Short answer
More precise, but not yet shown to be better for patients. Robotic assistance places knee implants with more accurate alignment, yet the largest randomised trial so far, RACER-Knee, found no meaningful difference in how the knee felt and worked at one year, while operations took longer and cost more.1 Whether greater precision leads to longer-lasting implants is unknown, because long-term data do not yet exist.13 The surgeon’s experience and the right choice of operation for your knee matter more than the tool.
Key points
- Robotic systems help the surgeon plan and make the bone cuts.
- Trials consistently show better implant alignment with robotic assistance.
- Pain, function and complication rates have been similar to conventional surgery.
- NICE allows use in the NHS only while further evidence is collected.
- Whether robotic knees last longer will not be known for many years.
What robotic knee replacement means
In any knee replacement, the damaged ends of the thigh and shin bones are removed and metal and plastic parts are fitted to form a new joint. The operation usually takes 1 to 2 hours under a general or spinal anaesthetic.4
In conventional surgery, the surgeon uses mechanical guides (jigs) to plan and make the bone cuts. In robotic-assisted surgery, a computer system builds a plan of the knee, often from a scan, and a robotic arm or handheld device helps make the cuts to that plan. Systems differ in how much of the cutting they guide or carry out. The trials discussed below compared these two approaches directly.13
What the trials show
RACER-Knee, published in The Lancet in 2026, is the largest double-blinded randomised trial of the question. It enrolled 339 people at 10 UK hospitals, under 33 surgeons, and compared conventional total knee replacement with robotic assistance using one widely used system.1 Neither patients nor assessors knew which operation had been done.
At 12 months there was no meaningful difference in the main measure, the Forgotten Joint Score (a questionnaire on how aware people are of their artificial joint in daily life). Walking, pain at 3 and 12 months, pain in hospital and the need for further surgery were also comparable. Robotic surgery was more precise and did not increase serious adverse events.1
Earlier evidence points the same way. A 2024 meta-analysis of seven randomised trials, covering 1,942 knees, found better alignment with robotic surgery but no significant differences in standard knee function scores or complication rates.3 Those trials mostly used older robotic systems, and follow-up varied widely.3
| Outcome | Robotic vs conventional | Evidence |
|---|---|---|
| Implant alignment and precision | Better with robotic assistance | Strong |
| Function and pain at one year | No meaningful difference | Moderate |
| Complications | No significant difference | Moderate |
| Operating time and cost | About 10 minutes longer and more expensive in RACER-Knee | Moderate |
| Implant lifespan | Not yet known | Insufficient |
Why better precision has not meant better results
Conventional knee replacement already works well for most people, so there is limited room to show a difference within a year. How satisfied someone is with a new knee may also depend on factors that alignment does not change, such as pain sensitivity, muscle strength, rehabilitation and expectations.
The main hope is that more accurate alignment could reduce wear and loosening over decades. That is plausible, but unproven. RACER-Knee will follow its patients for ten years to look for longer-term differences.1 For context on how long current implants last, see how long knee and hip replacements last.
What NICE says, and the question of cost
The National Institute for Health and Care Excellence (NICE) has assessed robot-assisted systems for knee and hip replacement through an early value assessment. It says they can only be used in the NHS while evidence set out in an evidence generation plan is collected, and once they have the appropriate regulatory approval.2 For total or partial knee replacement, NICE noted that the results suggest the technologies may not be cost effective, though this estimate was uncertain.2
RACER-Knee reached a similar conclusion: robotic operations took about 10.5 minutes longer on average, cost more, and were not cost-effective over the first year under current NHS thresholds.1 Outside the NHS, the presence of a robot is sometimes presented as a mark of quality. The evidence does not support treating it that way.
Common questions
Does robotic surgery mean a faster recovery?
Not on current evidence. In RACER-Knee, pain in hospital, pain at 3 months and walking ability were comparable between the two groups.1 The usual milestones are in the recovery timeline after knee replacement.
Is robotic knee surgery safer?
It appears to be about as safe, not safer. Neither RACER-Knee nor the meta-analysis found an increase or a reduction in complications.13
Does the robot do the operation?
Not on its own. The surgeon plans and oversees the operation; systems differ in how much they guide or carry out the bone cuts. The experience of the surgeon and team with whichever method they use remains central.
Should I choose a surgeon because they use a robot?
Current evidence does not support that. More useful questions are how many knee replacements the surgeon performs, their outcomes and revision rates, and why a partial or total replacement is being recommended. See partial vs total knee replacement.
Bottom line
Robotic knee replacement makes implant positioning more accurate, but so far that has not translated into less pain, better function or fewer complications in the first year. It costs more and takes slightly longer. Ten-year results from current trials will show whether precision pays off in implant lifespan.
Related reading
- Knee replacement: what it involves, risks and recovery
- Partial vs Total Knee Replacement: What Is the Difference?
- How Long Do Knee and Hip Replacements Last?
- What Is the Recovery Timeline After Knee Replacement?
Sources
- University of Warwick. Robotic-assisted knee replacement: RACER-Knee trial results (press release on the trial published in The Lancet; 2026; https://doi.org/10.1016/S0140-6736(26)00986-4). University of Warwick. https://warwick.ac.uk/news/pressreleases/robotic-assisted-knee Accessed 5 November 2026.
- National Institute for Health and Care Excellence. Robot-assisted surgery for orthopaedic procedures: early value assessment (HealthTech guidance 743). NICE. https://www.nice.org.uk/guidance/htg743/chapter/1-Recommendations Accessed 5 November 2026.
- Alrajeb R, Zarti M, Shuia Z, Alzobi O, Ahmed G, Elmhiregh A. Robotic-assisted versus conventional total knee arthroplasty: a systematic review and meta-analysis of randomized controlled trials. European Journal of Orthopaedic Surgery & Traumatology; 2024;34(3):1333–1343. https://doi.org/10.1007/s00590-023-03798-2
- NHS. Knee replacement: how it’s done. NHS website. https://www.nhs.uk/tests-and-treatments/knee-replacement/how-its-done/ Accessed 5 November 2026.
Article history
- — First published.