Magnetic resonance imaging (MRI) is being used as an outcome criterion in therapeutic trials in multiple sclerosis (MS) on the assumption that it, as a sensitive marker of biologic disease activity, could serve as a surrogate marker of disability. We evaluated the relation between MRI findings and disability in a quantitative follow-up study of 48 MS patients. Median duration of follow-up was 24 months (range, 10 to 42 months). Computer-assisted volume measurements employing a seed-growing technique yielded a standard error of measurement of 0.275 cm2. The median total area of the hyperintense lesions on the initial T2-weighted images was 8.4 cm2. The median increase was 0.76 cm2/yr (9%). In a subgroup (n = 19) with short-TR/short-TE spin-echo (SE) images, we measured the hypointense lesion load. The median total area of the lesions at entry was 0.70 cm2, with a median increase of 0.28 cm2/yr (40%). The total area of the hyperintense lesions on the initial T2-weighted images showed a weak correlation with the Expanded Disability Status Scale score at entry (Spearman rank correlation coefficient [SRCC] = 0.30; 0.02 < p < 0.05). The increase in disability showed a positive correlation (SRCC = 0.19) with the increase in hyperintense lesion load on the T2-weighted SE images, but this correlation did not reach statistical significance (p = 0.09), probably because of lack of clinical progression.(ABSTRACT TRUNCATED AT 250 WORDS)