Progressive Whole-Body Vertical Vibration with Conventional Rehabilitation for Upper Limb Function and Corticospinal Excitability in Subacute Stroke Patients

Summary

This randomized controlled trial aimed to examine the effects of progressive Whole-Body Vertical Vibration (WBVV) combined with conventional rehabilitation on upper limb motor function and corticospinal tract excitability in subacute stroke patients and to explore potential neurophysiological mechanisms.

Abstract

Stroke often results in upper limb impairment. Although many patients regain independent ambulation, upper limb function recovery remains incomplete, which limits activities of daily living (ADL) and reduces quality of life. Optimizing upper limb rehabilitation strategies continues to represent a major challenge in stroke care. This article presents a randomized controlled trial protocol investigating the effects of progressive WBVV training on upper limb function and corticospinal excitability in subacute stroke patients. Forty-eight subacute stroke patients are allocated to an experimental group (n = 24) and a control group (n = 24). The control group receives conventional rehabilitation training, while the experimental group receives additional progressive WBVV training. This intervention offers the advantages of easy implementation, low energy consumption, and favorable tolerability for subacute stroke patients. The WBVV protocol is gradually intensified over four weeks by adjusting frequency and amplitude. After four weeks of intervention, both groups showed significant improvements in upper limb function, ADL ability and neurophysiological indicators (p < 0.05). The experimental group achieved significantly better outcomes in Fugl–Meyer Assessment of Upper Extremity (FMA-UE), Wolf Motor Function Test (WMFT) and Modified Barthel Index (MBI) scores compared with the control group (p < 0.05). Both groups presented shortened motor evoked potential latency and increased peak-to-peak amplitude, but no significant between-group differences were found in neurophysiological measures (p > 0.05). These findings indicate that WBVV combined with conventional rehabilitation can effectively promote upper limb functional recovery and ADL performance in subacute stroke patients. Further research is required to verify its effects on corticospinal excitability.

Introduction

Stroke is a leading cause of disability and mortality worldwide¹. In China, the incidence and prevalence of stroke are rising continuously, largely due to population aging and shifting lifestyles, with a growing impact on younger adults². Approximately 80% of stroke survivors experience upper limb dysfunction, which severely restricts their performance of activities of daily living (ADL), impairs quality of life, and imposes a substantial burden on both families and society³. Restoring upper limb function remains a critical research priority and a major clinical challenge. Therefore, the development of safe and effective interventions to improve upper limb function represents a key focus of current rehabilitation research.

WBVV, first applied in the field of sports medicine⁴, has been widely adopted as an adjunctive rehabilitation intervention for stroke patients^5,6. Previous studies have demonstrated that WBVV significantly enhances muscle strength, explosive power, and bone density^7,8, especially among elderly individuals and clinical rehabilitation populations (e.g., patients with stroke and those after heart transplantation). Its therapeutic mechanism is thought to involve vertical mechanical stimulation that activates muscle proprioceptors and may modulate central nervous system excitability.

In stroke rehabilitation, WBVV has been proven effective in improving balance and lower limb motor function⁹; however, research focusing specifically on its effects on upper limb rehabilitation remains limited. Current established upper limb rehabilitation strategies still present notable practical limitations. Task-oriented training¹⁰ requires sufficient active motor control, which is often difficult for patients with severe impairment. Constraint-induced movement therapy (CIMT)¹¹ demands intensive use of the affected limb, frequently leading to fatigue and poor compliance during the subacute phase¹². Robotic-assisted therapy enables precise training but is restricted by high costs and limited accessibility¹³. In contrast, whole-body vibration (WBV) provides proprioceptive input¹⁴ without requiring sustained voluntary effort¹⁵. It is cost-effective, well-tolerated, and easy to integrate into routine clinical care, making it a promising adjunct therapy for upper limb recovery after stroke¹⁶.

The progressive WBVV protocol, characterized by gradual increments in vibration frequency and amplitude, was designed to deliver controlled sensorimotor stimulation while reducing the risk of neural habituation associated with fixed-parameter interventions¹⁷.The parameters used in this study (5–25 Hz; 1–8 mm) were selected based on prior clinical studies that confirmed safety and tolerability in neurological populations, while providing adequate mechanical stimuli to activate proprioceptive pathways¹⁸. Stroke-related upper limb impairment is closely tied to corticospinal tract damage and diminished neural excitability¹⁹. Although motor evoked potentials (MEPs) offer a reliable tool for assessing corticospinal function^{4,20­­–23}. The precise mechanisms by which WBVV improves upper limb recovery remain poorly understood, leaving a substantial gap in the existing literature^24,25.

Accordingly, this randomized controlled trial is designed to investigate the effects of progressive WBVV on upper limb motor function and corticospinal tract excitability in patients with subacute stroke. The protocol evaluates the effects of WBVV combined with conventional rehabilitation on upper-limb motor function and ADL performance, in comparison with conventional rehabilitation alone. It further assesses changes in corticospinal excitability by measuring MEP latency, amplitude, and motor threshold, and examines the relationship between functional outcomes and these neurophysiological parameters.

Protocol

Ethical approval was obtained from the Ethics Committee of Xuzhou Medical University (Approval No.: XYFY2024KL31601), and written informed consent was obtained from all participants prior to enrollment. The trial is registered prospectively in the Chinese Clinical Trial Registry (Registration No.: ChiCTR2500097167). Forty-eight inpatients diagnosed with stroke are recruited from the Department of Rehabilitation Medicine, Affiliated Hospital of Xuzhou Medical University between January 2024 and January 2025.

1. Participant inclusion and exclusion criteria

1. Inclusion criteria: Include patients meeting the following criteria: first-ever stroke confirmed by computed tomography (CT) or magnetic resonance imaging (MRI), including ischemic stroke or intracerebral hemorrhage, diagnosed according to World Health Organization (WHO) criteria; clinically stable condition; disease duration of 10 days to 6 months; age between 35 and 80 years; upper limb Brunnstrom stage II or higher; sitting balance grade 2 or higher; Mini-Mental State Examination (MMSE) score ≥ 26; Visual Analog Scale (VAS) score = 0 for shoulder pain; and voluntary willingness to complete the 4‑week training and assessment.
2. Exclusion criteria: Exclude patients with severe cardiovascular or cerebrovascular disease, osteoarthritis, or peripheral nerve injury; those with contraindications to transcranial magnetic stimulation (TMS); those intolerant to vibration or with orthostatic hypotension; and those unwilling to participate in the study.

2. Study design

1. Design a prospective, single-center, randomized, assessor-blinded controlled trial. Ensure all outcome assessors and statistical analysts remain fully blinded to the group allocation of participants and maintain blinding until the completion of all statistical analyses to minimize subjective bias in outcome evaluation.
2. Perform sample size calculation based on the primary outcome (FMA-UE), with a two-sided α of 0.05, a power of 0.80, and an effect size obtained from preliminary and published data. The total required sample size is 48 patients.
3. Generate a random sequence by an independent statistician. Use sequentially numbered, opaque, sealed, tamper-proof envelopes for allocation concealment. Randomly assign 48 eligible patients to the experimental group (n=24) and the control group (n=24) at a 1:1 ratio after baseline assessment (Figure 1).

3. Baseline assessment

Perform all baseline assessments before randomization by one rehabilitation therapist with at least 5 years of clinical experience in neurological rehabilitation and standardized training, including FMA-UE, MBI, WMFT, and MEPs.

1. FMA-UE
   1. Administer the FMA-UE, which consists of 33 items assessing shoulder, elbow, and distal joint movements. The FMA-UE is widely accepted for evaluating upper limb motor recovery after stroke and demonstrates high reliability (ICC = 0.96–0.99) and strong validity.²⁶
   2. Score each item from 0 to 2 points, with higher scores indicating better motor function. Instruct the participant to perform each standardized movement and score according to task completion.
2. MBI
   1. Administer the MBI to evaluate 10 activities of daily living items, with a total score of 100. Higher scores indicate greater independence. The MBI demonstrates high reliability (ICC = 0.90–0.95) and strong validity in stroke rehabilitation populations²⁷.
   2. Assess the participant’s functional independence and score based on actual task performance.
3. WMFT
   1. Administer the WMFT, which includes 15 tasks: items 0–6 assess simple movements and items 7–15 assess complex movements. The WMFT demonstrates excellent reliability (ICC = 0.88–0.97) and strong construct validity in stroke rehabilitation²⁸.
   2. Instruct the participant to complete each task quickly and accurately, with scoring based on time and movement quality.
4. Motor Evoked Potentials (MEPs) recording
   1. Record motor evoked potentials²⁹ from the unaffected abductor pollicis brevis (APB) muscle. Use a transcranial magnetic stimulator with a 70-mm figure-of-eight coil.
   2. Place surface EMG electrodes with the active electrode on the belly of the unaffected APB muscle, the reference electrode on the distal aspect of the unaffected thumb, and the ground electrode on the dorsum of the unaffected wrist.
   3. Position the coil over the unaffected primary motor cortex (C3/C4) according to the international 10–20 Electroencephalogram (EEG) system, with the handle pointing backward at 45° to the midline to identify the motor hot spot.
   4. Define resting motor threshold (RMT) as the lowest stimulus intensity able to evoke MEPs of ≥50 µV in at least 3 out of 5 consecutive stimuli. Set stimulation intensity at 120% of RMT.
   5. Filter signals at 20 Hz–2 kHz with a 100-ms analysis window. Deliver 8 to 10 stimuli at intervals of ≥3 s, and average five artifact-free trials.
   6. Define latency as the time from stimulus onset to the first positive peak. Define amplitude as the peak-to-peak value.

4. Intervention protocol

1. Control group: conventional rehabilitation
   1. Administer conventional rehabilitation in accordance with evidence-based guidelines for stroke rehabilitation³⁰.
   2. Provide all treatments by licensed rehabilitation therapists with at least 5 years of clinical experience and unified standardized training before study initiation.
   3. Include the following components in each treatment session:
      1. Health education: Deliver once at the first session, with a duration of approximately 20 min, covering stroke rehabilitation knowledge, safety precautions, and self-management guidance.
      2. Occupational therapy: Administer for 20 min per session. Adjust task difficulty according to individual functional levels, including peg placement, roller training, screw turning, pinching and clamping, and ring stacking tasks.
      3. Activities of daily living (ADL) training: Perform for 10 min per session, including dressing, feeding, bathing, and grooming training.
      4. Physical therapy: Perform passive range-of-motion exercises for shoulder, elbow, wrist, hand, hip, knee, and ankle joints. Conduct 5–10 repetitions per joint for 2 sets. Apply neuromuscular facilitation techniques, including Bobath³¹ and Brunnstrom³² for 10 min per session according to muscle tone and motor recovery stage. These techniques involve reducing abnormal muscle tension and guiding normal limb movements while using staged training strategies based on the patient’s motor recovery stage.
      5. Conventional acupuncture therapy: Administer for 30 min per session in accordance with the Standardized Acupuncture Protocol for Post-Stroke Rehabilitation issued by the China Association of Chinese Medicine.
   4. Administer treatment 6 days per week for 4 consecutive weeks
2. Experimental group
   1. Provide the same conventional rehabilitation protocol as described for the control group. Add WBVV training using a vertical vibrating platform
   2. General training protocol
      1. Administer each WBVV session for 15 min, once daily, 6 days per week, for 4 consecutive weeks.
      2. Conduct all sessions by therapists with standardized training and a minimum of 5 years of clinical experience.
   3. Equipment installation, calibration, and parameter setting
      1. Install the vibrating platform on a stable, non-slip surface prior to the study and adjust to ensure a horizontal position.
      2. Check frequency, amplitude, timer, and emergency stop function at the start of each session.
      3. Verify parameter accuracy using a calibrated frequency-amplitude meter.
      4. Administer continuous vibration for 15 min without scheduled rest intervals.
      5. Suspend vibration temporarily if significant fatigue or discomfort develops, and resume once symptoms subside.
      6. Set frequency and amplitude at the start of each session according to predefined weekly ranges.
      7. Initiate at the lower end of each range and increase gradually based on individual tolerance and motor control.
      8. Maintain fixed parameters throughout the session once established.
   4. Posture control
      1. Position the participant seated in front of the vibrating platform with feet shoulder-width apart and trunk slightly forward (Figure 2).
      2. Place the affected upper limb directly on the platform (Figure 3). Position the shoulder at 30°–45° flexion with mild abduction.
      3. Maintain the elbow at approximately 5° flexion to avoid bony contact. Maintain the wrist at 15°–20° dorsiflexion. Extend the fingers to allow uniform weight bearing.
      4. Instruct the participant to sustain active weight bearing through the affected limb throughout the session.
      5. Minimize trunk compensation. Monitor limb alignment and trunk position continuously. Provide minimal assistance only as required to maintain posture.
   5. Weekly progression protocol
      1. Week 1: Set frequency at 5–10 Hz and amplitude at 1–2 mm. Support the affected elbow with the unaffected hand and place the affected upper limb on the platform. Maintain continuous weight bearing.
      2. Week 2: Set frequency at 10–15 Hz and amplitude at 2–4 mm. Maintain the elbow at 5° flexion and wrist at 15°–20° dorsiflexion. Place the affected upper limb on the platform without support. Maintain continuous weight bearing. Provide verbal or light tactile cues as needed.
      3. Week 3: Set frequency at 15–20 Hz and amplitude at 4–6 mm. Maintain the same limb position as Week 2. Maintain continuous weight bearing without support. Use the unaffected hand to sequentially lift pegs from the pegboard upward for 5 min. Perform forward and backward ball control on the platform with the unaffected hand for 5 min (Figure 4). Monitor trunk alignment and upper limb support continuously.
      4. Week 4: Set frequency at 20–25 Hz and amplitude at 6–8 mm. Perform partial sit-to-stand training without fully standing. Maintain continuous weight bearing through the affected upper limb (Figure 5). Ensure therapist proximity for safety.

5. Safety monitoring and adverse event management

1. Monitor vital signs, including blood pressure and heart rate, before each session (BP < 140/90 mmHg; HR 60–100 bpm).
2. Terminate training immediately if dizziness, nausea, chest discomfort, excessive sweating, severe pain, muscle spasm, involuntary tremor, or postural instability occurs.
3. Record all adverse events with onset time, symptoms, duration, and interventions. Notify the attending physician immediately if symptoms persist or become severe.
4. Place the participant in a supine position and administer oxygen if hypotension (BP < 90/60 mmHg) or tachycardia (HR > 120 bpm) develops.

6. Follow-up assessment

1. Use the same assessment tools and methods applied at baseline.
2. Assess primary outcome using FMA-UE.
3. Assess secondary outcomes using MBI, WMFT, and MEP.
4. Perform all assessments at predefined post-intervention time points using the same assessor to ensure consistency.

7. Statistical analysis

1. Perform all statistical analyses using SPSS 26.0. Express continuous variables as mean ± standard deviation and categorical variables as numbers.
2. Assess normality using the Shapiro–Wilk test and homogeneity of variance using Levene’s test. Consider a two-sided P-value <0.05 statistically significant.
3. Compare baseline continuous data using independent-samples t-tests with Welch’s correction for unequal variances and categorical data using chi-square tests.
4. Analyze within-group changes using paired t-tests. Examine between-group differences using independent-samples t-tests.
5. Apply no adjustment for multiple comparisons to the primary outcome. Use the Bonferroni correction for secondary outcomes by adjusting the significance level according to the number of comparisons.
6. Calculate effect sizes using Cohen’s d based on mean differences and pooled standard deviations.

Results

A total of 48 stroke patients were recruited and randomized into a control group and an experimental group, with 24 participants per group. All patients completed the full course of assessments and interventions, with no loss to follow-up or dropouts reported, and no adverse events related to the intervention (including mild symptoms such as dizziness, local pain, fatigue, or abnormal vital signs) were observed in either group during the 4-week intervention period, demonstrating good safety and clinical tolerability of the progressive WBVV training protocol.

No statistically significant differences were noted between the two groups in baseline characteristics, including age, gender, disease course, and stroke subtype (P > 0.05, Table 1), confirming the balance and comparability of baseline data across groups.

Prior to intervention, there were no significant intergroup differences in FMA-UE, WMFT, and MBI scores (P > 0.05, Table 2). Following intervention, all three scores were significantly elevated in both groups compared with baseline (P < 0.001), and the experimental group exhibited significantly higher scores than the control group across all three measures (P < 0.05).

For MEP-related indices (latency, amplitude, motor threshold), no significant intergroup differences were detected at baseline (P > 0.05, Table 3). After intervention, these MEP indices were significantly improved in both groups relative to pre-intervention levels (P < 0.001), with no statistically significant differences observed between the two groups in the post-intervention measures (P > 0.05).

Figure 1: Patient screening and enrollment flowchart. This diagram outlines recruitment, eligibility assessment, randomization, intervention, and follow-up for 48 subacute stroke patients. Abbreviation: WBVV, Whole-Body Vertical Vibration. Please click here to view a larger version of this figure.

Figure 2: Posture control setup for WBVV. Participant seated in front of the vibrating platform, with the affected upper limb placed on the platform surface for continuous weight-bearing training. Please click here to view a larger version of this figure.

Figure 3: Upper limb positioning on the vibrating platform. The affected upper limb is positioned on the vibrating platform, with the shoulder at 30°–45° flexion, the elbow at approximately 5° flexion, and the wrist at 15°–20° dorsiflexion to maintain proper alignment. Please click here to view a larger version of this figure.

Figure 4: Dual-task training during WBVV. Participant performing pegboard lifting (left) and ball control (right) tasks with the unaffected hand, while maintaining continuous weight bearing through the affected upper limb on the vibrating platform. Please click here to view a larger version of this figure.

Figure 5: Partial sit-to-stand training during WBVV. Participant performing partial sit-to-stand movements, with continuous weight bearing maintained through the affected upper limb on the vibrating platform, under therapist supervision for safety. Please click here to view a larger version of this figure.

Group	Time point	FMA-UE	WMFT	MBI
Control group (n = 24)	Pre-treatment	18.54 ± 6.00	18.54 ± 6.00	36.46 ± 10.26
	Post-treatment	32.25 ± 10.33	23.63 ± 4.50	54.33 ± 11.62
Experimental group (n = 24)	Pre-treatment	18.79 ± 5.39	18.58 ± 6.00	38.38 ± 9.71
	Post-treatment	38.00 ± 5.28	27.54 ± 7.00	64.38 ± 13.60

Table 1: Comparison of general clinical characteristics between the control and experimental groups. Data are presented as mean ± standard deviation (x̅ ± s) or number (n). There were no statistically significant differences in sex, age, course of disease, or affected side between the two groups (p > 0.05).

Group	n	Sex (n)		Age (x̄±s, y)	Stroke course (x̄±s, d)	Stroke subtype (n)		Impairment side (n）
		Male	Female			Cerebral infarction	Cerebral hemorrhage	Left hemisphere	Right hemisphere
Control group	24	15	9	59.04±8.06	14.96±2.51	18	6	13	11
Experimental group	24	16	8	58.83±10.75	14.88±2.42	18	6	11	13

Table 2: Comparison of FMA, WMFT, and MBI scores between the two groups before and after intervention. Abbreviations: FMA-UE = Fugl-Meyer Assessment of Upper Extremity; WMFT = Wolf Motor Function Test; MBI = Modified Barthel Index; Data were presented as mean ± standard deviation.

Group	Time point	Latency (ms)	Amplitude (μA)	Motor threshold (%)
Control group (n=24)	Pre-treatment	20.38 ± 4.20	74.63 ± 50.76	43.63 ± 5.76
	Post-treatment	20.11 ± 4.16	76.55 ± 50.89	41.25 ± 5.50
Experimental group (n=24)	Pre-treatment	20.67 ± 4.20	74.89 ± 50.76	43.25 ± 5.50
	Post-treatment	19.65 ± 4.00	80.22 ± 51.00	39.96 ± 5.00

Table 3: Comparison of motor-evoked potential (MEP) related neurophysiological indices between the two groups before and after intervention. Latency (unit: ms): the time from stimulation to MEP waveform appearance (shorter latency indicates better neural conduction function); Amplitude (unit: µV): the peak-to-peak value of the MEP waveform (higher amplitude indicates stronger neural excitability); Motor threshold (unit: %): the minimum stimulation intensity required to induce MEP (lower threshold indicates higher excitability of cortical motor neurons). Data were presented as mean ± standard deviation.

Supplementary File 1: Raw experimental data. This file contains all raw clinical assessment and neurophysiological data collected in this study.Please click here to download this file.

Discussion

This study demonstrates that combining WBVV with conventional rehabilitation improves upper limb motor function and activities of daily living in subacute stroke patients, and may indicate a potential improvement in corticospinal tract excitability in this population. The protocol integrates progressive vibration parameters with controlled weight-bearing of the affected upper limb, providing a structured and reproducible stimulation framework.

Previous studies have documented the beneficial effects of vibration therapy on upper limb recovery after stroke^33,34,35. The present protocol builds on this evidence by standardizing parameter progression and positioning to improve consistency across sessions.

A central feature of the protocol is the staged adjustment of vibration frequency and amplitude. Lower frequencies (5–10 Hz) and amplitudes (1–2 mm) are applied initially to support sensory adaptation and limit excessive reflex activation. As patients tolerate stimulation, parameters are gradually increased (up to 20–25 Hz and 6–8 mm) to sustain sensorimotor responsiveness and reduce neural accommodation to mechanical stimulation. Progressive tuning helps sustain sensorimotor responsiveness and reduce accommodation that can occur with fixed-intensity stimulation. Although low-frequency vibration can effectively reduce for spasticity reduction^36,37, parameter selection may depend on therapeutic goals³⁸. Introducing higher frequencies following an adaptation phase may provide stronger stimulation while maintaining tolerability.

Proper and sustained weight-bearing through the affected upper limb represents a critical component of the protocol^39,40. The degree of mechanical loading directly influences proprioceptive input and muscle spindle activation. Proper alignment of the shoulder, elbow, and wrist, along with trunk stabilization, ensures consistent mechanical loading. Inadequate loading or compensatory movements may reduce effective stimulation. Therapist supervision during early sessions is therefore essential to standardize posture and loading conditions.

Within-group reductions in MEP latency and motor threshold, together with increases in MEP amplitude, were observed after treatment, suggesting enhanced corticospinal excitability during the rehabilitation period. However, no significant between-group differences were detected in post-treatment MEP latency (P = 0.698), amplitude (P = 0.804), or motor threshold (P = 0.399). These findings indicate that corticospinal excitability improved during rehabilitation in both groups, while the additional electrophysiological advantage of WBVV over conventional therapy was not statistically confirmed in this study. Whole-body vibration is thought to activate muscle spindle Ia afferents and modulate spinal and supraspinal circuits^{33,34,35,41,42,43}.

In clinical practice, tolerance to vibration varies widely. If transient increases in tone or fatigue emerge during parameter progression, modest reductions in frequency or amplitude can help maintain safety and continuity. For patients who struggle to sustain weight-bearing, partial support or shorter exposure intervals may be introduced before gradually restoring progression. When fatigue compromises posture, maintaining proper alignment and loading quality should take priority over higher stimulation intensity.

WBV offers clear clinical advantages over conventional upper-limb rehabilitation strategies. Unlike CIMT and mirror therapy⁴⁴, which require substantial patient participation and active effort, WBV modulates sensorimotor function through mechanical stimulation and is suitable for subacute stroke patients with limited motor control¹⁵. Robotic training remains limited by high costs and poor adoption in primary care settings¹³. WBV uses simple equipment with short setup time, and its progressive protocol can be adjusted based on individual tolerance⁴⁵. These features make WBV a practical and easily implemented adjunct therapy in routine stroke rehabilitation¹⁶.

Several limitations warrant mention. This was a single-center study with a relatively small sample size and short intervention duration, which may restrict generalizability. In addition, no sham vibration control was included; therefore, the specific physiological effects of WBVV cannot be fully distinguished from nonspecific influences. Future multicenter studies incorporating sham-controlled designs and longer follow-up are needed to further validate this protocol and refine stimulation parameters⁴⁶.

In conclusion, progressive WBVV combined with conventional rehabilitation is a safe, effective, and feasible intervention for improving upper limb motor function and activities of daily living in subacute stroke patients. Although improvements in corticospinal excitability were observed during rehabilitation, no significant between-group differences in MEP outcomes were identified. This protocol provides a standardized and reproducible clinical training method, which has important clinical application value for the rehabilitation of upper limb dysfunction in stroke patients.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
SPSS Statistics	IBM Corp		version 26.0
Transcranial Magnetic Stimulation Device	Nanjing Wiss Medical Technology Co., Ltd.		Magneuro R130
Whole-body Vertical Vibration Therapy Device	Taiwan BodyGreen		MF1