To the Editor.
—The report by Smolin et al (Archives 97:2181-2183, 1979) on the beneficial effect of vitamin A in herpes keratitis in (presumably nondeficient) rabbits is intriguing. Nonetheless, their suggestion that the same dose extrapolated to humans "would probably not be toxic" is in error. The proverbial 70-kg man is 28 times the weight of their rabbits, suggesting a dose of nearly 3 million IU/day. Oil-miscible preparations of this size, taken orally, have caused acute toxicity even in full-grown adults.1-3 Water-miscible parenteral preparations of the sort used by Smolin et al lead to still higher serum vitamin A levels and are likely to be far more toxic.4,5 Massive parenteral doses of vitamin A are clearly nonphysiologic; the amount converted to retinol and attached to retinol-binding protein (holo-RBP) will represent only a tiny fraction of the circulating vita