FDA’s Surgical Robot Guidance Prefers Robotic Comparator Data, Which Mostly Means Measuring Challengers Against Intuitive
The FDA’s first dedicated premarket guidance for surgical robots sets one bar for every challenger to Intuitive. Its most consequential line is a preference: when a new system reports clinical outcomes, the comparator should be robotic surgery, and two decades of US robotic surgery have run largely on one company’s systems.
The FDA issued draft guidance on premarket submissions for robotically-assisted surgical devices on 25 September 2026, with comments due by 24 November. It covers teleoperated, leader-follower systems used in open, minimally invasive and endoluminal surgery, and excludes remote teleoperation and autonomous robots.
For systems new to the US, the agency says clinical data are often needed. It lists nine surgical endpoints, from length of stay to operative time, and states that robotic comparator data are preferred over laparoscopic or open data.
An umbrella and covered procedure approach lets data from a complex procedure support authorisation for simpler procedures in the same specialty. Human factors testing must assess negative knowledge transfer from surgeons’ experience with other robots, and data from outside the US must reflect US patients, surgeons and operating rooms.
The bench bar is specific: at least three systems tested for latency, accuracy and workspace, 95% reliability with 95% confidence for grip and cutting, an Enhanced software documentation level, and cybersecurity testing that simulates a compromised hospital network.
It lands as challengers arrive. Medtronic’s Hugo was cleared for urology in late 2025, the first system from a large medtech company in US soft-tissue surgery since Intuitive entered more than two decades ago, and Johnson & Johnson filed a de novo request for Ottava in early 2026.
A reader can stop here with the full picture. The sections below are the detail.
The comparator clause makes the incumbent’s outcomes the reference
The guidance allows clinical endpoints to be compared against robotic, laparoscopic or open surgery. It then ranks them: robotic data are preferred, with laparoscopic data the fallback only when robotic data are unavailable.
That preference changes the question a challenger has to answer. The bar moves from whether a new robot does as well as conventional surgery to whether it does as well as existing robotic surgery. In the US soft-tissue market, where Intuitive worked essentially alone for two decades, most of that existing robotic evidence comes from its systems. That is Synopulse’s reading of the market, not a statement in the guidance.
The incumbent also shapes the evidence base directly. Intuitive funded database access for the bariatric surgery cost study Synopulse covered on 28 September, and three of its five authors were Intuitive employees. A regulatory preference for robotic comparators gives that kind of outcomes research more weight.
The bar moves from matching conventional surgery to matching existing robotic surgery.
Umbrella procedures lower the cost of a broad label
A company seeking authorisation for several procedures can use data from an umbrella procedure of higher complexity and risk to support covered procedures of lower complexity in the same specialty, and additional in vivo data for the covered procedures may not be needed. For 510(k) submissions, all procedures should appear in the indications for use or in the referenced labelling.
This matters most to challengers building labels one specialty at a time. Medtronic’s Hugo was cleared in urology, with general and gynaecologic surgery planned next, and J&J’s Ottava request covers general surgery in the upper abdomen. A single well-chosen complex procedure could now carry a set of simpler ones, which compresses the time and cost of each label extension.
Surgeons’ habits become a regulatory hazard
Human factors testing must consider the impact of no prior robotic experience and, separately, of experience with other robots, specifically whether negative knowledge transfer leads to use errors. Training must be validated for surgeons and for bedside assistants and scrub nurses, with criteria such as the number of procedures needed to reach competency.
In a market where most robotic surgeons trained on one platform, that clause turns a commercial switching cost into a documented regulatory one. A challenger must show that a surgeon’s habits from the incumbent’s console do not produce errors on its own.
The clinical data section adds a second filter. Data collected outside the US should come from patients with similar demographics and comorbidities, and from surgeons with similar training, experience and practice patterns. Systems that built their evidence abroad will need to show that their users resemble US users.
| Area | What the FDA recommends | Competitive effect |
|---|---|---|
| Comparator | Robotic data preferred over laparoscopic or open | Benchmarks entrants against the installed base |
| Umbrella procedures | Complex procedure data can support simpler ones in the same specialty | Cheaper label breadth for challengers |
| Human factors | Assess negative knowledge transfer from other robots; train OR staff | Formalises switching costs |
| Clinical data | Often needed for new systems; outside-US data must mirror US users | Raises the bar for systems proven abroad |
| Bench testing | At least 3 systems; 95/95 grip and cut reliability | Fixed costs that weigh more on small entrants |
| Software and cyber | Enhanced documentation; simulate a compromised hospital network | Longer submissions for connected systems |
| AI features | Risk-based testing, Q-Submission, change control plans | Case-by-case path for AI-enabled robots |
| Out of scope | Remote teleoperation, autonomy, intravascular robots | Next frontier left to direct negotiation |
Source: FDA draft guidance, Robotically-Assisted Surgical Devices, Premarket Submissions, 25 September 2026. The competitive-effect column is Synopulse analysis.
The bench bar is specific enough to price
Much of the document reads like a test plan. Latency must be measured from hand input to instrument motion to displayed image under worst-case load; accuracy, repeatability and resolution under software-driven commands along every axis; and singularity points identified on a kinematic diagram, with mitigations tested. Emergency removal must be timed.
For large manufacturers, that specificity makes review more predictable. For smaller entrants, recommendations such as testing at least three complete systems and demonstrating 95% instrument reliability with 95% confidence translate directly into capital and time before a first submission.
What the guidance leaves out is where the next fight sits
Remote teleoperation and autonomous robots that perform significant parts of a procedure are outside the scope, although some recommendations may apply, and the FDA points developers to its Q-Submission programme. MedTech Dive reports that the agency plans a workshop on robotic devices with remote teleoperation capabilities.
AI-enabled features such as critical structure identification or instrument tracking get a risk-based approach rather than fixed tests. Developers are directed to the agency’s AI device software draft guidance and to predetermined change control plans for post-authorisation updates. The features most likely to differentiate the next generation of robots are the ones the guidance leaves least defined.
The access angleUS entry is being standardised as other markets cap prices
Intuitive’s latest 10-Q describes the pressure outside the US. China’s 2023 quota allows 559 new surgical robots, open to all makers; Intuitive had placed 169 under it by 30 June. China’s National Healthcare Security Administration issued a national framework in the second quarter of 2026 requiring provinces to hold robotic surgery pricing within set limits.
Japan is moving the other way on volume, with incremental reimbursement for hospitals that exceed 200 qualifying robotic cases a year. Challengers entering the US under clearer rules will also be competing in markets where governments now set the price of a robotic procedure as well as approving it.
Source: FDA draft guidance, section IV.Q(2). The FDA notes that other uses, such as intravascular procedures, may warrant different endpoints.
- Late 2025Medtronic’s Hugo cleared for urologyFirst large medtech system in US soft-tissue surgery since Intuitive’s arrival.
- Early 2026J&J files de novo request for OttavaGeneral surgery in the upper abdomen.
- Q2 2026China sets national robotic surgery pricing limitsProvinces to implement within mandated caps.
- 25 September 2026FDA issues draft RASD guidanceFirst dedicated premarket framework for surgical robots.
- 24 November 2026Comment period closesIndustry responses will show where the comparator and umbrella provisions are contested.
What to watch
Comments on the comparator preference. Challengers have an interest in asking for laparoscopic comparators to count equally where robotic data are thin or come mainly from one platform.
How the umbrella pathway is used in the next label extensions from Hugo and Ottava, and whether the FDA’s appendix of worked examples survives into the final text.
The remote teleoperation workshop, which will show how the FDA intends to handle the robots this guidance leaves out.
The guidance sets one test for every surgical robot. By preferring robotic comparators, it also decides whose results that test is scored against.