A neuroradiological study on the influence of cerebral atrophy and white matter lesion on cognitive function in the elderly

We investigated the influence of brain atrophy and white matter lesions on cognitive function in elderly people. We selected 33 subjects (mean age, 79.2 +/- 5.1yrs) with a MMSE score from 14 to 30 who had no previous history of stroke from the outpatients in the Memory Clinic of our hospital. These subjects were divided into four groups on the basis of their MMSE score as follows: 14-20; moderate dementia (Moderate-D, n = 9), 21-23; mild dementia (Mild-D, n = 9), 24-27; mild cognitive impairment (MCI, n = 10), 28-30; normal (Normal, n = 5). Among these four groups, we compared the frequency of the associated risk factors for cerebral infarction (hypertension, diabetes mellitus, hyperlipidemia, heart disease), and the severity of brain atrophy and cerebral white matter lesion which were visually evaluated by MRI technique. Brain atrophy and white matter lesions were assessed by reviewing the cerebral cortex and hippocampus, and deep white matter lesion (DWML) and periventricular hyperintensity (PVH), respectively. Brain atrophy was divided into three grades (mild, moderate, severe) and white matter lesions were classified into four grades (0-3) using Fazekas's criteria. We performed statistical analysis to detect t parameters which correlate with and influence MMSE scores from among the MRI findings. The cases with dementia were all diagnosed as Alzheimer's disease. There were no significant differences among the four groups in mean age, the incidence of individual associated risk factors, the severity of cortical atrophy, or the grade of DWML (< or = 2) and PVH (< or = 2). However, the frequency of hippocampal atrophic change greater than a moderate grade increased in parallel with the exacerbation of reduced cognitive function (Normal; 20%, MCI: 40%, Mild-D; 56%, Moderate-D 89%), and approximately 76% with such a change were AD cases. Statistical analysis showed a significant negative correlation between the grade of hippocampal atrophy and MMSE score (r = -0.518, p < 0.005) and a great influence of hippocampal atrophy on that score (step-wise regression analysis: r = 0.518, p < 0.005). From the above results, it was suggested that more than moderate atrophic change in the hippocampus might possibly be related with cognitive impairment and that both DWML and PVH less than the second grade had little influence on the decline of brain function.