performing the task used in this study. There was no correlation between VAS-measured image vividness as a quality of MI and the excitability of spinal nerve function. Because Moriuchi et al.18) used hand movements as the imagery task, it is assumed that one of the factors was the difference in body parts used in the imagery task compared to the current study. Kaneko et al.30) measured the excitability of corticospinal tracts corresponding to the soleus and tibialis anterior muscles, as well as the vividness of imagery, while participants were walking. The excitability of the corticospinal tracts did not correlate with the vividness of the imagery. The authors believe that MI of the lower limb is unaffected by imagery saliency. Therefore, because this study also used calf raises with ankle plantar flexion as the imagery task, it is possible that the vividness of the imagery was unrelated to changes in the excitability of spinal nerve function. This study had several limitations. The delay controller in this study was programmed to provide ES to derive the H-reflex at 0 ms, 200 ms, 400 ms, 600 ms, and 800 ms after the onset of EMS for heel raise imagery. Therefore, more detailed changes in the excitability of spinal nerve function during heel raise imagery in this study task (e.g., ES intervals of 100 ms) are unclear. Furthermore, in order to discuss the factor that the excitability of spinal nerve function at 800 ms after the onset of EMS was the same as the resting condition, it is necessary to confirm the change in the excitability of spinal nerve function when ES is applied to derive the H-reflex between 800 ms and 1,000 ms. The study suggests that the imagery which the heels were maximally raised for 1 s from a standing position with EMS increases the excitability of the spinal nerve function corresponding to the soleus muscle from 0 ms to 600 ms after the onset of EMS, but decreases the excitability at 800 ms. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflict of interest None of the authors has any conflict of interest to disclose. REFERENCES