In the April 2007 issue of the Archives, Soheilian et al1 discussed prophylaxis of acute posttraumatic bacterial endophthalmitis. Although the study seems well designed, I was very surprised that the prophylactic treatment included the intravitreal injection of gentamicin sulfate. Conway and Campochiaro2 (in this same journal) discussed back in 1986 how gentamicin can cause macular infarction in the setting of trauma. Again in the Archives, this time in 1989, Conway et al3 showed in primate retinas that gentamicin is definitely toxic, implying that it should never be used intraocularly. Although Soheilian and colleagues discussed studies in rabbits where gentamicin is not toxic when used in an infusion fluid, as recently as December 2005 Hancock et al4 reiterated the significant toxicity of gentamicin in both rabbit and rat retinas. With a quick literature search, at least 9 journal articles can be found discussing the significant retinal toxicity of intraocular and periocular gentamicin. I understand the need to use an antibiotic effective against gram-negative bacteria for such a trial, but the inclusion of a known toxic compound in this study makes the findings not only useless but in fact dangerous to any physician unaware of the prior research who may try to emulate what is recommended in your pages. Those who do not know history are doomed to repeat it.