Whole-body vibration-based interventions and dynamic balance in older patients with stroke: a systematic review and multilevel meta-analysis

This systematic review and multilevel meta-analysis evaluated the effects of whole-body vibration (WBV) interventions on dynamic balance among older patients after stroke and examined potential moderators and dose-response relationships.

A comprehensive search was conducted across several databases, including PubMed, Web of Science, PsycINFO, the Cochrane Library, and the CNKI International Version, from inception to December 26, 2025. The inclusion criteria focused on randomized controlled trials involving patients who had experienced a stroke, with a mean age of at least 60 years at either the sample or arm level. Standardized mean differences (SMDs), adjusted using Hedges’ g correction, were computed from between-group changes from baseline to the immediate post-intervention period. A three-level random-effects model was employed, with additional sensitivity analyses, robust variance estimation, subgroup analyses, meta-regression, and exploratory dose–response modeling.

Twelve randomized controlled trials (RCTs) involving 486 participants and reporting 26 effect sizes were included in this analysis. WBV-based interventions were associated with improved dynamic balance in older patients after stroke, with an SMD of 0.69 (95% CI: 0.27 to 1.11, P = 0.002), despite substantial heterogeneity (I² = 72.6%). The pooled effect remained directionally consistent across sensitivity analyses and small-sample-corrected robust variance estimation, although the overall certainty of the evidence was rated as low. Subgroup analyses indicated positive effects for both WBV alone and WBV combined with other rehabilitation components, with no significant differences between the two intervention modes. Additionally, exploratory dose–response modeling suggested a possible intervention range of approximately six sessions per week, each lasting 20–22 min, over 5–6 weeks, yielding a cumulative exposure of approximately 7.6–10.6 h.

Low-certainty evidence suggests that WBV interventions may improve dynamic balance in older stroke patients. However, these findings should be interpreted cautiously given significant heterogeneity, concerns about risk of bias, and the limited number of trials. Furthermore, subgroup analyses and dose–response findings should be considered exploratory and hypothesis-generating rather than definitive clinical recommendations.