Efforts have been made to correlate traditional T1 and T2 weighted MRI findings with outcomes in cervical spondylotic myelopathy (CSM), though pre-operative MRI characteristics are relatively poor predictors of post-operative outcomes.1 Many MRI variables have been analyzed, including increased T2 signal intensity, decreased T1 signal intensity, number of segments with increased T2 signal intensity, transverse area of the spinal cord, and the anterior-posterior compression ratio (APCR, i.e. A-P diameter divided by the transverse diameter). In a study looking at clinical and radiographic predictors of surgical outcomes, transverse area of the spinal cord was not an independent predictor of outcomes, while many clinical factors (i.e. duration of symptoms, baseline mJOA score, age, smoking status, gait dysfunction, etc.) were.2 Given the poor predictive value of traditional MRI findings, investigators have sought out other imaging findings that may more reliably improve predictions of surgical and non-operative outcomes. In the May 1 issue, Dr. Li and his colleagues from Hong Kong evaluated diffusion tensor imaging (DTI) in predicting levels of involvement in CSM. They evaluated traditional MRI findings such as APCR and increased T2 signal intensity as well as orientation entropy (OE) on DTI. While the physics are complex, DTI works on the principle that white matter is anisotropic and water diffuses with greater ease parallel to the orientation of the nerve fiber as compared to perpendicular to it. When the neuron is damaged, this diffusion gradient diminishes and the OE increases. The authors used clinical exam findings to define the cranial-most myelopathic level and then compared the sensitivity, specificity, and accuracy of APCR, increased T2 signal intensity, and OE for defining that level. They found that OE was the most sensitive, specific, and accurate of the imaging findings, indicating that DTI is likely a useful imaging modality in CSM.
While the authors suggested that DTI might be useful in defining the levels that should be included in the decompression, DTI is far from being clinically useful at this stage. The conclusions that can be based on the current paper are limited by design issues in the study, namely that clinical exam is not considered the “gold standard” by which the levels affected by CSM are determined. There is generally no widely accepted “gold standard” to determine the affected levels, and such a determination is generally a gestalt based on both clinical and radiographic findings. Additionally, the authors arbitrarily defined abnormal APCR and OE based on two standard deviations from the mean as derived from 14 healthy volunteers. While this might be a starting point for analysis, there is nothing to indicate that such a threshold is clinically relevant. Despite the limitations of this paper, it is valuable for bringing DTI to the forefront and indicating it might be clinically useful at some point. Being able to predict surgical and non-operative outcomes is the holy grail of CSM investigation, as this would allow patients and surgeons to make well-informed decisions. In patients predicted to have no significant benefit from surgery compared to non-operative treatment, an invasive procedure could be avoided, along with the complications that are common in this elderly patient population. While we are not close to having such a prediction model—primarily due to the lack of data on the non-operative outcomes in CSM—it is possible that DTI will yield some prognostic information. DTI is not widely available in most centers, but hopefully it can be incorporated into clinical studies more frequently such that it’s utility for surgical planning and outcomes prediction can be evaluated.
Please read Dr. Li’s article on this topic in the May 1 issue. Does this article change how you view DTI in CSM? Let us know by leaving a comment on The Spine Blog.
Adam Pearson, MD, MS
Associate Web Editor
1. Tetreault LA, Dettori JR, Wilson JR, et al. Systematic review of magnetic resonance imaging characteristics that affect treatment decision making and predict clinical outcome in patients with cervical spondylotic myelopathy. Spine (Phila Pa 1976) 2013;38:S89-110.
2. Tetreault LA, Kopjar B, Vaccaro A, et al. A clinical prediction model to determine outcomes in patients with cervical spondylotic myelopathy undergoing surgical treatment: data from the prospective, multi-center AOSpine North America study. J Bone Joint Surg Am 2013;95:1659-66.