Evolution of specific cognitive subprofiles of mild cognitive impairment in a three-year longitudinal study.

The present work addresses one of the currently most controversial aspects of early detection of Alzheimer's disease (AD) and other dementias; that is, the identification of the Mild Cognitive Impairment (MCI) syndrome-in some cases, prior to AD-in a sample of older subjects who are healthy from the cognitive viewpoint. In a three-year longitudinal study, we classified the participants between 58 and 90 years of age in different cognitive profiles: healthy and MCI (amnestic MCI, non-amnestic MCI, and multi-domain MCI). We followed the evolution of each one by means of the administration on three occasions of an extensive battery of neuropsychological tests. We have found a high percentage of MCI in our sample. Although some of them were amnestic MCIs, this group was not the most frequent. The multi-domain MCI is the one that evolves directly into AD, not the amnestic MCIs. We have found diverse evolutional trajectories over the past three years, some expected, others somewhat unexpected. We also point out the methodological difficulties posed by the administration of certain episodic memory tests, which is not the most appropriate to detect subclinical MCI, due to the effect of practice.