A comparison of two cytotoxicity tests for predicting the ocular irritancy of surfactants

In vivo rabbit eye tests have attracted criticism on both scientific and ethical grounds. Consequently there is a need to introduce methods that provide relevant information whilst avoiding the use of whole laboratory animals. Cytotoxicity models have been used to predict the ocular irritancy of surfactants since the action of this class of chemicals on cell membranes is essentially common and involves membrane disruption. The aim of the present studies was to compare the predictive ability of two in vitro cytotoxicity tests (the K562 and the red blood cell lysis tests) in the assessment of the in vivo eye irritancy of surfactants. The results of these studies on 14 selected surfactant materials showed that the K562 assay was less likely to over-estimate effect (when assessed over a wide range of surfactant concentration) than had previously been observed. This increased specificity (86%) was accompanied by a decrease in sensitivity (57%), the ability to correctly identify 'irritant' surfactants. In contrast, the red blood cell lysis test was more predictive, correctly identifying all irritants tested. In addition, all non-irritant surfactants examined were predicted by this test and a high (89%) ability to rank irritant effect was demonstrated. The red blood cell lysis test could be a powerful addition to a testing strategy or pre-screen for the evaluation of surfactant chemicals.