Regional measures of water diffusion associated with impairment in chronic SCI
Ann S Choe1,2,3, Cristina L Sadowsky3,4, Seth A Smith5,6, Peter C.M. van Zijl1,2, Visar Belegu3,7, and James J Pekar1,2

1Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, 2F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, 3International Center for Spinal Cord Injury, Kennedy Krieger Institute, Baltimore, MD, United States, 4Physical Medicine and Rehabilitation, Johns Hopkins University School of Medicine, Baltimore, MD, United States, 5Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, 6Vanderbilt University Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States, 7Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, United States

Synopsis

Prior studies have shown that DTI allows for noninvasive assessment of the severity of spinal cord injury (SCI). The present study investigated whether subject-specific demarcation of injury (vs. anatomically-driven ROI placement) could enhance the specificity of diffusion measures, specifically, fractional anisotropy (FA). Results showed that FA averaged over the region inferior to the injury epicenter demonstrated significant associations with impairment, suggesting that FA measures in the region are sensitive to Wallerian degeneration in the descending ventrolateral motor columns. We conclude that in chronic SCI, regional analysis of water diffusion using subject-specific injury demarcation may be more specific to impairment.

Purpose

Analysis of spinal cord MRI requires placement of regions of interest (ROIs) along the length of the spinal cord (i.e., along the longitudinal axis). In one of the conventional methods, an entire neurological region (e.g., cervical region) is defined as the ROI1,2. The resulting spatial averaging of diffusion tensor imaging (DTI) metrics over anatomically-based regions of interest (ROIs) may lead to a loss of subject-specific information in persons with spinal cord injury (SCI). Accordingly, the primary objective of this study was to investigate whether subject-specific demarcation of the region of injury can yield diffusion metrics (measured using fractional anisotropy; FA) that are specific to impairment.

Methods

The image acquisition protocol and processing/analysis pipeline used in this study have been previously described in detail3, and are briefly summarized here. DTI of the spinal cord was obtained in 17 individuals with chronic SCI (20-66 years, mean 47, M/F ratio: 14/3) using a Philips 3T with 16-channel neurovascular coil (total scan time=28 min, b = 0 and 500 s/mm2, 16 diffusion weighting directions, TR/TE = 6300/63 ms, SENSE factor = 2, 96x96x40 image volume matrix, 1.5x1.5x3 mm3 resolution, zero-filled to 0.57x0.57x3 mm3). DTI fiber tractography was performed to identify white matter left and right lateral, dorsal, and ventral spinal cord columns. Once each spinal column was defined, vertebrae level of C2 and C6 were identified and corresponding column profiles4,5 were created. Next, subject-specific demarcation of the injury region was performed for each individual, in order to identify three regions relative to injury (RRI; regions superior to (SRRI), at (ERRI), and below (IRRI) the injury epicenter) as well as an “all levels” (AL) ROI (defined as the entire length of the cervical cord), as shown in Figure 1. Thus, the demarcation process yielded two classes of ROIs – spinal cord columns (dorsal, ventral, right, and left columns) and spinal cord regions relative to injury (SRRI, ERRI, and IRRI). A two-way ANOVA was performed to assess whether each ROI class was a significant source of variation in FA. Finally, the ability of FA (averaged over each RRI region) to describe the severity of SCI was assessed by the degree of association between neurological measures and FA. Specifically, the American Spinal Injury Association Impairment Scale (AIS)6,7 was used to quantify motor and sensory function, and linear regression analyses were performed to determine whether the choice of RRI had any significant effect on the association between AIS and FA.

Results

Result of the ANOVA showed that a significant effect of spinal cord RRI (p=0.004), but not spinal cord columns (p=0.258), existed for FA. Based on this result, further assessment was performed using a whole cord profile, created by averaging the profiles of the four spinal cord columns. Figure 2 shows the result of the series of linear regression analyses, which were performed to assess the effect of RRI on the relationship between FA and total AIS motor and sensory scores (AISms-tot). Prior to the analysis, four FA measurements were derived from the whole cord profile of FA: a) FAAL (average FA calculated from the AL region, b) FASRRI, c) FAERRI, d) and FAIRRI. Results show that a significant (p = 0.013) relationship exists between FAAL and AISms-tot. The observed slope is positive, indicating that as impairment becomes more severe (i.e., AISms-tot value gets smaller), FAAL values decrease. This positive linear trend between FA and AISms-tot is preserved when FAAL is separated into FASRRI, FAERRI, and FAIRRI. However, the association with AISms-tot was significant only for FAIRRI (p = 0.002). Furthermore, the adjusted R2 value for the FAIRRI (0.586) was larger than that for the FAAL (0.389), indicating a tighter association with AISms-tot.

Discussion/Conclusion

DTI metrics calculated using the whole cord region (FAAL) were less specific to the degree of injury than DTI metrics calculated using subject-specific injury demarcation – specifically, metrics calculated from the IRRI region (FAIRRI), suggesting contributions from Wallerian degeneration in the descending ventrolateral motor columns1,8-10. We therefore conclude that subject-specific demarcation of injury preserves subject-specific information associated with impairment, and that diffusion measures averaged over the region below the injury epicenter may reflect impairment in individuals with SCI.

Acknowledgements

The authors thank Ms. Terri Brawner, Ms. Kathleen Kahl, Ms. Ivana Kusevic, and Mr. Joseph S. Gillen for experimental assistance. This work was supported in part by grants from the Craig H. Neilsen Foundation (338419), DOD (W81XWH-08-1-0192), and NIH (P41 EB015909).

References

1. Cohen-Adad, J., El Mendili, M. M., Lehericy, S., Pradat, P. F., Blancho, S., Rossignol, S., and Benali, H. (2011). Demyelination and degeneration in the injured human spinal cord detected with diffusion and magnetization transfer MRI. Neuroimage 55, 1024-1033.

2. Samson, R. S., Ciccarelli, O., Kachramanoglou, C., Brightman, L., Lutti, A., Thomas, D. L., Weiskopf, N., and Wheeler-Kingshott, C. A. (2013). Tissue- and column-specific measurements from multi-parameter mapping of the human cervical spinal cord at 3 T. NMR Biomed 26, 1823-1830.

3. Choe, A. S., Belegu, V., Yoshida, S., Joel, S., Sadowsky, C. L., Smith, S. A., van Zijl, P. C., Pekar, J. J., and McDonald, J. W. (2013). Extensive neurological recovery from a complete spinal cord injury: a case report and hypothesis on the role of cortical plasticity. Front Hum Neurosci 7, 290.

4. Smith, S. A., Jones, C. K., Gifford, A., Belegu, V., Chodkowski, B., Farrell, J. A., Landman, B. A., Reich, D. S., Calabresi, P. A., McDonald, J. W., and van Zijl, P. C. (2010). Reproducibility of tract-specific magnetization transfer and diffusion tensor imaging in the cervical spinal cord at 3 tesla. NMR Biomed 23, 207-217.

5. Cohen-Adad, J., El Mendili, M. M., Morizot-Koutlidis, R., Lehericy, S., Meininger, V., Blancho, S., Rossignol, S., Benali, H., and Pradat, P. F. (2013). Involvement of spinal sensory pathway in ALS and specificity of cord atrophy to lower motor neuron degeneration. Amyotroph Lateral Scler Frontotemporal Degener 14, 30-38.

6. Marino, R. J., Ditunno, J. F.,Jr, Donovan, W. H., and Maynard, F.,Jr. (1999). Neurologic recovery after traumatic spinal cord injury: data from the Model Spinal Cord Injury Systems. Arch Phys Med Rehabil 80, 1391-1396.

7. Maynard, F. M.,Jr, Bracken, M. B., Creasey, G., Ditunno, J. F.,Jr, Donovan, W. H., Ducker, T. B., Garber, S. L., Marino, R. J., Stover, S. L., Tator, C. H., Waters, R. L., Wilberger, J. E., and Young, W. (1997). International Standards for Neurological and Functional Classification of Spinal Cord Injury. American Spinal Injury Association. Spinal Cord 35, 266-274.

8. Cohen-Adad, J., Leblond, H., Delivet-Mongrain, H., Martinez, M., Benali, H., and Rossignol, S. (2011). Wallerian degeneration after spinal cord lesions in cats detected with diffusion tensor imaging. Neuroimage 57, 1068-1076.

9. Buss, A., Pech, K., Merkler, D., Kakulas, B. A., Martin, D., Schoenen, J., Noth, J., Schwab, M. E., and Brook, G. A. (2005). Sequential loss of myelin proteins during Wallerian degeneration in the human spinal cord. Brain 128, 356-364.

10. Qian, J., Herrera, J. J., and Narayana, P. A. (2010). Neuronal and axonal degeneration in experimental spinal cord injury: in vivo proton magnetic resonance spectroscopy and histology. J Neurotrauma 27, 599-610.

Figures

Figure 1. Injury region demarcation. Using sagittal T2-weighted images, three RRIs (SRRI, ERRI, and IRRI) were identified, where ERRI was defined as the regions with hyper-intense T2 contrast, and superior and inferior RRIs were defined as one vertebrae level above and below the epicenter RRI.

Figure 2. Spatial dependence of the association between FA and total AISms-tot. Linear regression analysis was performed to observe the effect of RRIs on the relationship between AISms-tot and FA. Scatter plot and estimated linear trend line are shown for each RRI.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1452