Sex steroids play an important role in neuroprotection following brain injury. Although their protective action has been found to be effective in many experimental animal models, their use in humans remains controversial. Many authors have shown that the frequency and extent of post-ischaemic damage is sex-dependent, and even low concentrations of estrogens are neuroprotective, but not therapeutic. The authors have summarized the current knowledge of neuroprotection by sex steroids in experimental and clinical settings. They conclude that direct extrapolation from animals to humans is not possible, and so far, widespread clinical use of sex steroids in humans is not justified.