# Improving dynamic balance in relapsing-remitting MS: insights from robotic-assisted rehabilitation therapy

## TL;DR

A study found that combining robotic-assisted training with traditional physiotherapy helps improve balance in people with multiple sclerosis as effectively as traditional therapy alone.

## Contribution

The study demonstrates non-inferiority of robotic-assisted rehabilitation combined with conventional therapy for balance improvement in MS patients.

## Key findings

- Both robotic-assisted and traditional therapy groups showed significant improvements in balance and mobility.
- The robotic-assisted group showed additional improvements in dynamic balance metrics under unstable conditions.
- Improvements in the robotic-assisted group persisted for at least two months post-treatment.

## Abstract

Balance deficits affect over 75% of people with multiple sclerosis (PwMS), significantly limiting mobility and daily activities. To address these symptoms, technology-assisted rehabilitation has recently shown promise for restoring balance. This study aimed to test the non-inferiority of a rehabilitative protocol that combines robotic-assisted training with conventional physiotherapy (ROB) compared to a protocol consisting solely of traditional balance training (TRAD) for PwMS.

Forty-two PwMS with relapsing-remitting course were randomly assigned to either ROB or TRAD group. Participants had to undergo 20 sessions (2/week for 10 weeks, 45 min each), focusing on balance exercises targeting static and dynamic control. In the ROB group, traditional exercises were progressively integrated with robotic-based tasks using the hunova® system (Movendo Technology s.r.l., Genoa, Italy). The Berg Balance Scale (BBS) was the primary outcome measure, while the Two-Minute Walking Test (2MWT) and the Composite Score (COMP) from the Sensory Organization Test were selected as secondary outcomes. Additional robotic metrics from hunova® were analyzed to assess stability under static and dynamic conditions.

Both groups showed significant overall improvements in BBS (p = 0.003), 2MWT (p = 0.031), and COMP (p < 0.001), supporting the non-inferiority of the robotic-assisted protocol. However, participants in the ROB group demonstrated additional improvements in unstable conditions compared to TRAD group. Specifically, elastic balance tasks resulted in significant reductions in path length (p = 0.047) and trunk variability (p = 0.017). Additionally, reactive balance metrics showed significant decreases in the first oscillation time for both left and right directions (p = 0.027 and p = 0.029), as well as in the average oscillation time for both directions (p = 0.006 and p = 0.012).

Robotic-assisted rehabilitation combined with conventional physiotherapy is at least as effective as traditional therapy for improving balance in PwMS, demonstrating non-inferiority for the primary outcome. Additionally, the greater dynamic balance improvements observed in the ROB group suggest that robotic technology may provide added benefits by enhancing specific balance mechanisms. Since these improvements persisted for at least two months post-treatment, robotic-assisted training may serve as a complementary or alternative approach to conventional rehabilitation strategies for balance disorders in MS.

The online version contains supplementary material available at 10.1186/s12984-025-01856-w.