Postural adjustments result from the complex coactivity of the vestibular organs (otoliths and semicircular canals), cerebellar activity, and antigravity muscle proprioceptors (Sugita et al., 2004; Kitajima et al., 2006). Projections between vestibular nuclei and the spinal cord integrate neural responses that maintain balance control (Peusner et al., 2012). The lateral, medial and spinal vestibular nuclei project to the spinal cord through the vestibulospinal tract (VST). The VST is divided into two pathways: the lateral and medial vestibulospinal tracts (LVST and MVST, respectively). Particularly, the LVST provides inputs to the forelimb and hindlimb Rexed’s lamina VII and VIII, exerting excitation to extensor motor neurons and inhibition to flexor motor neurons, thus integrating vestibulospinal reflexes (VSR) (Shinoda et al., 1986).

Galvanic Vestibular Stimulation (GVS) is a non-invasive method that modulates the vestibular system firing activity of the semicircular canals and the otolith organs (Wardman et al., 2003). The influence of GVS on LVST can be studied by performing H-reflex recordings in the lower limbs. Typically, stimulation protocols use GVS as a conditioning pulse to modify H-reflex amplitude. For example, GVS applied 100 ms before H-reflex stimulation in 10 healthy subjects provoked a modulation effect with anodal and cathodal stimulation, decreasing and increasing H-reflex amplitude, respectively (Kennedy and Inglis, 2001). In another study, the H-reflex amplitude recorded 2 min after GVS anodal current decreased as compared to the control condition (Čobeljić et al., 2016). Matsugi et al. (2017) found that H-reflex facilitation induced by GVS increased after only 5 min of performing Gaze Stabilization Exercise (GSE) (Matsugi et al., 2017). Additional maneuvers have been added to evaluate the effect of GVS on the H-reflex, for example, it was found that posture influences H-reflex amplitude facilitated by GVS (Tanaka et al., 2021). The head position must also be considered while assessing H-reflex amplitude (Okada et al., 2018; Nepveu et al., 2020). In 10 healthy subjects, Kennedy and Inglis (2002) found that H-reflex modulation by GVS showed a larger response with anodal stimulation and subjects turning to the right, and cathodal stimulation with subjects turning to the left, than any other polarity or head position combination (Kennedy and Inglis, 2002).

By combining GVS and Transcranial Magnetic Stimulation (TMS), Sayenko et al. (2018) studied the effect of both stimuli on spinally evoked motor potentials (SEMPs). Although GVS and TMS showed a facilitator effect on SEMPs, the conditioning effects lasted longer with GVS (>200 ms) (Sayenko et al., 2018). Another study used GVS and TMS as vestibular and corticospinal input, respectively, to determine its influence at premotor (H-reflex) and motoneuron (electromyography) levels by analyzing amplitude and latency. The results suggested that modulation with GVS occurs at premotor neuron and motor neuron levels but is not influenced by cortical activity (Nepveu et al., 2020). Moreover, cerebellar repetitive TMS followed by noisy GVS facilitated the H-reflex ratio excitability recorded 100 ms after a vestibulospinal function test performed with squared GVS in healthy adults. The combination of these stimulation techniques contributed to inferring the cerebellar influence on vestibulospinal activity (Matsugi et al., 2020) and the vestibular modulatory activity on cervical interneurons by studying myogenic evoked potentials amplitude recorded at the biceps brachii muscles (Suzuki et al., 2017).

The P-AD of the H-reflex (P-AD hereinafter) is a well-documented phenomenon in animals and humans. The P-AD occurs when a peripheral nerve is electrically stimulated at frequencies >1 Hz, consequently, the amplitude of the following H-reflexes recorded at a selected muscle decreases as compared to the first one (Ishikawa et al., 1966; Crone and Nielsen, 1989; Hultborn et al., 1996). Although the mechanisms of P-AD are not fully understood, results in healthy subjects indicate that both presynaptic and postsynaptic mechanisms contribute to this phenomenon (Özyurt et al., 2020).

To our knowledge, there are no reports evaluating the effect of cathodal and anodal GVS on the P-AD. We hypothesized that, as GVS modulates the excitability at the premotor and motor neuron levels by changing the amplitude of the H-reflex, an altered, but not suppressed P-AD would occur provoked by the combination of stimuli. Excitability alterations have been seen previously when combining cutaneous stimulation with cathodal and anodal GVS (Lowrey and Bent, 2009). The main aim of this study was to evaluate the P-AD in the lower limbs before, during, and after cathodal and anodal bilateral GVS in healthy subjects. The secondary aim was to evaluate the excitability of the H-reflex during bilateral GVS.

This descriptive, cross-sectional study was performed on 19 healthy volunteers (9 women, 10 men) after an initial screening of 26 subjects (13 women, 13 men). The sample means for age, height, and weight were 22 ± 2 years, 165.6 ± 7.9 cm, and 62.9 ± 13.4 kg, respectively. All subjects showed blood pressure normotensive values (systolic 110 ± 12 mmHg, diastolic 70 ± 8 mmHg). Enrolled individuals reported no history of auditory, vestibular, neurodegenerative, or cardiovascular disease; no subject was under medical treatment. The research protocol was approved by the Faculty of Medicine Ethics in Research Committee of the Autonomous University of the State of Mexico (January 2022, Registration No. CEI/001/2022). Subjects signed an Informed Consent before the experimental session.

The effect of GVS on the H-reflex in healthy subjects has been studied previously to evaluate the vestibular influence on premotor and motor excitability in the spinal cord. However, no previous research was found evaluating the vestibular influence on H-reflex amplitude, nor in P-AD during GVS. The P-AD phenomenon has been studied typically by paired-pulse stimulation; however, it was recently described the importance of evoking >2 consecutive responses, as the second H-reflex may not represent the maximal depression compared to the first one (Worthington et al., 2021). For this reason, in our protocol, 10 consecutive pulses were applied to characterize the behavior of individual pulses (2nd–10th) and to determine the mean P-AD at 1, 5, and 10 Hz; for comparison, H-reflex was evoked by stimulating at 0.1 Hz.

The mechanisms producing the P-AD are not completely understood. Based on previous findings indicating that orthodromic action potentials generated by Primary Afferent Depolarization and recorded as Dorsal Root Reflexes can produce EPSPs in motoneurons; Hari et al. (2022) and Metz et al. (2023) proposed that the P-AD is produced by a mechanism of presynaptic inhibition (PSI). The authors proposed that the orthodromic potentials invading the Ia terminal and generated by the facilitation of sodium spikes via GABA-A receptors near Ranvier nodes, could reduce neurotransmitter release long enough to decrease the amplitude of subsequent H-reflexes and that postsynaptic inhibition mediated by GABA on GABA-A and GABA-B receptors on motoneurons may account for long depression during the P-AD. Figure 11 in Metz et al. (2023) depicts the circuits mediating P-AD through primary afferent depolarization evoked spikes that travels toward the Ia afferent terminal, and to glutamatergic neurons that activate GABAergic neurons and consequently GABA-B receptors in the same Ia afferent. Although additional information is necessary, this scenario shows multiple PSI mechanisms. In this context, by applying a sustained stimulation over the mastoid process while evoking P-AD, it was certain that the changes in excitability produced by GVS would be present at all tested frequencies of P-AD, including the stimulation frequency at 0.1 Hz. For instance, in our protocol, the longest period with GVS at 0.1 Hz lasted 100 s (10 s × 10 repetitions), while the longest period of excitability increase produced by conditioning pulses found in the literature was 300 ms (Sayenko et al., 2018).

In this study, the effect of GVS on the P-AD was described for the first time. We found that the P-AD was not significantly affected during (Ipsi, Contra) and post-GVS (P Ipsi, P Contra) conditions compared to Control, i.e., the mean AUC decrease from 2nd to 10th H-reflex potentials was ≈ 60, 40, and 30%, at 1, 5, and 10 Hz stimulation frequencies, respectively, showing that P-AD was preserved (Figures 4, 5). As expected, an overall increase in the amplitude of individual H-reflexes was observed during (Ipsi/Contra) and post-GVS (P Ipsi/P Contra) in all stimulation frequencies compared to control (Figures 6–8), however, no previous study was found analyzing H-reflex response during GVS. Additionally, we analyzed the changes in variability provoked by GVS on the PA and AUC H-Reflex, and the P-AD. Interestingly, H-reflex (first response pooled for all stimulation frequencies) variability decreased during and after GVS, while P-AD variability incremented with stimulation frequency, although no clear effect of GVS was identified.

Overall, our results confirm previous findings indicating that GVS increases the excitability at the neural elements involved in the H-reflex (Lowrey and Bent, 2009; Okada et al., 2018; Nakamura et al., 2021); however, this increase in excitability did not affect the P-AD in healthy individuals, suggesting preservation of inhibitory spinal mechanisms. In this context, it has been described that a reduction or loss of the P-AD in pathological conditions occurs, such as in multiple sclerosis (Cantrell et al., 2022), spinal cord injury (Nielsen et al., 1995; Knikou and Murray, 2019), and recently, in diabetic neuropathy (Lee-Kubli and Calcutt, 2014; Marshall et al., 2017; Worthington et al., 2021) and prediabetic, overweight and obese subjects (Salinas et al., 2022).

It is noticeable the high variability of PA and AUC in P-AD, so its origin must be discussed to understand our findings. Regarding GVS, variability has been attributed to experimental characteristics such as current intensity and polarity (Nakamura et al., 2015; Sprenger et al., 2020), electrode montage (bilateral or unilateral) (Pliego et al., 2021; Tanaka et al., 2021), body posture (Iles et al., 2007; Hannan et al., 2021), and neck position (Osler and Reynolds, 2012). The current intensity is associated with the modified level of excitability, with the most notable amplitude increase observed in H-reflex amplitude using a conditioning GVS pulse between 2–3 mA (Okada et al., 2018). Our results show an amplitude increase of the H-reflex using a stimulation intensity in the range of 2.0 ± 0.6 mA, so we may assume that the changes in excitability are equivalent to those previously reported with larger current intensities (Lowrey and Bent, 2009; Okada et al., 2018; Tanaka et al., 2021).

The origin of H-reflex variability must also be considered. First, the individual cutaneous, motor, and movement perception thresholds using GVS activate separate nervous pathways, thus, provoking dissimilar effects on excitability (Wuehr et al., 2018; Mikhail et al., 2021). Another important factor about H-reflex is that variability increases at the intensity where both M and H waves are present, stressing the need for PA and AUC normalization (Brinkworth et al., 2007). Additional variability elements are diverse, going from body posture, plantar flexion, electrode position, and current intensity, to the excitability state of the motor pool at a particular time of the muscle contraction (Funase and Miles, 1999). Variability is present depending on the recorded muscles (vastus medialis, lateralis, and rectus femoris), neck position, and eyes closed or open (Kameyama et al., 1989). Although our subjects were indicated to lie prone, facing forward, with eyes open, no restraining device was provided to prevent head movement, allowing eyes opening or closing without supervision. We do not discard other variability sources such as differences in anatomical features related to electrode montage or individual body composition (Bolfe et al., 2007). Our results show that H-reflex PA decreased in 4 subjects (2 females, 2 males) with GVS (Ipsi, Contra) (Figure 6). Nevertheless, after analyzing the participants’ records, we could not identify a possible clinical or anthropometrical parameter causing this result, for example, Body Mass Index (BMI) was in the normal range (18.5–25 kg/m²). However, a more profound analysis of the impact of metabolic parameters or preclinical conditions on motor function could determine excitability alterations of H-reflex using GVS.

It is presumed that, while the cathodal electrode has a depolarizing effect on the underlying structures, the anodal electrode is hyperpolarizing. However, we found that both, cathodal and anodal stimulation provoked H-reflex facilitation. This increase in excitability was also observed in a previous study using short interstimulus intervals (Lowrey and Bent, 2009). In addition, anodal current of 8 mA showed no inhibitory effect on short-latency and long-latency electromyographic response of the soleus muscle (Bacsi and Colebatch, 2003). Our finding appears to be in accordance with what has been described extensively at the postural level, for we found a more pronounced increase in excitability with Contra stimulation, this is, with the cathode contralateral to the H-reflex stimulation site, meaning anodal GVS current. Similarly, during walking or standing, the subject veers toward the anode (left or right) with GVS, instead of moving forward, even while following a visual target (Fitzpatrick et al., 1999; Bent et al., 2004; Reimann et al., 2017). To our knowledge, only one publication showed that GVS polarity did not provoke the opposite effect on a motor task, with subjects lying prone, legs extended, and moving a force platform with plantar flexion. Both anodal and cathodal stimulation increased motor unit activity threshold, recorded with needle electrodes, from 8.7 N to ≈ 10.5 N with GVS (Kennedy et al., 2004). The disparity in the observed responses in posture and motor tasks, compared to H-reflex, points to a separate effect of GVS provoked over premotor neurons and the neuromuscular unit. Under our experimental conditions, we assume that the GVS is activating descending vestibulospinal pathways such as the LVST and reticulospinal tract that affect interneuron circuits mediating motoneuron responses (Kennedy et al., 2004).

Altogether, our study adds evidence to continue exploring the use of GVS as a valuable means to evaluate the integrity of the vestibulospinal pathway, for it has become a useful tool in basic, clinical and rehabilitation research (Tax et al., 2013; Welgampola et al., 2013; Čobeljić et al., 2016; Schniepp et al., 2018). In this context, it would be of great interest to study P-AD in combination with GVS in pathological populations that have previously been shown to respond separately to control subjects using GVS alone, as seen in asymptomatic patients with Human T-lymphotropic virus 1 and patients with the associated myelopathy spastic paraparesis (Matos Cunha et al., 2013). Furthermore, GVS in combination with other stimulation techniques, such as TMS (Matsugi et al., 2020) and transcutaneous spinal cord stimulation (Sayenko et al., 2018) has shown to become a promising intervention to understand the modulation of supraspinal structures on the motor output and to evaluate residual function, particularly after spinal cord injury. In summary, we posit that the use of P-AD and GVS can be applied to evaluate vestibulospinal pathways, in pathological and non-pathological conditions.

Galvanic Vestibular Stimulation did not produce a significant change in the P-AD as compared to Control in healthy subjects, i.e., for this spinal inhibitory process was preserved in all tested conditions (Ipsi, Contra, P Ipsi, and P Contra). Interestingly, an overall increase of excitability of the H-reflex was observed, being statistically different between Control and Contra. However, this increase was observed in 14/19 subjects when responses were averaged across subjects. GVS concedes to study the influence of the vestibular pathway on the spinal circuitry and evaluate associated phenomenon such as P-AD and H-reflex excitability. Our research adds to the available methods that study the integrity of the vestibulospinal tract and its mechanism, putting forward to extend the test to pathological population with altered neural excitability.

The datasets presented in this article are not readily available because they are not in a online database, but are available from the corresponding author on reasonable request. Requests to access the datasets should be directed to CC, carlos.cuellarra@anahuac.mx.

The studies involving humans were approved by the School of Medicine Ethics in Research Committee of the Autonomous University of the State of Mexico (CONBIOETICA-15-CEI-002-20210531, No. CEI/001/2022). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

MA-N, AP-C, and CC organized and performed the experiments and analyzed and interpreted the data. MA-N built the experimental data base. AP-C and CC wrote the manuscript. AP-C, CC, and CL-R contributed to the design and discussion of the study. All authors reviewed and approved the submitted version.