Virtual Reality-assisted Physiotherapeutic Training for Patients With Knee Osteoarthritis: A Systematic Review and Meta-analysis
Clinical Snapshot
PICO Framework
| P — Population | Adults with knee osteoarthritis (KOA) |
| I — Intervention | Virtual reality (VR)-assisted physiotherapy / rehabilitation |
| C — Comparator | Traditional / conventional rehabilitation (described as 'consistency rehabilitation') |
| O — Outcomes | Pain relief, physical function, and balance (all measured as standardised mean differences) |
Bottom Line
This systematic review and meta-analysis of five RCTs reports statistically significant, small-to-moderate improvements in pain (SMD 0.40), physical function (SMD 0.37), and balance (SMD 0.44) with VR-assisted physiotherapy compared to conventional rehabilitation in knee osteoarthritis. While the direction of effect is encouraging and consistent across all three outcomes, the evidence base is critically limited. Only five trials were included, heterogeneity statistics are unreported, risk of bias assessment is not described in the abstract, and no GRADE certainty rating is provided. The conclusions overstate the strength of the evidence. Adverse events — including cybersickness and fall risk during VR use — are not addressed. For Australian clinicians, exercise-based physiotherapy remains the evidence-based standard per RACGP guidelines. VR-assisted rehabilitation may be considered as an adjunct in motivated patients with access to appropriate technology, but cannot yet be recommended as a routine alternative. Larger, well-conducted RCTs with longer follow-up, standardised VR protocols, and patient-reported outcome measures are needed before practice change is warranted.
Key Findings
P Value: Pain: P = 0.002; Physical function: P = 0.003; Balance: P = 0.002
Effect Size: Pain: SMD = 0.40 (small-to-moderate); Physical function: SMD = 0.37 (small-to-moderate); Balance: SMD = 0.44 (small-to-moderate)
Primary Outcome: Pain relief in knee osteoarthritis patients receiving VR-assisted physiotherapy versus traditional rehabilitation
Nnt Or Sensitivity: NNT not calculable from SMD data without baseline event rates. SMD values of 0.37–0.44 correspond to small-to-moderate effects per Cohen's conventions (small ≥0.2, moderate ≥0.5). The lower bound of the physical function CI (0.05) approaches clinical non-significance.
Confidence Interval: Pain: 95% CI 0.14–0.65; Physical function: 95% CI 0.05–0.70; Balance: 95% CI 0.16–0.72
Clinical Application
VR-assisted physiotherapy requires access to VR hardware and software, trained physiotherapists, and appropriate clinical space. Cost, technology literacy, and patient tolerance (including cybersickness risk) are practical barriers. Scalability in community physiotherapy settings remains uncertain. The evidence base is insufficient to support widespread implementation over established physiotherapy protocols. Knee osteoarthritis is highly prevalent in Australia, affecting approximately 9% of the population and representing a major burden on the healthcare system. The RACGP Clinical Guideline for the Diagnosis and Management of Knee and Hip Osteoarthritis recommends exercise therapy as a core treatment. VR-assisted physiotherapy is not currently listed on the Medicare Benefits Schedule (MBS) as a distinct reimbursable item, nor are VR devices subsidised through the Prostheses List or NDIS for KOA specifically. The TGA does not currently have specific regulatory pathways for therapeutic VR software that would affect clinical adoption. Australian physiotherapy practice would require higher-certainty evidence — ideally from larger RCTs or meta-analyses with GRADE-rated outcomes — before VR-assisted rehabilitation could be recommended as a standard adjunct. The current evidence may support cautious use in specialist rehabilitation settings with appropriate patient selection and monitoring. Adults with diagnosed knee osteoarthritis who are candidates for physiotherapy-based rehabilitation. Applicability across KOA severity grades, age groups, comorbidity profiles, and VR technology types (immersive vs. non-immersive) cannot be determined from the available data.
Abstract
OBJECTIVE: To evaluate the effectiveness of virtual reality (VR)-assisted physiotherapy in improving pain, physical function, and balance in patients with knee osteoarthritis (KOA) compared with traditional rehabilitation treatments. DESIGN: Systematic review and meta-analysis of randomized controlled trials (RCTs). RESULTS: This analysis included 5 randomized controlled trials (RCTs). The pooled results showed that, compared with consistency rehabilitation, virtual reality (VR)-assisted physiotherapy produced significantly better outcomes in patients with knee osteoarthritis (KOA) for pain relief (SMD = 0.40, 95% CI: 0.14-0.65, P = 0.002), physical function (SMD = 0.37, 95% CI: 0.05-0.70, P = 0.003), and balance (SMD = 0.44, 95% CI: 0.16-0.72, P = 0.002). CONCLUSIONS: Virtual reality-assisted physiotherapy is an effective intervention for patients with knee osteoarthritis, showing superior outcomes in pain reduction, functional improvement, and balance enhancement compared to traditional rehabilitation approaches.
References
- 1.Liu, H.-W., & Tsai, T.-L. (2026). Virtual reality-assisted physiotherapeutic training for patients with knee osteoarthritis: A systematic review and meta-analysis. American Journal of Physical Medicine & Rehabilitation. Advance online publication. https://doi.org/10.1097/PHM.0000000000002914
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service