Objective: To examine associations between dementia severity and quantitative magnetic resonance imaging measures of cortical gray matter volume and abnormal white matter volume in 52 patients diagnosed with probable Alzheimer disease.
Design: Analysis of the relationship between magnetic resonance imaging volume measures and dementia severity using multiple regression and Pearson correlations.
Setting: Alzheimer's Disease Research Center, University of California, San Diego.
Participants: Twenty-three men and 29 women with probable Alzheimer disease (average age, 71.7 years; average education, 13.3 years).
Main outcome measures: The Mattis Dementia Rating Scale (MDRS) and the Mini-Mental State Examination.
Results: Using simultaneous multiple regression, magnetic resonance imaging volumetric measures of cortical gray matter and abnormal white matter were independently associated with dementia severity measured by either the MDRS or the Mini-Mental State Examination. Cortical gray matter volume and abnormal white matter volume also made independent contributions to performance in 4 of 5 cognitive domains assessed by the MDRS. Regional analysis indicated that limbic cortical gray matter volume and nonlimbic cortical gray matter volume were also correlated with the MDRS score; however, in the regression analysis the individual gray matter measures were not independently associated with MDRS performance. A similar analysis revealed statistically independent relationships of limbic gray matter volume and abnormal white matter volume, but not nonlimbic cortical gray matter volume, to Mini-Mental State Examination performance.
Conclusions: Quantitative magnetic resonance methods provided strong evidence that cortical gray matter volume, which may reflect atrophy, and abnormal white matter volume are independently related to dementia severity in probable Alzheimer disease: lower gray matter and higher abnormal white matter volumes are associated with more severe dementia.