Medical-Grade Silicone: USP Class VI, ISO 10993 and the Paper Trail Behind Both

What medical grade actually means

There is no single medical grade of silicone. There are materials that have been tested against a set of standards, processed under a quality system, and documented well enough for a device manufacturer to include them in a regulatory submission.

The confusion comes from the fact that most suppliers quote USP Class VI, which is one test, and buyers treat it as a complete qualification. It is a screening test for acute biological response, and it says nothing about long-term implantation, extractables at temperature, or the consistency of the next batch.

What actually qualifies a silicone for a medical device is: a tested material, a validated process, a sterilisation route that has been shown not to degrade it, and a change control system that prevents silent substitution. Three of those four are about documentation.

USP Class VI versus ISO 10993

USP <88> Class VI is a set of in vivo tests: systemic injection in mice, intracutaneous reactivity in rabbits, and implantation in rabbits. It is a pass or fail screening test that indicates the material does not cause an acute reaction. It does not characterise the material.

ISO 10993 is a framework, not a test. Part 1 requires a biological evaluation plan that identifies what tests are needed based on the nature and duration of contact. Part 5 covers in vitro cytotoxicity, Part 10 covers sensitisation and irritation, Part 6 covers implantation, Part 4 covers blood interaction.

The difference matters practically. A silicone intended for a 30-day blood contact device needs a different test set from one intended for a 24-hour skin contact device, and neither is covered by quoting USP Class VI. The evaluation plan is the document that decides, and it belongs to the device manufacturer, not the material supplier.

Table 1 — What each standard covers. A device submission needs a plan that selects among them, not a single certificate.

The test matrix for a typical device

For a silicone component in a short-term, indirectly contacting device, the usual set is cytotoxicity (10993-5), sensitisation and irritation (10993-10), and a chemical characterisation (10993-18) to identify extractables. For a long-term implant, add implantation (10993-6), genotoxicity, and a full extractables and leachables study.

Extractables testing is where most programmes underestimate the work. It requires extracting the finished, sterilised device in multiple solvents at multiple temperatures, identifying the compounds, and then assessing toxicological risk for each. For silicone, the compounds that appear are cyclic siloxanes, residual catalyst fragments and antioxidant residues.

The result of that assessment drives the material choice. A cheaper peroxide-cure compound may produce aromatic fragments that fail the toxicological risk assessment, while a platinum-cure compound at twice the price passes. The material cost difference is small against the cost of repeating a submission.

Sterilisation: four routes and what each does

Silicone tolerates all four common sterilisation routes better than any other elastomer. That is one of its main advantages, and it is why a single material can serve devices sterilised by different methods in different markets.

• Steam at 134 °C. The default for reusable devices. Silicone is unaffected, but the cycle count matters: 100 cycles is a different qualification from 1,000.
• Dry heat at 160 °C. Longer cycle, no moisture. Silicone tolerates it; the packaging may not.
• Ethylene oxide. The standard for single-use devices that cannot take heat. EtO residues must be aerated out, and silicone absorbs EtO more than most polymers, so the aeration time is longer.
• Gamma at 25 kGy. Fast and penetrating. It can cause some crosslinking and a slight colour shift in silicone, and it can affect the cure state of an under-cured part.

The qualification requirement is that the sterilisation route is validated against the specific part, not against the material family. A part that passes after three steam cycles may fail after fifty, and the number of cycles in the intended use is what has to be tested.

The documentation that has to exist

Table 2 — The six documents a medical silicone component needs. The last one is the one most often missing and the one that causes recalls.

Change control is where medical silicone programmes fail. A compound supplier changes a filler source, or a moulder changes a pigment supplier, and the device is now made from a material that was never tested. Nothing in the part looks different, and the regulatory position is invalid.

The control is a written agreement that any change to material, process, site or sterilisation requires notification and re-qualification. It sounds bureaucratic and it is the only thing that prevents a silent change.

Common gaps

The first gap is a certificate for the wrong article. A biocompatibility report on a 2 mm test sheet is not a report on a 0.5 mm moulded seal with a different cure state.

The second is a cure state that was never verified. Under-cured silicone has different extractables from fully cured silicone, and the difference shows in a leachables study rather than in a visual inspection.

The third is a sterilisation route that was validated on a prototype made in a different tool. A change of tool changes the surface and the cure, and the sterilisation validation has to follow.