White Matter Hyperintensities on MRI in the Neurologically Nondiseased Elderly

Abstract

Background and Purpose We undertook this study to evaluate the frequency and risk factors of white matter hyperintensities seen on T 2 -weighted MR imaging. We examined cohorts of neurologically nondiseased elderly subjects participating in a general-community study, the Helsinki (Finland) Aging Brain Study. Cohorts of consecutive subjects aged 55, 60, 65, 70, 75, 80, and 85 years (n=20, 18, 20, 18, 19, 18, and 15, respectively; total, n=128) were divided into a young-old (age <75 years, n=76) group and an old-old (age ≥75 years, n=52) group. Methods Frequency of hyperintensities seen on T 2 -weighted axial and coronal MR images (0.02 T) was rated using a four-point scale in periventricular and centrum semiovale areas. Results The majority of the subjects showed only mild white matter hyperintensities, which were more frequent in the periventricular areas. Age was the most important factor to explain the presence of hyperintensities. A logistic regression analysis related periventricular hyperintensities in the entire group to central atrophy (odds ratio [OR], 4.7; 95% confidence interval [CI], 1.7 to 12.9) and silent infarcts (OR, 5.6; 95% CI, 1.0 to 19.8); among the young-old, hyperintensities related to diabetes (OR, 17.0; 95% CI, 1.9 to 154.2) and central atrophy (OR, 14.7; 95% CI, 3.5 to 61.8). Centrum semiovale hyperintensities related in the entire group to cardiac arrhythmia (OR, 4.0; 95% CI, 1.0 to 15.5), central atrophy (OR, 3.9; 95% CI, 1.2 to 12.4), and silent infarcts (OR, 3.6; 95% CI, 1.0 to 12.5). Conclusions These mild white matter hyperintensities in the neurologically nondiseased elderly related especially to age and also to concomitant silent infarcts, atrophy, and some vascular risk factors. The known factors, however, explained only part of the variation. The young-old and old-old groups showed different associations. In contrast to former assumptions, the presence of white matter hyperintensities among the aged is likely to be linked to other as yet unidentified age-related factors.