Upper cervical spinal cord cross-sectional area in relapsing remitting multiple sclerosis: application of a new technique for measuring cross-sectional area on magnetic resonance images.

To measure accurately the upper cervical cord cross-sectional area (CSA) in patients with relapsing remitting multiple sclerosis (RRMS), and normal control subjects, to address the paradox that longitudinal reduction in CSA has been detected in RRMS while reduction compared to controls has not. We hypothesized that a lack of statistical power and/or measurement sensitivity due to partial volume averaging in previous studies contributed to this paradox. Using a technique that corrects for partial volume averaging, we measured the CSA in 35 normal controls and 35 RRMS patients. We used the total intracranial volume (TICV) to normalize the CSA and therefore reduce the normal variance and improve the statistical power. The mean TICV did not differ between groups. Statistical power analysis indicated that a 5% reduction in CSA in the patients could be detected with an estimated power of 0.74 before normalization and 0.9 after. The mean CSA in the patients was not reduced compared to controls after (P = 0.928) or before (P = 0.881) normalization. Using a sensitive analysis method, and apparently appropriate statistical power, we did not detect reduced CSA in RRMS patients. We hypothesize that this may be due to inflammation. Copyright 2007 Wiley-Liss, Inc.