FDA Has Drafted What It Wants to See Before Clearing a Surgical Robot. Comments Close Nov. 24.
A 50-page draft guidance on robotically-assisted surgical devices sets out nine clinical endpoints, validated training for the whole OR team and new labeling on learning curves. It is not final, and comments on docket FDA-2026-N-9505 are due Nov. 24.
On Sept. 25, the Food and Drug Administration announced in the Federal Register a draft guidance titled "Robotically-Assisted Surgical Devices—Premarket Submissions." It describes what manufacturers should put in front of FDA before a surgical robot is cleared or approved. Comments are due Nov. 24.
The 50-page draft is addressed to manufacturers and FDA reviewers, and it is not for implementation. Nothing changes for a system already in your operating room. But several of its recommendations land on the people who credential surgeons and buy these systems.
What the draft covers
The draft covers teleoperated, software-controlled systems used under the direct control or supervision of a qualified practitioner for open, minimally invasive or endoluminal procedures. It excludes pre-operative planning and stereotaxic navigation systems, robots that place interventional devices such as catheters, remotely teleoperated systems and autonomous robots that perform significant parts of a procedure on their own.
Those last two categories are the subject of a separate FDA public workshop on Dec. 2 and 3 on robotic devices with autonomous or remote teleoperation capabilities.
The clinical evidence FDA wants
For a new system, the draft says clinical data are often needed. For a modified one, FDA may ask for them when there is a new indication, a special population such as children, different technology or a claim of superiority. It recommends nine surgical endpoints: length of stay, intraoperative adverse events, transfusion rate, conversion rate, 30-day postoperative adverse events, readmission and reoperation, mortality and operative time.
Those endpoints should be analyzed against comparators, with robotic data preferred over laparoscopic and laparoscopic over open. Under an "umbrella and covered procedure" approach, data from a more complex, higher-risk procedure can support use in simpler procedures in the same specialty. FDA may also require postmarket data on outcomes by surgeon experience and patient demographics where premarket evidence leaves uncertainty.
The draft treats a surgical robot as a system that performs surgical tasks, not a treatment for disease. That distinction belongs in how hospitals read vendor claims.
Training and labeling
Each manufacturer should have a training program validated to bring users to competency before clinical use, the draft says, and it should cover bedside assistants and scrub nurses as well as surgeons. FDA gives as an example criterion the number of procedures generally needed for a surgeon to reach competency, and says existing training should be re-evaluated when a manufacturer changes the user interface, adds instruments or adds user groups.
Labeling would include information on the learning curve and on patient selection, prominent emergency withdrawal procedures and results on the nine endpoints. Where applicable, it would also state that outcomes related to treatment of the underlying disease were not evaluated as part of the authorization.
How to comment
Electronic comments go to docket FDA-2026-N-9505 on Regulations.gov and are posted unchanged, including any name in the body. Confidential material, such as contract terms, must be filed on paper in two copies, one marked "THIS DOCUMENT CONTAINS CONFIDENTIAL INFORMATION" and one redacted for posting. FDA also asks for comments on costs or savings the guidance may generate.
FDA accepts comments on guidance at any time, but Nov. 24 is the date that ensures consideration before work on the final version begins. As with the generative AI devices docket, specific comments are more useful than general ones.
What to do this week
Assign one person on the credentialing or value-analysis committee to read sections IV.H, on training, and IV.Q, on clinical data, against what you already ask vendors. Surgeons who have trained teams on a new system can say whether a procedure count is a sound competency criterion, and whether training for the bedside team matches what the draft describes.
Decide early whether comments come from the institution, a department or individual surgeons, since each will be posted under the name on it. Describe experience in aggregate, with no patient details.
Reported, fact-checked and edited by The Script Pad's editorial staff.
This article is for professional education and does not replace clinical judgment. Treatment decisions should be based on the individual patient and current guidelines.
