Quantitative diffusion weighted magnetic resonance imaging, cerebral atrophy, and disability in multiple sclerosis

OBJECTIVES To investigate the relations between quantitative diffusion coefficient MRI histograms, clinical variables, and cerebral atrophy. METHODS Twenty two patients with clinically definite multiple sclerosis and 11 healthy volunteers were studied. Histograms of apparent diffusion coefficient (ADC) from a volume of interest that included multiple slices encompassing the lateral ventricles were processed from diffusion weighted MRI. In addition, total lesion load was measured on T2 weighted dual echo images, and cerebral volume from 3D magnetisation prepared rapid acquisition gradient echo scans. All patients underwent neurological assessment, including disability on the expanded disability status scale (EDSS). RESULTS Histograms from the patient group showed a reduced peak height and a "right shift" compared with healthy controls. Peak height of the diffusion histogram correlated with both EDSS (r=-0.54, p=0.0101) and disease duration (r=-0.52, p=0.0140), but not with age. Brain volume correlated with peak height of the ADC histogram (r=0.55, p=0.0129), but not with disability. Total lesion load also correlated moderately with EDSS (r=0.46, p=0.03). CONCLUSIONS This study shows for the first time that quantitative MRI measures of diffusion correlate with clinical variables (disability, disease duration) and cerebral atrophy in multiple sclerosis. Cerebral atrophy and fixed neurological deficit may be attributed to axonal loss, which would be expected to have a significant effect on ADC. Extension of this method to more patients and longitudinal studies will further elucidate its sensitivity, reproducibility, and potential role in clinical practice and treatment trials.