Effects of Whole-Body Vibration Therapy on Body Functions and Structures, Activity, and Participation Poststroke: A Systematic Review

Bone turnover.

The review revealed that the effect of WBV on bone metabolism in individuals with stroke is far from conclusive, as only one study⁴² measured biochemical markers of bone turnover and no significant results were identified. Examining the literature in older adults also provides little insight as to what WBV protocols may be the best in inducing favorable bone outcomes. A number of studies showed that WBV training did not induce any significant effects on bone turnover rate compared with other exercise training or no-intervention control.^53–55 Only Turner et al⁵⁶ showed that their 8-week WBV protocol (12 Hz, 0.3 g, 20 minutes per session with interspersed rest periods) resulted in a significant reduction in level of bone resorption marker (N-telopeptide X-linked) in women postmenopause compared with sham vibration exposure. Their protocol used a WBV frequency (12 Hz) that was lower than that used by Pang et al⁴² and in other studies (25–40 Hz) in this review. Studying the effect of WBV on bone metabolism is an important question, as it is well documented that people with stroke sustain accelerated bone loss in the paretic limbs,⁵⁷ elevated bone resorption, and reduced bone formation marker levels.⁵⁸ More research on WBV and bone health poststroke is definitely needed.

Muscle structure and function.

Although Tihanyi et al³⁸ (level 2 study) demonstrated that WBV stimulation has additional effect on increasing knee muscle strength transiently after a single treatment session, no conclusion could be drawn because it was the only study that assessed this issue. In addition, out of the 4 studies that measured muscle strength or thickness after multiple WBV sessions, none showed significant results.^41–43,45 These findings may indicate that the vibration stimulation itself may not confer additional benefits on muscle strength or structure after stroke, although it cannot be ruled out that their protocols used may not be optimal to facilitate gain in these outcomes. The frequency range used in these 4 studies was 5 to 30 Hz. A meta-analysis by Marín and Rhea⁵⁹ claimed that WBV frequencies of 35 to 40 Hz are more effective than other frequencies (30–35 Hz and 40–45 Hz) in inducing gain in muscle power. However, it is not clear whether the meta-analysis was preceded by a systematic review. The criteria for selection of articles also were not explicitly specified. For example, studies of different populations (eg, young adults, athletes, older adults) or comparison groups might have been mixed together. It is not known whether only RCTs were included in their analysis. Inclusion of studies with poor scientific rigor may compromise the validity of the meta-analysis. Additionally, the effects of different vibration frequencies may depend on the muscle group being stimulated.^12,13

Spasticity.

Previous studies in healthy individuals and people with spinal cord injury suggested that WBV may decrease the excitability of the spinal motoneuron pool, as reflected by the amplitude of the H-reflex or Hmax/Mmax ratio.^17,60 Based on our review, the evidence on the effect of WBV on spasticity poststroke is somewhat conflicting.

The evidence related to the transient effect of a single WBV session on spasticity is based on one level 2 study and is thus not conclusive.⁴⁴ Although the authors claimed that WBV significantly reduced spasticity,⁴⁴ the reported improvement in MAS and VAS scores was not accompanied by other measurements of spasticity (Tab. 4). In addition, the VAS is a subjective measure, and its improvement can easily be explained by the placebo effect of the added WBV, as the study participants were not blinded.

Of the 2 studies that measured spasticity after multiple sessions of WBV treatment,^42,43 only Pang et al⁴² (level 1 study) reported some beneficial effects on knee spasticity. This finding is somewhat intriguing, as spasticity at the ankle joint, which is typically more severe than that at the knee, was not modified by their WBV protocol. The more apparent reduction in spasticity at the knee than the ankle may be related to the fact that the exercise protocol used heavily involved the knee musculature (eg, semi-squat, deep squat, forward lunge). Taken together, there is no consistent evidence to show that WBV stimulation can reduce spasticity. A common drawback of these 2 studies is that the MAS was the only measure used to evaluate spasticity. The MAS may not be the best assessment tool because it is ordinal in nature, with only moderate reliability and correlation with muscle activity and resistance in response to passive movements,^61,62 making it difficult to detect significant changes in spasticity level. The modified Tardieu scale may be a better option to assess the effects of WBV on spasticity in future studies.⁶³

Postural control and falls.

The beneficial effects of a single WBV session on postural control were supported by Chan et al (level 2 study) only.⁴⁴ However, postural control was assessed with only a single measure (weight distribution between the 2 legs). However, it is well known that there are different domains of balance function (ie, biomechanical constraints, sensory orientation, walking balance, and so on). It remains elusive as to whether a single session of WBV can induce changes in other aspects of balance function. Additionally, the placebo effect of WBV could not be ruled out, as the participants were not blinded.

The evidence also is insufficient to support the use of longer-term WBV training in improving balance. Of the 3 studies, none demonstrated a significant between-group difference in balance outcomes after a training period of 6 weeks to 3 months,^40,43,45 suggesting that WBV has no real effects on postural control in people with stroke. An alternative explanation of the nonsignificant results may be related to the psychometric properties of the outcome measure used. The Berg Balance Scale (BBS) was used as the main balance outcome in these 3 chronic stroke trials. Although the BBS is a commonly used balance measure in clinical practice, its ceiling effect is well documented.⁶⁴ In all 3 studies, the balance ability of the participants was quite good before treatment, as confirmed by the baseline data showing a mean BBS score varying from 46.1 to 51.2 points.^41,43,45 This finding was probably due to the inclusion criteria used in these studies (eg, able to remain standing without external support for at least 30 seconds,⁴⁵ ambulate independently for >100 m⁴³) (Tab. 2). Thus, the BBS may be unable to detect changes in balance ability for these individuals who have only mild impairments in balance performance, thereby contributing to the negative results.

Only one study measured incidence of falls and showed negative results.⁴¹ This finding is not surprising, given the lack of significant effects on neuromotor outcomes and the fact that only 10% of the participants had experienced at least one fall within 3 months before the training period. No recommendation can be made on the use of WBV to reduce fall rate after stroke.⁴¹

Functional mobility.

No firm conclusion can be derived from the available evidence to determine whether a brief WBV session has significant transient effect on mobility, as this topic was addressed by only one level 2 study.⁴⁴ Despite the positive results reported, the WBV group was substantially more impaired than the control group, as reflected by the considerably more time required to complete the TUG (mean difference=22 seconds) and 10-Meter Walk Test (mean difference=7 seconds) at baseline. The different mobility status of the participants between the 2 groups may partially explain the difference in outcomes, as there may be more room for improvement in individuals with more severe limitations in mobility. The evidence also is inadequate to support the use of longer-term WBV training to improve mobility function poststroke.^41,43 Based on the 2 studies that incorporated mobility outcomes, WBV stimulation was shown to confer no additional benefit on mobility function after chronic stroke. This finding is reasonable, as the various measures of body functions and structures that are highly related to mobility (eg, muscle strength, postural control) were not influenced by WBV stimulation, as discussed above. Additionally, the WBV training did not involve any gait-related activities. Based on the principle of task specificity, it is not entirely surprising that no improvement in mobility was induced by WBV.

Participation.

No conclusion can be drawn concerning the effects of WBV on particpation,⁴³ as it was evaluated in one level 2 study only using the Stroke Impact Scale, with unremarkable results compared with sham vibration.

Muscle strength.

Out of the 3 studies that addressed muscle strength,^37,39,46 Tihanyi et al³⁹ and Tankisheva et al⁴⁶ (both level 2 studies) reported better outcomes in the WBV group, whereas van Nes et al³⁷ (level 1 study) reported comparable gains in muscle strength in the 2 groups. Several reasons may explain the discordance in results. First, the outcomes may be influenced by the interaction of many different factors, such as WBV protocols, participant characteristics, and outcome measures used. As shown in Table 2, these factors demonstrated substantial diversity across the different studies. Upon closer examination of the data, we could not identify any specific trend that would explain the discrepancies in results. Second, the activities in the comparison group for the 3 studies were different, involving exercise with music,³⁷ conventional exercise training,³⁹ and habitual physical activity,⁴⁶ respectively. Third, the total treatment time may be a confounding factor. For the 2 studies that demonstrated results in favor of WBV, the intervention group might have had additional treatment time due to WBV training.^39,46 This conclusion is in contrast to the study by van Nes et al,³⁷ in which the total treatment time was the same in the 2 groups. Based on the finding of van Nes et al,³⁷ one can argue that WBV exercise training as a whole may induce beneficial effects on muscle strength that are comparable to exercise with music. However, it cannot be determined whether the improvement in muscle strength detected in both groups was induced by the conventional exercise program (which both groups received) or the added WBV training or exercise with music.³⁷ Hence, it remains elusive as to whether WBV exercise training is a viable alternative to other forms of rehabilitative training to improve muscle strength poststroke.

We do not have sufficient evidence to determine whether WBV has differential effects in improving the strength of various types of muscle contraction (eg, isometric, concentric, eccentric). As demonstrated by Tankisheva et al,⁴⁶ the outcome also may be highly dependent on other factors, including functional role of the muscle (eg, flexor versus extensor), the type of exercise training done while receiving WBV stimulation, baseline muscle strength, and contraction speed.

Spasticity.

There is insufficient evidence to support or refute the notion that WBV is beneficial in reducing spasticity compared with other forms of exercise, as only one level 2 study addressed this issue and no significant change in leg muscle spasticity was found in both groups after the 6-week intervention period.⁴⁶

Postural control.

Two studies showed that WBV training yielded similar results on postural control compared with other types of physical activity.^37,40 However, the WBV group had received more treatment time, which might have confounded the results.^37,40 Superiority of WBV training over habitual physical activity was reported by Tankisheva et al,⁴⁶ in which the WBV group had more improvement in equilibrium scores when standing on a sway-referenced platform. The authors, however, offered no convincing explanation why improvement was observed only in this variable, out of the many balance outcomes used. Thus, it remains uncertain whether WBV is a useful alternative treatment to enhance postural control poststroke.

Sensation.

Although the WBV group and exercise with music group in the study by van Nes et al,³⁷ were shown to have comparable improvements in somatosensory threshold, we cannot conclude the WBV is effective because the improvement could have been due to the conventional rehabilitation program that both groups received. Additionally, factors that are common in both groups, such as maturation effects, may account for the observed improvement.

Functional mobility.

Of the 2 studies that compared WBV with other exercise approaches,^37,40 Merkert et al,⁴⁰ but not van Nes et al,³⁷ demonstrated better outcomes in the WBV group.⁴⁰ As aforementioned, the WBV group in the former study received more treatment time than the comparison group, which may partially explain the better outcomes.

Activities of daily living.

The Barthel Index was used in 2 studies, which compared WBV with other forms of exercise, but the results were conflicting.^37,40 The additional treatment time from WBV training in the study by Merkert et al⁴⁰ may have contributed to the signficant results, as opposed to the study by van Nes et al,³⁷ in which the total treatment time for both groups was even. Due to the limited number of studies and conflicting findings, no conclusion can be made regarding the therapeutic effects of WBV on this domain of function.