Medical and assistive systems
Surgical robots and the evidence behind a clinical claim
Understand the surgeon's control, the scope of device authorization and the difference between a technical result and a patient outcome.
A surgical robot should be identified by its device, intended procedure and the person controlling its instruments. A hospital delivery cart, a therapy machine and an instrument-positioning system answer different questions even when all three are called medical robotics.
This guide focuses on robot-assisted surgery. Rehabilitation, prosthetic devices and worn assistance have separate guides because their tasks and evidence differ.
Identify the surgeon's role before calling a system autonomous
The FDA describes robotically assisted surgical systems with a surgeon's console, a bedside instrument arrangement and supporting equipment. In the systems discussed by that guidance, the surgeon controls the instruments while viewing the surgical field. A mechanical arm in the operating room is not evidence that the procedure runs without direct human control. [1]
A technical description should identify where commands originate and what the system does with them. Instrument positioning, visual navigation and planning assistance can be separate functions. A video of one automated movement should not be expanded into a claim about an entire operation.
For a research report, retain the task and test material. Completing a movement on a bench model answers a different question from clinical use in a patient. The evidence needs to name the setting before readers can interpret the result.
Read authorization for the named device and intended use
The FDA's guidance ties cleared uses to trained physicians, specified surgical procedures and operating-room use. It also warns that device models differ and may require different training. Authorization for one intended use does not establish benefit for every procedure or patient group. The FDA example concerns the United States. [1]
When comparing a claim, retain the region, exact model and intended use stated in the supporting document. A general manufacturer page or a photograph of a hospital installation cannot substitute for that record.
A claim about a patient's recovery needs an outcome study. A claim about instrument movement needs a technical measurement. Keeping these questions separate avoids using one kind of evidence to answer another.
Scroll sideways for all columns.
| Claim | Evidence to examine | Question left open by a movement video |
|---|---|---|
| Instrument positioning | Defined task, reference measurement and error | Whether the measured result transfers to clinical use |
| Shorter procedure time | Matched procedure, timing definition and comparison group | Which preparation and operating stages were included |
| Patient benefit | Patient population, outcome and follow-up | Whether the outcome differs from an appropriate alternative |
| Authorized use | Exact device, region and intended-use document | Whether the proposed use is within that scope |
Keep the comparison group and follow-up period visible
Imagine a hypothetical report stating that 90 of 100 patients reached a defined outcome. Without an appropriate comparison, that percentage does not establish how much of the result came from the device. The definition of the outcome and the time at which it was measured are also necessary.
Now suppose 20 patients have no follow-up result. Reporting 90 successes out of the original 100 would be impossible unless the missing-data handling explains those counts. A reader should be able to reconcile enrollment, follow-up and analysis denominators.
This is an example of reading a report, not a clinical result. Do not use these numbers to estimate the risks or benefits of surgery. The clinical team needs evidence for the relevant procedure and patient circumstances.
A device report does not establish an event rate
The FDA explains that medical-device reports can be incomplete, inaccurate or duplicated. A submitted report does not by itself prove that a device was defective or caused an event. The agency evaluates these reports alongside other evidence. [1]
To calculate a rate, the numerator and exposure denominator must be defined for the same population and period. A collection of reports without that information cannot be turned into a reliable probability for an individual patient.
Use questions about device experience, alternatives and the relevant outcome evidence in discussion with the treating team. This page does not rank hospitals or determine suitability for a procedure.
Sources and scope
Sources were consulted for this revision. Manufacturer descriptions are identified in the text. Worked examples are illustrative calculations, not measurements from a TechniaHQRobot test.
- FDA computer-assisted surgical systems
Surgeon control, intended-use and training context, and limitations of device-event reports. No individual treatment recommendation is made.