ROSA Knee System (with OptimiZe)
Made by Zimmer Biomet (NYSE: ZBH)
Photo: Zimmer Biomet (NYSE: ZBH)
- robotic arm
- Stäubli six-axis arm, 6 DoF
- active track
- Added with OptimiZe — keeps resections on-plane without pinning the cut guide to bone
- data platform
- ZBEdge Analytics; post-op recovery through mymobility and WalkAI
- imaging workflow
- Imageless or image-based from plain pre-op X-rays reconstructed by X-Atlas 2D-to-3D; the surgeon can fall back to imageless mid-case if registration fails. No CT required.
- cutting mechanism
- Semi-autonomous. The arm positions a cut guide to the plan and stops; every bone cut is made by the surgeon with a hand-held saw through that guide. The robot does not cut.
- cart configuration
- Two carts, one on each side of the table: Robotic Unit (arm plus touchscreen) and Optical Unit (tracking camera plus touchscreen)
- arm working radius mm
- 850
- optical unit weight kg
- 140
- robotic unit weight kg
- 320
- positioning accuracy mm rms
- 0.75
Who's exposed
Jobs in the threat radius
- surgical first assistant (limb positioning and retraction portion of the role)
- orthopedic surgical technologist (scrub tech, via leaner team sizes as cases move to ambulatory surgery centers)
- hospital operating room circulating nurse (arthroplasty service lines losing volume to ASCs)
- orthopedic surgeon running a non-robotic arthroplasty practice (referral volume and hiring)
Deployment status
Shipping and placing. The original ROSA Knee TKA indication cleared FDA 510(k) on 24 January 2019 (K182964); the OptimiZe build cleared 13 November 2025 (K251314) and reached full commercial launch at AAOS in February 2026. On installed base, Zimmer Biomet's 8-K exhibit filed with the SEC on 7 October 2025 states the ROSA platform is 'rapidly approaching 2,000 installations worldwide' and describes itself as a leader outside the United States. That is the last installed-base figure the company has put in an SEC filing; the FY2025 10-K describes ROSA qualitatively with no unit count, and nothing has been updated in 2026, so any 2026 ROSA installed-base number is unverified. Cumulative ROSA procedures have never been disclosed and we do not publish a figure. Penetration comes from CEO Ivan Tornos on the 5 August 2026 Q2 call: of roughly 29,000 US orthopedic surgeons, the share using robotics 'remains 20%,' leaving 80% who do not, and only 10% outside the US. The same quarter brought record capital sales, 21.5% growth in the Technology and Data segment, and 50% growth in US technology sales. The competitor discloses a different metric: Stryker said in July 2026 that Mako has passed 2.5 million cumulative procedures across 47 countries. Zimmer Biomet is pushing its commercial organization at this — 200 additional robotic clinical sales representatives, and a stated plan to grow specialized US reps 3.5x from 2024 to 2027. The successor track is Monogram, acquired October 2025: the semi-autonomous CT-based mBos targets early 2027, a fully autonomous version late 2027 to early 2028.
When this hits the labor market
This is the smallest threat in the catalog, and that is the finding. ROSA does not cut. It positions a guide; the surgeon saws. There is no path to surgeon displacement here and none visible in three to five years. Three real changes are underway instead. First, work is redistributed inside the OR rather than removed: a 60-case prospective study of robot-assisted TKA found the robotic arm did not raise surgeon stress but shifted load, with scrub technicians reporting the greatest workload from the added robotic tasks — up, not down. A separate 20-case workflow study measured retractor-holding down from 13.0 to 7.8 minutes and power-tool holding from 7.7 to 2.7 minutes, so the first assistant's physical burden genuinely falls, with no evidence of headcount falling with it. Second, a new role appears — robotics technician / clinical application specialist — and a meaningful share of it is supplied by the vendor's own field organization, which is what those 200 new clinical reps actually are. Third, and the one to watch: cases migrate from hospitals to ambulatory surgery centers, the team follows, and ASC staffing is leaner. That plays out over three to seven years and the constraint is reimbursement, not technology — Medicare pays $13,837 for a TKA in a hospital outpatient department against $10,339 in a freestanding ASC.
Is this robot coming for your job?
Press releases speak in averages — LostJobs talks about your role. Find out which parts of your job hold up and which don't. Free.
Talk to LostJobs about my future
The robot does not cut
Start with the thing most coverage gets wrong. ROSA Knee is semi-autonomous, and the entire job of the robotic arm is to position a cut guide according to the pre-op plan and then hold still. Every bone cut is made by the surgeon, by hand, with a saw, through that guide. Zimmer Biomet’s own user manual says so.
That makes ROSA the smallest threat in this catalog. We are cataloging it anyway, because it is the second pole of the orthopedic robotics market, and because the answer to “who loses work when a robot enters the OR” turns out to be stranger than most people expect.
What it is
A Stäubli six-axis arm, 6 DoF, positioning accuracy under 0.75 mm RMS, maximum working radius 850 mm. Two carts roll into the room and sit on either side of the table: a Robotic Unit at about 320 kg and an Optical Unit at about 140 kg.
It runs imageless, or image-based from plain pre-op X-rays reconstructed into 3D by X-Atlas. No CT. If registration fails mid-case the surgeon can drop back to imageless and keep going. The OptimiZe build cleared FDA in November 2025 and added Active Track, which holds resections on-plane without pinning the cut guide into bone. Full commercial launch came at AAOS in February 2026.
There is exactly one citable installed-base number: “rapidly approaching 2,000 installations worldwide,” in an SEC filing dated 7 October 2025. Nothing since. The FY2025 10-K carries no unit count at all, and cumulative procedures have never been disclosed, so we publish neither.
What actually changes in the operating room
A 60-case prospective study measured workload separately for surgeons, scrub techs, circulators and technicians. The result runs against intuition. The arm did not raise surgeon stress; it redistributed the work, and scrub technicians reported the greatest workload from the added robotic tasks. The robot inserts several steps that did not exist in the conventional workflow, and those steps land on the scrub tech.
A 20-case workflow study measured the other end. The arm cut retractor-holding from 13.0 to 7.8 minutes, power-tool holding from 7.7 to 2.7 minutes, and leg-holding from 4.7 to 3.4 minutes. The first assistant’s physical burden really does drop. In that same study, total OR time showed no significant difference. A separate prospective study built for this question took 75 timestamps per case across 24 cases and found no statistically significant difference in any OR interval, turnover included.
One sentence: the robot redistributes the work in the room without reducing the people in the room.
Why we care for LostJobs
The story worth tracking is not inside the OR. It is the address of the OR.
Of roughly 29,000 US orthopedic surgeons, 20% use robotics. Eighty percent do not. Meanwhile ambulatory surgery centers are buying robots while knowing the math is against them: Medicare pays $13,837 for a TKA in a hospital outpatient department and $10,339 in a freestanding ASC. Becker’s reporting names the reason, and it is a labor reason. Residents and fellows finishing training today spent their formative years on robotic platforms, many have never done certain joint replacements without one, and whether a center has a robot decides where they affiliate.
So the endpoint here is not a machine replacing a surgeon. It is arthroplasty volume moving from hospitals to ASCs, the team moving with it, and the ASC running a thinner roster. For a hospital orthopedic OR’s circulating nurses and scrub techs, the threat is not the arm on the cart. It is the schedule relocating to a different building. Squeezed from the other side is the 80% of arthroplasty surgeons not operating robotically, who are losing referral volume and the ability to recruit young surgeons at the same time. It is a shape of displacement this catalog rarely sees: the machine takes no one’s work, it just changes where the work happens and who is qualified to sign for it.