The course, treatment response, and recovery potential after acute traumatic spinal cord injury (SCI) have been shown to differ depending on the neurological level of injury. There are limited data focused on thoracic-level injuries, however. A cohort of 86 patients from the prospectively maintained North American Clinical Trials Network SCI registry were identified and studied to characterize the patterns of neurological recovery and to determine rates of acute hospital death and pulmonary complications. Regression analyses were used to examine the relationship between timing of surgery and administration of methylprednisolone on neurologic and clinical outcomes. Neurological conversion (≥1 American Spinal Injury Association Impairment Scale [AIS] grade improvement) was poorest for AIS grade A patients; 14.3% converted at last available follow-up (mean eight months). While rates of conversion were more optimistic for AIS-B patients (54.5%) and AIS C injuries (77.8%) at the same time point, none of the AIS grade D patients converted to AIS E. At last available follow-up (mean eight months), the magnitudes of lower motor extremity score (LEMS) change were highest for AIS C injuries (21.9 points), then AIS B (17.7 points), AIS D (16.4 points), and finally AIS A (2.5 points) (p < 0.05). Early surgical intervention (<24 h post-injury) was independently associated with an additional seven points in motor recovery and a 60% decreased incidence of pulmonary events (p < 0.05). Methylprednisolone administration was not an independent predictor of neurological outcome or pulmonary complications. Evaluation of this cohort obtained from a modern multi-center SCI registry provides an update on the natural history, acute death, and incidence of pulmonary complications after traumatic thoracic SCI. Although small sample size limited the extent of analyses possible, early surgical treatment was associated with significantly larger motor recovery and lower rates of pulmonary complications.