Interlaboratory validation of the in vitro eye irritation tests for cosmetic ingredients. (3) Evaluation of the haemolysis test.

The haemolysis test using sheep red blood cells (RBC) was evaluated as an alternative method to the Draize rabbit eye irritation test (Draize test) by six to nine laboratories. The participating laboratories performed the test according to the standard operating procedure (SOP). Thirty-eight cosmetic ingredients and isotonic sodium chloride solution were used as test substances in this validation study. The concentrations of the test substances that induced 50% haemolysis (HC(50) value) was obtained to serve as a toxicological index and compared with in vivo Draize scores. HC(50) values were not obtained for coloured or water-insoluble (turbid) substances. Three acids caused denaturation of haemoglobin leaked from RBC and consequently interfered with the determination of the HC(50) value. Interlaboratory reproducibility was relatively good except in the case of water-insoluble substances. The average values of coefficient of variation (CV) was 37%. The correlation coefficient and Spearman's rank correlation between the HC(50) value and maximum average Draize total score (MAS) were -0.631 and 0.641, respectively. The equivalence ratio between the haemolysis test and MAS was 70.0% when MAS 15 was set as the in vivo cut-off point. On the other hand, strong irritants (MAS50) could be correctly classified by this method. These results suggest that the haemolysis test might be applied to cosmetic ingredients as a screening method to distinguish strong irritants that directly affect the cell membrane permeability and do not disturb spectrophotometrical determination of haemoglobin. In order to evaluate the potential for eye irritation of cosmetic ingredients, a combination of haemolysis with other methods based on different mechanism should be employed to improve the predictability.