Objectives Dizziness is estimated to affect 49%–75% of people with multiple sclerosis and has a significant impact on quality of life. Current management includes medications, lifestyle modifications and exercise therapies such as vestibular rehabilitation. The aim of this study was to evaluate the efficacy of vestibular rehabilitation in people reporting dizziness with multiple sclerosis.

Methods Six electronic databases were searched from earliest date until October 2016. Search terms were based on three concepts; multiple sclerosis, dizziness and vestibular rehabilitation. All studies were screened independently by two authors according to pre-determined eligibility criteria. Standardised mean differences (effect sizes) and 95% CI were calculated to describe differences between groups (eg, intervention and control) and pooled in a meta-analysis where two or more studies were clinically homogeneous. The quality of evidence of the body of evidence was assessed using the GRADE approach.

Results Two randomised controlled trials were identified with a total of 65 participants. In both studies, dizziness was measured using the patient reported Dizziness Handicap Inventory. Low quality evidence indicates there is no significant effect for vestibular rehabilitation on improving symptoms of dizziness compared with a control group of people with multiple sclerosis (SMD (95% CI) 0.76, (–0.19 to 1.70)). The quality of the meta-analysis was limited due to imprecision of results and risk of bias. The two studies identified only used a short period of vestibular rehabilitation (six to eight weeks).

Conclusions There is low quality evidence that vestibular rehabilitation has no effect on dizziness in the multiple sclerosis population. Given the efficacy of vestibular rehabilitation in other causes of dizziness, further high quality studies with longer rehabilitation periods are required.