Digital rehabilitation earned a qualified place in knee osteoarthritis, and not yet after replacement

Design

systematic review with meta-analyses, stratified by immersion depth and clinical mechanism, GRADE assessed

Population

21 trials - 12 in knee osteoarthritis, 9 after total knee arthroplasty; samples 36 to 306, mostly ≤3 months follow-up

Primary outcome

pain-related functioning and pain intensity

Effect

knee osteoarthritis interactive digital rehabilitation pooled SMD -0.59 (95% CI -1.11 to -0.06) for functioning and -0.46 (-0.92 to 0.00) for pain at 6-8 weeks, low certainty; post-arthroplasty pain intensity -0.12 (-0.75 to 0.52)

Extended reality covers a wide span of things, from a phone-based exercise programme with feedback to a fully immersive headset, and reviews that pool them tell you little. This systematic review deliberately stratified by depth of immersion and by clinical mechanism, searching to December 2025, and found 21 eligible trials - 12 in knee osteoarthritis and 9 in recovery after total knee arthroplasty - with samples from 36 to 306 and mostly three months of follow-up or less.

For knee osteoarthritis, interactive digital rehabilitation may improve pain-related functioning (pooled standardised mean difference -0.59, 95% CI -1.11 to -0.06, from 5 studies) and lower pain intensity at six to eight weeks (-0.46, -0.92 to 0.00, from 4 studies), both at low certainty. Virtual reality digitally augmented exercise gave inconsistent results across two studies. After knee arthroplasty, interactive digital rehabilitation may improve pain-related functioning but made little or no difference to pain intensity (pooled SMD at 3-4 months -0.12, -0.75 to 0.52). Single studies suggested benefit from virtual reality psychoeducation and distraction. Adverse events were seldom reported, and where they were, interactive digital rehabilitation was not associated with them.

The authors' conclusion is appropriately split: interactive digital rehabilitation may be ready to integrate into knee osteoarthritis care, while its use after arthroplasty needs more evidence. Note what is actually being recommended - guided, interactive exercise delivered digitally, not headsets. In Indian practice, where access to supervised physiotherapy is the binding constraint for most patients with knee osteoarthritis, a remotely delivered interactive exercise programme is the version of this with real reach, and the evidence for it is the strongest part of the review.

• Consider interactive digital rehabilitation as an adjunct in knee osteoarthritis where supervised physiotherapy access is limited
• Do not extend the same conclusion to recovery after knee replacement, where pain intensity did not improve
• Distinguish interactive digital rehabilitation from immersive virtual reality - the evidence differs, and it is the former that is better supported
• Note follow-up was three months or less in most trials; durability is unknown
• Ask about adverse effects anyway, since they were reported in only 5 of 21 trials

Why it matters

It separates the version of this technology that works from the version that is marketed.

Don't overread it

Low certainty evidence with wide prediction intervals, short follow-up and sparse adverse event reporting - this supports trying it, not adopting it as standard.

The statistics, in plain English

The prediction intervals here are more honest than the confidence intervals and worth reading: the knee osteoarthritis functioning result has a confidence interval of -1.11 to -0.06 but a prediction interval from -1.72 to 0.55, meaning a new trial could plausibly find harm. A confidence interval describes the average effect across these studies; a prediction interval describes what the next study might find, and it is much wider. GRADE low certainty means the estimate is likely to change with further research.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.