The Spine Blog

Friday, October 25, 2013

Cervical spondylotic myelopathy: Has our understanding improved since the 1950s?

Much of our understanding of cervical spondylotic myelopathy (CSM) goes back to the classic papers of the 1950s and 1960s that followed CSM patients treated non-operatively over the long-term and showed that the majority of them had progression of their disease. In Clarke and Robinson’s classic paper, they followed patients with myelopathic symptoms severe enough to justify hospitalization and reported that 75% of them had a pattern of disease that progressed in a sudden, stepwise fashion and that two-thirds of this group continued to progress over the decade following their presentation.1 Not surprisingly, these classic studies did not use validated outcome measures and are narrative rather than quantitative in their description of the results. Nonetheless, these studies have formed the basis for the modern approach to treatment of CSM which views it as a surgical disease. Given the lack of a single, definitive, modern study on the natural history of CSM, Dr. Fehlings and his Canadian co-authors elected to perform a systematic review to evaluate the current state of the literature. They included 8 studies from the 1980s through 2012 that evaluated the natural history of CSM over 3-6 years as described by changes in JOA or Nurick scores and found that 20-62% of patients deteriorated. Due to the substantial heterogeneity of inclusion criteria, outcome measures, and length of follow-up, the authors felt that quantitative meta-analysis was not appropriate, so they simply reported the range of deterioration rates across the studies. They also evaluated the strength of evidence behind factors predicting progression and found the data to be quite limited. The only predictive factor that reached even a “low” strength of evidence was circumferential cord compression that showed a significantly increased risk of progression compared to partial cord compression. Gender, age, severity of disease, duration of symptoms, range of motion, and other radiographic parameters had insufficient evidence to determine if they predicted progression. In conclusion, the authors suggested that there was a relatively high rate of progression in CSM and that the likelihood of progression was unpredictable.

Does this review change any of the classic concepts of CSM? Given the conclusion that CSM frequently progresses and that we cannot predict which patients are going to progress, it seems as though we should continue to view CSM as a surgical disease. Similar to most other spinal disease processes, CSM presents with a broad spectrum of disease severity, signs, and symptoms across a wide patient population, and we know that not all CSM patients benefit from surgery. Given that approximately 50% might not progress and that many do not see substantial improvements following surgery, there is clearly a group of patients who would be better off with non-operative treatment. Unfortunately, we are currently unable to identify this group. Are younger patients with mild symptoms as likely to progress as the patients with more severe CSM who were included in prior studies? Do the risks of progression outweigh the risks of surgery in the elderly population with more severe disease? We still do not know the answers to these questions, and until we do, we are likely going to continue to offer surgery to all CSM patients. Given that the state of knowledge around CSM is incomplete and patients need to consider different sets of risks and benefits associated with surgical and non-operative treatment, it represents a good opportunity for shared decision making in which the risks of progression are weighed against the risks and recovery associated with surgery. Due to the ethical challenges surrounding a large-scale modern trial comparing surgery to non-operative treatment for a potentially progressive, irreversible disease process, we may never have Level 1 evidence to support what we do. Hopefully we will be able to continue to collect observational data on patients who choose not to have surgery so that we may gain a better understanding of the natural history of CSM and determine if there are predictors of progression that can allow us to better tailor our treatment recommendations to individual patients.

Please read Dr. Fehlings’s article on this topic in the special focus issue on CSM. Does it change how you approach the CSM patient? Let us know by leaving a comment on The Spine Blog.


Adam Pearson, MD, MS

Associate Web Editor


1.            Clarke E, Robinson PK. Cervical myelopathy: a complication of cervical spondylosis. Brain 1956;79:483-510.