Robot-Assisted Knee Surgery Is More Precise. Patients Did Not Feel the Difference
What happened
One of the biggest questions in modern surgery has just been given an unusually clear answer, and it is not the answer many people expected. Results from the RACER-Knee trial were published in The Lancet in August 2026. The trial compared robot-assisted total knee replacement with conventional knee replacement done by a surgeon without robotic assistance. It involved 339 patients with advanced knee osteoarthritis, treated by 33 surgeons across 10 NHS hospitals in the UK. It was co-sponsored by the University of Warwick and University Hospitals Coventry and Warwickshire NHS Trust. The design was rigorous. Patients were randomly assigned to one type of surgery or the other, and both the patients and the people assessing their recovery were kept unaware of which operation they had received. That matters, because knowing you had "the robot" could easily colour how you report your own recovery. Twelve months later, the two groups looked remarkably similar. Pain levels, mobility, and how aware patients were of their new joint in daily life were all comparable. Rates of reoperation and serious adverse events were similar too. There were differences, but not the ones hoped for. Robot-assisted operations took around 10.5 minutes longer and cost roughly £950 more per patient. The robotic approach was more precise in placing the implant, but that extra precision did not translate into patients feeling better within the first year. Professor Andrew Metcalfe, one of the researchers, said: "Technology such as this could really help the care we give patients, but for robotic-assisted knee replacements, there is still work to do before we see benefits such as less pain or better movement."
Why this matters
Knee replacement is one of the most common major operations in the NHS. Tens of thousands of people have one every year, usually because osteoarthritis has made walking, sleeping or climbing stairs genuinely miserable. Robotic systems have been spreading through orthopaedic surgery for years. They are expensive to buy, expensive to run, and they carry an obvious intuitive appeal. A machine that positions an implant to within a fraction of a degree simply sounds better than a human eye and a steady hand. That intuition is exactly what this trial tested. And it found that the machine really was more precise — but that the extra precision did not show up in how patients felt a year later. This is important for two reasons. The first is money. The NHS has a fixed budget, and £950 extra per patient, multiplied across thousands of operations, is a large sum that could be spent elsewhere. If the benefit is not there, that spending needs justifying. The second reason is more interesting. It is a reminder that precision is not the same as benefit. Surgeons have long debated how much implant positioning actually affects how a knee feels afterwards. This trial suggests that, within the range a good surgeon already achieves, being even more exact may not be what patients notice. That is a genuinely useful thing to learn. It means the next improvement in knee replacement probably has to come from somewhere other than positioning accuracy.
What the evidence actually says
This is a strong piece of evidence, and it is worth understanding both its weight and its limits. On weight: this was a randomised controlled trial, which is the most reliable way we have of comparing two treatments. Randomising removes the risk that healthier patients end up in one group. Masking both the patients and the assessors removes a large source of bias. Running it across 10 hospitals with 33 surgeons makes the result more likely to hold in ordinary practice rather than just in one expert centre. The researchers also placed their findings alongside the other randomised trial in this area, ROAM. Combining the two trials, covering 439 participants in total, showed no difference in patient-reported outcomes at either 3 or 12 months. Two independent trials pointing the same way is more persuasive than one. Now the limits, which matter just as much. This result covers one year. Knee replacements are meant to last decades. It remains genuinely possible that better implant positioning pays off over 10 or 20 years, through implants that loosen less or last longer. The researchers are continuing to follow these patients for 10 years precisely because that question is still open. It is also a trial of one robotic system for one operation. It does not tell us about robotic assistance in hip replacement, spinal surgery or other fields, where the trade-offs may be entirely different. And "no difference on average" does not rule out that particular patients, or particularly complex knees, might benefit. The researchers have suggested robotic precision could yet prove useful in more personalised surgical approaches.
Practical advice
**If you are waiting for a knee replacement, this is not bad news.** Both operations in the trial worked. Patients in both groups improved. The finding is that one is not better than the other, not that either is ineffective. **Do not choose a hospital purely because it has a robot.** On this evidence, the presence of a robotic system is not a good reason to travel further or wait longer. Surgeon experience, hospital infection rates and how quickly you can be seen are more useful things to ask about. **Do ask questions at your appointment.** Reasonable ones include: how many of these do you do a year, what does recovery realistically look like, and how long should this implant last? **Take prehabilitation seriously.** Strengthening the muscles around the knee before surgery, and stopping smoking if you smoke, are both linked to better recovery. These are things you can influence; implant positioning is not. **Expect recovery to take months, not weeks.** Most people are walking early, but full benefit typically takes considerably longer. **Follow the physiotherapy plan.** It is dull and it is uncomfortable, and it is also one of the strongest predictors of how well you do. **Be patient with the evidence.** If someone tells you robotic surgery is definitely better, or definitely pointless, they are ahead of what the data currently supports.
What to know
The RACER-Knee trial, published in The Lancet in August 2026, compared robot-assisted and conventional total knee replacement in 339 patients across 10 UK NHS hospitals, with 33 surgeons involved. At 12 months, pain, mobility, joint awareness, reoperation rates and serious adverse events were similar between the two groups. Robotic surgery was more precise, took about 10.5 minutes longer, and cost around £950 more per patient. It was not found to be cost-effective within one year. Combined with the earlier ROAM trial, covering 439 participants in total, there was no difference in patient-reported outcomes at 3 or 12 months. The open question is the long term. Follow-up is continuing for 10 years, and it remains possible that greater precision matters over decades even if it does not show up in year one. The wider lesson is worth keeping: in medicine, a technology that is measurably better on a technical measure is not automatically better for the person on the table. That is what trials are for, and it is a credit to the NHS and the research teams that this question was asked properly rather than assumed. Sources: The Lancet, "Robotic-arm-assisted versus conventional total knee replacement (RACER-Knee): a pragmatic, multicentre, participant-masked and assessor-masked, superiority, randomised controlled trial", August 2026, https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(26)00986-4/fulltext | University of Warwick, "Robotic assisted knee replacement gives the same patient outcomes as a surgeon only procedure", August 2026, https://warwick.ac.uk/news/pressreleases/robotic-assisted-knee/ | Royal Orthopaedic Hospital NHS Foundation Trust, "Robotic assisted knee replacement shows no improvement in largest clinical trial of its kind", August 2026, https://roh.nhs.uk/news-events/robotic-assisted-knee-replacement-shows-no-improvement-in-largest-clinical-trial-of-its-kind This article is for general information and does not replace advice from a doctor, surgeon, physiotherapist or other qualified healthcare professional. Decisions about surgery should always be made with your clinical team.
Related articles
Robot Tomato Pickers and Hoverflies: £13m Backs Farming's Next Ideas
On 1 September 2026 the government announced £13 million for the latest round of its Farming Innovation Programme. The money goes to researchers, farmers and ag…
New Rules Proposed for How the UK Licenses Medical Devices and Shares Safety Data
On 1 September 2026 the government tabled three amendments to its Health Bill. All three are about how the MHRA, the UK's medicines and medical devices regulato…
A New Space Telescope Has Launched to Hunt Down the Universe's Missing Pieces
On 30 August 2026, NASA launched the Nancy Grace Roman Space Telescope. It lifted off at 7:26am Eastern Time on a SpaceX Falcon Heavy rocket from Launch Complex…