STUDY DESIGN: Systematic review. OBJECTIVES: To assess efficacy and safety of spinal cord stimulation in patients with chronic leg and back pain and failed back surgery syndrome and to examine prognostic factors that predict spinal cord stimulation outcome. SUMMARY OF BACKGROUND DATA: A previous systematic review of spinal cord stimulation in patients with chronic back and leg pain and failed back surgery syndrome by Turner et al in 1995 identified 39 case studies and no controlled studies. METHODS: A number of electronic databases were searched through January 2002. Citation searching of included papers was undertaken, and gray literature was sought through contact with clinical experts. No language restrictions were applied. All controlled and noncontrolled study designs were included. Study selection was carried out independently by two reviewers. Prognostic factors (age, sex, duration of pain, time post surgery, follow-up duration, publication year, data collection year, indication, data collection country, study setting, and quality score) responsible for pain relief outcome across case series were examined using univariate and multivariate metaregression. RESULTS: One randomized controlled trial, one cohort study, and 72 case studies were included. The randomized controlled trial reported a significant benefit (P = 0.047) in the proportion of patients with failed back surgery syndrome reporting 50% or more pain relief with spinal cord stimulation (37.5%) compared with patients undergoing back reoperation (11.5%). There was evidence of substantial statistical heterogeneity (P < 0.0001) in the level of pain relief following spinal cord stimulation reported across case series studies. The four principal prognostic factors found to be predictive of increased level of pain relief with spinal cord stimulation were poor study quality score, short follow-up duration, multicenter (versus single center) studies, and the inclusion of patients with failed back surgery syndrome (versus chronic back and leg pain). Overall, 43% of patients with chronic back and leg pain/failed back surgery syndrome experienced one or more complications following a spinal cord stimulation implant, although no major adverse events were reported. CONCLUSIONS: Despite an increase in the number of studies over the last 10 years, the level of evidence for the efficacy of spinal cord stimulation in chronic back and leg pain/failed back surgery syndrome remains "moderate." Prognostic factors found to be predictive of the level of pain relief following spinal cord stimulation were study quality, follow-up duration, study setting, and patient indication.