Neurodegenerative disorders are a large group of diseases characterized by progressive regionally specific pathologic changes. The most common neurodegenerative disorder among the elderly is AD. AD typically presents with cognitive decline, most notably in verbal and visual memory and in the visuospatial domain, as a result of progressive deterioration of the posterior association cortices and the hippocampal formations.
Unlike brain atrophy, PET metabolic changes are not considered to be a normal part of aging. Our study included 19 cognitively normal subjects who were initially evaluated and re-evaluated approximately 2 years later with cognitive examinations and PET imaging. The longitudinal changes localized to the lateral posterior association cortices, the posterior cingulate and precuneus bilaterally, and the left middle and inferior frontal gyrus. While both the decliners and nondecliners experienced relative metabolic decline, the former demonstrated greater changes in the areas discussed above (e.g., 5–15% vs. 2.5–5%). ApoE4 carriers showed greater metabolic decline in posterior brain regions, as is typically seen in patients with early AD. The quantitative differences between decliners and nondecliners and ApoE4 carriers and noncarriers did not however reach statistical significance likely because of limitations imposed by our low sample size.
So far, several research groups have reported that metabolic decline in the temporal [5
] and/or parietal [5
] areas is predictive of clinical diagnosis of AD within a follow-up period ranging from 12 to 36 months on average. Posterior hypometabolism in ApoE4 carriers relative to noncarriers was also noted in several studies [10
]. One group conducted 3-year follow-up of cognitively normal individuals and reported entorhinal hypometabolism in normals that converted to MCI in follow-up and newly developed hippocampal and lateral and inferior temporal hypometabolism in follow-up (e.g., in the MCI stage) [12
]. By using a sensitive computational anatomy technique to map progressive metabolic changes, our study extends these findings by reporting more widespread progressive changes encompassing the association cortices of the posterior temporal, superior, and inferior lateral parietal, occipital, precuneus, and posterior cingulate. The lateral but not the medial cortical changes also showed linkage with the presence of an ApoE4 allele in the present study.
A major strength of this study is its prospective longitudinal design coupled with a sensitive surface mapping technique well suited for assessing intra-individual longitudinal change. Surface feature-guided cortical alignment improves the comparison of corresponding cortical areas between individuals and results in greater statistical power to detect disease-induced changes. Similar to several other longitudinal studies in normal elderly and MCI [8
], we normalized the individual average PET intensity to the group average global brain intensity, which may lead to underestimation of the regional changes [40
]. The diagnostic criteria for cognitive decline used in the present study are similar to those used in some [5
] but not all [12
] longitudinal PET studies of cognitively normal elderly and mild cognitive impairment subjects.