Comparison of in vitro cell toxicity with in vivo eye irritation.

The effects of 26 different cosmetic ingredients (e.g., permanent wave and hair dye compounds, emulsifiers, resins, and detergents such as quats) were assessed by four end points indicative for qualitatively and quantitatively different cytotoxicity: (1) neutral red uptake reduction after 24 h of treatment (NR-90 and NR-50); (2) cell detachment from culture dish after 4 h of treatment (CD-25); (3) growth inhibition after 48 h of treatment (GI-50); and (4) membrane permeability measured by fluorescent dye retention (fluorescence shift FS-25) and dye exclusion (viability ratio VR-25). The cytotoxicity potentials of the test agents were ranked for each in vitro test and compared with the in vivo eye irritation in guinea pigs (Draize test) after application of 5 or 2.5% (w/v) solutions of the same test batches. Strong irritants could be easily detected by most of the in vitro tests, but the neutral red uptake assay (especially NR-50) was the only one that was able to distinguish the minimally irritating test agents from strong irritants as well as from nonirritants. (I) All three extremely irritating quaternary ammonia compounds were identified as the strongest cytotoxic agents. (II) Nine out of 12 minimally irritating substances (mainly emulsifiers and resins) were ranked in the intermediate group. (III) Eight out of 11 non-or practically nonirritating chemicals (mainly permanent wave compounds) showed cytotoxic effects at very high concentrations only. The distinction of these three groups was better by means of NR-50 than by NR-90 data. At least two of the other cell tests (CD-25, GI-50, FS-25, and VR-25) had to be considered to allow an adequate interpretation of in vitro cytotoxicity.