Purpose of study: Total costs of low back pain to society are very high. They have been reported in two European studies to be 1.7% of the gross national product and in another study to be 0.5% to 2% for US conditions. Surgical treatment of chronic low back pain might be regarded as costly, but because the main purpose of economic evaluations is to help set funding priorities, the “gold standard” is that both the treatment costs and the treatment effects of two or more alternative interventions are compared (a full economic evaluation). We performed such an evaluation and investigated the cost-effectiveness of lumbar fusion surgery compared with nonsurgical treatment in selected patients with chronic low back pain during the first 2 years after treatment start.Methods used: In a randomized controlled trial 284 patients with chronic low back pain were allocated to lumbar fusion (n=217) or to nonsurgical treatment (control, n=67). The direct costs (SEK 1999, 1 US*$=SEK 8.27) for using primary care, hospital care (including fusion) and back-related drugs were investigated. Family costs were evaluated. Societal costs were identified as the sum of indirect (production loss) and direct costs (health-care costs). Treatment effects using patient global assessment, return to work, back pain (visual analog scale 0 to 100) and disability (Oswestry 0 to 100) were assessed. Marginal cost-effectiveness analyses were performed when appropriate. Sensitivity analyses were performed. The ethics committees at all participating universities approved the study. All patients gave informed consent and pretreatment questionnaires, and protocols were completed before randomization.of findings: Treatment costs 2 years: In the societal perspective the costs per patient in the surgical group was SEK 711,000±48,500 per patient, compared with SEK 626,000±81,000 in the control group (not significant). In the health-care perspective, the cost difference was significant, SEK 131,000±9,500 versus SEK 55,000±6,500 (<.0001). Treatment effects: 63% of the patients in the surgical group assessed themselves as improved, compared with 29% in the control group. The comparison figures for return to work were 36% versus 13%. In the surgical group, pain was reduced by 33% compared with 7% in the control group, and disability by 25% versus 6%. The marginal cost-effectiveness in a health-care perspective for getting one more patient improved using surgery instead of nonsurgical treatment was SEK 223,000±62,000, and for getting one more patient back to work, SEK 329,000±122,000.Relationship between findings and existing knowledge: The proportion of the direct costs in relation to the total costs in the surgical group was 18.4% compared with 8.8% in the control group, which is comparable with the literature. The costs of SEK 329,000±122,000 for society to get one more patient back to work if surgery was used instead of nonsurgical treatment (direct costs), could be compared to the mean production loss per patient (indirect costs) in the nonsurgical group, which in this study was SEK 464,000±52,000. The comparison is valid if the surgical patient returned to a full time job.Overall significance of findings: We consider that the treatments provided in this study resembled a close to “real life situation.” In a societal perspective, there was no significant difference in costs between fusion surgery and nonsurgical treatment in this patient category, whereas treatment effects were significantly better in the surgical group. This information may be of value when planning the treatment strategies for highly selected patients with chronic low back pain.Disclosures: Device or drug: VSP (Variable Screw Placement). Status: approved.Conflict of interest: Peter Fritzell, grant research support; Olle Hagg, grant research support; Anders Nordwall, grant research support.