A number of drugs commonly used for a variety of clinical indications have been found recently to have substantial neuroprotective properties, raising the potential for rapid translation into human clinical trials of spinal cord injury (SCI). In this study we compared the neuroprotective efficacy of erythropoietin and a derivative of it, darbepoetin, in an acute model of thoracic SCI. Sprague-Dawley rats were randomized to receive erythropoietin (5000 IU/kg), darbepoetin (10 mug/kg), or saline, as a single intravenous injection 1 h after a thoracic contusion SCI. The animals were evaluated for behavioral recovery over 6 weeks, which included BBB locomotor testing, horizontal ladder testing, video-analysis of gait, and hindlimb monofilament sensory testing. At 6 week post-injury, the spinal cords were evaluated histologically to measure white and grey matter sparing at and around the epicenter of injury. We found that neither erythropoietin nor darbepoetin led to improved behavioral recovery over saline controls, with no significant differences observed in BBB scores, BBB subscores, footfall errors on horizontal ladder testing, width of hindlimb base of support, or threshold for paw withdrawal on sensory testing. Furthermore, no differences were observed in grey or white matter sparing between the three experimental groups. Using doses of erythropoietin and darbepoetin that other investigators have reported to be beneficial in SCI and stroke models, we were unable to demonstrate a neuroprotective effect when administered 1 h after injury. Further preclinical investigation is necessary to refine the treatment strategy of using erythropoietin or darbepoetin in acute spinal cord injury.