Virtual reality-based interventions enhance functional status, balance and muscle strength in patients with ankle instability: a systematic review with meta-analysis of randomized controlled trials.

Only 12 trials were included, and some analyses rested on two trials (muscle strength) or four studies (hopping and jumping), so those estimates are imprecise. The included trials averaged 5.8 out of 10 on the PEDro quality scale. In all of them the patients and the treating therapists knew which treatment was given, and in 7 of 12 the person measuring the outcome knew as well. That kind of unblinding tends to make a treatment look better than it is. 89% of participants were male, the mean age was 21.8 years, and everyone had a previous ankle sprain, so the results may not carry over to older, less active patients or to those with ankle problems from other causes. Outcomes were measured right after the programme in every trial, and only two trials assessed anyone afterwards (4 weeks later), so nothing is known about whether the improvements last. The virtual reality systems and protocols differed (Wii Fit versus head-mounted displays, 3 weeks to 3 months, different outcome tests), so the best device and dose are unknown and gait could not be analysed at all. Possible publication bias was flagged in the ankle function and lower limb performance analyses, and 7 of the 12 included trials received external funding.

The authors state that they have no conflict of interest with any financial organisation regarding the material discussed. Seven of the 12 included trials received external funding to carry out their research, though the review does not say who funded them.

Do not read these large effects as a reason to replace conventional ankle strengthening with games. Patients and therapists in every included trial knew which treatment was being given, which tends to inflate the difference between groups, and the strength results favoured virtual reality only for plantar flexion and eversion, with no improvement in hopping or jumping.