One of--if not the most—controversial topics in spine surgery is fusion for chronic low back pain (LBP) without symptoms of nerve compression in the setting of disk degeneration. There have been multiple prior systematic reviews on this topic that have generally summarized the results of Level 1 trials comparing surgery to non-operative treatment. These reviews have generally concluded that while fusion does lead to improvement, it is not clear that surgery results in better outcomes than an intensive non-operative program including cognitive-behavioral therapy. These prior reviews tended to focus narrowly on the Level 1 trials comparing surgery to non-operative treatment and did not include other studies that evaluated the results of fusion for chronic LBP such as trials comparing fusion and disk replacement or different fusion techniques. Given the lack of a review that included studies from the whole universe of fusion papers, Dr. Phillips and his colleagues undertook a systematic review including all studies of one or two level fusion for LBP as long as the studies included at least 20 patients, had at least 12 months of follow-up, and reported at least 2 validated, patient-reported outcome measures (generally a visual-analog pain scale [VAS], Oswestry Disability Index [ODI], SF-36 Physical Component Scale [PCS] or patient satisfaction). They pooled data from over 3,000 fusion patients and demonstrated an average improvement of 36.8 points on the VAS, 22.2 points on the ODI, and 12.5 points on the SF-36 PCS. Patient satisfaction was 71%. Given these results, the authors concluded that fusion for chronic LBP is effective with outcomes similar to those for other widely accepted orthopadedic interventions.
The conclusion of this review is obviously different from previous reviews on the topics, and the most likely cause for the discrepancy is the difference in the included studies. The authors noted that there were not differences in outcomes between studies with different levels of evidence, though they do not report the details of this analysis. Scrutiny of the results demonstrates the fusion patients in the RCTs comparing surgery to non-operative treatment improved an average of 22.8 points on the VAS for back pain, 13.9 points on the ODI, and 9.4 points on the SF-36 PCS. This compares to improvements of 40.3 points on the VAS for back pain, 27.2 points on the ODI, and 13.7 points on the SF-36 PCS for the RCTs comparing different surgical techniques. This comparison demonstrates that the patients in the RCTs comparing different surgical techniques had absolute improvements on the VAS and ODI that were nearly twice the degree of improvement observed in the RCTs comparing surgery to non-operative treatment. While there are many possible explanations for these differences (i.e. patients in the RCTs comparing different techniques had worse baseline scores, included many American patients, were typically involved in industry-sponsored trials, etc.), the bottom line is that the fusion results from the RCTs comparing surgery to non-operative treatment were worse than those for the other studies included in the analysis. Another important difference is that prior systematic reviews comparing surgery to non-operative treatment did not include the 2011 study from Ohtori et al. that reported a degree of improvement following surgery that was markedly better than that reported in prior studies (i.e. improvements of 55 points on the VAS, 50 points on the ODI). Again, there are many possible explanations for why this group did so well (i.e. included mostly ALIF patients, patients had high baseline scores, it was a Japanese population, all patients had positive discograms and “discoblocks”, etc.), and including this study in a review comparing surgery to non-operative treatment could affect the conclusions. The authors of the current review appropriately criticize studies comparing surgery to continued non-operative care (i.e. the methodology used in the Fritzell and Ohtori studies) as standard non-operative care has by definition failed by the time patients are considered for enrollment in a surgical trial. The most appropriate study design to address this question is probably an RCT comparing surgery to an intensive rehab program including cognitive-behavioral therapy (i.e. similar to the Brox and Fairbanks studies) as these are really the two competing treatment options for patients who have failed traditional non-operative care. No sufficiently powered trial comparing a standardized fusion technique to a clearly defined intensive rehabilitation program has ever been performed, and no Level 1 study comparing surgery to non-operative treatment for chronic low back pain has included American patients. It remains to be seen if such a trial could be performed successfully, though it seems clear that this topic will remain controversial until such a study is done. The current systematic review provides a comprehensive description of the available literature on the topic and allows one to conclude the following:
1. Spinal fusion for chronic low back pain in the setting of one or two level disk degeneration results in significant improvement, though the degree of improvement varies depending on the population included in the study.
2. Intensive rehabilitation including cognitive-behavioral therapy results in nearly the same degree of improvement as fusion surgery with an obviously lower cost and complication rate.
3. Continued standard non-operative care in patients who have already failed to improve with such treatment is significantly less successful than surgery.
Please read Dr. Phillips’s article on this topic in the April 1 issue. Does this article change how you view outcomes following lumbar fusion for chronic LBP? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor