Lesion load quantification in serial MR of early relapsing multiple sclerosis patients in azathioprine treatment. A retrospective study.

Azathioprine (AZA) has a slight but consistent effect on clinical outcome in multiple sclerosis (MS), but very few data are available on magnetic resonance imaging (MRI) changes. We performed a retrospective study aimed to quantify changes of lesion load in two serial proton density weighted MRI sequences (TR 2500, TE 30, 1.5 T) at a mean interval of 2.5 years in 36 relapsing-remitting (RR) MS patients: 19 had been treated with AZA, beside steroids after relapses (AZA group), and 17 had been treated with steroids only (control group). All but 3 patients were in the early phase of the disease. Total lesion area (TLA) was measured by manual outlining method and the arbitrary score proposed by Ormerod (total score) was also calculated from the number and diameter of lesions. Lesion load was the same at baseline, but median percentage difference of TLA between first and second scan was + 15.6% in control, -43.7% in the AZA group (p < 0.05, Mann-Whitney test). The distribution of patients according to TLA change, assuming that an increase or decrease was significant if larger than 50%, was found to be significantly different in favor of AZA-treated patients (chi(2) = 35.92, p < 0.001). These results suggest an effect of AZA treatment on MRI lesion load in early RR MS: a larger prospective study is worthwhile.