2001 Volvo Award Winner in Clinical Studies: Lumbar fusion versus nonsurgical treatment for chronic low back pain: a multicenter randomized controlled trial from the Swedish Lumbar Spine Study Group.

STUDY DESIGN A randomized controlled multicenter study with a 2-year follow-up by an independent observer. OBJECTIVES To determine whether fusion of the lower lumbar spine could reduce pain and diminish disability more effectively when compared with nonsurgical treatment in patients with severe chronic low back pain (CLBP). SUMMARY OF BACKGROUND DATA The reported results after fusion surgery on patients with CLBP vary considerably, and the evidence of treatment efficacy is weak in the absence of randomized controlled studies. PATIENTS AND METHODS A total of 294 patients referred to 19 spinal centers from 1992 through 1998 were randomized blindly into four treatment groups. Patients aged 25-65 years with CLBP for at least 2 years and with radiologic evidence of disc degeneration at L4-L5, L5-S1, or both were eligible to participate in the study. The surgical group (n=222) included three different fusion techniques, not analyzed separately in this study. Patients in the nonsurgical group (n=72) were treated with different kinds of physical therapy. The surgical group comprised 49.5% men, and the mean age was 43 years. The corresponding figures for the nonsurgical group were 48.6% and 44 years. The patients had suffered from low back pain for a mean of 7.8 and 8.5 years and been on sick leave due to back pain for a mean of 3.2 and 2.9 years, respectively. The Visual Analogue Scale (VAS) was used to measure pain. The Oswestry Low Back Pain Questionnaire, the Million Score and the General Function Score (GFS) were used to measure disability. The Zung Depression Scale was used to measure depressive symptoms. The overall result was assessed by the patient and by an independent observer. Records from the Swedish Social Insurance were used to evaluate work disability. Patients who changed groups were included in the analyses of significance according to the intention-to-treat principle. RESULTS At the 2-year follow-up 289 of 294 (98%) patients, including 25 who had changed groups, were examined. Back pain was reduced in the surgical group by 33% (64 to 43), compared with 7% (63 to 58) in the nonsurgical group (P=0.0002). Pain improved most during the first 6 months and then gradually deteriorated. Disability according to Oswestry was reduced by 25% (47 to 36) compared with 6% (48 to 46) among nonsurgical patients (P=0.015), according to Million by 28% (64 to 46) compared with 8% (66 to 60) (P=0.004), and accordingtoGFS by 31% (49 to 34) compared with 4% (48 to 46) (P=0.005). The depressive symptoms, according to Zung, were reduced by 20% (39 to 31) in the surgical group compared with 7% (39 to 36) in the nonsurgical group (P=0.123). In the surgical group 63% (122/195) rated themselves as "much better" or "better" compared with 29% (18/62) in the nonsurgical group (P<0.0001). The "net back to work rate" was significantly in favor of surgical treatment, or 36% vs. 13% (P=0.002). The early complication rate in the surgical group was 17%. CONCLUSION Lumbar fusion in a well-informed and selected group of patients with severe CLBP can diminish pain and decrease disability more efficiently than commonly used nonsurgical treatment.