Although administration of glucocorticoid steroids is one of the most widely used therapeutic modalities for the clinical management of acute trauma of the central nervous system (CNS), controversy continues regarding their effectiveness. In essence, two viewpoints concerning their use exist. Some believe that despite their poor clinical record, the steroids nevertheless have a place in the treatment of human CNS trauma. In general, this group of clinical investigators uses the steroids primarily out of tradition, feeling that steroid therapy may be of some benefit. Unfortunately, confusion remains as to what constitutes an appropriate dose or regimen. In this regard, it has been suggested that the steroid dose be increased and the regimen intensified. Others believe that steroids should not be used. They contend that in view of their poor clinical record, it is unlikely that increasing the steroid dose or changing the dosing regimen will improve clinical efficacy, since steroids have already failed at what may be considered huge doses by glucocorticoid standards. Furthermore, it is contended that the side effects associated with large steroid doses reduce the margin of safety so as to make the steroids unsafe. Complicating these arguments is a body of experimental evidence that by and large strongly supports the utility of steroids for the acute treatment of CNS trauma. The intent of this article is to evaluate the current use of steroid therapy for CNS trauma from a purely pharmacological perspective, and to compare the steroids' experimental use with their clinical application.