Cognitive consequences of subcortical magnetic resonance imaging changes in Alzheimer's disease: comparison to small vessel ischemic vascular dementia

Abstract

Objective: The objective of this study was to compare psychometric profiles of Alzheimer's disease (AD) patients with subcortical magnetic resonance imaging (MRI) signal abnormalities to those of AD patients without such MRI findings (normal subcortical MRI) and to those of patients with ischemic vascular dementia (IVD) associated with small and primarily subcortical ischemic changes.

Background: The cognitive significance of MRI white matter and other subcortical abnormalities in AD is unknown. Prior studies comparing AD patients with white matter changes on MRI have not included IVD patients with comparable MRI findings. If white matter/subcortical changes in AD reflect vascular abnormalities, they might be associated with cognitive profiles similar to those seen in subcortical IVD.

Method: We studied 15 AD patients with normal subcortical MRIs, 22 AD patients with subcortical MRI hyperintensities, and 18 IVD (NINCDS-ADRDA and NINDS-AIREN criteria) at the Alzheimer's Disease Research Center of the Baylor College of Medicine. IVD patients had predominantly small and subcortical signal abnormalities, and none had large cortical infarcts. AD patients had only nonspecific subcortical signal abnormalities with or without atrophy (atrophy was not analyzed). We compared the AD group with abnormal MRIs to the AD group with normal subcortical MRIs and the AD group to the IVD group using ANCOVA planned comparisons (dementia severity and education covaried).

Results: AD patients with abnormal MRIs did not differ significantly from AD patients with normal subcortical MRIs on any of the neuropsychological measures. AD patients exhibited significantly better attention/concentration, visuospatial/visuoconstructional performance, letter fluency, motor programming, and simple motor speed than IVD patients as well as significantly worse delayed verbal recognition memory. Because MRI changes were generally more extensive in IVD, a subset of AD patients with abnormal subcortical MRIs was compared to a subset of IVD patients matched for degree of MRI signal abnormalities. These subsets of AD and IVD patients still showed distinctive neuropsychological profiles.

Conclusions: AD patients with or without MRI subcortical signal abnormalities have similar neuropsychological profiles, and they differ from IVD patients with comparable MRI changes. Although MRI signal abnormalities in AD patients who have no history or examination findings of cerebrovascular disease overlap with those seen in IVD patients, they do not seem to have the same cognitive significance. Periventricular hyperintensities (PVHs) and deep signal hyperintensities, especially those of a mild to moderate degree, may reflect a different pathophysiologic process in AD than in IVD and do not necessarily have cognitive consequences in AD patients.