Objectives: To determine whether postinjury methylprednisolone could reduce the generation of known mediators of secondary neurological injury. Methods: Intrathecal microdialysis probes were used to sample cerebrospinal fluid (CSF) for measurement of PGE 2 , glutamate, and citrulline (a byproduct of nitric oxide generation), before and after spinal cord injury in anesthetized pigs. The spinal cord was removed at the end of the study for measurement of myeloperoxidase and methylprednisolone concentrations. Animals were randomly allocated to receive intravenous methylprednisolone (30 mg/kg bolus then 3.4 mg/kg/h), intrathecal methylprednisolone (5 mg bolus then 5 mg/h), or saline, beginning 30 min after the spinal cord was injured by using a modification of the Allen weight drop technique. Results: Spinal cord injury significantly increased the amount of glutamate, PGE 2 , myeloperoxidase, and citrulline, recovered from the CSF dialysates. However, neither intravenous nor intrathecal methylprednisolone administered after injury had any effect on the magnitude of the increase in any of the measured biochemicals. Intrathecal methylprednisolone administration produced a spinal cord methylprednisolone concentration that was eight times greater, and a plasma concentration that was 32 times less, than that achieved with intravenous administration. Conclusions: Contrary to earlier animal studies in which methylprednisolone was administered either before or immediately after spinal cord injury, we found no effect of intravenous or intrathecal methylprednisolone on any of the parameters measured when administered 30 min postinjury.