Problems with diagnostic criteria for multiple sclerosis

• Published in: The Lancet (British edition) 2001-11, Vol.358 (9295), p.1746-1747
• Main Authors: Poser, Charles M, Brinar, Vesna V
• Format: Article
• Language: English
• Subjects: Humans; Lesions; Magnetic Resonance Imaging; Medical diagnosis; Multiple sclerosis; Multiple Sclerosis - diagnostic imaging; NMR; Nuclear magnetic resonance; Radiography; Reproducibility of Results; Spine
• ISSN: 0140-6736; 1474-547X
• DOI: 10.1016/S0140-6736(01)06842-8

The new MRI criteria proposed may be an improvement over previously published ones, yet they will not differentiate between MS and other relapsingremitting neurological diseases, such as recurrent and multiphasic encephalomyelitis and recurrent neuromyelitis optica, and the ones in which MR images mimic those of MS. The new MRI criteria provide quantitative and anatomical guidelines (eg, nine lesions in the periventricular white matter, or a spinal-cord lesion) but lack qualitative and descriptive ones (eg, large confluent lesiosn), which are much more difficult to delineate. The provision of images as examples of what is not MS would be helpful. The discussion of spinal lesions is important but could have underlined the importance of routinely obtaining MRIs of the cervical spine to rule out possible cord compression by herniated disks or spondylotic ridges, disorders that are generally remediable and whose symptoms are sometimes mistaken for those of MS. A valuable note made in the paper describing the new set of criteria is that a spinal-cord lesion exceeding two vertebral segments in length is a point against the diagnosis of MS. The discussion of the criteria add valuable guidance on the use of the MRI in determining dissemination in time. The authors also emphasise the near impossibility of differentiating between MS and disseminated encephalomyelitis on the basis of a biopsy, a procedure to be undertaken only very rarely for the diagnosis of MS.