Edited by Benjamin Davies Research Summary on MRI Methods for Predicting Functional Recovery from Surgery in Patients with Cervical Myelopathy.
Rao A et al., Diffusion Tensor Imaging in a Large Longitudinal Series of Patients With Cervical Spondylotic Myelopathy Correlated With Long-Term Functional Outcome. Neurosurgery. Epub ahead of print Feb 23, 2018
Reason for the study
Currently, the assessment of the impact of cervical myelopathy is based largely on patient reported symptoms and commonly quantified using an assessment scale called the mJOA. Patient reported symptoms are very important to take into account when discussing disease severity and function, but imaging measures which explain symptoms are also helpful. Currently there has been limited success in developing imaging measures which explain symptoms and, perhaps more importantly, predict future symptoms and potential recovery following surgery. This last part will be especially important for patients and surgeons determining who will benefit from surgery.
This study examined a different magnetic resonance imaging (MRI) technique than normal, called diffusion tensor imaging. This technique quantifies how water naturally flows (diffuses). The specific measure they chose (fractional anisotropy) measures the degree to which water flows in a single direction on a scale of 0 to 1 with values closer to 1 indicating diffusion more strongly in 1 direction. In axons of the spinal cord, for example, it would be expected that water would flow consistently along the path of axons rather than perpendicular to the axons. If a group had a worse score, however, it might indicate that they had damage to the neurons allowing water to move more freely. So, the purpose of this study to use fractional anisotropy as a measure of integrity of neurons in the spinal cord and see if this correlated with function pre-surgically and change in function following surgery.
Methods
This study enrolled patients who were diagnosed with cervical myelopathy over a 5 year period (age range 33 – 81, 18 male, 26 female) and followed these patients for 2 years following surgery. They took their MRI scans and determined mJOA score pre-operatively. For this study they focused their imaging analysis at the spinal level with the greatest compression of the spinal cord. For this analysis they included the whole cord except for the border around the cord because the data they could gain from this area might be incorrect due to the surrounding cerebrospinal fluid. They then compared the MRI scans (fractional anisotropy) and mJOA pre-operatively, and 6-, 12-, and 24-months after surgery. They also compared fractional anisotropy values to those from a group of healthy control participants.
Results
As expected fractional anisotropy (the MRI measure of neuron integrity) was lower in patients with cervical myelopathy than controls. It was also associated with the severity of myelopathy before surgery, as assessed by the mJOA (a measure of patient function).
Also, fractional anisotropy inversely predicted change in mJOA score at 12 months, but was less strongly predictive of change in mJOA at 6 and 24 months. Baseline mJOA also was inversely predictive of change in mJOA at 12 months. This means that in this study, those participants with lower fractional anisotropy (neural integrity) or mJOA (function) scores pre-operatively improved the most following surgery.
Why is this important?
At present, we are unable to predict accurately the response to surgery, and therefore markers which help this will be useful for doctors and patients. Whilst fractional anisotropy, has been investigated before previous studies using this technique were less successful in drawing a relationship with function. This may be due to improving the methods of collecting and analyzing the MRI data.
The identified relationship between baseline mJOA and change in mJOA with surgery are therefore promising, but require further research to understand the meaning for patients; for example it is recognized that patients with greater disability typically do improve more as measured by the mJOA, but that does not mean that they achieve a better functional outcome.The mJOA is not a linear scale, where each point gain is equally as important as the next.
We look forward to watching our understanding of Fractional Anisotropy improve and if it can be of benefit to doctors and patients.