Dose-response studies with chemical irritants in the albino rabbit eye as a basis for selecting optimum testing conditions for predicting hazard to the human eye.

Abstract Twenty-one chemicals, solutions, and mixtures with different degrees of recognized irritant potential to the human eye were tested in albino rabbit eyes at dose volumes of 0.003, 0.01, 0.03, and 0.1 ml. Materials were administered directly to the central corneal surface. Irritation that developed was followed up to 21 days and graded by the Draize scale. Maximum irritation scores and median number of days for eyes to return to normal were compared with available data on human experience as a basis for selecting the dose volume that would best predict human response. Using irritation categories of negligible = clearing within 24 hr, moderate = clearing within 7 days, substantial = clearing within 21 days, and severe or corrosive = persisting beyond 21 days, a dose volume of 0.01 ml most often gave results that were consistent with information on effects of human exposures. This dose of hexane, triethanolamine, 0.1% benzalkonium chloride, 0.5% sodium hydroxide, 5% hydrochloric acid, and 20% sodium chloride caused negligible irritations. Granular sodium chloride, 10% sulfuric acid, 5% sodium hypochlorite, isopropyl alcohol, cocount soap powder and 10% solution, a powdered laundry soap, two phosphate laundry detergents, sodium carbonate-sodium sulfate mixtures, 3% acetic acid, 1% silver nitrate, and 10 and 40% sodium lauryl sulfate gave moderate irritation. Sodium lauryl sulfate, 29%, in another experiment was substantially irritating, and 10% acetic acid, 10% sodium hydroxide, formaldehyde, and calcium hydroxide were severely irritating or corrosive. Corneal application of 0.01 ml of material is proposed as a more realistic test of eye hazard than is the Draize test.