Robotic knee replacement cost more and did not help, in a properly masked trial

Design

pragmatic, multicentre, participant-masked and assessor-masked superiority randomised controlled trial

Population

339 patients with advanced knee osteoarthritis at ten British hospitals, 33 surgeons; median age 68.8 years

Primary outcome

Forgotten Joint Score at 12 months

Effect

49.2 robotic against 50.2 conventional; adjusted mean difference -1.5 (95% CI -7.5 to 4.5, P=0.62) against a prespecified target of 12 points; 16 serious adverse events in each arm

RACER-Knee randomised 339 patients with advanced knee osteoarthritis at ten hospitals in Great Britain, across 33 surgeons, to total knee replacement using the Mako robotic-arm system or conventional instruments. What distinguishes it from most surgical trials is the masking: sham incisions, additional draping and masked operation notes kept both participants and assessors blind. Patients with inflammatory arthropathy, previous fracture or a need for complex implants were excluded. The primary outcome was the Forgotten Joint Score at 12 months, with a prespecified target difference of 12 points.

At 12 months, mean Forgotten Joint Score was 49.2 in the robotic group and 50.2 in the conventional group - an adjusted mean difference of -1.5 (95% CI -7.5 to 4.5, P=0.62), pointing slightly towards conventional surgery. Serious adverse events were identical: 16 participants in each arm. The robotic approach was more costly.

For a rheumatologist this matters at the point of referral and of counselling, because patients arrive having read about robotic surgery and having been offered it at additional cost in private practice. The trial answers that question cleanly for routine primary knee replacement: no measurable patient benefit at one year, same safety, higher cost. It does not address complex primaries, revision, or unicompartmental replacement, all of which were excluded, and long-term follow-up continues - an effect on implant survivorship would take years to appear and is not excluded by this.

• Tell patients asking about robotic knee replacement that a masked randomised trial found no patient-reported benefit at one year
• Do not let cost differences for robotic assistance drive a referral decision for routine primary knee replacement
• Note the exclusions - inflammatory arthropathy, previous fracture and complex implants were not studied, which covers many rheumatology patients
• Keep the longer-term question open; implant survivorship was not the endpoint and follow-up continues
• Use the masking as the reason to trust this over observational comparisons - sham incisions are rare in surgical trials

Why it matters

It answers a question patients are already asking, with a trial design that makes the answer hard to argue with.

Don't overread it

Routine primary knee replacement only - inflammatory arthropathy, previous fracture and complex implants were excluded, and implant survivorship was not measured.

The statistics, in plain English

The trial was designed to detect a 12-point difference on the Forgotten Joint Score and found -1.5, with a confidence interval from -7.5 to 4.5. That interval excludes the difference the trial set out to find, which makes this a properly informative negative rather than an underpowered one - a benefit of the size that would justify the extra cost has been ruled out. A small benefit of two or three points remains possible and would not matter. The masking matters because the Forgotten Joint Score is patient-reported, and knowing you had the robot would plausibly move it.

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.