Two guidelines, a single difference in figures, and the same biological model: it's no wonder that OECD492 and OECD492B are frequently confused. Yet, they don't answer the same question, and using the wrong method generally results in a second attempt and several weeks of delay.
This article explains what really separates them, and provides a decision rule applicable to your product.
Table of Contents
What the two methods have in common
Both guidelines are based on the same biological principle: a reconstituted human cornea - like epithelium (RhCE ). This is a three-dimensional tissue obtained from human corneal epithelial cells, cultured at the air-liquid interface until they form a stratified structure that reproduces the corneal barrier.
In both cases:
- the product is applied directly to the surface of the fabric;
- Tissue damage is assessed by measuring cell viability using the MTT test ;
- no animals are used — neither in vivo, nor in the form of tissue samples, unlike the BCOP (bovine corneas) or ICE (chicken eyes) methods;
- the results feed into the classification under the CLP regulation.
Up to this point, everything brings them together. The difference lies elsewhere.
The fundamental difference: the question asked
This is the key point to remember above all else. The two methods are not trying to address the same issue.
OECD 492 poses a closed question
Entitled "Method of test on reconstituted human corneal epithelium for the identification of chemicals not requiring classification and labeling for eye irritation or serious eye injury", it answers a single question with yes or no: can this product be excluded from any classification?
- Viability above the threshold → product identified as unclassified. The case is closed for this reason.
- Viability below the threshold → the product cannot be ruled out. And that's all. The method doesn't indicate whether it's a simple irritation or a serious injury.
This is called an exclusion method, used in a bottom-up. A negative result closes the case; a positive result opens it.
OECD 492B poses an open question
Entitled "test method on reconstituted human corneal epithelium for theidentification of ocular hazards ", it answers: in which hazard category does this product fall?
His three possible answers cover all cases:
- Category 1 — serious eye damage (H318);
- Category 2 — eye irritation (H319);
- Unclassified.
It is an autonomous method : whatever the result, it concludes.
How does the 492B manage to discriminate?
The answer can be summed up in three words: time-to-toxicity, or "time until toxicity".
OECD 492 measures viability after a single exposure period. It therefore obtains a measurement point — sufficient to say "the tissue is fine", insufficient to characterize the severity when it is not.
OECD 492B exposes tissue for several increasing durations and measures viability after each. The biological intuition is sound: a product causing severe eye damage destroys tissue quickly, while a simple irritant only alters it after prolonged exposure. It is the kinetics of the damage that carry the classification information, not just its final extent.
Specifically, the validated model (SkinEthic™ HCE TTT) includes two protocols:
- liquids : three exposures — 5 minutes on pure product, then 16 and 120 minutes on 20% (w/v) dilution;
- solids : two exposures on finely ground pure product — 30 and 120 minutes.
The combination of viability values obtained at each time point is then read into a prediction model which assigns the category.
The complete comparison
| OECD 492 |
||
|---|---|---|
| Object | Identify products that do not require classification | Identify the eye hazard |
| Question asked | Can we disregard the classification? | What category of danger? |
| Tissue model | RhCE — EpiOcular™, SkinEthic™ HCE and other validated models | RhCE — SkinEthic™ HCE TTT |
| Exposure time | Just one | Several (2 or 3 depending on the shape) |
| Possible conclusions | Unclassified / undetermined | Cat. 1 / Cat. 2 / without classification |
| Distinguish between Cat. 1 and Cat. 2? | No | Yes |
| Autonomous method? | No — further testing if positive result | Yes |
| Physical forms | Liquids, solids, semi-solids, waxes | Liquids, solids, semi-solids, waxes |
| Adoption | 2015 | 2022 |
| Current version | June 2025 | June 2024 |
Are you looking for an analysis?
So, which one to choose?
The right method depends on what you already know about your product and what you need to demonstrate.
Choose OECD 492 if…
- You have good reason to believe that your product is non-irritating — similar formulation history, existing data, well-characterized ingredients;
- Your objective is to quickly confirm a lack of classification ;
- you test a large number of samples in screening, where the unit cost is paramount;
- Your laboratory or historical record already relies on this method and a given model.
Choose OECD 492B if…
- the behavior of your product is unknown or uncertain ;
- You already know that it is probably classified, and what you are looking for is the category ;
- you need to label and therefore must decide between H318 and H319;
- you want to avoid a second attempt and secure your regulatory schedule;
- You are testing a complex formulated mixture, semi-solid or waxy.
The simple rule
Ask yourself just one question: if the result is positive, will I need to know the category?
If the answer is yes — which is the case as soon as labeling is required — go directly to 492B. Following up a 492 with a complementary method almost always costs more, both in money and time, than using 492B from the outset.
If you are reasonably confident that there will be no classification, the 492 remains the fastest and most economical route.
Three common mistakes
Interpreting a positive result of 492 as "irritant product"
This is the most common mistake, and it appears in many technical documents. Viability below the threshold simply means that the product cannot be excluded from classification. It may fall under category 2, category 1, or be found to be unclassified after further testing. Writing "viability < threshold = eye irritant" in a report is an overinterpretation.
To believe that the 492B replaces the 492
The two guidelines coexist and both remain in effect. 492B does not supersede 492; it addresses a different need. For large-scale exclusion screening, 492 remains entirely relevant.
Ignoring interference with the MTT test
A highly colored product, or one capable of spontaneously reducing MTT, skews the viability reading—in both methods. This is a known limitation, managed by additional controls specified in the guidelines, but it must be anticipated before the study begins, not discovered upon reviewing the results.
Where do these methods fit within the family of eye tests?
Eye irritation is not limited to these two guidelines. The OECD proposes several, which can be used depending on the chosen strategy:
- OECD 405 — acute irritant or corrosive effect on the eyes, in vivo , now very regulated and prohibited in cosmetics;
- OECD 437 (BCOP) and OECD 438 (ICE) — ex vivo methods on bovine corneas and chicken eyes, used in particular to identify category 1;
- OECD 491 (STE) — short-term exposure test on cultured cells;
- OECD 492 and OECD 492B — the two RhCE methods compared here;
- OECD 496 — in vitro macromolecular method .
These methods are combined within an integrated testing and evaluation approach (IATA), which aims to reach a conclusion with the minimum number of tests required. This is where the initial choice is most crucial: a well-constructed strategy avoids two or three unnecessary studies on a complete dossier.
Frequently Asked Questions
Is there an OECD 492A?
No. There are only two guidelines: OECD 492 and OECD 492B. The suffix "B" was added to the new method to distinguish it from the original, without renaming the latter.
Is OECD 492B accepted in Europe?
Yes. This is an adopted OECD guideline, recognized under the mutual acceptance of data. Its use in a given regulatory file depends on the applicable regulations and the evaluation strategy adopted.
Can we test a finished cosmetic product?
Yes, both methods cover mixtures. Method 492B is often preferred for complex, semi-solid, or waxy formulations because it provides a definitive conclusion without further testing.
How much time should be allowed?
The experimental phase lasts for both methods, measured in days. The real difference in timeframe arises in the event of a positive result: the 492 then requires a second study, with a new schedule and a new waiting period.
Is it still necessary to conduct an in vivo trial?
In cosmetics, no: Regulation (EC) No 1223/2009 prohibits it. In other sectors, in vitro are preferred by regulators, andin vivo remains only in residual, duly justified situations.
Have your tests done with YesWeLab
The choice of method is made before the trial, not after. We work with you to define the most cost-effective strategy in light of your regulatory objective, and then we mobilize the appropriate laboratory within our network.
Discover our two dedicated fact sheets: OECD 492 – In vitro eye irritation (RhCE) and OECD 492B – Identification of eye hazards, or consult our entire catalogue of eye irritation tests.
