The Spine Blog

Sunday, May 13, 2012

How Little We Know: Systematic Review of Nonoperative Treatment for Spinal Stenosis

The spine surgery literature is frequently criticized for its low quality evidence. In the May 1 issue, Dr. Ammendolia and his colleagues from Toronto reported the results of their systematic review evaluating nonoperative treatments for spinal stenosis, which demonstrated the even worse state of the evidence behind medications, epidural steroid injections, and physical therapy for stenosis. After an extensive literature search, they found 21 RCTs in which a nonoperative treatment modality was evaluated, and only 4 of these were judged to have “low risk of bias”. Meta-analysis was only possible for the comparison of Oswestry Disability Index scores between surgery and poorly defined “multi-modal” nonoperative care based on two comparable studies. The authors concluded that in no case was the quality of evidence for any nonoperative treatment better than “low” or “very low”. As such, the vast majority of care provided for spinal stenosis—including medications, injections, and physical therapy—is not based on a strong foundation of evidence.

Given that there were few actual results that the authors could comment on in the discussion, this portion of the paper was primarily focused on the barriers to performing high quality studies in this field and possible solutions to these problems. They noted that there is no clear definition of spinal stenosis, so the studies included very heterogeneous populations (i.e. patients with disk herniations, degenerative spondylolisthesis, degenerative scoliosis, etc.) Additionally, there is no standardization of epidural steroid injection techniques or physical therapy regimens, making comparisons across studies impossible. What is somewhat surprising about the poor state of the literature for medications and injections is that these interventions should be able to be studied using rigorous placebo-controlled, double-blinded designs, but this was rarely done successfully. Studying physical therapy and surgery presents greater logistical challenges, but clearly the field can improve on the current state of the literature. Given the recent focus on improving the quality of healthcare by improving outcomes and/or decreasing costs, it is possible that payors are going limit coverage for some of these interventions without evidence supporting their efficacy, especially those that are more costly like epidural steroid injections. Spinal stenosis patients need nonoperative treatment options that are evidence-based, and the spine community has a responsibility to stop wasting resources on treatments that are ineffective. While this article is unlikely to change practice given that the authors were unable to make any treatment recommendations, hopefully it will serve as an impetus to improve the quality of the evidence underlying nonoperative treatment for spinal stenosis.

Please read Dr. Ammendiola’s article and accompanying commentary. What type of studies do you feel would be most beneficial in this field? Let us know by leaving a comment on The Spine Blog.

Adam Pearson, MD, MS

Associate Web Editor