Modulatory effects of repetitive peripheral magnetic stimulation on the soleus H-reflex: comparison between peripheral nerve and motor point stimulation

Abstract

[Purpose] We compared the effects of repetitive peripheral magnetic stimulation (rPMS) applied to the peripheral nerve and motor points on the H-reflex pathway in the soleus muscle. [Participants and Methods] Thirty-six healthy adult males were randomly assigned to one of three groups: peripheral nerve stimulation, motor point stimulation, or control (n=12/group). The excitability of the monosynaptic reflex pathway was assessed using the H-reflex of the soleus muscle. In each session, 24 H-reflexes were recorded 10 min before and after either rPMS or rest (control groups). The final analysis was carried out by comparing the mean amplitudes of the resulting 12 measurements. [Results] The H-reflex amplitude increased following stimulation in the peripheral nerve stimulation group, and a significant interaction effect was observed among the three groups (before and after intervention in the peripheral nerve and motor point stimulation groups, and 10 min before and after rest in the control group). [Conclusion] The increase in H-reflex amplitude after peripheral nerve stimulation may be attributed to effective stimulation of the tibial nerve, with signals ascending through Ia sensory fibers. Future research should clarify the mechanism by which rPMS influence spinal neural circuits via peripheral nerves and muscles.

INTRODUCTION

Exercise therapy based on central nervous system (CNS) plasticity (i.e., a neurophysiological approach) is the primary strategy for restoring motor function in patients with CNS disorders, such as stroke and spinal cord injury. However, systematic reviews^1,2,3) have reported that many studies on neurophysiological approaches are limited by low levels of evidence, often affected by bias and small sample sizes¹⁾, or show no significant differences in outcomes from those of other forms of exercise therapy^{2, 3)}. Therefore, strengthening the evidence base for neurophysiological approaches and developing new rehabilitation methods are essential to enhance exercise therapy effectiveness.

Stefan et al.⁴⁾ and Popovic et al.⁵⁾ have shown that combining transcranial magnetic stimulation with voluntary movement and electrical stimulation can enhance corticospinal tract excitability, thereby promoting brain and spinal cord plasticity associated with functional recovery. More recently, repetitive peripheral magnetic stimulation (rPMS) has gained attention as an alternative to electrical stimulation for targeting peripheral sites. In physical therapy for CNS disorders, rPMS serves as both a complementary and alternative approach to exercise therapy based on CNS plasticity. However, the effects of rPMS on spinal nerves remain controversial; some studies have reported that spinal nerves can be stimulated or blocked by rPMS^6,7,8,9), while others have found it difficult to achieve spinal nerve stimulation using this method^{10,11,12,13,14)}.

In a 2013 review, Beaulieu and Schneider⁶⁾ examined the effects of magnetic stimulation in healthy adults and individuals with stroke or spinal cord injuries. Additionally, a 2015 review¹⁵⁾ focused on the therapeutic effects of magnetic stimulation for pain and peripheral neuropathy. These reviews reported enhancements in motor functions, such as changes in electromyography (EMG) activity, muscle strength, and activities of daily living (ADL), and reductions in muscle tone, as assessed using the Ashworth Scale and tendon reflexes⁶⁾. Improvements in pain (based on VAS [visual analogue scale] and algometer values) and peripheral neuropathy were also noted¹⁵⁾. Kamo et al.¹⁶⁾ conducted a systematic review of the Cochrane Library in 2022, evaluating the effects of rPMS on functional and disability outcomes post-stroke. This review included four randomized trials involving 139 participants who received rPMS. The authors concluded that rPMS did not produce clear improvements in ADL, muscle strength, upper limb function, or spasticity and emphasized the need for further research to establish its efficacy as an evidence-based intervention.

Zhu et al.¹²⁾ and Piponnier et al.¹³⁾ reported that the H-reflex tends to be underestimated under single magnetic stimulation. These studies suggest that single magnetic stimulations result in a brief duration of sensory nerve activation, making it difficult to adequately stimulate the sensory nerves. Behrens et al.¹⁰⁾ found no significant changes in H-reflex amplitude following rPMS applied to various sites on the soleus muscle belly. Similarly, Nito et al.¹⁴⁾ measured motor evoked potentials using transcranial magnetic stimulation, along with H-reflex and M-wave responses, before and after rPMS applied to the wrist extensor muscles of healthy adults. Their findings showed that rPMS at 25 Hz or higher increased motor cortex excitability but did not alter spinal cord excitability.

Little is known about the mechanism by which rPMS influences the spinal nerve circuit. We believe that the following issues must be resolved to clarify precisely how magnetic stimulation affects spinal nerves through peripheral nerves and muscles: (1) the distinct effects of single-pulse versus repetitive magnetic stimulation on spinal nerves, (2) the differing impacts on spinal nerves depending on the stimulation site (specifically, peripheral nerve stimulation versus motor point stimulation), and (3) the influence of reciprocal Ia inhibition and α-motor nerve facilitation within spinal nerve circuits during magnetic stimulation.

Our research focuses on the changes in spinal nerve mechanisms mediated by peripheral nerves and muscles through rPMS, aiming to develop methods to modify neural circuit function. We have previously reported the effects of rPMS applied to the common peroneal nerve on the soleus muscle H-reflex¹⁷⁾, demonstrating that stimulating the nerve controlling an antagonist muscle with rPMS inhibits the spinal nerve controlling the agonist muscle.

Based on the above studies, we hypothesized that rPMS applied to the muscles (motor points) and peripheral nerves would differentially influence spinal cord excitability. This study aimed to investigate the effects of rPMS applied to two different sites (soleus muscle belly and tibial nerve) on the excitability of spinal cord circuits using H-reflex measurements.

PARTICIPANTS AND METHODS

This study was approved by the Tokyo International University Research Ethics Committee (approval number: 23-3) and was conducted in accordance with the fundamental ethical principles outlined in the Declaration of Helsinki (1964). The study included 36 healthy male university students (aged 18–22 years; height: 171.8 ± 5.7 cm; weight: 64.5 ± 6.8 kg) with no history of trauma or disorders affecting the musculoskeletal or nervous systems. Participants with internal metal implants or those using electrical life support devices, such as pacemakers, were excluded. Participants were randomly assigned to one of three groups: peripheral nerve stimulation (n=12), motor point stimulation (n=12), and control (n=12) (Table 1). All participants received a detailed explanation of the study’s purpose, procedures, benefits, risks, data privacy, and their right to refuse or withdraw, and provided written informed consent. The participants were positioned prone with the hip and knee joints at 0°, and the ankle joint was fixed in a neutral position between plantarflexion and dorsiflexion. The feet were sandwiched between the bed and the wall and compressed to fix them (Fig. 1). During H-reflex recording and rPMS, participants rotated their necks to the right and maintained visual fixation on a target while white noise (intensity 40–50 decibels, in a soundproof shielded room) was played to minimize auditory distractions and sustain alertness. The participants were instructed to remain relaxed throughout the measurements. Surface electromyography electrodes (Nihon Kohden Corporation, Tokyo, Japan, F-150S) were placed on the right soleus muscle with an inter-electrode distance of 20 mm. The electromyographic signals were amplified (×1,000) using a multi-channel high-sensitivity amplifier (EMG-6108; Nihon Kohden Corporation) and band-pass filtered between 5 Hz and 3 kHz. These analog signals were digitized using an A/D converter (PowerLab16/35; ADInstruments, Bella Vista, Australia) at a sampling frequency of 10 kHz. The H-reflex was elicited by electrical stimulation of the right tibial nerve at the popliteal fossa, with the anode placed over the center of the patella and the cathode positioned directly over the tibial nerve. Stimulation was delivered using a constant current stimulator (DS7A; Digitimer Ltd., Welwyn Garden City, UK) with 1 ms pulses. First, the H-reflex recruitment curve was obtained by gradually increasing stimulation intensity, and the intensity producing 60% of the maximal H-reflex amplitude was identified (Fig. 2). This intensity was used for test stimulation in subsequent H-reflex measurements (Table 2). The electrical stimulation rate of the H-reflex was set at 0.2 Hz after reviewing previous research¹⁸⁾. In each session, 24 H-reflexes were recorded before and after intervention in the peripheral nerve and motor point stimulation groups, as well as before and after 10 min of rest in the control group. Data from the final 12 measurements out of the total 24—when waveforms had stabilized—was averaged using the peak-to-peak H-reflex amplitude.

Table 1. Participant characteristics.

	Age (years)	Height (cm)	Weight (kg)
Control group (N=12)	20.1 ± 1.1	170.3 ± 4.8	61.9 ± 5.3
Peripheral nerve stimulation group (N=12)	20.1 ± 1.2	172.6 ± 7.5	64.9 ± 8.7
Motor point stimulation group (N=12)	20.2 ± 1.0	172.6 ± 4.6	66.9 ± 5.4

Values are expressed as mean ± standard deviation (SD).

There were no significant differences among the three groups in terms of age, height, or weight at baseline.

Fig. 1.

Block diagram of the experiment.

The participants were positioned prone with the hip and knee joints at 0°, and the ankle joint was fixed in a neutral position between plantarflexion and dorsiflexion. The feet were sandwiched between the bed and the wall and compressed to fix them. During H-reflex recording and repetitive peripheral magnetic stimulation (rPMS), participants rotated their necks to the right and maintained visual fixation on a target while white noise (intensity 40–50 decibels, in a soundproof shielded room) was played to minimize auditory distractions and sustain alertness.

Fig. 2.

H-reflex recruitment curve.

First, the H-reflex recruitment curve was obtained by gradually increasing stimulation intensity, and the intensity producing 60% of the maximal H-reflex amplitude was identified. This intensity was used for test stimulation in subsequent H-reflex measurements.

Table 2. Amplitude and stimulation current values for H-max and 60% H-max.

	H-max amplitude (mV)	H-max stimulation current (mA)	60% H-max amplitude (mV)	60% H-max stimulation current (mA)
Control group (N=12)	3.84 ± 3.37	9.54 ± 6.71	2.35 ± 2.11	7.35 ± 5.93
Peripheral nerve stimulation group (N=12)	3.85 ± 1.65	6.68 ± 7.08	2.31 ± 0.99	5.70 ± 5.69
Motor point stimulation group (N=12)	3.91 ± 1.81	7.41 ± 7.91	2.37 ± 1.11	6.27 ± 6.83

Values are expressed as mean ± standard deviation (SD).

rPMS were applied for 10 min using a multiple-pulse skeletal muscle magnetic stimulator (Pathleader; IFG, Sendai, Japan) with a pulse width of 350 µs, frequency of 50 Hz, stimulation intensity set at 70%, and maximum magnetic flux density of 0.85–0.94 T (100%). The duty cycle consisted of 2 s of stimulation, followed by 8 s of rest. For the peripheral nerve stimulation group, rPMS was applied to the right tibial nerve at the center of the right popliteal fossa, whereas for the motor point stimulation group, rPMS was applied to the soleus muscle motor point (Fig. 3). H-reflex amplitudes before and after the intervention within each group were compared using a two-way analysis of variance (ANOVA) with a significance level of 5%. Post-hoc comparisons were performed using paired t-tests with Bonferroni correction, setting the significance threshold at p<0.016.

Fig. 3.

Magnetic stimulation and electromyography (EMG) measurement points.

For the peripheral nerve stimulation group, repetitive peripheral magnetic stimulation (rPMS) was applied to the right tibial nerve at the center of the right popliteal fossa. For the motor point stimulation group, rPMS was applied to the soleus muscle motor point. Surface electromyography electrodes (Nihon Kohden, F-150S) were placed on the right soleus muscle with an inter-electrode distance of 20 mm.

RESULTS

A two-way ANOVA of H-reflex amplitudes before and after the intervention across the control, peripheral nerve site stimulation, and motor point stimulation groups revealed a significant interaction between amplitude and group/stimulation site (degrees of freedom=2, F=5.22, p=0.011, effect size-eta-squared=0.24). Post hoc tests showed that the H-reflex amplitude increased significantly from 2.42 ± 0.94 mV before stimulation to 3.10 ± 1.44 mV after stimulation (p=0.003; Table 3) in the peripheral nerve stimulation group. In the motor point stimulation group, the H-reflex amplitudes decreased from 2.53 ± 1.23 mV before to 2.46 ± 1.22 mV after stimulation, and the difference was not significant (p=0.647). In the control group, the amplitudes increased before (2.18 ± 1.54 mV) and after rest (2.45 ± 1.78 mV), and the difference was not significant (p=0.122).

Table 3. Changes in H-reflex amplitude before and after the 10-minute repetitive peripheral magnetic stimulation (rPMS) session.

	Pre-intervention	Post-intervention
Control group (mV)	2.18 ± 1.54	2.45 ± 1.78
Peripheral nerve stimulation group (mV)	2.42 ± 0.94	3.10 ± 1.44*
Motor point stimulation group (mV)	2.53 ± 1.23	2.46 ± 1.22

Values are expressed as mean ± standard deviation (SD).

*p<0.016, compared to pre-intervention (paired t-test).

Two-way analysis of variance (ANOVA) revealed a significant interaction effect (p<0.05).

DISCUSSION

In this study, the H-reflex amplitude—an indicator of spinal cord excitability—was measured before and after stimulation in three groups: a control group without rPMS, a peripheral nerve stimulation group receiving rPMS at the nerve site, and a motor point stimulation group receiving rPMS at the motor points. A significant increase in H-reflex amplitude was observed only in the peripheral nerve stimulation group following rPMS. This increase may have resulted from short-term continuous rPMS delivered at 50 Hz, which likely traveled up Ia sensory fibers to the spinal cord and stimulated motor neurons through temporal summation at spinal synapses. Funase¹⁹⁾ reported that the H-reflex amplitude reflects changes in the number of motor neurons firing in response to membrane potential shifts within the spinal motor neuron pool controlling the muscles. At rest, without external stimulation or voluntary commands, an action potential and corresponding H-reflex occur when the membrane potential of a small motor neuron exceeds the threshold due to Ia input, consistent with the size principle²⁰⁾. Under additional excitatory input, the excitatory postsynaptic potential raises the membrane potential of the next motor neuron with the lowest threshold beyond those already firing, thereby increasing the number of active motor neurons²¹⁾. In this study, repeated peripheral nerve magnetic stimulation resulted in temporal summation, bringing the membrane potential of motor neurons closer to the firing threshold, increasing motor neuron firing, and resulting in a higher H-reflex amplitude.

In the motor point magnetic stimulation group, the neuromuscular junction and α-motor nerve fibers were targeted. Visual observation during stimulation at both the motor point and nerve site using the same current intensity showed stronger muscle contractions during motor point stimulation than during peripheral nerve stimulation. This observation indicates the activation of more muscle fibers, suggesting a stronger stimulus ascending toward the spinal cord. However, the H-reflex amplitude in the motor point stimulation group did not change significantly after rPMS; in fact, the average amplitude decreased. This result is likely due to the magnetic stimulus ascending the α-motor nerve to the spinal cord and descending via the same pathway. α-motor nerve fiber stimulation is thought to have little effect on the spinal cord. Fok et al.²²⁾ reported that the motor point is an effective peripheral neuromuscular stimulation site in paired associative stimulation, which synchronizes peripheral nerve electrical stimulation with transcranial magnetic stimulation of the primary sensorimotor cortex. Additionally, Fok et al.²²⁾ and Nakagawa et al.^{23, 24)} have discussed the mechanistic differences between motor point and peripheral nerve stimulation, suggesting that peripheral nerve stimulation can effectively activate Ia sensory nerves, whereas motor point stimulation is less effective. Furthermore, Kaneko et al.²⁵⁾ investigated the effects of peripheral nerve and motor point stimulation on Ia sensory nerve activation in the soleus muscle. In their study, peripheral nerve and motor point stimulation were applied to the soleus muscle as conditioning stimuli, and transcutaneous spinal cord stimulation was used as the test stimulus, applied to the skin between the L1 and L2 spinous processes to evaluate Ia sensory nerve activation. The results demonstrated that peripheral nerve stimulation was more effective than motor point stimulation in activating Ia sensory nerves. Kaneko et al.²⁵⁾ reported that peripheral nerve stimulation and motor point stimulation differ in their effects on neuroplasticity; peripheral nerve stimulation activates a broader range of nerves with a preference for recruiting large-diameter axons, whereas motor point stimulation primarily activates motor units located in superficial layers and has a limited ability to reach deeper, larger nerves. The differences between the nerve site and motor point magnetic stimulations observed in the present study are consistent with those reported in previous studies^{17, 19,20,21,22,23,24,25)}.

In CNS disorders, such as stroke, motor paralysis of agonist muscles and spasticity of antagonist muscles often co-occur, creating a significant barrier to motor recovery. The inhibition of spasticity is believed to result from input stimuli via peripheral nerves that induce Ib inhibition of agonist muscles²⁶⁾ and reciprocal Ia inhibition of antagonist muscles²⁷⁾. The therapeutic effects of functional electrical stimulation on motor paralysis may involve several mechanisms: enhanced recruitment through activation of motor nerve axons in agonist muscles, facilitation of spinal interneuron firing, neuroplastic changes in central neurons through modulation of inhibitory circuits (such as recurrent inhibition mediated by spinal Renshaw cells)^28,29,30), and carry-over effects³¹⁾. Hirabayashi et al.^{32, 33)} and Kagamihara et al.^{34, 35)} reported that impairment of the reciprocal Ia inhibitory circuit contributes to foot dorsiflexor paralysis during gait and abnormal hypertonia of antagonistic plantar flexors.

Based on these findings, peripheral nerve stimulation is effective in facilitating α-motor neurons and inducing reciprocal Ia inhibition in antagonist muscles¹⁷⁾. We suggest that rPMS applied to nerve sites may enhance agonist muscle function through excitatory (motor function recovery) and inhibitory (spasticity management) mechanisms. In contrast, motor point stimulation is considered effective primarily for targeting α-motor nerves to directly activate muscle activity. Given these differences, continued analyses of neuromuscular modulation strategies using repetitive peripheral magnetic stimulation are warranted.

Some limitations should be considered. Firstly, the generalizability is limited due to the restriction of participants to young healthy males. Second, the evaluation was limited to the short-term only. Third, the sample size was relatively small and was not calculated before the start of the study. Fourth, the stimulation parameters were fixed. Lastly, the study relied on physiological rather than functional outcomes.

In the future, it is deemed necessary to further investigate these aspects and to advance their application to actual CNS disorders.

Funding

This research was supported by the Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research^© (Grant Number: 23K10485).

Conflict of interest

The authors declare that there are no conflicts of interest.