OBJECTIVE In a retrospective study, the extent of functional recovery and the merits of surgical exploration versus conservative management for spinal cord injuries were evaluated for 145 casualties from the front lines of the Iran-Iraq conflict. METHODS Eighty-seven patients who underwent surgical exploration and 58 patients who were conservatively treated were monitored for 6 to 140 months (average, 57 +/- 31 mo) for any changes from their baseline neurological status, using the Frankel Scoring System. RESULTS Twenty-two of 90 patients (24.4%) with complete injuries (Frankel score A) and 53 of 55 (96.4%) with incomplete injuries (Frankel scores B, C, and D) experienced changes in their neurological status as they recovered. Improvement was noted for 42 of 87 surgically treated patients (13 of 55 with complete injuries and 29 of 32 with incomplete injuries). Among conservatively treated patients, improvement was noted for 32 of 58 (55%), including 25.7% of those with complete injuries and 100% of those with incomplete injuries. Independent walking (Frankel scores D and E) was achieved by 10 of 90 patients with complete injuries and by 52 of 55 patients with incomplete injuries. Surgery did not affect the final outcome. Thirteen of 17 (76%) cerebrospinal fluid fistulas, 13 of 15 (87%) meningitides, and 4 of 6 (67%) local septic complications were encountered in the surgically treated group. CONCLUSION Surgery did not enhance functional recovery from spinal cord missile injuries but did correlate with an increased prevalence of complications (fistulas, meningitis, and local sepsis). With or without surgical exploration, patients with seemingly complete cauda equina injuries demonstrated better functional outcome than did those with spinal cord injuries.