Throughout the world, stroke is indeed the second leading cause of death and the third leading cause of disability (1). According to statistics, about 5.5 million people die from stroke every year, and the risk of stroke in adults is as high as 24.9% (2). As of 2019, China ranked first in stroke incidence globally, with ~17 million people over 40 years of age affected by stroke (2). Spasticity is the most frequent after-stroke consequence, affecting about 40% of patients within days or weeks, and resulting in adverse effects, such as pain, immobility, and muscular contracture, which severely reduce the quality of life (3). Direct finance due to post-stroke spasticity (such as hospitalization, drugs, and health professional services) is about four times more than those who do not have spasticity (4). Thus, spasticity management is an essential component of stroke rehabilitation.

Spasticity can be treated in various ways, such as by administering botulinum toxin injections (5), which is considered the most widely used local treatment of spasticity. Rehabilitation interventions such as stretching of the muscles (5), aquatic exercise (6, 7), mirror therapy (8, 9), ultrasound therapy (10), electrical stimulation (11), and extracorporeal shock wave therapy (12) have also been shown to be useful. Recently, whole-body vibration (WBV) was suggested as a potential therapeutic approach to address spasticity in stroke survivors (13, 14). WBV was recognized as a non-invasive, easily accepted and operated, and well-tolerated technique in which an individual stands on a vibratory board that emits sinusoidal oscillations through the feet to the entire body (15).

Several possible mechanisms have been proposed to explain decreased post-stroke spasticity during exposure to WBV. One proposed mechanism (16) hypothesized that WBV may suppress the synaptic transmission between Ia afferent and motor neurons by inducing presynaptic inhibition, then inhibiting the excitation of spastic muscles, thereby reducing muscle tension. Another way in which vibration has this effect was through the “busy hypothesis,” in which the Ia discharge becomes locked to vibration and was subsequently unable to accurately convey the stretch-induced volley due to the entrained action potentials and the high vibration frequency in the Ia fibers (17). Nevertheless, the mechanism of action on spasticity is complicated, and the straightforward monosynaptic reflex or a single route cannot fully account for the anti-spasticity effect of WBV. Many transcranial magnetic stimulation studies (18–20) have demonstrated that WBV also has an impact on the brain's central nervous system. By altering the brain's central nervous system or reducing spinal cord excitability, spasticity of the affected lower limb may alleviate (21). Miyara et al. (22) found that WBV could increase cortical excitability by functional near-infrared spectroscopy. Thus, WBV is thought to be a potential approach for treating spasticity in patients with stroke.

Despite some randomized controlled experiments claiming that WBV reduces post-stroke spasticity (23–25), some other researchers reported it was not observed a benefit of WBV in reducing muscular spasticity (26). Therefore, how WBV impacts post-stroke spasticity is still a debate. This study aims to compile randomized controlled studies that have already been published, analyze the efficacy and safety of WBV systematically, and offer more thorough and rigorous proof for the application of WBV to treat post-stroke spasticity.

This meta-analysis was performed in accordance with the PRISMA guidelines (27). The protocol has a PROSPERO registration (registration code CRD42022293951). Every study was based on previously published research; therefore, neither written consent nor ethical clearance was required.

This meta-analysis was intended to assess the effectiveness and safety of WBV in treating individuals with post-stroke spasticity, and simultaneously, an appropriate application scheme was also explored by analyzing some possible influencing parameters or factors. Overall, the analysis comprised a total of 11 trials with 475 individuals. We identified moderate-quality proof that WBV was regarded as a safe and effective adjunctive therapy in patients with post-stroke spasticity, especially when used at a vibration frequency below 20 Hz for 10 min, for patients with a stroke under the age of 60 years who have post-stroke spasticity in their upper and lower limbs.

As mentioned in a prior review (24), it came to the conclusion that there was weak proof that short-term WBV therapy lowers lower limb spasticity in patients suffering neurological disorders, which was consistent with our conclusion. Lucrezia et al. (46) also found in contrast to chronic patients, those who were acute or subacute appear to benefit from vibration therapy more. However, the appropriate WBV treatment parameters for individuals with post-stroke spasticity have not been identified by prior investigations. Compared with early reviews (24, 46), this review conducted a more comprehensive subgroup analysis of the factors that may affect treatment efficacy, including age, stroke course, and vibration parameters (frequency, time, and location). Given the available research data, our study offered safe and effective WBV settings to relieve post-stroke spasticity despite the lack of high-quality evidence. However, a different systematic review (47) found inadequate data to either support or disprove the claim that WBV can relieve spasticity in patients with stroke. The short number of studies included the low number of studies that produced statistically significant outcomes, and the wide range of intervention strategies may be responsible for this conclusion. As the studies evaluating how WBV affects muscular spasticity post-stroke have significantly increased, we believe that the conclusions of this study need to be further updated with increasing evidence on the effectiveness of WBV for spasticity today.

As far as we are aware, this is the first meta-analysis to thoroughly compile and evaluate the effectiveness and safety of WBV in the treatment of post-stroke spasticity; besides, some influencing parameters and factors were also explored, offering support for the clinical application of WBV. Furthermore, from the results of the subgroup analysis, we filled the current gap in the clinical use of WBV in post-stroke spasticity treatment by offering a reliable and secure prescription for it. The Cochrane Collaboration's guidelines and criteria were strictly adhered to in this meta-analysis (31). In addition, as determined by the strict inclusion and exclusion standards, the most relevant randomized clinical studies were included. To prevent conclusions from being biased or misleading, we evaluated the quality of the evidence using GRADEpro GDT (48). Subgroup analysis of the moderate quality of evidence revealed WBV can reduce spasticity in the upper and lower limb after stroke, which was consistent with the conclusion of previous systematic reviews (49, 50). However, current studies have focused on the upper and lower limbs, and there are few studies on WBV for trunk muscle spasticity, and the effectiveness of trunk spasms is difficult to conclude at present. Therefore, to better comprehend how WBV affects post-stroke spasticity in other clinically relevant body parts, especially the trunk, additional well-designed randomized clinical trials are required.

This review found low-quality evidence that patients with both acute and subacute stroke can benefit from WBV for post-stroke spasticity. Considering that spasticity gradually increased within 1 month of onset, while stroke survivors often present with limb weakness within 3 months (5), in survivors with acute and subacute stroke, the spasticity reduction would have been more considerable. According to certain studies (50, 51), patients with an acute stroke who receive low vibration frequency (20 Hz) had stronger muscles than those in the control group. However, this conclusion required validation with more and higher quality RCTs due to the low quality of the literature evidence. Due to the low quality of the literature data, this finding needed to be validated with more randomized controlled trials of higher quality. Moreover, varying degrees of impairment may have different effects on soft tissue's properties, such as how much muscle atrophy, how much muscle turns into connective tissue, and how much sarcomere was lost (3), thus limiting the application of WBV.

The present study found moderate-quality evidence that WBV could effectively improve post-stroke spasticity at frequencies below 20 Hz. Some studies (41, 52) suggested that an enhanced vibration signal and possible negative effects can emerge from vibrations at frequencies below 20 Hz because of a significant resonance effect, such as internal organ damage. However, three of the included studies (37, 39, 44) used vibrations below 20 Hz, among which only a few produced mild adverse effects, and spasticity alleviation outperformed that of the control group by a large margin. Considering the outcomes of this study, WBV at frequencies < 20 Hz can be applied to ameliorate post-stroke spasticity. We did not decide to carry out a subgroup analysis for the study with a frequency higher than 30 Hz because of the small number of inclusions. Since their greater peak acceleration values, high-frequency vibrations (>30 Hz) have the potential to cause damage (45), which may not be appropriate for patients with chronic stroke, many of whom have frail bones (53). Furthermore, the only included study (41) reported that some patients showed fatigue and discomfort at high frequency vibration. Some studies (54, 55) have shown that spasticity can be effectively treated by inducing frequencies between 20 and 30 Hz. Therefore, to strengthen the evidence for therapeutic application, more study is needed to confirm the impact and mechanism of different vibration frequencies on post-stroke spasticity.

Regarding the WBV duration, post-stroke spasticity can be effectively alleviated for the last 10 min. Four included trials (40, 42–44) that adopted a 10-min vibration time reported significant effects on spasticity improvement, and no withdrawal from the trial due to adverse effects. Hence, the present review suggests that 10 min of WBV as an adjunctive therapy was safe and effective for improving post-stroke spasticity.

Another important parameter is the posture of the patient. Due to the limited number of inclusions, only a systematic evaluation of patient posture was performed. Patient postures in 10 studies were standing posture, nine of which measured lower limb spasticity, and one measured trunk spasticity. In the other two studies, patients were in a seated posture, which studied upper limb spasticity. In research involving WBV training in a standing position, the majority of patients were asked to perform static or dynamic semi-squat training instead of a static upright standing station. Studies have found that static upright stations were prone to transmitting vibration to the head, which can lead to adverse reactions such as dizziness (56). There is plenty of evidence that knee flexion angles may impact the transmission of vibration to the head, which should be avoided. Abercromby et al. (57) demonstrated that when the knee flexion angle rises from 10 to 30, it appears that head transmissibility decreases, using a fixed setup (30 Hz and 4 mm). It can also be utilized with dynamic motions for WBV. The response to dynamic exercise was equivalent to static postures, according to a recent study (58) that looked at transmissibility to the head (frequencies between 20 and 50 Hz) during dynamic squats. Therefore, a certain knee flexion Angle can reduce the transmission of vibration to the head to reduce adverse reactions. Moreover, Boo et al. (59) found that chronic stroke patients with whole-body vibration training in a seated position had increased muscle tone and upper limb function. This may be related to an improvement in the postural control ability. Improving postural control can make arm movement more flexible and improve activities of daily living (60). Verheyden et al. (61) reported that sitting training was effective in improving postural control. Hence, to enhance upper limb function, it was advised to carry out rehabilitation activities while seated on shaky or moving surfaces, such as the vibration platform.

There are certain restrictions on the current meta-analysis. First, all the participants of this study were patients with mild to moderate spastic stroke. There has not been enough research done on the effectiveness of WBV in patients suffering from severe spasticity, thus, this conclusion may not be applied and extended to all patients with post-stroke spasticity. Second, the potential long-term effects of WBV cannot be determined because only two studies had a 3–6-month followed-up. Third, due to a variety of mixed factors, the results can be biased to some extent. These limitations, to some extent, limit the applicability of WBV and the credibility of the conclusions, and the conclusions drawn should be treated with caution. To assess how different vibrational parameters affect post-stroke spasticity, create optional treatment protocols, and provide scientific justification for the purpose of WBV therapeutic usage in the management of post-stroke spasticity, additional prospective studies are required.

Based on current moderate evidence, it seems when used as an adjuvant therapy for 10 min at a frequency of < 20 Hz, whole-body vibration has been proven to work best for upper and lower limb spasticity in patients with acute and subacute strokes under the age of 60 years. Studies of better quality are required in the future to examine its long-term safety and effectiveness, as well as the mechanism of action.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

QZe, JZ, and XZ: conception and design and typographical logic of the article. YZ, LC, and GL: literature selection and acquisition of data. QZh, SZ, and ShuL: analysis and interpretation of data and editing the article. ShiL, LH, and SC: study supervision and revising the article. All authors contributed to the article and approved the submitted version.