Spinal Cord Atrophy in Multiple Sclerosis: A Systematic Review and Meta-Analysis

Courtney Casserly, Estelle E. Seyman, Paula Alcaide-Leon, Melanie Guenette, Carrie Lyons, Stephanie Sankar, Anton Svendrovski, Stefan Baral, Jiwon Oh*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


BACKGROUND AND PURPOSE: Spinal cord atrophy (SCA) is an important emerging outcome measure in multiple sclerosis (MS); however, there is limited consensus on the magnitude and rate of atrophy. The objective of this study was to synthesize the available data on measures of SCA in MS. METHODS: Using published guidelines, relevant literature databases were searched between 1977 and 2017 for case-control or cohort studies reporting a quantitative measure of SCA in MS patients. Random-effects models pooled cross-sectional measures and longitudinal rates of SCA in MS and healthy controls (HCs). Student's t-test assessed differences between pooled measures in patient subgroups. Heterogeneity was assessed using DerSimonian and Laird's Q-test and the I 2-index. RESULTS: A total of 1,465 studies were retrieved including 94 that met inclusion and exclusion criteria. Pooled estimates of mean cervical spinal cord (SC) cross-sectional area (CSA) in all MS patients, relapsing-remitting MS (RRMS), all progressive MS, secondary progressive MS (SPMS), primary-progressive MS (PPMS), and HC were: 73.07 mm2 (95% CI [71.52-74.62]), 78.88 mm2 (95% CI [76.92-80.85]), 69.72 mm2 (95% CI [67.96-71.48]), 68.55 mm2 (95% CI [65.43-71.66]), 70.98 mm2 (95% CI [68.78-73.19]), and 80.87 mm2 (95% C I [78.70-83.04]), respectively. Pooled SC-CSA was greater in HC versus MS (P <.001) and RRMS versus progressive MS (P <.001). SCA showed moderate correlations with global disability in cross-sectional studies (r-value with disability score range [−.75 to −.22]). In longitudinal studies, the pooled annual rate of SCA was 1.78%/year (95%CI [1.28-2.27]). CONCLUSIONS: The SC is atrophied in MS. The magnitude of SCA is greater in progressive versus relapsing forms and correlates with clinical disability. The pooled estimate of annual rate of SCA is greater than reported rates of brain atrophy in MS. These results demonstrate that SCA is highly relevant as an imaging outcome in MS clinical trials.

Original languageEnglish
Pages (from-to)556-586
Number of pages31
JournalJournal of Neuroimaging
Issue number6
StatePublished - 1 Nov 2018
Externally publishedYes


  • Spinal cord atrophy
  • magnetic resonance imaging (MRI)
  • meta-analysis
  • multiple sclerosis
  • systematic review


Dive into the research topics of 'Spinal Cord Atrophy in Multiple Sclerosis: A Systematic Review and Meta-Analysis'. Together they form a unique fingerprint.

Cite this