Abstract Information


Spinal cord atrophy after spinal cord injury, a meta-analysis and replication.

1Linnman C, 2Dahlberg L, 3Morse L
1Boston Children's Hospital and Spaulding Rehab Hospital, Boston, MA, USA; 2Boston Children's Hospital, Boston, MA, USA; 3Craig Hospital, Denver, CO, USA

After a spinal cord injury (SCI), there is rapid atrophy of the spinal cord above the site of the lesion. Here, we review the literature, and provide new data on spinal cord cross-sectional area in a preliminary analysis of 17 patients and 11 healthy controls.
Meta-analysis: We identified eight publications (1-8) on cervical spinal cord cross-sectional area in patients with traumatic SCI. A total of 141 patients and 152 healthy controls were included in the meta-analysis. On average, patients showed a 14 mm2, or 17%, reduction in cross-sectional area of the spinal cord at the C2 level, corresponding to a Cohen’s D effect size of 1.76 (range 0.78 to 5.69).
The initial rate of atrophy, evaluated in subacute patients tracked longitudinally, corresponded to a linear spinal cord cross section reduction of about 0.4 mm2 (or 0.7%) per month in the first 12 months (1), and an 8% reduction in spine cross-section was reported in patients approximately 9 weeks post-injury, with larger reductions associated with poor motor recovery (8). Lundell et al (6) did not find a relation between atrophy and duration. Our meta-analysis, however, suggests such a logarithmic relation, with cord atrophy plateauing years after injury.
There are mixed findings with regard to the impact of lesion level and cord atrophy: Lundell et al (6) report a modest correlation between lesion level and atrophy, and Jutzeler et al (5) report greater atrophy in tetraplegia compared to paraplegia, while Freund et al (1) found no relation between lesion level and the rate of atrophy.

Present study: We report initial results from a cohort of 17 subjects with traumatic SCI, and 11 healthy controls. Methods: The spinal cords of 17 spinal cord injury patients, and 11 healthy controls were segmented and analyzed with the Spinal Cord Toolbox (9), a semi-automated method employing spinal cord straightening and warping to a common template.
We found that C2 area was 58.13 mm2 (± 6.52) for patients, and 67.99 mm2 (± 11.37) for control subjects, p=0.003. We also observed a significant correlation between numerical level of injury and C2 cross sectional area, r = 0.628, p=0.003. We did not, however, observe a relationship between C2 cord thickness and time since injury (mean 10 years, range 1 to 37 years).

Conclusions: Meta-analysis and our present data confirm extensive atrophy of the cervical spinal cord after SCI. Atrophy appears to occur within the first years post-injury, and may be more extensive the higher the lesion. Given the magnitude and consistency of the effect, cord cross-sectional area is a sensitive marker for injury, but large studies, preferably longitudinal and long-term, are needed to define individual contributions of time, lesion level and completeness. Ongoing studies on rehabilitation efforts will determine if rehab can slow the rate of atrophy, thereby preserving the cord for when restorative therapies become available.


You Must Be Logged In

This Reminders tab will display for your scheduled lectures 15 minutes before start time and for new messages waiting.