Article Text

Download PDFPDF

Cervical cord and brain grey matter atrophy independently associate with long-term MS disability
  1. U Bonati1,2,
  2. L K Fisniku1,2,
  3. D R Altmann1,3,
  4. M C Yiannakas1,2,
  5. J Furby1,2,
  6. A J Thompson1,4,
  7. D H Miller1,2,
  8. D T Chard1,2
  1. 1MS Nuclear Magnetic Research Unit, UCL Institute of Neurology, London, UK
  2. 2Department of Neuroinflammation, UCL Institute of Neurology, London, UK
  3. 3Medical Statistics Unit, London School of Hygiene and Tropical Medicine, London, UK
  4. 4Department of Brain Repair and Rehabilitation, UCL Institute of Neurology, London, UK
  1. Correspondence to Dr Declan Chard, Department of Neuroinflammation, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK; d.chard{at}

Statistics from

In MRI studies of multiple sclerosis (MS), brain grey matter (GM) and spinal cord atrophy correlate with disability.1 2 However, it is uncertain whether they independently associate with long-term disability. We measured upper cervical cord cross-sectional area (UCCA) in a cohort of patients ∼20 years after presenting with a clinically isolated syndrome (CIS) suggestive of MS. Associations of brain GM atrophy and white matter (WM) lesion load with disability had been previously observed in the cohort.1 We now report UCCA measurements and independent associations of the cord and brain measures with disability.

Seventy patients presenting with CIS had analysable brain and MRI scans acquired on a 1.5 T scanner after a median of 20 years (range 18–27). Clinically definite MS (CDMS) was diagnosed clinically.3 Disability was assessed with the expanded disability status scale (EDSS)4 and MS functional composite score (MSFC) plus its three components, paced serial auditory attention test (PASAT; 3 s interval), nine-hole peg test (9HPT) and 25-foot timed walk (T25FW). CDMS subgroups were defined as relapsing–remitting (RR) or secondary progressive (SP) MS; those with an EDSS ≤3 were classified as benign.5

At the 20-year follow-up, 27 remained CIS, 32 had RRMS (21 benign MS, 11 non-benign RRMS (EDSS>3)), and 11 had SPMS. UCCA and brain MR parameters were measured as previously described.1 2 UCCA was also measured in 17 healthy volunteers.

Statistical analysis was performed using the SPSS Version 11.0 (SPSS, Chicago, Illinois) and Stata 9.2 (Stata Corporation, College Station, Texas). Subgroup comparisons of UCCA were performed using linear regression with total intracranial volume (TICV), age and gender as covariates. The Spearman rank-correlation (rs) evaluated the correlation of UCCA with EDSS, MSFC, WMF, GMF and lesion load.

The association of UCCA, GMF, WMF and lesion load with long-term disability was assessed using ordinal logistic regression (for EDSS) and linear regression (for MSFC). EDSS (as an ordinal scale) and MFSC (as a continuous variable) were modelled as response variables, with UCCA, GMF, WMF, lesion load (from T2-weighted MRI scans), TICV, age and gender as potential associations. No significant effects of CIS type were found, so this was not included. Lesion load was log-transformed to improve normality before including it in the model (lesion load=zero was assigned a nominal log volume value of 0.01).

The R2 obtained from linear regression analysis was used to compare the proportion of variability in MSFC scores explained by each MRI measure alone (from regressions with the single MRI measure) and by the composite of MRI measures (in the multiple regressions).

The median age of patients was 51 years (range 39–68) and controls 41 years (31–59). Forty-eight/70 patients and 10/17 controls were female. The median EDSS was 2.5 (0–8) for all patients and 3.5 (1–8) for CDMS.

The mean UCCA was 83.17 mm2 in controls (SD 8.32), 79.22 mm2 (10.19) in those who remained CIS, 75.07 mm2 (8.32) in RRMS (78.52 mm2 (5.63) in benign and 68.08 mm2 (7.20) in non-benign RRMS) and 66.76 mm2 (5.99) in SPMS. UCCA did not differ significantly between controls and CIS or benign MS but did between controls and SPMS (−15.88, −26.24 to 12.80, p<0.001) and non-benign RRMS (−9.36, −15.07 to −3.64, p=0.002).

UCCA correlated significantly with EDSS and MSFC z–scores for all patients (table 1), and similarly for those with CDMS (data not shown).

Table 1

Spearman correlations between clinical and MR measures in all patients

In all patients, UCCA correlated with WMF (rs=0.32; p=0.007) and lesion load (rs=−0.31; p=0.009) but not GMF (rs=0.20; p=0.097). In CDMS only, UCCA correlated with WMF (rs=0.30; p=0.052) but not GMF (rs=0.24, p=0.120) or lesion load (rs=−0.11; p=0.471).

In all patients, GMF, UCCA and gender were independently associated with EDSS: there was an estimated 64% (95% CI 38 to 79, p<0.001) reduction in odds of a patient having a higher EDSS per 1SD greater GMF and a 61% (95% CI 35 to 76, p<0.001) reduction in odds of a patient having a higher EDSS per 1SD greater UCCA. Males showed higher EDSS scores compared with females (p=0.031). GMF and UCCA were independently associated with MSFC: there was a 0.55 (p<0.001) increase in MSFC per 1SD greater GMF and a 0.31 (p=0.010) increase per 1SD greater UCCA. The R2 for UCCA was 0.15, and for GMF 0.30; the combined R2 was 0.37.

In CDMS only, UCCA and lesion load were independently associated with EDSS: there was an estimated 70% (95% CI 41 to 85, p=0.001) reduction in the odds of a patient having a higher EDSS per 1SD higher UCCA. A 10% increase in lesion load was associated with an 8% (95% CI 2 to 13%, p=0.007) increase in odds of a higher EDSS. GMF and UCCA were independently associated with MSFC: there was a 0.55 (p=0.005) increase in MSFC per 1SD greater GMF and 0.43 (p=0.024) increase in MSFC per 1SD greater UCCA. R2 for GMF was 0.26, and for UCCA 0.20; the combined R2 was 0.35.

In a previous report of this cohort, brain GM atrophy was found to be associated with disability independently of WM lesion load at 20 years.1 In the present study, UCCA was associated with disability, independently of brain GM atrophy. UCCA was associated with the 9HPT and T25FW (which measure motor function). As previously reported,1 brain GM atrophy was also associated with the 9HPT and T25FW. However, the regression analyses suggest that the functional effects of pathology in these two regions are in part independent. This may indicate that mechanisms of spinal cord and brain GM atrophy differ, or that they are linked but temporally dissociated. These results are mechanistically plausible, with arm (9HPT) and leg (T25FW) reliant on intact spinal cord sensorimotor tracts and neocortical and cerebellar neuronal integrity.

Spinal cord and GM atrophy measures in combination only explained about a third of the variance in disability, suggesting that atrophy measures do not capture all clinically relevant pathology in MS. Inclusion of other MRI data, such as magnetisation transfer and diffusion imaging measures, may increase the correlations in future studies.

The observation that cervical cord atrophy and brain GM atrophy independently associate with long-term disability in MS should encourage further research on imaging and pathology of both brain grey matter and spinal cord in MS.


We would like to thank J Jackson and K Miszkiel for their help with the analysis of the brain MRI measures.


View Abstract


  • Funding The NMR Research Unit is supported by the MS Society Great Britain and Northern Ireland, and Department of Health's NIHR Comprehensive Biomedical Research Centre at UCLH; UB is supported by the MS Society of Switzerland.

  • Competing interest AJT has received honoraria for consulting services, speaking, and serving on scientific advisory boards from Novartis, Eisai, Weleda/Society for Clinical Research, Hoffman La Roche, UCB Pharma, Serono Foundation, Sanofi-Aventis and the MS Society of Great Britain and Northern Ireland. He is Editor-in-Chief of Multiple Sclerosis for which he receives an honorarium from Sage Publications. DHM has received honoraria from UCB Pharma, Schering, Biogen Idec, GSK and Wyeth for consulting services, speaking and serving on a scientific advisory boards. He has received reimbursement for work as co-Chief Editor of Journal of Neurology, and research grant support from the MS Society of Great Britain and Northern Ireland, Wellcome Trust, Medical Research Council UK, Biogen Idec, Novartis, GlaxoSmithKline and Schering.

  • Ethics approval Ethics approval was provided by the National Hospital for Neurology and Neurosurgery and UCL Institute of Neurology Joint Research Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.