Study design: A multicenter randomized study with a 2-year follow-up period and an independent observer was conducted.
Objective: To compare three commonly used surgical techniques to achieve lumbar fusion primarily in terms of their ability to reduce pain and decrease disability in patients with severe chronic low back pain.
Summary of background data: Lumbar fusion can be used to reduce pain and decrease disability in patients with chronic low back pain, and different surgical techniques are available. The reported results after fusion surgery vary considerably, but most studies are retrospective and/or performed on heterogeneous patient groups. Spine surgeons commonly use the techniques presented in this report. However, in the absence of randomized trials it is difficult to know whether any one of them is better than the others for these patients.
Methods: From 1992 through 1998, 294 patients referred to 19 spinal centers were blindly randomized into four treatment groups: three surgical groups (n = 222) and one nonsurgical group (n = 72). The sociodemographic and clinical characteristics did not differ among the groups. Eligibility included patients 25 to 65 years of age with therapy-resistant chronic low back pain that had persisted for at least 2 years and radiologic evidence of disc degeneration (spondylosis) at L4-L5, L5-S1, or both. Only patients randomized to one of three surgical groups were analyzed in the current study: Group 1 (posterolateral fusion; n = 73), Group 2 (posterolateral fusion combined with variable screw placement, an internal fixation device; n = 74), and Group 3 (posterolateral fusion combined with variable screw placement and interbody fusion; n = 75). The "circumferential fusion" in Group 3 was performed either as an anterior lumbar interbody fusion (n = 56) or as a biomechanically similar posterior lumbar interbody fusion (n = 19). The groups were composed of 51%, 58%, and 40% men respectively, and the respective mean ages were 44, 43, and 42 years. The patients had experienced low back pain for at least 2 years (mean, approximately 8 years). They had been on sick leave for approximately 3 years. Pain was measured by a visual analog scale, and disability was assessed by the Oswestry Low Back Pain Questionnaire, the Million Visual Analogue Score, and the General Function Score. Depressive symptoms were measured by the Zung Depression Scale. The global overall rating of the result was assessed by the patient and an independent observer after 2 years. Hospitalization data including operation time, blood loss, blood transfusion, and days of hospitalization in connection with surgery were reported, along with complications and the fusion rate. Records from the Swedish Social Insurance Board providing information on sick leave and economic compensation for Swedish residents were used to evaluate the patients' work status.
Results: An independent observer examined 201 (91%) of 222 patients after 2 years, after 18 "group changers" and 3 dropouts had been excluded from the analyses. All surgical techniques were found to reduce pain and decrease disability substantially, but no significant differences were found among the groups. In all three groups, the patients rated the overall outcome similarly, as did the independent observer. The more demanding techniques in Groups 2 and 3 consumed significantly more resources in terms of operation time, blood transfusions, and days in hospital after surgery. The early complication rate was 6% in Group 1, 16% in Group 2, and 31% in Group 3. The fusion rate, as evaluated by plain radiograph, was 72% in Group 1, 87% in Group 2, and 91% in Group 3.
Conclusions: All the fusion techniques used in the study could reduce pain and improve function in this selected group of patients with severe chronic low back pain. There was no obvious disadvantage in using the least demanding surgical technique of posterolateral fusion without internal fixation.