Given the absence of a detectable contaminant in the poppers vials examined to date, it is likely that visual symptoms were directly linked to NO intake. However, the putative mechanisms linking poppers to retinal toxicity remain elusive. There is little knowledge regarding the pharmacological effects of inhaled alkyl nitrites on neural tissues.7,8 At physiological doses, NO modulates photoreceptor metabolism and function,9,10 in particular through activation of guanylate cyclase, a key enzyme of phototransduction.11 The presence of photopsias in many patients suggests permanent activation of central cones rather than their inhibition, which would be expected if only guanylate cyclase activation was involved. Accordingly, an increased ERG after NO administration was described in rats,12 and another study suggested that NO potentiates the light response of cones, while it decreases that of rods.13 At higher doses, it has been shown that photoreceptors are among the most sensitive retinal neurons to the toxic effects of NO, both in vitro and in vivo.14,15 Nitric oxide is also known to decrease the threshold of light toxicity.16,17 Yet, these studies were performed in retinas that do not have a fovea; thus, their relevance to the clinical toxicity described here is questionable. Accordingly, the elective targeting of the fovea in our patients suggests light-induced damage, although patients denied having stared at bright lights. Moreover, in addition to their effect on neuronal metabolism, it has been reported that NO interacts with the macular pigment zeaxanthin,18 which protects the fovea against light damage. In our patients, the presence of a central increase in autofluorescence and a central window defect and the absence of pigmentary changes even after months of exposure suggest a defect of macular pigment that may potentiate light toxicity. Measuring the concentration of macular pigment in these patients may thus be of interest to understand the physiopathology of the affection.