GB/T 14233: Linking the Chemical and Biological Test Routes

A method standard does not give acceptance requirements, and that is where the linkage starts

If you are running registration testing on syringes, infusion sets, transfusion sets or intravascular catheters, anything with a fluid path in contact with the patient, GB/T 14233 is unavoidable. Yet the test request companies send us often says nothing more than "test to GB/T 14233" and stops there. What the engineer has to establish on receiving the samples is precisely what that sentence leaves out: which product standard is the acceptance judgement made against? Is the work the chemical analysis part, the biological test part, or both?

GB/T 14233 is a method standard. It answers "how to measure", not "does the result pass". It is split by method category into a chemical analysis part and a biological test part, and wherever this article refers to a specific method it will say which part it belongs to. The standard numbers, titles and scopes of the two parts are governed by the current valid version of the standard text or the official standard information channels. Do not cite them from memory, and do not settle for writing only the parent standard number on the test request.

The acceptance requirements are distributed across the individual product standards: GB 15810 for sterile single-use syringes, GB 8368 for single-use infusion sets, YY 0285 for intravascular catheter products. The scope of each of those product standards, whether it is published in parts, and which part your product falls under, has to be confirmed one by one against the product construction and intended use, subject to the current valid version of the standard text. The product standard names the items to be performed and gives the corresponding requirements; the method standard supplies the operational detail. The two are used together, and with either side missing no valid conclusion can be issued.

Let us also set the scope of this article: it deals only with methodology and with how the two routes connect. It does not lay out the test-item list for any single product. Which items a particular product needs, and what the requirements are, is set by its own product standard. We have separate articles broken down by product line, and we do not enumerate them here so as to avoid any divergence from what the product standards actually say.

That structure determines every linkage question that follows. On the chemical analysis route, the items come from the product standard. On the biological route, the items are derived from the evaluation framework formed by the ISO 10993 family and the GB/T 16886 family, within which ISO 10993-1 gives the overall principles for evaluation and test selection, deciding which endpoints must be answered on the basis of the nature and duration of body contact. The two routes work from different briefs. Many companies plan their submission as if it were a single job, and that is where the trouble begins.

Extraction is not just soaking: the two routes have different logic

Almost every item in the chemical analysis part of GB/T 14233 is built on an extract: the fluid-path portion that contacts the patient is extracted in a specified medium under specified conditions, and the extract is then measured for acidity and alkalinity, reducing substances, metal ions, evaporation residue, ultraviolet absorbance and so on. The biological route is equally dependent on extracts: cytotoxicity, haemolysis and acute systemic toxicity are all run using an extract or by direct contact.

Both look like "liquid produced by soaking something", but the two extraction schemes are designed for different purposes. Chemical extraction exists to compare batch consistency and material cleanliness under a uniform set of conditions, with the medium, the vessels and the blank control all specified. Biological extraction has to cover leachables across a range of polarities, so it normally requires both a polar and a non-polar medium; the extraction ratio is set from sample surface area or mass, and the choice of conditions also has to match the clinical contact situation of the product.

A frequent misunderstanding is that one batch of extract will serve both sides, so only enough sample for the chemical items gets shipped. The biological tests then have to prepare their own extract at their own ratio, the sample runs out, more has to be sent, and the schedule slips. Worse still, some companies extract in their own plant and post the liquid to us. In most cases that cannot be used to issue a report: the extraction vessels, the container material and the standing time all affect the result, and the laboratory needs to carry out the preparation itself starting from the finished product in its final state.

Do not memorise the extraction numbers, memorise how to look them up

Extraction temperature, extraction time, and the ratio between sample surface area and the volume of extraction medium are the conditions most easily copied wrongly in the chemical items. This article gives no numerical values, not for reasons of confidentiality but because these conditions vary with product category, material system and the standard being cited. Apply one condition wrongly and the whole data set has to be discarded. The right habit is to remember the lookup path:

1. Fix the product standard you intend to work to. It determines which chemical items are performed and against what requirements they are judged.
2. Find the requirement clause for that item in the product standard and see where it points for preparation of the test solution.
3. Follow that pointer into the chemical analysis part of GB/T 14233 to the corresponding method clause. The extraction medium, the extraction conditions and the sample handling are all there.
4. The relationship between surface area or mass and the volume of extraction medium is specified in the same clauses. Where a product comes in several sizes, calculate separately for the fluid-contacting portion of each; do not let the ratio for one size cover the whole range.

The extraction conditions on the biological route do not come from the product standard. They are set by the method the evaluation relies on and must match the clinical contact situation. It is normal, not contradictory, for the same product to end up with different extraction conditions on the two routes.

One practical suggestion: before placing the order, send the laboratory the product standard text (current valid version) together with the construction drawing and let the engineer check the extraction location and conditions. That is far more reliable than filling in the test request from memory.

One table showing the two routes the same batch of samples has to travel

The two "sample condition" cells say the same thing, and that is not a layout error. It is precisely because the two routes ask for the same sample condition that companies so easily assume one shipment will do, overlooking the fact that the quantity logic behind each route is different. As for the actual value of every extraction condition in that table, go back to the text of the standard being cited. This article only records who specifies what.

Ethylene oxide residue: one item, two sets of limits

For a fluid-path product sterilised with ethylene oxide, the residue determination sits on the chemical analysis route, but its significance lands mainly on the biological risk side: the toxicological concerns associated with the residue relate directly to irritation, sensitisation and systemic toxicity. This is one of the most tangible crossing points between the two routes, and also one of the easiest places to mix up the frame of reference.

The first thing to get straight is which system the limit comes from. The two frames cannot be blended:

• The product standard frame. The product standard gives a residue limit in its own requirement clause, and the test report judges conformity against it. Different product categories and different product standards give different limits, so it has to be checked against the current valid version of the actual standard you are citing.
• The biological evaluation frame. The part of the ISO 10993 family that deals specifically with ethylene oxide residues expresses the limit as an exposure dose acceptable to the patient, tied to the contact-duration category, and serves biological risk assessment rather than product release. Which part specifies it and which edition applies is subject to the current valid version of the standard text and to your own evaluation route.

The two frames have different denominators: one is residue per device, the other a dose normalised over exposure time. They cannot be converted into one another, still less used to judge on each other's behalf. A typical error we have seen is checking a product-standard item against the dose limit from the evaluation system, concluding that there is plenty of margin, and only discovering when the report arrives that the product-standard requirement is exceeded. The reverse happens too. This article deliberately gives no numbers, precisely so that readers do not memorise a figure from one frame as a universal limit. Any value under either frame has to be verified against the corresponding standard text before use.

Beyond the frame question, a few operational points are worth noting. First, the sample submitted must be a fully aerated finished product in its original packaging. Some companies, chasing a deadline, ship samples straight off the sterilisation line; the residue naturally reads high, production has to be repeated and aeration waited out again, and the round trip costs far more time than the schedule they were trying to save.

Second, aeration conditions belong to sterilisation process validation and cannot be rescued at the testing stage. If residue repeatedly fails the product-standard requirement, go back to the sterilisation process and look at loading pattern, packaging breathability and aeration environment. That work can be arranged together with sterilisation validation and residue testing rather than by repeatedly submitting samples and hoping.

Third, ethylene oxide sterilised products often bring ethylene chlorohydrin into scope as well. Whether it has to be determined alongside depends on the product standard requirement and the material system, and it should be confirmed at the order stage so that a missing item is not discovered after the report is issued.

How chemical results feed back into the choice of biological items

Under the principles of ISO 10993-1 and the evaluation logic of the GB/T 16886 family, biological evaluation is not a matter of running every test. It means completing material characterisation and gathering existing data first, then judging which endpoints genuinely need new testing to answer. Chemical information carries real weight as an input here.

Evaporation residue and ultraviolet absorbance are often treated as nothing more than pass/fail indicators, but they carry more information than that. A high evaporation residue says the total non-volatile content of the extract is high. An anomaly in ultraviolet absorbance at particular wavelengths often points to organic species with conjugated structures, such as certain antioxidants, light stabilisers or their degradation products. If a signal of that kind departs noticeably from the historical data for comparable products, it is a prompt to go one step further towards chemical characterisation: qualitative and quantitative work along the lines of ISO 10993-18, with the outcome handed to toxicological assessment to decide whether the risk is acceptable.

The reverse also holds. Where chemical characterisation is done systematically and the toxicological argument is sound, some biological endpoints may be answerable through evaluation rather than new animal testing. Whether that route is viable depends on the completeness of the characterisation data, the change history of the materials and the specific requirements of the submission pathway. It is a case-by-case judgement and should not be planned from the outset as a cost-saving device. For the overall arrangement, see the item descriptions on the biocompatibility testing service page.

One linkage is easy to miss: when a haemolysis result is abnormal, do not stop at the base material. Look back at processing aids, mould release agents and lubricating coatings introduced by the process. On the chemical route these may show up as elevated evaporation residue, yet the chemical item alone may still sit within requirements. Only by reading both routes together do you get a direction for investigation.

Extraction seen from materials and mass transfer (mechanistic analysis, not a statistical conclusion)

Extraction is essentially a mass-transfer process: small molecules diffuse from inside the polymer matrix to the interface, then dissolve into the extraction medium. The amount released is governed jointly by the diffusion coefficient within the matrix, how well the polarity of the analyte matches that of the medium, the accessible surface area, and the contact time. Understanding this explains a lot of what you see on the floor.

For instance, within the same formulation the thin-walled sizes usually approach release equilibrium earlier than the thick-walled ones, simply because the diffusion path is shorter. Similarly, a non-polar additive has limited release into an aqueous medium, so a "very clean" result measured in an aqueous system does not mean the product would look equally clean under extraction with an organic medium. That is exactly why the biological route calls for media of differing polarity.

From this follows a practical warning: do not tighten the extraction conditions on your own initiative to be "on the safe side". Raising the temperature does speed release, but it may also change the state of the matrix, drawing in components that would never have migrated, and even introducing thermal degradation products. The result then conforms neither to the method as specified nor to real clinical exposure. Any tightening of conditions should be supported by a standard or by argument, not decided from experience. The reasoning above comes from general principles of materials science and mass transfer; it is there to help interpret observations and set the direction of an investigation, and it does not replace test data generated to the standard.

Documentation and sample condition to prepare before submission

Incomplete documentation blocks jobs more often than the samples do. Get the following in order before placing the order:

• A statement of intended use, spelling out the site of body contact, the mode of contact and the duration of contact. This is the basis on which biological items are selected.
• The product standard you intend to work to (with the current valid version noted), plus your own product technical requirements. Without this, chemical testing can only produce data, not a judgement.
• A materials list covering base materials, colour masterbatch, plasticisers, antioxidants, lubricants and coating materials. The more specific it is, the easier the results are to interpret.
• Sterilisation method and process information; for ethylene oxide sterilised products, state whether aeration is complete.
• A list of the components in contact with the fluid path and a construction sketch, used to fix the extraction location and the surface area calculation.
• The samples themselves: sterilised finished product, in original packaging, from one batch, with margin left for retest and retains.

Do not calculate sample quantity on the basis of "enough to run once". Each route needs its own samples, and once retests and retains are added the real requirement is usually well above a company's initial estimate. Checking the sample list with the laboratory before ordering is much less trouble than sending more later. For a more detailed breakdown see the testing requirements page, and for scheduling see the testing process page.

Situations that commonly lead to rework

• The test request names only the method standard, not the product standard. The report can carry data but no acceptance conclusion, and in registration files that is often not accepted. The request should also state whether the chemical analysis part, the biological test part, or both are required.
• Self-checking against the wrong limit frame. Ethylene oxide residue especially: the product standard frame and the biological evaluation frame are not interchangeable, and using the wrong one shifts the whole conclusion.
• Submitting unsterilised samples or semi-finished parts. Sterilisation itself changes the state of the material and the leachables profile, so data from unsterilised samples does not represent the finished product.
• Repeating only the chemical items after a formulation or supplier change. Whether a change triggers biological re-evaluation is judged under the evaluation framework; passing the chemical items does not automatically keep the biological conclusion valid.
• A stated contact duration that does not match the submission. The contact-duration category directly determines the scope of biological work, and an inconsistency will be raised during technical review.
• Mixing sizes in one shipment but calculating the extraction ratio for only one. Different sizes have different contact areas, so ratios are calculated separately; using a representative size to cover the rest requires justification.
• Treating in-house data as a formal conclusion. In-house data is fine for development screening and internal control, but issuing a report still requires the work to be repeated under controlled conditions.

What SUNGO Lab can do

SUNGO Lab (Shanghai Shage Medical Technology Co., Ltd.) operates laboratories in Shanghai and Hefei and is accredited by CNAS, CMA and IAS (USA). It should be stated clearly that an accreditation mark only demonstrates that the laboratory holds the corresponding technical competence within its accredited scope; it does not constitute a commitment as to the outcome of market access in any target market. Registration and market-access conclusions remain subject to the review opinion of the receiving authority.

For injection and infusion products we can carry out testing under both the chemical analysis part and the biological test part of GB/T 14233, issue acceptance conclusions against product standards such as GB 15810, GB 8368 and YY 0285, and arrange biocompatibility evaluation within the framework of the ISO 10993 and GB/T 16886 families, supporting toxicological risk assessment with chemical characterisation along the lines of ISO 10993-18 where required. Sterilisation residue, packaging and shelf-life items can be planned onto the same batch of samples to cut down on repeat submissions. Our full scope is set out on the services page.

If you have a product and are unsure how to split the items, send us the construction drawing and the materials list. An engineer will first work through the items on both routes, where the extraction conditions come from and how much sample is needed, and only then discuss pricing. Call +86 132 4819 8029 or request a quote.