R. Bakshi et al., MAGNETIC-RESONANCE-IMAGING FINDINGS IN 22 CASES OF MYELITIS - COMPARISON BETWEEN PATIENTS WITH AND WITHOUT MULTIPLE-SCLEROSIS, European journal of neurology, 5(1), 1998, pp. 35-48
We reviewed the magnetic resonance imaging (MRI) database of the Dent
Neurologic Institute to study the abnormal findings in myelitis. We id
entified 22 patients, and compared non-MS-related acute transverse mye
litis (ATM, n = 9), to myelitis associated with multiple sclerosis (MS
-myelitis, n = 13). The ATM patients were significantly older than MS
patients at the time of the myelitis diagnosis (mean age 46 vs 35, p <
0.05). ATM appeared as a ''longitudinal myelitis'', with fusiform cor
d expansion on T-1-weighted images and intramedullary increased signal
on T-2-weighted images, each involving multiple spinal levels (mean =
7-8). However, MS-myelitis lesions appeared focal, involving signific
antly fewer spinal levels (mean = 1-2, p < 0.001), although the lesion
s were equally likely to expand the cord. Four (42%) of the ATM lesion
s showed abnormal, variable enhancement, whereas none of the MS myelit
is lesions enhanced. Cranial MRI was more likely to be normal in ATM (
78%) than in MS-myelitis patients (15%, p < 0.001). Although readily d
istinguishable from lesions due to MS, the various etiologies for ATM,
including post-infectious (n = 2), post-vaccination (n = 3), and idio
pathic (n = 4) were indistinguishable on MRI. The MRI findings of an e
xtensively lesioned, swollen cord, suspicious for an intramedullary tu
mor and providing a temptation for a biopsy, may reflect a non-neoplas
tic inflammatory disorder.