determination of eye irritation potential of low-irritant products: comparison of in vitro results with the in vivo draize rabbit test

In an attempt to build the evaluation strategies to assess the human eye irritation, a reassessment of some in vitro tests is necessary, particularly concerning the non-irritants, mild and moderate irritants. Thus, the correlations between results obtained from the Draize test with the in vitro methods HET-CAM (Hens Egg Test-Chorion Allantonic Membrane) and RBC haemolysis assay to assess the ocular irritancy potential of 20 eye drops were examined. Parameters such as accuracy (%) and specificity (%) were determined. All results were correlated with RBC correctly with the results obtained in the Draize test. The HET-CAM presented four false-positive results, showing a tendency of data overestimation. Despite the high specificity provided mainly by the RBC, it would be necessary to test a wider range of products representing all the scales of irritation to confirm its ability to be used as a first alternative to test products that could be presumptive non-irritants.