A multimodal training with visual biofeedback in subacute stroke survivors: a randomized controlled trial.

Background: Early interventions maximizing patient's involvement are essential to promote gait restoration and motor recovery after stroke.

Aim: The aim of this study is to evaluate the effects of a multimodal biofeedback training involving cycling augmented by functional electrical stimulation (FES) and balance exercises on walking ability and motor recovery.

Design: Randomized controlled trial (NCT02439515).

Setting: Inpatient rehabilitation facility.

Population: Subacute stroke survivors (less than 6 months from the first event) aged up to 90 years old.

Methods: Sixty-eight participants were randomly allocated to an experimental group, performing 15 sessions of biofeedback FES-cycling training followed by 15 sessions of biofeedback balance training (20 minutes each) in addition to usual care (70 minutes), and a control group performing 30 sessions (90 minutes) of usual care. Participants were evaluated before training, after 15 sessions, after 30 sessions, and at 6-month follow-up through: gait speed (primary outcome), spatiotemporal gait parameters, Six-Minute Walking Test, Functional Independence Measure, Motricity Index, Trunk Control Test, Berg Balance Scale, and Fall Efficacy Scale.

Results: Both groups significantly improved over time, but no group and interaction effects were found for any outcomes. The 73% of the experimental group achieved a clinically meaningful change in gait speed compared to the 38% of the control group (P=0.048). These percentages were even more unbalanced for patients with a moderate to severe gait impairment at baseline (91% versus 36%; P=0.008).

Conclusions: The multimodal biofeedback training was not statistically superior to usual care, showing only a positive trend in favor of the experimental group on locomotion recovery. Patients initially not able to walk might be the best candidates for such a training.

Clinical Rehabilitation Impact: The multimodal biofeedback training is a task-specific, repetitive and intensive training requiring a minimal supervision, which might result in a lower staff to patient ratio if organized in group sessions. Therefore, it can represent a good alternative for early stroke rehabilitation.