MRI of the hippocampus in Alzheimer's disease: sensitivity, specificity, and analysis of the incorrectly classified subjects

Neurobiol Aging. 1998 Jan-Feb;19(1):23-31. doi: 10.1016/s0197-4580(98)00006-2.


In this study, magnetic resonance imaging (MRI) of the hippocampus for the diagnosis of early Alzheimer's disease (AD) is evaluated. We measured hippocampal volumes and the area of the medial hippocampus with a 1.5 T MR imager in 160 subjects: 55 patients with probable AD according to the NINCDS-ADRDA criteria, 43 subjects fulfilling the NIMH criteria of age-associated memory impairment (AAMI), 42 cognitively normal elderly controls, and 20 controls younger than 50 years. Three methods for normalization were compared. The hippocampi were atrophied in the AD patients, but not in the AAMI subjects or the elderly controls. There was no significant correlation between hippocampal volumes and age in the nondemented subjects. The discrimination based on volumetry resulted in an overall correct classification of 92% of AD patients vs. nondemented elderly subjects, whereas discrimination based on hippocampal area was less accurate, producing a correct classification in 80% of the subjects. We conclude that the hippocampus as assessed by MRI volumetry is atrophied early in AD, and spared by aging or AAMI. A brief critical review of previous studies is in concordance with the presented data: all the previous studies that have used volumetry, have similarly ended up with a good classification, whereas simpler or subjective measurements, subject to various sources of bias, have produced most variable results.

Publication types

  • Clinical Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Alzheimer Disease / classification*
  • Alzheimer Disease / pathology*
  • Female
  • Hippocampus / pathology*
  • Humans
  • Magnetic Resonance Imaging
  • Male
  • Middle Aged
  • Neuropsychological Tests
  • Predictive Value of Tests
  • Reference Values
  • Sex Characteristics