Quantitative cerebral MR in rheumatoid arthritis.

PURPOSE To determine the presence of hyperintense white matter lesions and atrophy reflecting cerebral vasculitis in rheumatoid arthritis.

METHODS Thirty-three patients with rheumatoid arthritis and 48 control subjects were examined with MR. Mean age was 45.1 years (range, 26 to 55 years) for the patients and 42.2 years (range, 25 to 55 years) in the control group. To determine atrophy we measured the area of corpus callosum, the cerebrum, and the cerebellum on midline sagittal sections. On transverse images, the ventricle-to-brain ratio, the bifrontal ratio, and the bicaudate ratio were selected as atrophy parameters. Area and signal intensity were measured for the biggest and the smallest lesions in both groups.

RESULTS Nine patients (27%) had hyperintense lesions compared with 15 (31%) of the control subjects. Mean numbers of hyperintense lesions were 1.3 in patients and 2.1 in control subjects. Mean area of the largest lesion in each patient was 27.4 mm2 for the patients and 29.8 mm2 in the control group. In patients with long disease duration (> 15 years) the mean ventricle-to-brain ratio was 0.09 compared with 0.08 in the control subjects. The midsagittal area of the cerebellum was 1349.8 mm2 in the patients with long disease duration and 1573.3 mm2 in the control group. No difference in number of hyperintense white matter lesions was detected between patients with long disease duration and the control subjects. Comparing the total group of patients with the control subjects, no significant differences in atrophy parameters or hyperintense white matter lesions were found. Also, there were no significant differences in relative signal intensity of the hyperintense lesions and corpus callosum between the two groups. We were not able to detect differences between treated versus untreated patients.

CONCLUSION This study indicates a tendency of more cerebral and cerebellar atrophy in patients with severe rheumatoid arthritis. The number and size of the white matter lesions were not significantly different in the two groups and do not support a higher frequency of even clinically silent infarcts caused by vasculitis in the patients with rheumatoid arthritis compared with control subjects.