Gray matter atrophy pattern in elderly with subjective memory impairment

Alzheimers Dement. 2014 Jan;10(1):99-108. doi: 10.1016/j.jalz.2013.05.1764. Epub 2013 Jul 15.

Abstract

Background: Individuals with subjective memory impairment (SMI) report worsening of memory without impairment in cognitive tests. Despite normal cognitive performance, they may be at higher risk of cognitive decline compared with individuals without SMI.

Methods: We used a discriminative function (a support vector machine) trained on an independent data set of 226 healthy control subjects and 191 patients with probable Alzheimer's disease (AD) dementia to characterize the baseline gray matter patterns of 24 individuals with SMI and 53 control subjects. We tested for associations of these gray matter patterns with SMI presence, cognitive performance at baseline, and cognitive decline at follow-up.

Results: Individuals with SMI showed greater similarity to an AD gray matter pattern compared with control subjects without SMI. In addition, episodic memory decline was associated with an AD gray matter pattern in the SMI group.

Conclusions: Our results indicate a link between the gray matter atrophy pattern of patients with AD and the presence of SMI. Furthermore, multivariate pattern recognition approaches seem to be a sensitive method for identifying subtle brain changes that correspond to future memory decline in SMI.

Keywords: Cognitive decline; Early diagnosing; Magnetic resonance imaging; Pattern recognition; Prediction; Subjective memory impairment.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Atrophy / etiology
  • Atrophy / pathology
  • Brain / pathology*
  • Cognition Disorders / diagnosis
  • Cognition Disorders / etiology
  • Cross-Sectional Studies
  • Female
  • Humans
  • Imaging, Three-Dimensional
  • Longitudinal Studies
  • Magnetic Resonance Imaging
  • Male
  • Memory Disorders / complications*
  • Memory Disorders / pathology*
  • Middle Aged
  • Neuropsychological Tests
  • ROC Curve
  • Support Vector Machine