Although high dose glucocorticosteroid therapy has become the standard of care for hyperacute management of non-penetrating spinal cord injury, it does not enjoy the same prestigious status in traumatic brain injury (TBI). A Brain Trauma Foundation standard concluded that steroids fail to improve outcome and lower intracranial pressure in severe TBI, but in the absence of interventions able to improve the functional outcome, glucocorticosteroid use continues into the present day. This paper summarizes the pathobiology of TBI and provides a methodological critique of the major studies of steroids in TBI. Given the growing appreciation of the complexity and heterogeneity of brain injury, corticosteroid use may find a rational place in management. Alone or in combination with other agents, corticosteroids may have a role in the treatment of severe injury, which so far has remained intractable to all manner of intervention.