The PMA clock doesn't stop at approval

A PMA record in openFDA carries one date the public actually sees: the date FDA approved it, or the date FDA approved whatever supplement changed it since. Nothing in that record marks the report a PMA holder owes FDA every year the approval stays in force — a filing the regulation makes a condition of keeping the approval at all, and one the public dataset has no field, no number, and no date for.

The condition attached to “approved”

21 CFR 814.84(a) makes continued approval of a PMA contingent on the applicant filing a periodic report— at intervals of one year, unless FDA specifies a different interval for a given device — for as long as the approval stands. The report is not paperwork adjacent to the approval; the regulation ties the approval’s own continuation to it.

21 CFR 814.84(b) sets what has to be in it: a summary and bibliography of unpublished reports of clinical and nonclinical data involving the device, known or reasonably knowable to the applicant, plus published literature about it; changes made under 21 CFR 814.39(a) that did not require a supplement in the first place, and changes reported under 814.39(b); and, for reports filed after September 24, 2014, every device identifier currently in use for the device and every one discontinued since the last report. An approved device’s active identifiers get reported to FDA a second time through this filing — not only through GUDID — in a record that never becomes public either way.

A filing the supplement record doesn’t carry

openFDA’s PMA dataset logs a supplement_number and a decision date for every change filed against an approved device — the record a competitor’s device history is usually read from. A periodic report under 814.84 is not a supplement. It gets no supplement number, because 814.39 and 814.84 answer different questions: one changes the device, the other keeps FDA current on a device that hasn’t changed enough to need a supplement at all. FDA’s own field reference for the PMA endpoint has nothing that marks whether, or when, a given PMA’s annual reports have actually been filed.

A PMA’s continued approval depends on a filing that never earns a supplement number, a decision date, or a row in the same dataset that shows the approval itself.

The practice

Reading a Class III competitor’s device off openFDA’s PMA record shows two kinds of events reliably: the original approval, and any supplement FDA later decided on. It does not show whether the required annual reports have kept coming, because a periodic report that changes nothing about the device leaves nothing in that record to find. A long gap between supplements is consistent with a stable device under current periodic reporting — and it is equally consistent with a sponsor that has let its own reporting lapse, a fact the public record has no way to surface either way. Confirming which one describes a specific device is a question for the sponsor’s own compliance history, not for the supplement trail alone — this is regulatory intelligence and method, never regulatory advice about any device’s current standing.

PMA data sits outside FDA Radar’s current source coverage — per what we monitor and how often, the pipeline watches 510(k) clearances, not PMA approvals, their supplements, or the periodic reports that keep them in force.