ISO 10993-1:2025 is here. What changed, and how to bring your documentation in line.

Spot it

Open the BEP. If it cites the 2018 Table A.1, the categorization is stale by definition. Pay closest attention to devices near a boundary, circulating-blood contact or duration close to 24 h or 30 d, which are the ones most likely to cross into a new bracket.

Fix it

Recategorize against the four § 6.4 tables before rewriting anything else, and update the BEP to cite the 2025 edition. Starting the rewrite from the old panel is the common way to lose weeks of editing. Where the panel grows, plan the new biological effects before commissioning tests: an unjustified gap is easier to defend in remediation than a partial test that hints at a missing effect.

Spot it

Look at how the BEP derives its exposure category. If it comes straight from the IFU with no consideration of off-label or foreseeable-misuse exposure, the file is incomplete by the new standard.

Fix it

Add a short section to the BEP naming the foreseeable-misuse cases considered, the exposure each would imply, and either the extra biological effects run to cover them or the argument that the original panel already does. The slow part is usually getting the clinical and regulatory teams to agree on the list, not the edit itself; post-market complaints and competitor IFUs are a good way to seed it. A covered scenario or a reasoned exclusion is acceptable; silence is not.

Spot it

Open the BER, pick three biological effects, and for each find the matching hazard ID in the risk-management file. If you cannot, the file fails the new test. Many 2018-era BERs leave that mapping implicit, which is exactly what no longer passes.

Fix it

Add a single hazard-to-effect cross-reference table to the BER, and update the risk file so each biological hazard has a control entry pointing at the relevant 10993-series report. The trace usually already exists, held implicitly across the risk file, BER, and supporting reports and in the team's heads; pulling it into one table is a week of careful work for whoever knows the file best, not an evening. Resist inventing retroactive hazards to fill the trace: a test run with no hazard behind it is itself a gap to discuss.

Spot it

Search the BER for "extractables", "leachables", "ISO 10993-18", or "ISO 10993-17". If they are absent on a Class IIb or III device with anything beyond limited exposure, the file is likely short, and a missing genotoxicity rationale on any prolonged- or permanent-contact device is the same kind of gap. Equivalence arguments that lean on the absence of chemical characterization are especially exposed.

Fix it

For long-term and implant devices, plan an E&L study under exhaustive extraction conditions paired with a toxicological risk assessment per ISO 10993-17:2023, and book it early: the bottleneck is rarely the study but getting the toxicologist and sample stock lined up at once, which is months, not weeks. For lower-risk devices, write the rationale for not running E&L into the BEP, referencing the categorization that drives the decision. Either path is defensible; neither being documented is not.

Spot it

Compare the BER's revision history against the device's change log. If the last biological-evaluation update predates the most recent design, supplier, or material change, the file is not lifecycle-current.

Fix it

Define a written re-evaluation trigger list, design, supplier, material, manufacturing, complaint, and post-market signal, and add a short re-evaluation log to the BER that records each trigger and its outcome, even when the outcome is "no change required". Wire the list into the QMS change-control SOP rather than leaving it standalone: if the SOP forces a biocomp checkbox on every design or supplier change, the log fills itself; if it does not, it quietly falls behind. The log is the artifact the reviewer looks for.