Eye Irritation Testing: The Way Forward

The Report and Recommendations of ECVAM Workshop 341,2

Reprinted with minor amendments from ATLA 27, 53-77.

Appendix 1

An Evaluation of the OECD Proposal for the Harmonised Classification of Eye Irritants and Corrosives

Menk K. Prinsen

TNO Nutrition and Food Research Institute, Division of Toxicology, 3700 AJ Zeist, The Netherlands

Summary - Classifications of eye irritation/corrosion were assigned to 59 of the chemicals used in the European Commission/British Home Office (EC/HO) validation study by applying both European Union (KU) criteria and the harmonised criteria proposed by the OECD. It was found that the application of the two classification systems to the 59 chemicals resulted in comparable classifications: all of the chemicals class)fied as R36 or R41 according to EU criteria were classified as Category B or Category A, respectively, according to the proposed OECD criteria. Only two of the 59 chemicals, ethanol and ethyl-2-methylacetoacetate, were classified differently by the two systems, they were unclassified by the EU system, and were classified as Category B by the OECD system. It is concluded that: a) the proposed OECD classification system is broadly equivalent to the EU classification system, and b) future validation studies on alternatives to the Draize test would benefit from the application of the OECD classification system to the test chemicals.

Introduction

The failure of recent validation studies (1, 2) to find a suitable replacement to the Draize test is partly a result of the statistical method chosen for evaluating the performance of the in vitro tests The relevance of these methods was assessed by correlating the in vitro test scores with the Modified Maximum Average Draize Test Score (MMAS), which is problematic because the MMAS shows considerable variability, particularly in the middle of the irritancy range. The use of the MMAS as the in vivo endpoint in an international validation study could also be regarded as inappropriate on the grounds that most regulatory systems (for example, the European Union [EU], US Environmental Protection Agency, US Food and Drug Administration, and Canadian workplace systems) do not classify chemicals on the basis of their MMAS values, but according to their effects in individual tissues of the eye (conjunctive, cornea and iris), taking into account the recovery from or irreversibility of these effects. However, it would be difficult to accommodate all of the different classification systems when conducting a validation study. Fortunately, this should soon be unnecessary, since an OECD proposal for the global harmonisation of criteria for the classification of eye irritants (3, 4) is in its final phase of acceptance. The aim of this study was to examine the effect of applying the OECD criteria to chemicals which have already been classified according to EU criteria (5).

Materials and Methods

The animal data for 59 of the European Commission/British Home Office (EC/HO) chemicals were taken from the ECETOC reference chemicals data bank (6). These data were used to classify the 59 chemicals according to both EU criteria (5) and the proposed OECD criteria (3, 4), which are summarised in Tables I and II, respectively. Chemicals which could not be classified as irritant or corrosive were arbitrarily chosen as means of classify to the eye were classified as non-irritant (NI).

Table I: European Union Classification System for Eye Irritation/Corrosion

  R36
(irritating to eyes)		 R41
(risk of serious damage to eyes)1
Effect Three animalsb Six animalsc Three animalsb Six animalsc
Corneal opacity > 2.0, but < 3.0 > 2.0, but < 3.0 > 3.0 > 3.0
Iris lesion > 1.0, but < 2.0 > 1.0, but < 1.5 > 2.0 > 1.5
Conjuctival redness > 2.5 > 2.5    
Conjuntival chemosis > 2.0 > 2.0    

aA classification of R41 is also assigned if one of the eye effects has not reversed at the end of the observation period in at least one animal.

bThe classification is assigned if the mean tissue effect (averaged over the 24-hour, 48-hour and 72-hour periods) is greater than or equal to the threshold value in at least two of the three animals. In this study, the same criteria were applied if four rabbits were used.

cThe classification is assigned if the mean tissue effect (averaged over the three periods and over the six animals) is greater than or equal to the threshold value.

Table II: Proposed OECD Classification System for Eye Irritation/Corrosion

Effect Category Ba Category Ab
Corneal opacity > 1.0 > 3.0
Iris lesion > 1.0 > 1.5
Conjuctival redness > 2.0
Conjuntival chemosis > 2.0

aAll effects have to be reversible within 21 days of treatment. The subcategory of B1 can be used for chemicals considered to be mildly irritating to the eyes, i.e. chemicals whose eye effects are reversible with 7 days of treatment.

bCategory B is also applicable if an eye effect has not reversed, or is expected to reverse, within 21 days of treament in at least one animal.

In addition, the 59 chemicals were divided three-fold categorization of chemicals was into three groups: MMAS < 25; 25 < MMAS < 59; and MMAS > 59. These cut-off values were arbitrarily chosen as means of classifying chemicals into three groups. This three-fold categorisation of chemicals was compared with the EU classification of chemicals (NI/R36/R41).

Results

The EU and OECD classifications for the 59 chemicals, ordered in terms of decreasing MMAS, are given in Table III.

Table III: Classification of Eye Irritation/Corrosion for 59 Chemicals Obtained by Applying European Union (EU) and OECD Criteria

  In Vivo Data Classificationa
No. Test Chemical Referenceb MMAS EU OECD
1. Sodium hydroxide (10%) 82 108 R41 A
2. Benzalkonium chloride (10%) 186 108 R41 A
3. Trichloroacetic acid (305) 36 106 R41 A
4. Cetylpyridium bromide (10%) 193 90 R41 A
5. Cetylpyridium bromide (6%) 191 86 R41 A
6. Benzalkonium chloride (5%) 184 84 R41 A
7. Captan 90 concentrate 170 83 R41 A
8. Chlorhexidine 231 82 R41 A
9. Cyclohexanol 77 82 R41 A
10. Quinacrine 230 82 R41 A
11. Promethazine hydrochloride 229 72 R41 A
12. Parafluoroaniline 105 70 R41 A
13. Triton X-100 (10%) 207 69 R36 B
14. Acetone 157   R36 B
15. Hexanol 74 65 R36 B
16. 1-Napththalene acetic acid, sodium salt 168 64 R41 A
17. Sodium oxalate 147 61 R41 A
18. Isobutanol 70 60 R36 B
19. Imidazole 124 59 R41 A
20. Sodium lauryl sulphate (15%) 176 59 R36 B
21. 2-Ethyl-1-hexanol 68 51 R36 B
22. 4-Carboxybenzaldehyde 79 50 R36 B
23. Methyl ethyl ketone 155 50 R36 B
24. Pyridine 123 48 R41 A
25. 1-Naphthalene acetic acid 166 47 R41 A
26. Benzalkonium chloride (1%) 180/182 34/56 R41 A
27. 2,2-Dimethylbutanoic acid 34 45 R41 A
28. y-Butyrolactone 228 43 R36 B
29. Thiourea -c -c -c -c
30. Octanol 66 41 R36 B
31. Methyl acetate 28 40 R36 B
32. L-Aspartic Acid 33 37 R36 B
33. Benzoyl-L-tartartic acid 32 37 R41 A
34. Triton X-100 (5%) 203/205 32/34 NI/R36 NI/B
35. Potassium cyanate 146 31 R36 B
36. Isopropanol 64 30 R36 B
37. Sodium perborate 144 30 R41 A
38. Dibenzyl phosphate 161 30 R36 B
39. 2,5-Dimethylhexanediol 63 28 R41 A
40. Methyl cyanoacetate 27 28 R36 B
41. Sodium hydroxide (1%) 80 26 R36 B
42. Ethanol 62 24 NI B
43. 2,6-Dichlorobenzoyl chloride 49 24 R36 B
44. Ammonium nitrate 143 18 R36 B
45. Ethyl-2-methylacetoactetate 26 18 NI B
46. Sodium lauryl sulphate (3%) 174 16 NI NI
47. Ethyl acetate 24 15 NI NI
48. Maneb 164 14 R36 B
49. Fomesafen (acid form) 163 14 NI NI
50. Tetraaminopyrimidine sulfphate 122 10 NI NI
51. Toluene 101 9 NI NI
52. Butyl acetate 20 8 NI NI
53. Trichloroacetic acid (3%) 30 7 NI NI
54. Methyl isobutyl ketone 149 5 NI NI
55. Tween 20 201 4 NI NI
56. Ethyl trimethyl acetate 18 4 NI NI
57. Methylcyclopentane 138 4 NI NI
58. Cetylpyridinium bromide (0.1%) 187 3 NI NI
59. Glycerol 56 2 NI NI
60. Polyethylene glycol 400 195 0 NI NI

aNI = non-irritant; R36/category B = irritating to eyes; R41/category A = risk of serious damage to eyes.

bData from reference 6

cAcutely toxic, and therefore discarded from study.

MMAS = Modified Maximum Average Draize Test Score.

Bold type denotes different classifications obtained by applying the EU and OECD systems.

R41 classifications

Twenty two of the 59 chemicals were classified as R41, of which sodium hydroxide (10%; chemical 1) and benzalkonium chloride (10%; chemical 2) had the highest MMAS values (108), whereas 2,5-dimethylhexanediol (chemical 39) had the lowest MMAS (28). The R41 classification for the latter chemical is due to the persistence of eye effects in at least one rabbit.

R36 classifications

Twenty chemicals were classified as R36, of which the non-ionic surfaetant Triton X-100 (10%; chemical 13) had the highest MMAS (69) and the pesticide Maneb (chemical 48) had the lowest MMAS (14).

Nl classifications

Seventeen chemicals were class)fied as NI; Triton X-100 (5%, chemical 34) had the highest MMAS (32) and polyethylene glycol 400 (chemical 60) had the lowest MMAS (0).

Harmonised OECD classifications

With the exception of two chemicals, all of the chemicals classified as R41 were also classified as Category A, and all of the chemicals classified as R36 were also classified as Category B. The two exceptions, ethanol and ethyl-2-methylacetoacetate, were classified as NI on the basis of EU criteria, but as Category B on the basis of the harmonised OECD criteria. The subcategory B1 (mildly irritating) was not taken into consideration. The difference in the classification of ethanol under the two systems is interesting, given that ethanol is used as the positive control in the bovine corneal opacity/permeability (BCOP) assay.

Application of MMAS cut-offs

The comparison between the EU classifications and the classifications obtained by applying the MMAS cut-offs of 25 and 59 is summarised in Table IV. The results show that the cut-off values cannot be used to classify chemicals reliably.

Table IV: Classification of Eye Irritants on the Basis of Modified Maximum Average Draize Test Score (MMAS)

  European Union Classification  
  NI R36 R41 Total
MMAS < 25 16 3 0 19
25 < MMAS < 59 1 13 8 22
MMAS > 59 0 4 14 18
Total 17 20 22 59

NI = non-irritant; R36 = irritating to eyes; R41 = risk of serious damage of eyes.

Conclusions

It is concluded that: a) the EU and proposed OECD systems for the classification of eye irritants/corrosives are broadly equivalent; b) the MMAS cut-offs of 25 and 59 are not appropriate for classifying chemicals according to the two systems; and c) future validation studies on alternatives to the Draize test would benefit from the use of classifications based on the proposed OECD harmonised system.

Acknowledgement

The author gratefully acknowledges Ingrid Gerner (BgVV, Berlin, Germany) for her constructive comments regarding this work.

References

  1. Balls, M., Botham, P.A., Bruner,L.H. & Spielmann, H. (1995). The EC/HO international validation study on alternatives to the Draize eye irritation test. Toxicology In Vitro 9: 871-929.
  2. Brantom, P.G. et al (1997). A summary report of the COLIPA international validation study on alternatives to the Draize rabbit eye irritation test. Toxicology In Vitro 11: 141-179.
  3. OECD (1996). Final Report of the OECD Workshop on Harmonization of Validation and Acceptance Criteria for Alternative Toxicological Test Methods, 62pp. Paris, France: OECD.
  4. OECD (1998). Revised Proposal for the Harmonizahon of Hazard Classification Based on Eye Irritation/Corrosion. ENV/MC/CHEM/HCL(98)5, 10pp. Paris, France: OECD.
  5. EC (1993). Commission Directive 93/21/EEC of 27 April 1993 adapting to technical progress for the 18th time Council Directive 67/548/EEC on the approximation of laws, regulations and administrative provisions relating to the classification, packaging and labelling of dangerous substances. Official Journal of the European Communities L11OA: 1-86.
  6. ECETOC (1998). ECETOC Technical Report No. 48, Eye Irritation: Reference Chemicals Data Bank, 2nd edn, 236pp. Brussels, Belgium: ECETOC.

Return to Main Article