Modifications to the McDonald MRI dissemination in space criteria for use in Asians with classic multiple sclerosis: the Taiwanese experience

Objective: The study is to examine the accuracy of modified MRI dissemination in space criteria for prediction of conversion from clinically isolated syndrome to definite multiple sclerosis according to patients' baseline MRI. Methods: From 2001 to 2007, we recruited 67 patients with clinically isolated syndrome. They had been followed-up until development of definite multiple sclerosis or remaining as clinically isolated syndrome for more than 2 years. The non-converters were taken as negative cases. The 67 patients' baseline MRIs were evaluated by two radiologists and determined as either fulfilling or not fulfilling McDonald and modified MRI criteria for dissemination in space. Results: Thirty-two patients converted to definite multiple sclerosis and 35 did not. The modified criteria are slightly more sensitive (53.1% vs. 50.0%) and accurate (77.6% vs. 76.1%) as compared with McDonald criteria. However, further reduction of the cutoff of abnormal MRI criteria from three of four to two of four criteria yields best sensitivity (71.9%) and accuracy (83.6%). Conclusions: Modifications to the McDonald MRI dissemination in space criteria (by using fewer T2 lesions, removal of the restriction on the spinal cord lesion and reduction of the cutoff of MRI criteria) are more appropriate for use in the Taiwanese population for the diagnosis of classic multiple sclerosis.