Forty-eight spinal cord injury victims were implanted with an epidural spinal cord stimulation system to treat spasms that had not satisfactorily responded to medical therapy. All the patients were at least 6 months after the injury. The protocol included assessment by independent examiners preoperatively and at 3, 6, 12 and 24 months after the implant. Pre- and postoperative data collection included the frequency and severity of the spasms. Combining the frequency and intensity scores into a 'severity' score provided a more accurate clinical picture. No patient observed neurological deterioration following the surgical procedure or the neurostimulation treatment. A statistically significant reduction in the severity of the spasms was observed in the follow-up evaluations, with results that progressively increased in time. It is appears that spinal cord stimulation is an effective and safe alternative in the management of spasms in spinal cord injury victims. Its exact role in relation to intrathecal baclofen infusion and ablative procedures remains to be defined.