Robotic assistance in arthroplasty has spread faster than the evidence for it, and the central question — whether more accurate component positioning translates into better outcomes for patients — has not been answered by a randomised trial. This double-blinded randomised trial reports interim two-year results in 107 patients with symptomatic medial compartment osteoarthritis, randomised to conventional jig-based unicompartmental knee arthroplasty with navigational control (52 patients) or robotic arm-assisted arthroplasty (55). Both arms had CT-based planning, a standard medial parapatellar approach and identical rehabilitation.
The robotic arm won on everything proximate to the operation. Accuracy in executing planned femoral and tibial component positions was better (both p<0.001). Inpatient pain scores were lower (p<0.001), opioid consumption was lower (p=0.008), and hospital stay was shorter (p=0.004).
It did not win on the outcomes that patients live with. At two years the Oxford Knee Score (p=0.299), KOOS (p=0.261) and WOMAC (p=0.281) showed no difference between groups. Only the Forgotten Joint Score — which measures how often a patient is aware of the joint at all — favoured robotics, at six months (p=0.002) and two years (p=0.021).
That pattern deserves a careful reading rather than a verdict. Accurate positioning plausibly matters most for long-term survivorship, which two years cannot assess, and the Forgotten Joint Score is a more sensitive instrument than the Oxford Knee Score in well-functioning knees where the others hit a ceiling. Equally, the honest summary today is that the established outcome measures did not separate.
One detail in the paper is worth more attention than the results. Recruitment declined during the trial because eligible patients increasingly wanted the robot and operating surgeons lost clinical equipoise. That is a description of how a technology becomes standard before the evidence arrives, and it is the reason definitive trials in this area are becoming difficult to run. For Indian practice, where robotic systems are being installed rapidly in private hospitals and marketed directly to patients, this trial supports faster early recovery and shorter stay — real benefits — and does not yet support a claim of better function at two years.
- Robotic assistance improved component positioning, early pain, opioid use and length of stay.
- Oxford Knee Score, KOOS and WOMAC were no different at two years.
- Only the Forgotten Joint Score favoured robotics, at six months and two years.
- Two years cannot assess implant survivorship, which is where positioning accuracy would be expected to matter.
- Do not let marketing outrun the evidence: promise faster recovery, not better two-year function.
The statistics, in plain English
This is an interim analysis of a trial that stopped recruiting well, so the sample of 107 is smaller than planned and the null findings at two years are correspondingly less certain — absence of a detected difference here is weaker evidence than it would be in a completed trial. The Forgotten Joint Score result is worth understanding: it was designed to avoid the ceiling effect that afflicts the Oxford Knee Score, where most successful arthroplasty patients cluster near the top and further improvement cannot be measured. So a difference appearing only on that instrument may reflect better discrimination rather than a different underlying effect — or may reflect one positive result among four instruments tested.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for orthopaedics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free