Effect of Individualized Whole-Body Vibration Exercise on Locomotion and Postural Control in a Person with Multiple Sclerosis: A 5-Year Case Report

Abstract

The present study aims to investigate the multi-year effects (5 years) of individualized whole-body vibration (WBV) on locomotion, postural control, and handgrip strength in a 68-year-old man with relapse remitting multiple sclerosis (PwRRMS). The dose–response relationship induced by a single session was quantified by determining the surface electromyographic activity (sEMG) of the participant. The participant wore an orthosis to limit the lack of foot dorsiflexion in the weakest limb during walking in daily life. The gait alteration during walking was assessed at 1, 2 and 3 km/h (without the orthosis) through angle–angle diagrams by quantifying the area, perimeter and shape of the loops, and the sEMG of leg muscles was recorded in both limbs. The evaluation of postural control was conducted during upright standing by quantifying the displacement of the center of pressure (CoP). The handgrip strength was assessed by measuring the force–time profile synchronized with the sEMG activity of upper arm muscles. The participant improved his ability to walk at higher speeds (2–3 km/h) without the orthosis. There were greater improvements in the area and perimeter of angle–angle diagrams for the weakest limb (Δ = 36–51%). The sEMG activity of the shank muscles increased at all speeds, particularly in the tibialis anterior of weakest limbs (Δ = 10–68%). The CoP displacement during upright standing decreased (Δ = 40–60%), whereas the handgrip strength increased (Δ = 32% average). Over the 5-year period of intervention, the individualized WBV improved locomotion, postural control and handgrip strength without side effects. Future studies should consider the possibility of implementing an individualized WBV in PwRRMS.

1. Introduction

Multiple sclerosis (MS) is an inflammatory autoimmune demyelinating disease of the central nervous system [1] affecting 2.8 million people worldwide [2]. Inflammation is the main cause of brain and spinal cord damage, which causes the destruction of myelin and axons [3]. The etiology of MS is largely unknown; however, it is assumed to be triggered by environmental factors in individuals with complex genetic risk profiles [2,3]. Nevertheless, it is well known that the disruption of the electrochemical coupling of neurons, due to nerve damage, leads to dysfunction of the sensory and motor systems, appreciable through several symptoms [2]. The most common symptoms of physical disability may include fatigue susceptibility, motor weakness, sensory loss and/or ataxia [1]. This multitude of neurological symptoms affects basic motor tasks (i.e., gait and postural balance), thus impairing the quality of life of people with MS (PwMS) [4].

Motor impairments are the main causes of disability in PwMS; overall, 85% of patients report mobility problems [5,6], and more than 50% of them require assistive devices within 15 years of disease onset [1]. PwMS motor impairment is typically associated with a reduced muscle strength [7] due to the following disease-related morphological and nervous factors: the cross-sectional area (type I, II, and IIa fibers are smaller); the tendency of the leg muscles to a higher fat percentage and a lower lean mass [7]; altered neural motor drive [8], causing a deficit of motor unit recruitments [9]; and factors related to postural control [10].

Gait alteration is perceived by PwMS as the most invalidating impairment across the spectrum of physical deficits [11]. The current literature reports that PwMS exhibit slower walking, take shorter steps, spend a greater percentage of the gait cycle in ‘double support’, and exhibit high joint kinematic alteration in the knee, and ankle joints [5,6,10,11,12,13]. The kinematic parameters appear to be predictive of the level of gait impairment in PwMS [10] and could be explained by altered patterns of leg muscle activation during walking [5,13]. The functional meaning of the EMG abnormalities, of the muscles involved in ankle control, is unclear but highlights impaired motor control in PwMS [5].

Postural control impairment in PwMS is relevant and well-known; however, there are not conclusive results that enable us to clarify the neuropathological mechanisms underlying balance disorders. Considering the damage variability of the central nervous system in PwMS, it is hypothesized that the impairment of postural control has different causes that vary from one person to another [14]. It has been suggested that the postural balance deficit is due to a slowing conduction of proprioceptive afferent along demyelinated tracts of the spinal cord, resulting in an altered central integration input (visual, vestibular, and somatosensory) [15]. Other studies identify the cause as focal or diffuse damage to the cerebellum or cerebellar peduncles and associative regions [16].

In the literature, there is clear evidence of the effectiveness of exercise to counteract the motor impairments in PwMS. Different types of exercise (i.e., moderate resistance or endurance training, aerobic activity, hydro kinesis) have been proposed to reduce the motor symptomatology of MS disease [17,18,19,20,21]. Following these kinesiological interventions, it is possible to appreciate neuroprotective adaptations (i.e.,: increase in the thickness of some area of the cortex) [22], probably linked to an increase in the release of neurotrophic factors [20,23,24,25] and inhibition of demyelination [26].

In the past two decades, whole-body vibration (WBV) has proven to be an alternative exercise intervention in neurological disease [27,28] and it has been suggested as a valid non-invasive form of exercise [29]. WBV generates a neuromuscular response that appears to mediate motor unit recruitment [30], involving not only reflex mechanisms but also cortical processes [31]. These neural mechanisms could explain the benefits of WBV intervention on muscle strength, gait, and balance [29,32,33,34]. The central nervous system’s capacity to recruit motor units is closely correlated with functional gains. The nervous system can generate and modulate a greater amount of force when a greater number of motor units are activated, which aids in the better control of fine movements [35,36]. This results in more precise and rapid postural adjustments and more well-regulated gait patterns [37]. The ability to perform functional daily activities through purposeful and coordinated movements is contingent upon the possession of adequate muscle strength (or force).

Furthermore, some studies have found changes in the hormone profile following WBV training, thus showing increased secretion of brain-derived neurotrophic factor (BDNF) [38] testosterone and growth hormone, and decreased cortisol levels [32,39,40,41].

Several studies have investigated the effects of WBV in PwMS [42,43,44,45,46,47,48,49]. Claerbout et al. [43] reported that a 3 weeks of WBV training improved muscle strength but not the functional mobility (i.e., 3 min walk test) of people with MS. On the other hand, Broekmans et al. [42] showed that 20 weeks of WBV did not positively affect the muscular performance of the lower limbs or the functional ability of the lower limb. Mason et al. [46] pointed out that the WBV training improves standing balance and walking time without negative effects. Other recent studies suggest that WBV has a beneficial impact on the functional and physical indices of PwMS [50] and improves gait parameters [51]. Thus, WBV exercise has been indicated in PwMS that report prevalent symptoms of motor impairment [52]. However, the results of these studies do not show conclusive evidence regarding the efficacy of WBV [42,43,46,47,48,49].

WBV has been applied without considering the individual response of the PwMS to the magnitude of the vibration load; in other words, a dose–response relationship has not been determined [39,53,54,55,56] and the WBV intervention did not follow the principle of load progression over time [47,48]. In many cases, WBV has been administrated by using a fixed value of magnitude for the entire duration of the intervention (i.e., frequency and/or amplitude) for all the people involved. In addition, no studies have investigated a multi-year WBV intervention in people with relapse remitting (RR) MS. Therefore, the present study aims to apply a five-year individualized WBV intervention, in which a dose–response relationship was determined, to a 68-year-old man with RRMS. The effectiveness of the individualized stimulus was quantified by determining the acute and residual biochemical profile of specific markers (growth hormone, testosterone, cortisol and creatine kinase).

Additionally, the handgrip strength assessment was included in the current study because it can be regarded as a comprehensive measure of physical impairment [57], even though the WBV stimuli are not directly applied to the upper limbs. This is due to its association with neuromuscular integrity and overall brain health [58]. Adjustments in grip strength may be linked to a decrease in independence in activities of daily living (e.g., drinking, eating, and buttoning a shirt) [59] due to the impairment in the upper limb function observed in PwMS [58]. Low grip strength may serve as an indicator of neuromuscular impairment on a global scale in individuals with multiple sclerosis [60].

The 5-year follow-up was monitored by assessing locomotion, handgrip strength and postural control synchronized with the EMG activity. We hypothesized that a multi-year strategy of individualized WBV improves gait, balance, and handgrip strength in a person with RRMS.

2. Materials and Methods

2.1. Participant

The study reports the case of a 68-year-old man with MS (body mass 70 kg, height 1.8 m, and BMI: 21.6 kg/m²). The participant had an RR course of the disease, the prognosis of which was considered benign. The first episode of the disease occurred when he was 27 years old and a second episode ocurred when he was 42 years old. Following the relapses, there was hypoesthesia and hypotrophy, more marked on the left side (weakest side), which caused impairment of postural control and walking. The weakest limb of the participant was the nondominant limb. The participant showed a score of 4.5 on the Expanded Disability Status Scale (EDSS) (the score was measured by the neurologist). The participant showed an inability to perform ankle-dorsiflexion with left ankle, resulting in a ‘footdrop’; for this reason, after the second relapse, he needed to use a Codivilla spring (dynamic walk ankle–foot orthosis—Orthogea Officine Ortopediche Group, Italy), that limits the ‘dragging’ of the forefoot on the ground during gait. The participant leads an active lifestyle, and despite limitations, performed long walks (3–4 km) 2 or 3 times per week. In addition, he has not undergone any type of medical or pharmacological treatment in the last five years. The participant voluntarily took part in this research project; he was aware of the purpose of this study and provided written informed consent.

2.2. Study Design and Procedures

The multi-year effects of individualized WBV were investigated in a single-person case report. The case report design was selected in accordance with logistical and methodological considerations. The participant consented to participate in a multi-year, individualized WBV training program and agreed to undergo routine motor assessments. Over the course of five years, this high level of adherence enabled the collection of structured and continuous data. Although the participant conditions (benign RRMS, active lifestyle, and use of foot orthoses) are not entirely representative of the broader RRMS population, the consistent data collected, which encompass a broad spectrum of clinical outcomes, offer valuable insight into the longitudinal effects of individualized WBV.

The participant underwent the WBV sessions at the hospital demyelinating disease center. The WBV intervention had a duration of about five years, during which 5 follow-ups were performed (T0–T4) (Figure 1). On the first lab visit, physical and clinical information were collected. The participant performed baseline assessments (T0) of gait and postural control; he was instructed to assume the correct positions during the WBV exercise, and finally the optimal vibration load was determined to individualize the appropriate exercise dose of WBV. The handgrip strength assessment was performed at T1 (baseline). The WBV intervention started 5 days after the initial lab visit (T0). At T0, to assess the effect of the single individualized WBV session, a venous blood sampling was executed to determine acute and residual hormonal (growth hormone, testosterone and cortisol) and biochemical (creatine kinase) responses to WBV. Gait was assessed by quantitative parameter (area, perimeter and shape) of the angle–angle diagram while walking on a treadmill at a standardized speed (1, 2, and 3 km/h). The first 2 assessments of gait (T0, T1) were performed exclusively at 1 km/h, as the participant was unable to walk at higher speeds. Gait assessment was synchronized with the surface electromyography (sEMG) activity. Balance was assessed during static and dynamic conditions, with eyes closed and eyes open on the force plate. The handgrip strength was also assessed as a functional task representative of the capacity to fulfill activities of daily living [54,56] with synchronized sEMG. To reduce the variability in measures between individuals and within individuals during testing, assessments were consistently conducted by the same expert evaluators. The results of the MRI, carried out by the participant as part of routine clinical examinations (T0 and T4), were also reported. The gold-standard assessment in MS management is magnetic resonance imaging (MRI), which can monitor the changes in the central nervous system (CNS) associated with the disease and the response to treatments over time [61].

A total of 182 WBV sessions were performed (42 before interruption, 140 after interruption). The protocol was interrupted for about 1 year due to the pandemic emergence of COVID-19. Considering the participant’s daily and work routines, the WBV intervention and assessment sessions were usually performed between 9.00 and 12.00 AM. The weekly frequency of WBV sessions did not always have the same constancy over the time, as the participant sometimes had health-related (i.e., seasonal coolness) or work-related constraints. This case report was carried out in accordance with the ethical standards of the Helsinki Declaration and has been approved by the Internal Review Board of the University (Prot. No. 13/2021).

2.3. Individualized WBV Load

The participant assumed an upright isometric high squat position with the heels raised, and the trunk tilted forward (elbow angle = 45°, hip angle = 130°, knee angle = 120° and ankle angle = 130°) on the vibrating platform (Figure 2A) [54]. Seven trials were performed in random order in the following conditions: no vibrations or 0.00, 0.1, 0.29, 1.10, 1.92, 5.70, and 8.43 g (expressed as a multiple of gravity, where 1 g is equal to 9.81 m∙s⁻²). There was a 2 min pause between trials, and each trial lasted 30 s. To maintain consistent positioning of the feet across the trials and sessions over time, the appropriate positions were marked on the platform. The sEMG for the vastus lateralis (VL), biceps femoris (BF), lateral gastrocnemius (LG), and tibialis anterior (TA) was recorded during each trial. The acceleration load (magnitude) that corresponded to the highest sEMG response (“Optimal”) [53,54,55,56] was applied during the WBV sessions. The participant obtained the highest muscle response at an acceleration of 1.10 g in both limbs (Figure 2B).

2.4. Whole-Body Vibration Sessions

The participant underwent individualized WBV exercise two times per week at the frequency of 28.9 Hz (corresponding to 1.10 g) through a vibrating platform (Bosco System, Nemes Lsb, Rieti, Italy). During the WBV, the workload was gradually increased according to the principle of load progression. The participant’s workload may be increased in the subsequent session if they successfully completed the prescribed workload for two or more repetitions [62] (

Table 1. Workload progression during WBV intervention.

Variable	1st Workload	2nd Workload	3rd Workload	4th Workload	5th Workload	6th Workload	7th Workload	8th Workload
Frequencies (Hz)	28.9	28.9	28.9	28.9	28.9	28.9	28.9	28.9
Displacement (mm)	0.6	0.6	0.6	0.6	0.6	0.6	0.6	0.6
WBV series × reps	1 × 7	1 × 8	2 × 5	2 × 5	2 × 5	2 × 5	2 × 5	2 × 5
WBV duration (s)	30	30	30	60	60	60	30	60
Rest time from reps. (s)	120	60	30	60	60	60	30	60
Rest time from series (s)	/	/	240	180	240	120	160	120
Weekly frequency	2	2	2	3	2	3	3	3
Position assumed	High squat	High squat	High squat	High squat	High/Half alternation	High/Half alternation	Half squat	Half squat *
Total session	4	4	32	40	20	14	12	16

Abbreviation: whole-body vibration (WBV); repetitions (reps.). * Shifting the body weight from weakest-Leg (30 s) to wtrongest-Leg (30 s).

). During the individualized WBV, the participant assumed an isometric squat position (high or half) with the heels raised [54] (Figure 2A). During the WBV exposure, the participant wore only socks to eliminate any vibration-dampening caused by the footwear.

2.5. Surface Electromyography (sEMG)

sEMG was recorded using bipolar electrodes (Ambu Neuroline 720 00-S/25, electrode diameter 45 × 22, interelectrode distance 2.5 cm, Ambu A/S, Baltorpbakken, Denmark) in 4 muscles: VL, BF, TA, and LG. The electrodes were placed on both legs, according to the international surface EMG for non-invasive assessment of muscles (SENIAM) recommendations [63]. To minimize the impedance (<5 kΩ) before the electrodes’ application, the skin was shaved, gently abraded with sandpaper (P320) without producing redness, and cleaned with alcohol. The exact location on the muscle belly was marked to ensure correct sensor positioning across the inter-day EMG recordings. In addition, the electrodes and the modules were fixed with elastic strips (Flexa Elast, Pic Solution, Pikdare S.p.A.) to prevent motion artifacts. The wireless data synchronization unit (MuscleLab 6000, Ergotest-innovation, Porsgrunn, Norway) characteristics were as follows: a built-in radio frequency (RF) module ML6RFM02 with 2.4 GHz and 1 mW and a typical wireless range in an open space of 20 m. The modules of the MuscleLab characteristics were as follows: RF characteristics of 2.4 GHz, 1 mW; sample rate of 1.2 kHz; high-pass filter of 20 Hz; bandwidth of 20–500 Hz; input signal range of ± 5 mVp-p; noise of 14 µVp-p (2.2 µV_RMS); and resolution of 0.33 µV/bit. Utilizing a preamplifier that was situated near the pickup point and directly connected to the electrodes, the raw sEMG signal was amplified and filtered at 1.2 kHz. The frequencies of 20 Hz and 500 Hz were used to band-pass filter the signals using a fourth-order Butterworth filter. The preamplifier’s common mode rejection ratio was 100 dB. In accordance with the standards for reporting sEMG data (International Society of Electrophysiology and Kinesiology, https://isek.org/resources/, accessed on 24 July 2025), the EMG signal was converted to a root-mean-square (RMS) signal using a hardware circuit network (frequency response 20–500 kHz, integrating moving average filter with 100 ms width, total error ± 0.5%).

2.6. Gait Analysis and sEMG Activity

The kinematic analysis of gait was performed using the SMART Motion Capture System (BTS, by Bioengineering, Milano, Italy) equipped with 4-optoelectronic cameras able to perform a 3D reconstruction of the trajectories, provided by reflective markers placed on the body according to a previous protocol [13,64,65]. Specifically, the markers were placed on the anterior superior iliac spines, greater trochanters, lateral femoral condyles, calcaneus, lateral malleoli, and fifth metatarsal. The cameras were calibrated in static and dynamic modality by determining the shooting volume around the treadmill; data were collected with a sampling rate of 60 Hz. The participant walked on a treadmill for 60 s at each selected speed (1, 2, and 3 km/h). The participant walked without plantar orthosis and the gait pattern was determined by hip–knee and knee–ankle angle cyclograms [13,64,66,67]. The data of three angles, plotted by the sagittal motion plane over time, were considered in the analysis of the lower limbs using the SMART Analyser (BTS, Bioengineering). The absolute hip angle is formed by the thigh axis and the vertical hip passing through the pelvis, the relative knee angle is formed by the extension of the thigh axis and the shank axis, while the relative ankle angle is formed by the shank axis and axis through the foot. The angle–angle diagrams were drawn by interpolating five points representative of the gait cycle: toe off (TO), maximum knee flexion (MKF), maximum hip flexion (MHF), foot support (FS), and maximum flexion of the knee during the contact phase (MKF-C) [66]. For TO and FS, the frames considered corresponded to those in which the value of the X vector of velocity changed from negative to positive and vice versa [68]. AutoCAD 2023 (Autodesk, San Rafael, CA, USA) enabled us to calculate the area and the perimeters of the angle–angle diagrams, as reported in previous studies [13,64,65]. The shape of the loops formed by the angle–angle diagrams was calculated by dividing the perimeter by the square root of the area ($Perimeter/\sqrt{Area}$) [69]. The kinematic data were synchronized with the wireless sEMG system (MuscleLab 6000, Ergotest-innovation, Porsgrunn, Norway), recording the muscle activity of the VL, BF, TA, and LG. The sEMG_RMS considered a full wave (Figure 3) of stance phase in two consecutive steps (extrapolated in the central part of the recording) for the weakest and strongest leg. The peak sEMG_RMS value of the respective muscle during the stance phase of walking was used to normalize the sEMG_RMS signals of the muscles during walking [70].

2.7. Postural Control

Postural control was assessed by standing upright on a force platform (Muscle-Lab 4000e, Ergotest Technology, Langesund, Norway), facing 1.5 m away from the wall of a room. All walls, around the participant, were draped white, and a 1 cm red square was placed on the opposite wall at eye level. The feet were placed such that the medial malleoli were approximately 1.5 cm apart. The appropriate toe and heel positions were marked on the force plate to ensure a consistent position among the trials and over follow-up. During the tests, the arms were extended at the sides, and the knees were fully extended. The participant performed four trials in random order with a two-minute rest between each trial, and each trial lasted 30 s. The four trials were performed in static and dynamic conditions [45] with their eyes open (gazing at the red square) and with their eyes closed. The dynamic condition was represented by an external stimulus generated by a pendulum system. The pendulum, placed behind the subject, impacted the dorsal part in correspondence with the area between the inferior angle of the two scapulae, generating a slight perturbation. The stimulus was proposed three times every 7 s (7 s, 14 s, and 21 s) within the entire trial duration of 30 s [64]. Body sway (sampled at 100 Hz) was quantified by the center of pressure (CoP) displacements. For the analysis, the total length sway path and the right–left (R-L) and forward–backward (F-B) components of CoP were considered [45,71]. The total path length was calculated using the following equation:

$$Sway path lenght= \sum _{n=1}^{N=1}\sqrt{{\left({FB}_{n+1}-{FB}_n\right)}^2+{\left({RL}_{n+1}-{RL}_n\right)}^2}$$

(1)

where N is the total number of data points in the CoP displacement time series and n is a data point of interest [72]. Analysis of the data was performed using Muscle-Lab 4000e software (Bosco-System, Ergotest Innovation, Stathelle, Norway). In the static conditions, the path displacements of the CoP (in millimeters) were analyzed during the entire trial (30 s). In the dynamic conditions, three windows, each of 1.5 s, were analyzed; each of them began when the ground reaction force changed due to the external stimulus. The values obtained in the three windows were averaged and considered for analysis of the horizontal components of the displacement (R-L and F-B).

2.8. Handgrip Strength Test (HGS) and Synchronized sEMG Activity

The HGS strength test was performed with a customized hand dynamometer equipped with a strain gauge (Muscle-Lab, Bosco-System 4000e, Ergotest Innovation, Stathelle, Norway). HGS was measured on the participant, who was seated with a straight back and the feet parallel and completely in contact with the floor. The shoulder was abducted, the elbow flexed to 90°, with the forearm placed on the table, and the wrist was positioned in the midrange of pronation and supination (neutral wrist posture). The dynamometer was adjusted to fit the participant hand and standardized. The base of the HGS was placed on a table. The position of the hands, body, and sensor was marked to standardize all tests over time. The participant was instructed to “squeeze” with maximal isometric contraction and maintain this for approximately 3–4 s. Grip force should be applied smoothly, without rapid jerking motion [73]. Three trials were performed and the best trial of the three was selected. The muscle activity of Deltoid Anterior (DA), Biceps Brachii (BB), and Palmaris Longus (PL) was measured during the HGS, using an integrated sEMG system. The electrodes were placed in the mentioned muscles following guidelines for the non-invasive assessment of sEMG [63], and the skin was prepared, to minimize the impedance (<5 kΩ). The electrodes and EMG cables were secured with elastic strips (Flexa Elast, Pic Solution, Pikdare S.p.A.) to prevent motion artifacts. EMG activity was recorded by using the triode electrode (T3402 M, nickel-plated brass, electrode diameter = 1 cm, interelectrode distance = 2 cm; Thought Technology Ltd., Montreal, QC, Canada). The sEMG detection technique entailed full-wave true RMS conversion of the signal from the preamplifier. The size of the averaging window was 100 ms, with a resolution of a 16-bit A/D converter. The EMG preamplifier characteristics were as follows: voltage supply, ±5 VDC; input impedance, 2 GΩ; common mode rejection rate, 100 dB; gain at 100 Hz, 500; 3 dB low-cut frequency, 8 Hz; and 3 dB high-cut frequency, (Muscle-Lab 4000e, Bosco-System, Ergotest Innovation, Stathelle, Norway). During the HGS, the force–time relationship was recorded through the Muscle-Lab software. The trial with the highest values in each hand were used for analysis. To analyze the sEMG_RMS (mV) of the upper limb’s muscles, a window of 400 ms around the peak of the maximal voluntary isometric contraction (200 ms before and 200 ms after) was opened. The rate of force development (RFD) in the initial portion of the force–time relationship (the first 0.200 ms) was also determined.

The HGS differences between the hands were calculated as the ratio of the maximal strongest and weakest HGS (Strongest HGS/Weakest HGS). The HGS asymmetries could be considered in the following way: symmetric (less than 15% difference, <1.15), mild asymmetry (15–25% difference, or 1.15–1.25), and moderate asymmetry (more than 25% difference, >1.25) [57].

2.9. Serum Collection

On the first WBV session (5 day after T0), the participant was acclimated in the lab for 30 min before baseline blood collection. Blood samples were drawn into vacutainers from the antecubital forearm vein using a 20-gauge needle (without additives) for measurement of creatine kinase (CK), total serum testosterone, growth hormone and cortisol concentration. After the baseline measurement, a WBV exercise session was performed. Measurements were repeated immediately after intervention (Post), 1 and 2 h after the end of the intervention, and for CK was also repeated 12 and 24 h after the end of the intervention. Blood was collected into serum-separator tubes (Greiner Bio-One), allowed to clot for 30 min at room temperature, and then centrifuged at 2000× g for 10 min. Sera were then harvested in aseptic conditions and frozen at −80 °C for storage. Serum samples were allowed to thaw in ice and then mixed thoroughly before assaying. Enzyme-linked immunosorbent assays (ELISAs) for testosterone (cat#ab108666; Abcam, Cambridge, UK), cortisol (cat#ab108665; Abcam) and growth hormone (GH, cat#ab190811; Abcam), and the Reflotron assay for creatine kinase (cat#11126695; Roche, Basel, Switzerland) were carried out following the manufacturer’s instructions. The sensitivities of these assays were 0.07 ng/mL, 2.44 ng/mL, 1.4 pg/mL, 0.7 pg/mL, and 24.4 U/L, respectively. The intra- and inter-assay variability coefficients for the ELISAs were 5.8% and 10.5% for testosterone, 9% and 9.8% for cortisol, 3% and 1.6% for GH. To calculate sample concentrations in ELISA, the optical density (OD) of the standards was zero-corrected for blank value and then regressed against their concentration using a 4-parameter logistic (4-PL) model. Once the standard curve had been obtained, the unknown sample concentration was calculated by substituting their zero-corrected OD value in the standard curve equation. For Reflotron assays, the concentration was automatically determined by the Reflotron Plus device.

2.10. Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) of the brain was carried out acquired using a clinical 1.5T scanner (Siemens Healthcare, Erlangen, Germany; software version syngo MR XA20; XJ gradients with 33 mT/m at 125 T/m/s) equipped with a dedicated 32-channel coil, as well as the Turbo Suites Essential (iPAT, tPAT, PAT2, CAIPIRINHA VIBE, CAIPIRINHA SPACE) and Exclerate (SMS EPI for DWI/DTI/BOLD, SMS TSE, CS TOF, CS SPACE). Simultaneous multi-slice imaging for DWI and TSE sequences and Compressed Sensing for MRCP and 3D time-of-flight sequences were integrated into clinical protocols. Furthermore, automated coil selection and patient positioning were applied.

2.11. Subjective Comments

Weekly, it was asked to the participant to report any effects on his symptoms to highlight potential subjective changes that may have been undetected by the chosen outcome measurements.

2.12. Statistics and Data Analysis

To evaluate the long-term effects of individualized WBV, a linear regression model was constructed using the Statistics and Machine Learning Toolbox and executed in the MATLAB software interface (version R2024b; MathWorks, Natick, MA, USA). During the follow-ups, this model examined the trend of the dependent variables. The significance level (α) was set to 0.05.

The differences between baseline measurement (T0) and the different follow-up (T1, T2, T3, T4) were compared for gait, postural control, handgrip strength and synchronized sEMG activity. The following formula was used:

$$∆\mathrm{\%}=\left[{\displaystyle \frac{\mathrm{f}\mathrm{o}\mathrm{l}\mathrm{l}\mathrm{o}\mathrm{w} \mathrm{u}\mathrm{p}-\mathrm{T}0}{\mathrm{T}0}}\right]\times 100$$

(2)

Symmetry Index (SI) of sEMG during gait was calculated as an interlimb difference measure. SI for sEMG activity of stance phase was calculated through the following equation:

$$\mathrm{S}\mathrm{I}\mathrm{\%}={\displaystyle \frac{(StS-StW)}{0.5\times (StS+StW)}}\times 100$$

(3)

where SI = Symmetry Index; StS = value of sEMG during stance phase for the strongest side; and StW = value of sEMG during stance phase for the weakest side. SI corresponds to the percentage of asymmetry observed for one side in relation to the other [74], an SI = 0 indicates the existence of perfect symmetry, while positive and negative values refer to the asymmetry for the strongest (S) and weakest (W) sides, respectively.

The same equation was used to compare the parameters (P) of angle–angle diagrams (area, perimeter and shape) between the W-Leg and S-Leg during gait:

$$\mathrm{S}\mathrm{I}\mathrm{\%}={\displaystyle \frac{(PS-PW)}{0.5\times (PS+PW)}}\times 100$$

(4)

where PS = value of area, perimeter or shape of the strongest side; and PW = value of area, perimeter or shape for the weakest side.

3. Results

3.1. Gait Analysis and sEMG

The hip–knee diagrams showed an increase in the overlap of loops between weakest leg and strongest leg (from T0 to T4) at all speeds (Figure 4). Table S1 summarizes the values of angle–angle diagrams parameters (area, perimeter and shape).

The hip–knee and knee–ankle diagrams exhibited significant changes over time only in the strongest leg, perimeter, and shape at a speed of 1 km/h (p < 0.05). There were no significant changes in any of the other variables (p > 0.05). The perimeter exhibited a higher Δ in the weakest limb at T4 (hip–knee = 17% and knee–ankle = 36%).

At T2, the participant became able to walk on the treadmill at 2 km/h and 3 km/h, unfeasible speeds for the first two test times (T0 and T1). There were no significant changes in the parameters of the angle–angle diagrams in both limbs at 2 and 3 km/h over time (p > 0.05).

Table S2 summarizes the values of normalized sEMG during the stance phase of gait, the differences over the time and SI for each muscle. The normalized sEMG activity did not exhibit any significant changes over time at any of the velocities (p > 0.05). In contrast, the weakest limb muscles at T4 exhibited an increase in Δ at all velocities for BF (Δ average = 22%), TA (Δ average = 40%), and LG (Δ average = 46%).

Table S3 reports the SI over time for the perimeter, area and shape of angle–angle diagrams between W-Leg and S-Leg.

3.2. Postural Control

The body sway values that were measured during postural control showed significant improvements over time in the static with open eyes and dynamic with closed and open eyes conditions (p < 0.05). In the static condition with closed eyes, there was no significant change over time in any component (total path, R-L, and F-B displacement) (p > 0.05) (Table S4). The dynamic condition with eyes closed exhibited the most significant reductions (Δ average = −60%), even though all trials exhibited an average Δ of −50% between T0 and T4.

3.3. Handgrip Strength Test and sEMG

Table S5 reports the handgrip strength values, and Δ differences between T1 and the other test times (T2, T3, T4). There were no significant changes in the mechanical variables (peak force and RFD) over time (p > 0.05). However, the RFD of the weakest limb appeared to exhibit a trend in the direction of significance (p = 0.079). The peak force symmetry ratio between the weakest and strongest hand demonstrated a substantial increase over time (p = 0.024), achieving the symmetry values at T4 (1.09). The BB muscle of the weakest limb exhibits a substantial increase in sEMG activity over time (p = 0.046). Other muscles did not change significantly over time (p > 0.05).

3.4. Hormonal and Creatine Kinase Acute and Residual Responses

The cortisol concentration decreased by 29% after a single WBV session (Pre–Post) (Figure 5A). The decrease was maintained after 1h and 2h, respectively, by 24% and 36%. On the other hand, testosterone showed an increase of 5% (pre-post) and 8% one hour after the end of the WBV session. After 2h, the concentration returned close to the basal value (Figure 5B). The testosterone–cortisol (T-C) ratio increased by 48%; the increase was maintained for the following 1 h and 2 h (42% and 57%, respectively) (Figure 5C). The GH response showed an increase of 72% over basal concentration, although the increase persisted with only 11% after 1 h, and at 2 h after the end, the concentration increased about 9-fold the basal level (827%) (Figure 5D). Creatine kinase concentration increased by 15% (pre–post) and the value remained constant (about 10%) until 48h after the WBV session (Figure 5E).

3.5. Magnetic Resonance Imaging

The brain MRI post WBV intervention (T4) was compared with the brain MRI pre-intervention (T0), with no changes in the clinical condition. Specifically, the multiple (>10) areas of altered signal affecting the bi-hemispheric white matter, with slight predominance in the periventricular region (Figure 6), appeared stable in terms of distribution, number, and size.

3.6. Subjective Comments

Subjective comments from the participant were favorable to WBV exercise, reporting excellent tolerance, which made it easy to integrate the sessions into a weekly routine. The main comments expressed weekly, albeit anecdotal, included the following: more energy during the day, better mood, improved sleep quality, reduced leg cramps at night, improved ability to perform long walking, and improvement in the weakest ankle’s dorsiflexion and the ability to better perceive the support surface. In addition, the participant asserted that despite the one year of interruption due to the COVID-19 pandemic, he did not perceive any decline in the improvements achieved with the WBV sessions prior to the interruption.

4. Discussion

Both the duration of sessions and the ease of stimulus administration were well received by the participant within the context of the individualized WBV protocol. Its potential for clinical applications or integration into rehabilitation programs may be facilitated by its adaptability. The participant reported an overall improvement in his ability to perform activities of daily living, such as climbing stairs, manual dexterity, removing the plantar orthosis, and feeling more stable during gait, which suggest that the improvement in the various outcomes has a positive impact on his functional independence.

The present case report investigated the effects of five-year individualized WBV intervention on locomotion, postural control and handgrip strength in PwRRMS.

The main results during walking include the ability to walk at higher speeds (2 km/h and 3 km/h) compared to the baseline assessment in which the participant was unable to walk at speeds greater than 1 km/h; and the ability to walk effectively without the plantar orthosis by improving his ankle dorsiflexion. The angle–angle diagrams revealed an improvement in the locomotion at all speeds of both legs (W-Leg and S-Leg).

The results of the angle–angle diagrams indicated specific adaptations of the walking modalities in relation to the speed. At low speed, the participant increased the range of motion (ROM) of the lower limbs over time. Considering that a decreased ROM characterizes the gait in PwMS [5,12], this adaptation is appropriate to counteract their gait alterations. Conversely, the ROM decreased over time at a higher speed, but the effect was counterbalanced by an increase in the step frequency, which was consistent with the increase in the imposed speeds on the treadmill. This adaptation over time could be also related to an improvement in conjoint coordination revealed by the shape of the angle–angle diagrams. Although the absolute value of the shape cannot be associated with a specific geometry, it has the potential to underline the shape changes described by an angle–angle diagram. Thus, when the coordination in a movement improves, the value quantifying the shape tends to decrease [69].

These kinematic adaptations are reflected in the normalized sEMG activity recorded in the leg muscles during the stance phase. Although the changes have not been significant over time, at the end of the WBV intervention (T4), the VL-normalized sEMG reduced its activity in the weakest leg (Δ average = −48.3%) more than in the strongest leg (Δ average = −28.6%), whereas the BF-normalized sEMG values had higher Δ in the strongest leg (Δ average = 81.3%) than in the weakest leg (Δ average = 22%). The shank muscles (TA and LG), involved in ankle flexion–extension, increased mostly in the weakest leg; in particular, the activation of the TA muscle increased in the weakest leg (Δ from 10 to 68%). The improvements in muscle activation could explain the reason why the participant began to walk without the orthosis as the TA muscle is directly involved in the ankle dorsiflexion function.

Although the participant was able to walk at higher speeds over time, the SI values seem to indicate that during walking at a speed of 3 km/h, motor impairments persisted. The preferable gait speed for the participant seems to be 2 km/h, as the overall SI of the area, perimeter and shape was <10%; additionally, the normalized sEMG activity showed a better SI than the other speeds’, with nearly perfect symmetry of the LG muscles. Improved locomotion facilitates daily activities, including the ascent of stairs and the traversing of uneven terrain, thereby enhancing independence and decreasing the necessity for assistance [75]. Moreover, the improvements encourage increased engagement in recreational and social activities [76], which, in turn, promotes psychological well-being and reduces social isolation. Generally, individuals with multiple sclerosis experience enhanced functional outcomes and a higher quality of life as a result of gait improvements.

In the literature, the WBV, as an alternative exercise to attenuate the gait impairments in MS, has shown conflicting results [42,44,46,47,48,49,51]. The heterogeneity of the results is partially due to the metrics used that are characterized by low sensitivity of the test procedure to assess the specific adaptations induced by WBV exercise on locomotion. These studies generally assessed the effect of WBV on PwMS through several functional tests (timed get-up-and-go, 6 or 3 min walking test, 10 m walk) [42,43,44,46,47,48]. None of these investigations monitored gait changes over time by using angle–angle diagram and recording the synchronized sEMG. This means that our results cannot be directly compared with those of other studies. Our method seems appropriate to provide specific information in PwMS [13], as quantification of area, perimeter and shape of the loops provides detailed information concerning the co-joint ROM (hip–knee or knee–ankle) and co-joint coordination.

In the present study, the ability to control the standing posture significantly improved over time in the different conditions (static, dynamic, open and closed eyes). Following 5 years of individualized WBV intervention, all CoP sway path parameters (total length and R/L and F/B components) showed a Δ reduction of between 40% and 60%, thus suggesting an improved postural control. In the literature, the studies that have investigated the chronic effect of WBV exercise on postural control show conflicting data; some of them confirm the positive influence on postural control following a few weeks of WBV sessions (8–10 weeks) [46,47,50], whereas others did not report significant improvements [43]. Recently, Krause et al. [45] reported that during the WBV intervention period, in PwMS with severe forms, the balance does not degenerate, and until six weeks after the end of WBV interventions, the balance degeneration does not seem to progress.

Most of the cited studies assessed balance through scored tests used in clinical practice, such as the Berg Balance Scale (BBS) [42,43,50] and the standing balance test [46]. The latter methods are not be able to capture mild impairments, and their poor reliability could limit the monitoring of the disease progression [77]. The current available technology to assess the postural control uses force plates by measuring the CoP displacement during a standing position [78]. Only one study has assessed the effect of WBV training by detecting CoP displacement on force plate [45], whose results suggest that during the WBV intervention in PwMS, an increase in CoP compared to the control was not observed. In our study, we measured the body sway by using a force plate under different conditions to assess the multisensory integration ability by removing the visual analyzer (eyes closed) and the proprioceptive integration response by applying an imbalance from behind. The improvements in our study suggest better function in activities of daily living in terms of improved coordination and stability of movements, as the adaptive motor behavior is based on adequate sensory integration or processing to organize an appropriate response [79]. Functional independence and enhanced stability are positively impacted by the enhancement in postural balance, which reduces the risk of falls and enhances safety during daily activities that necessitate an upright position [80]. The quality of life of individuals with MS is positively influenced by these benefits, resulting in a decreased requirement for assistance [80].

Grip strength showed some improvements over time. Peak force values, while they did not significantly change over time, showed a Δ of 41% in the weakest hand and 23% in the strongest hand. Although the absolute values of RFD were higher in the strongest hand than in the weakest hand, the weakest showed a greater increase over time tended to significance (at T4 Δ = 151%). These results seem consistent with the increased sEMG activity on the upper limb muscles. In addition, the participant significantly improved the symmetry ratio of the peak force between the weakest and the strongest hand that was moved to the range of symmetry (1–1.15) [57].

Grip strength is a widely used outcome that is used to assess neurocognitive and neuromotor degeneration; thus, it is an indicator of upper limb function or overall physical impairment [57,60]. Grip strength deficit is often associated with a loss of independence in activities of daily living, as appropriate grip strength value plays an important role in drinking, eating and caring [58]. However, recent research states that low grip strength values would be not associated with a reduced corticospinal excitability in people with MS with low disability [60]. Some studies that have investigated the chronic effect of WBV on functional parameters in an elderly population have not shown changes on grip strength [81,82,83]. However, it must be considered that the vibrations are minimally propagated to the upper limbs when the participants stand upright on the vibrating platform. In our study, the increase could be related to the longer duration of the WBV intervention in comparison to that of others. These results suggest that individualized WBV may also induce cortical adaptations over time [84].

Enhanced hand grip strength is a critical factor in the independent completion of daily tasks, including dressing, personal hygiene, and object grasping, as it facilitates the control of fine motor movements and increases manual dexterity, thereby reducing the necessity for assistance [85]. These benefits directly contribute to an enhanced quality of life and support the preservation of self-independence [59].

The subjective comments in parallel with the results of the analysis suggest that the participant’s overall condition improved after the WBV intervention. In addition, following the WBV intervention, the EDSS score improved; this was defined by a two-point reduction from baseline. However, the improvements at the different test times did not show a linear trend as the participant did not maintain the same constancy over the 5 years due to occasional work commitments or unexpected health events not related to the disease.

One noteworthy aspect concerns the results of the T2 follow-up after the interruption period (1 year). Although the improvements at T2 indicate a slight decline with respect to T1, the results were nevertheless better than baseline (T0). If we also consider participant’s subjective comments, albeit anecdotal, these results might suggest that the adaptation produced by the individualized WBV sessions before interruption was maintained over time. Although this represents an interesting result, it is not completely unexpected; in fact, as reported in the literature, the WBV stimulus produces a neuromuscular response due to a tonic vibration reflex (TVR) [30,86], similar to the reflex induced directly by applying vibration to the muscle–tendon unit [87]. The TVR seems to mediate the recruitment of reflexive and selective motor units (MUs) in which the recruitment threshold of low-threshold MUs increases after WBV, while the opposite effect is elicited on higher-threshold MUs [30]. Mileva et al. [31] have shown that acute WBV exposure increases the excitability of the corticospinal pathways related to the tibialis anterior muscle activity, highlighting that the effects are not limited to the periphery but also involve corticospinal and intra-cortical processes. Experiments carried out on the muscle–tendon unit have shown that muscle vibration is a strong proprioceptive stimulus that reaches both the primary somatosensory (SI) and motor (M1) cortices directly [88] and that the frequency exerts a dependent effect on the Ia afferent firing rate that is reflected in differential frequency-dependent effects on corticospinal motor neuron excitability [89]. These neural mechanisms explain the benefits of acute and chronic WBV intervention on muscle strength, power, gait, and balance [33,34,84,90]. Furthermore, these enhancements can be further demonstrated by the enhanced activation and synchronization of MUs [30] and the enhanced sensorimotor integration that results from the repeated stimulation of muscle spindles and Ia afferents [84,85]. Postural control and intermuscular coordination are likely to be more efficient during dynamic tasks as a result of this neuromuscular tune-in [31].

Furthermore, these enhancements can be further demonstrated by the enhanced activation and synchronization of MUs [30] and the enhanced sensorimotor integration that results from the repeated stimulation of muscle spindles and Ia afferents [84,85]. Postural control and intermuscular coordination are likely to be more efficient during dynamic tasks because of this neuromuscular tune-in [31].

The MRI results of the participant indicate that there were no reductions in CNS injury. This is consistent with a recent study by Kjølhede et al. [91], in which the injury of MS did not change, despite those subjects showing improvements in motor assessment. This is consistent with the specific nature of the treatment, which is not disease-modifying but supportive of recovery. Previous studies suggest that such improvements in individuals with MS following exercise may depend on increased serum levels of specific hormones [24]. Our results report an important increase in GH (827% two hours after the end of a single WBV session) and an increase in testosterone (of 8% after one hour). It has been suggested that in MS, hormones like GH are involved in the neuro-reparative process; in fact, GH exerts trophic effects on neuronal regeneration in peripheral and CNS, stimulating the protein synthesis in neurons, glia, oligodendrocytes, and Schwann cells. In addition, GH favors the survival of myelin and CNS cells, inhibiting apoptosis [92,93]. Furthermore, the effect of GH could be mediated by or be overlapped with that of other hormones, such as testosterone [92]. In this regard, Sicotte et al. [94] suggested that exogen testosterone has potential neuroprotective effects in men with RRMS because of its ability to increase the grey matter in the frontal cortex [92]. Furthermore, our results show a reduction in the cortisol level (36% after 2 h), which is consistent with other studies [39,95,96,97]. This result could have specific relevance because people with multiple sclerosis exhibit a failure to suppress cortisol release, correlated with high scores on depression and anxiety scales [98]. Interestingly, we also observed an increased testosterone–cortisol ratio post WBV session as the time under tension, in the isometric high squat position, was not brief (each repetition lasted 30–60 s × total of 7–10 reps) [39]. These hormonal changes may contribute to functional improvements by modulating neurotransmission and enhancing neuroplasticity. Elevated levels of GH and testosterone, with a decrease in cortisol, may help establish a favorable neuroendocrine environment that enables the repair of myelin and the restoration of motor function in individuals with MS [24,88,89].

Finally, the WBV effects that we found and the creatine kinase concentrations are in line with those induced by applying high-intensity eccentric exercise associated with prolonged muscle damage [39].

5. Limitations

Although the effects of long-term individualized WBV appear to mitigate motor impairments in people with RRMS, this study has some limitations. Firstly, the results have no relevance for statistics when considering a single participant, which limits the generalizability of the findings to the larger RRMS population. Additionally, the interruption caused by the pandemic may pose challenges to replicating the protocol. Nevertheless, it provided an opportunity to observe the preservation of some adaptations even after the interruption period. Finally, muscular fatigue represents an important parameter in MS symptomatology, and we believe that future investigations need to assess whether long-term individualized WBV could have a potential effect on muscle fatigue in this population.

Further research into a larger RRMS sample and balanced with control groups is necessary to support our conclusions. However, our findings should encourage future clinical trials to explore the effects of individualized WBV interventions.

6. Conclusions

This is the first study that has assessed the effect of five-year individualized WBV intervention on locomotion, postural control and handgrip strength in a person with RRMS. The kinesiological protocols of adapted physical exercise should consider the possibility of using the individualized WBV exercise as an alternative exercise strategy to counteract the motor alteration in individuals with RRMS, as it seems to suggest a potential neuroplasticity and neuromuscular adaptations.

From a statistical perspective, our results are irrelevant. Additional investigations will be necessary to verify our findings in RRMS if they are conducted with the same experimental design and acceptable statistical power.