Cervical spinal cord stimulation: an analysis of 23 patients with long-term follow-up.

BACKGROUND For more than 3 decades, spinal cord stimulation has successfully been employed to treat neuropathic pain. Cervical spinal cord stimulation, despite now being standard in many hospitals, has only rarely been subjected to a critical review within the literature. OBJECTIVES The aim of this study was to determine the efficacy of cervical spinal cord stimulation (SCS) in a representative clinical sample. We also wanted to evaluate how factors such as stimulation parameters, unwanted paresthesia of the trunk and legs, and changes in paresthesia status due to head movement and how they affect SCS effectiveness. STUDY DESIGN Retrospective study. SETTING Academic university interdisciplinary pain center. METHODS We reviewed the records of patients who had been treated at our institution with cervical neurostimulators from November 1, 2001 through October 31, 2011. Information regarding age, gender, diagnosis, age at time of implantation, duration of disease, lead position, hardware in use, revision operations, and stimulation parameters were recorded. In addition, a short telephone interview was conducted, which contained the following items: pain scores on the numeric analog scale (NAS) with and without stimulation, time intervals of stimulation, paresthesia coverage, changes in paresthesia coverage by head movements, unwanted paresthesia of the trunk and legs, treatment satisfaction, and medication intake. RESULTS Twenty-three patients were treated. Eighteen patients proceeded to an implantable pulse generator (IPG) implant. In one patient, the system was removed after 4 years despite optimal function, because the patient was no longer experiencing pain. Average NAS pain scores were 6.8 (range 5.5 - 10.0, standard deviation [SD] 1.7) without, and 2.8 (range 0 - 7.5, SD 2.2) with neurostimulation. Fourteen revisions (5 due to lead dislocation, 5 due to lead breakage and 4 IPG revisions) were necessary in 9 of the 18 patients during a mean follow-up of 6.2 years. Most patients reported complete paresthesia coverage. Four patients reported unwanted paresthesia of the trunk or lower limb and 11 patients reported changes in paresthesia with head movements. In both instances, pain reduction was not affected. LIMITATIONS Retrospective study. CONCLUSIONS Cervical spinal cord stimulation appears to be effective in the treatment of neuropathic upper limb pain. Complications are not significantly more frequent than in SCS for lower limb pain. Changes in paresthesia with head movements and unwanted paresthesia did not affect the outcome.