Animal studies have shown that insulin eyedrops containing an absorption-enhancing agent can have a significant effect on blood glucose levels. When formulated as a topical solution, insulin might potentially be used to treat or augment the treatment of diabetes mellitus in humans. We sought to investigate the feasibility of using insulin eyedrops in humans by studying the local toxicity and efficacy of insulin administered without surfactant to the eyes of healthy volunteers. A prospective, randomized, placebo-controlled, single-masked study was conducted in which 8 subjects were given 50 μ1 of sterile normal saline containing varying insulin concentrations randomized to one eye, and 50 μl of placebo (sterile normal saline) to the fellow eye. Subjective ocular irritation was evaluated, and the eyelids, conjunctiva, cornea, and anterior chamber were examined objectively with slit lamp biomicroscopy. Subjects were evaluated for 2 hours following administration of a single dose of insulin. There was no statistically significant difference (P > 0.05) in toxicity observed by any parameter evaluated between eyes receiving insulin and placebo. No systemic absorption of insulin was observed; blood glucose levels and serum immunoreactive insulin levels were unchanged. The results of this study suggest that single-dose insulin in concentrations up to 100 U/ml formulated in saline has no detectable clinical toxicity to the anterior structures of the human eye. © 1994, Mary Ann Liebert, Inc. All rights reserved.