The most dangerous phrase in medical device supply chains is "it's an equivalent grade." A moulder switches from one polypropylene to another because the original is on allocation, the price moved, or a distributor consolidated stock. The parts look identical, they pass dimensional checks, and production sees no reason to escalate. Eighteen months later a clinical complaint arrives: cracking under a specific cleaning agent, or irritation traced to a patient-contact surface.
The gap between those two events is where most material change failures live. Biocompatibility is not a property of the drawing or the specification; it is a property of the exact formulation, its additives, its processing, and its extractables profile. Change any one of those and the ISO 10993 test reports sitting in the technical file describe a material that no longer exists.
Regulators understand this well, and so should every supplier quality engineer with medical or drug-delivery exposure. The discipline required is not more testing — it is a defined retesting logic, contractual change control that reaches the resin bag, and a documentation chain that connects the substitution to the biological conclusion.
What Equivalent Actually Means
Material equivalence has a precise meaning that salespeople routinely blur. A datasheet match on tensile strength, melt flow index and density tells you the material will probably mould and perform mechanically. It tells you nothing about the additive package: antioxidants, slip agents, plasticisers, colourants, process aids. Two resins with near-identical mechanical datasheets can carry completely different stabiliser systems, and extractables testing will find the difference immediately.
When reviewing a supplier's change notification, ask for the full formulation declaration, not the datasheet. If the supplier is a distributor and cannot disclose the additive package, treat the change as a new material until proven otherwise. If the distributor has no contractual obligation to notify you of plant or country-of-origin changes, that is the hole through which a substitution enters your device silently.
Check whether the change stays within the same polymer family but moves to a different manufacturing site. Even identical grades produced at different plants can differ in residual monomers and catalyst residues. Then read the change notification letter itself: does it declare "no change to formulation", or only "no change to nominal properties"? Those are not the same statement, and the second is a red flag.

Retesting Triggers: A Practical Hierarchy
A change notification should trigger a defined amount of biocompatibility work, not a debate. A change of polymer grade, manufacturer or manufacturing site triggers extractables and leachables work plus a cytotoxicity screen at minimum. A change to colourant, additives or masterbatch carrier triggers the same, plus sensitisation and irritation consideration, because pigments and their carriers are a classic source of contact reactions.
A change to the moulding process itself can also change the extractables profile without touching the resin at all. Higher melt temperatures, a new mould release agent, or a different sterilisation cycle all shift the chemistry. Thermal degradation products from an overheated melt, or residue from a new release agent, will show up in a chemistry screen long before they show up in any mechanical test.
Process changes on patient-contact components therefore deserve the same chemical review as material changes. The evaluation should also confirm the contact category — tissue, blood, or intact skin with limited exposure — because the ISO 10993 endpoint set differs. Retesting scope should follow the biological risk, not a blanket re-run of everything.
Change-to-test decision path
- 01Classify the changeGrade, manufacturer, site, additive, masterbatch, or process parameter — each has a different trigger.
- 02Confirm contact categoryTissue, blood or intact skin determines the ISO 10993 endpoint set.
- 03Run chemistry comparisonExtract incumbent and candidate under identical conditions; compare by GC-MS and LC-MS.
- 04Assess new extractablesIdentify unknown peaks and have a toxicologist judge relevance in writing.
- 05Set retesting scopeDocument the reduced or full package in the biological risk assessment.
Where Notifications Get Lost
Most material changes become clinical issues not because the change was inherently dangerous, but because the notification never reached the right people. Distributors substitute a grade at the point of sale with no notification at all, on the argument that the part number on the resin bag is unchanged. Contract manufacturers sometimes accept a "drop-in equivalent" from a machine operator responding to a stockout, and the deviation record never mentions the resin.
Supplier agreements must explicitly define a material change as a notifiable event requiring written approval before use, with the formulation declaration attached. Anything softer — "notify significant changes", "inform of major revisions" — leaves the judgement to the party with the least incentive to interrupt production.
Audit for this behaviour directly. Walk to the presses and read the resin bags against the approved material list; this is where undocumented substitutions surface. Check incoming inspection records: is anyone verifying the resin grade and lot, or just the part dimensions? Most documented substitutions I have found in audits were standing in plain sight next to the machine.
Characterise Before You Test
Before committing to a full biological evaluation, run a chemistry-based comparison. Extract both the incumbent and the candidate material under identical conditions — same solvent polarity range, same temperature, same time — and compare the extractables profiles using GC-MS and LC-MS. If the profiles match at a practical level, you have a defensible scientific argument for a reduced retesting package, documented in the biological risk assessment.
If new peaks appear, identify them and assess toxicological relevance. Do not assume the supplier's certificates carry this work: a certificate of compliance states what the supplier tested against their own release specification, which rarely includes extractables. It is a statement about conformity to a spec, not a statement about biological risk.
Verify the stabiliser system against migration risk as well. Some antioxidants are well characterised and unremarkable; others raise flags depending on contact duration and tissue type. When in doubt, involve a toxicologist who can judge whether a new extractable is a genuine concern or analytical noise. That judgement, in writing, is what a notified body or reviewer will expect to see.
A part can be dimensionally perfect and biologically different. Dimensional inspection will never close that gap.
Verification, Validation and the Paper Trail
Once the change proceeds, the verification package must connect three things: the material declaration, the processing evidence, and the biological conclusion. Dimensional and functional testing alone will not carry the argument. Include moulding window evidence showing the new resin is processed within a comparable melt and thermal regime, because processing variation can shift extractables independently of the resin itself.
Where sterilisation is involved, re-verify that the new material survives the validated cycle without generating new degradation products. A resin that moulds cleanly and passes dimensions can still produce extractables that only appear after an ethylene oxide or gamma cycle. The sterilisation step is part of the material system, not an afterthought to it.
Keep the whole chain in one place: the change notification, the formulation comparison, the chemistry data, the toxicological rationale, and the signed decision on retesting scope. When an inspector asks "how do you know the material in the field matches what you tested", you should be able to answer with one document. I have watched otherwise competent companies fail on exactly this question, because the biological evaluation sat in a separate folder from the change record and nobody connected them after the fact.
The verification gap
What conventional checks confirm
- Dimensions within tolerance
- Mechanical properties per datasheet
- Cosmetic and functional acceptance
- Conformity certificate from supplier
What the change actually affects
- Additive package and stabilisers
- Extractables and leachables profile
- Post-sterilisation degradation products
- Traceability to the tested formulation
Closing the Loop with Traceability
The final discipline is lot-level traceability from the finished device back to the resin lot. When a complaint arrives, it means you can determine within hours whether the affected units were moulded before or after a change, from which resin lots, and at which site. Without it, you face a full complaint investigation across unknown material populations while the regulatory reporting clock runs.
Treat field intelligence as a change detector. A cluster of complaints about cracking, discolouration or irritation on a product that has been stable for years is often the first signal that an unnotified change occurred upstream. Investigating the complaint means investigating the material, not just the device: demand the resin lot records, the moulding parameters, and every deviation logged in the suspect period.
More than once in my career, that line of questioning has uncovered a substitution nobody had declared — and turned a mysterious clinical issue into an identifiable, correctable material change. That is the real test of a change control system: not whether it catches the notified changes, but whether it can reconstruct the ones that bypassed it.
