ISO 10993-23:2021 Biological evaluation of medical devices – Part 23: Tests for irritation assesses the possible contact hazards from medical devices which can produce irritation when in contact with human tissue.

In our topical article below, our Head of Operations Bastian Perroset summarises the key areas of the document and shares his thoughts on what this standard subpart signifies for the industry on a broader level regarding biological testing for medical devices moving forward…

What’s the context behind this latest 10993 subpart?

The subpart ISO 10993-10: 2010: Biological evaluation of medical device – Part 10: Tests for irritation and skin sensitization has this year been split into:

• ISO 10993-10:2021: Biological evaluation of medical devices – Part 10: Tests for skin sensitization and
• ISO 10993-23:2021: Biological evaluation of medical devices – Part 23: Tests for irritation

This article specifically looks at ISO 10993-23:2021, outlining some key elements regarding what’s included in the standard. We will look at the types of materials and testing methods covered and consider the impact of this document in terms of how it marks a continued paradigm shift from in vivo to in vitro testing.

What’s included in the standard?

ISO 10993-23:2021 specifies the procedure for assessing medical devices and their constituent materials regarding their potential to produce irritation. It aims to be the basic document for the selection and conduct of tests enabling evaluation of irritation responses relevant to the safety of medical materials and devices. It’s worth noting however, that skin sensitization is not addressed in this standard.

The tests outlined in the standard are designed to predict and classify the irritation potential of medical devices, materials, or their extracts according to ISO 10993-1 and ISO 10993-2, including:

• Pre-test considerations for irritation, including in silico and in vitro methods for dermal exposure
• Details of in vitro and in vivo irritation test procedures
• Key factors for the interpretation of the results

The standard describes a stepwise approach that includes one or more of the following:

• Chemical characterization, supplemented where required with chemical testing of samples in accordance with the general principles specified in the ISO 10993 family
• Literature review, as indicated in ISO 10993‑1, including an evaluation of chemical and physical properties, and information on the irritation potential of any product constituent as well as structurally-related chemicals and materials
• In vitro alternative testing using validated RhE as per the methods outlined in sections 6.2 to 6.12 of the standard
• In vivo animal testing

It also notes that clinical studies according to ISO 14155 and ethics principles governing human clinical research are not to be performed before the irritancy potential of a device has been established through one or more of the evaluations described above.

Which types of material are included?

As the standard does not cover an exhaustive list of possible materials used within medical devices that may cause adverse effects, it’s important to note that the obligation to evaluate each individual device for potential adverse effects prior to marketing sits with the individual device manufacturers.

Furthermore, additional chemical components that may be used as processing aids, e.g., lubricants or mould-release agents should be taken into consideration. Bear in mind that in addition to the chemical components of the starting material and manufacturing process aids, adhesive/solvent residues from assembly, sterilant residues or reaction products resulting from the sterilization process can be present in a finished product.

The following types of materials are often used in medical devices and could introduce risks for irritation:

• Ceramics, metals, and alloys (which are normally less complex than polymers and biologically derived in terms of the number of chemical constituents)
• Polymers (which often pose a higher risk than ceramics, metals, and alloys, because they’re composition is normally more chemically complex – several reaction products, impurities and additives can be present, and the completeness of polymerization can vary)
• Biologically derived materials (which are inherently complex in their composition and can be inconsistent from sample to sample)

Information on chemical composition is therefore especially important. To this end, the standard notes that qualitative and quantitative information on the starting material composition, and qualitative information on any additional processing additives needs to be obtained from the relevant parties involved in the manufacturing chain. And in the absence of any data on composition, the standard recommends a literature search.

Testing methods

The irritation potential of a medical device or its components can be predicted either with an in vitro irritation test (if qualified for use with medical devices) or through an in vivo animal irritation test.

Before we start outlining the testing methods included in the standard, it’s important to emphasise that ISO 10993-23 also states that pre-test considerations can result in the conclusion that testing for irritation is not necessary. For example, if the pH of the test sample is ≤2,0 or ≥11,5 the material shall be considered an irritant and no further irritant testing is required.

The standard describes tests to determine the irritancy of medical devices, materials, or their extracts either by in vitro or in vivo tests, with in vitro tests preferred when they can be appropriately validated through the provision of information equally relevant to what can be obtained from in vivo testing.

In recent years, in vitro tests have been added as adjuncts or alternatives. For the testing of skin irritation from neat chemicals, in vitro tests were developed using reconstructed human epidermis (RhE) models. The results of a large round robin study that tested two types of RhE models showed that these models can also be used to detect the presence of irritant chemicals extracted from polymeric materials.

This method was found to be equally sensitive in the detection of low concentrations of some strong irritant compounds when compared to the human patch testing and intracutaneous rabbit test. As such, the standard outlines that a stepwise approach for irritant testing can start with the in vitro RhE model.

For more information on the test methods within the standard, see:

• Annex A-C for instructions on the preparation of materials specifically in relation to the above tests
• Annex D for a description of several special in vivo irritation tests for the application of medical devices in areas other than skin
• Annex E for information on conducting human skin irritation testing

In vivo irritation tests

Regarding in vivo irritation tests, the standard notes that “irritation testing of either medical devices, components of medical devices or biomaterials, or all, can be performed with either the devices, device components, biomaterials themselves or extracts thereof, or all.”

Solid materials can be applied directly onto the skin, whilst for liquid and porous materials, pads should be used. For medical devices that are contacting breached or compromised surfaces, externally communicating, or used as an implant, intracutaneous (intradermal) administration is instructed.

Human skin irritation tests

ISO 10993-23 indicates that potential for irritation risk assessments may also be assessed using data from human skin patch tests, with clinical tests to be performed in accordance with ISO 14155.

Human studies can serve several purposes including:

• directly identifying hazards for humans by testing chemicals in humans rather than in laboratory animals
• facilitating extrapolation of human data obtained previously from laboratory animal studies

However, materials should not be tested on humans if:

• they have been shown to be an irritant or corrosive in a predictive assay, either in vitro or in vivo
• potential corrosivity for human skin can be predicted either based on structure/activity relationships or physicochemical properties such as strong acid or alkaline reserve, or both
• they present a risk of skin or respiratory tract sensitization
• they present any acute toxicity hazard under test conditions
• they present any genotoxic, reproductive, or carcinogenic hazard

A continued paradigm shift from in vivo to in vitro

Lifting from the detail, on a broader level this standard subpart supports a continued change in thinking for the industry from in vivo to in vitro testing, by demonstrating a continued progression towards a preference for in vitro tests:

“Preference for in vitro tests instead of in vivo tests in accordance with ISO 10993-2 shall be considered, with replacement of the latter as new in vitro tests are scientifically validated and qualified for use with medical devices and become reasonably and practicably available.”

This reflects the industry’s gradual move away from in vivo as the first method of choice, building on an increased emphasis on animal welfare reflected in the ISO 10993-2:2006 requirement regarding justification for animal tests and minimization of harm to animals. The uplift in the implementation of the 3Rs (reduce, replace, refine) and other guiding principles for more ethical use of animals in testing is also evidence of this progression. Furthermore, with skin sensitization being one of the most dynamic fields in in vitro toxicology, the advances being made in cell culture methods are promising.

We’re also starting to see an improved regulatory acceptance of RhE models globally. The EU considers ISO 10993-23:2021 as “state of the art”, with in vitro testing considered the first-choice method. In Japan, in vitro testing with the RhE model is already included in the MHLW guidance, with the testing method considered the primary choice.

That said, the US and China are behind the curve. The standard is only partially recognised by the FDA since Clause 4, Clause 6, Annex B, Annex C conflict with an existing FDA published guidance. And in China the official YY medical device standard version of ISO-10993 needs to be published by the NMPA (former CFDA) before acceptance of in vitro irritation data becomes commonplace.

Whilst the primary role of this standard subpart is to serve as a framework for planning a biological evaluation, its secondary aim is to utilise scientific advances to minimise animal testing by giving preference to in vitro models and chemical, physical, morphological, and topographical characterization testing (when these methods can yield equally relevant information as obtained from in vivo models). The standard is another “step in the right direction” – what’s needed now is its global acceptance and adherence.

Should you have a challenge relating to this standard, or Biocompatibility more generally, our Operations team is ready and happy to help. Simply get in touch to start the conversation.