Most implanted devices in this country never went through a clinical trial. They reached the market through the 510(k) pathway, which asks a manufacturer to show that its product is substantially equivalent to something already being sold — not that it is safe, and not that it works. A device cleared in 2024 can trace its equivalence back through a chain of predecessors to a product designed decades earlier, some of which were themselves recalled. Patients are rarely told any of this. They are told the device is FDA cleared, and they reasonably hear that as tested. When the failures start appearing in revision surgeries, the manufacturer points to the clearance as evidence of diligence. The litigation turns on what the company knew from its own bench testing, its complaint files, and its post-market surveillance, because that is usually where the failure rate first became visible. Clearance is a regulatory status. It is not a study, and it is not a defense.
Two words that do different work
Patients hear that a device is “FDA approved” and reasonably conclude that a federal agency examined evidence that it is safe and effective in people. For a large share of the devices in American operating rooms, that is not what happened. They were cleared, and clearance answers a narrower question.
The 510(k) premarket notification pathway asks whether a new device is substantially equivalent to a device already legally on the market — the predicate. If the agency agrees it is, the device may be marketed. The governing regulations sit at 21 CFR Part 807. Clinical trial data is not ordinarily part of that submission.
Predicate chains
Because a cleared device can itself serve as the predicate for the next one, a family of implants can descend from a common ancestor through many generations of incremental change — each step defensible on its own, the cumulative distance from the original considerable. The National Academies examined this pathway in a 2011 report commissioned by the FDA and concluded that the 510(k) process was not designed to evaluate safety and effectiveness, and largely cannot be expected to do so (Medical Devices and the Public’s Health).
Substantial equivalence is a comparison, not a conclusion about a patient.
Where the failures actually surface
If clearance is not where an implant’s safety record is established, the record has to be built somewhere. It gets built in use — which means the people generating the data are patients, and the surveillance is retrospective.
- Adverse-event reporting, including the MAUDE database of device reports, which is voluntary for some reporters and known to be incomplete
- Revision-surgery rates captured in national and institutional registries
- Published case series from surgeons who saw the same failure repeatedly and wrote it up
- Recalls and safety communications — which typically come after a signal, not before it
A revision surgery is the honest measure. It is a documented, dated, billed event in which a device was removed because it did not do what it was implanted to do.
What this changes about a claim
A patient who has had a device explanted usually assumes the case is about the operation. It generally is not. It is about the design, the testing that was or was not done, what the manufacturer learned once the device was in circulation, and what it told surgeons and regulators after it learned it.
- The regulatory file: the 510(k) submission, the predicate identified, and the differences the manufacturer chose not to treat as significant
- Design history and bench testing — including wear, corrosion and fatigue testing, and the endpoints the manufacturer set for itself
- Complaint files and internal adverse-event handling, which is often where a company’s own count of failures first appears
- Sales and training material, measured against what the design file supported
- The explanted device itself, where it has been retained
One practical step
If a revision is scheduled, ask in writing that the hospital retain the explanted hardware rather than discard it. It is a small request, and the pathology or metallurgical analysis it later permits cannot be reconstructed from records.
Drugs travel a different route — most reach the market on clinical data under 21 CFR Part 314 — but they share the same after-market logic: the safety picture keeps developing once the population using the product stops resembling the population that was studied.
This article is general information about how these cases work. It is not legal advice, it does not create an attorney-client relationship, and no two claims are alike. If something here sounds like your situation, the useful next step is a conversation about the facts.