Biological safety has to be demonstrated for every device in direct or indirect contact with the patient or the user. ISO 10993-1 asks for this to be done not with a list of tests but through an evaluation carried out within risk management. The nature and duration of contact determine which biological effects have to be addressed. The scope of evaluation for a device touching the skin briefly is very different from that of a permanent implant.
How is the evaluation planned?
First the device is categorised: contact type (surface, externally communicating, implant), the tissue contacted and the duration of contact. Those three determine which biological effects are addressed. The plan (BEP) states that reasoning and where existing data is sufficient.
The evaluation starts with the material. Data from chemical characterisation (ISO 10993-18) is interpreted through a toxicological risk assessment; that route often avoids unnecessary animal testing. What remains unresolved goes to testing.
Which tests come into play?
Depending on contact type and duration, the relevant ones are selected from:
- Cytotoxicity, sensitisation and irritation — for almost any device in contact
- Acute systemic toxicity and material-mediated pyrogenicity
- Genotoxicity — for prolonged contact and implants
- Implantation and haemocompatibility — for blood-path and implant devices
Before commissioning tests, sample preparation, extraction conditions and acceptance criteria are agreed with the laboratory. A result obtained under the wrong extraction conditions is an expensive thing to repeat.
How are the results linked to the technical documentation?
Test reports are not enough on their own. The biological evaluation report (BER) brings the plan, the data collected, the justification for the tests and the interpretation of the results into a single document. That report is linked to the risk management file: for each biological hazard, the acceptability of the remaining risk is demonstrated.
In the GSPR checklist, the items related to biocompatibility should name that report and its revision. If the material, the supplier, the manufacturing process or the sterilisation method changes, the evaluation is revisited.
What it covers
- Biological evaluation plan (BEP) and report (BER)
- Gathering material and manufacturing process information
- Test strategy, laboratory selection and agreeing the test plan
- Appraisal of test reports and linking them to the risk management file
Chemical characterisation and toxicological risk assessment
ISO 10993-1 expects the evaluation to start from the material. Chemical characterisation (ISO 10993-18) identifies the substances that can come out of the finished, sterilised device: extractables, and what is released under real conditions of use. The threshold below which the analysis will not report is set up front, and it is calculated from the duration of contact and the amount the patient is exposed to.
The sample sent for characterisation has to be manufactured and sterilised under production conditions; a result obtained from a prototype has to be repeated once production starts. Extraction solvents and times are chosen to ISO 10993-12 so that they reflect the real conditions of contact.
The substances found are interpreted through a toxicological risk assessment (ISO 10993-17). A tolerable exposure is derived for each substance and set against the estimated dose coming from the device. That route gains two things: unnecessary animal testing is avoided, and the evaluation can be updated quickly when a material changes. Endpoints that remain open go to testing.
The standards changed: irritation now has its own part
In 2021 ISO 10993-10 and ISO 10993-23 were separated. Sensitisation testing stayed in ISO 10993-10:2021, while irritation testing moved to ISO 10993-23:2021. The new part describes in vitro methods using reconstructed human epidermis (RhE) models in detail and expects that route to be tried before going to animal testing.
The practical consequence: if an older file covers irritation with an animal test alone, that may no longer be accepted as current on its own. ISO 10993-18 and ISO 10993-17 were revised in the same period. The file should state, part by part, which version of which standard the work followed; it is one of the first things a notified body looks at.
Standard versions are also kept current in the GSPR checklist in the technical documentation.
How do you choose a test laboratory?
The first thing to check when buying testing is the laboratory’s accreditation for the relevant standard, the second its experience with that device type. Sample numbers, extraction conditions, control materials and acceptance criteria should be settled in writing at the quotation stage.
The second point is the timetable. For prolonged contact and implantable devices some tests take months; if that time does not enter the project plan at the start, it will set the certification timetable for you. The third is the content of the report: a results table is not enough, deviations from the method and the raw data belong in it too — a notified body will ask for them.
Before samples are shipped, decide which configuration of the device will be tested. The worst-case variant within the product family is selected and the selection is justified in writing; otherwise the result covers only the variant that was tested.
Common problems
Relying only on a biocompatibility declaration from the raw material supplier is not enough. The evaluation has to be based on the finished, sterilised device. Processing aids, mould release agents and sterilisation residues can change the result. If the material or the supplier changes, the evaluation is updated as well.
