Alteration of medial temporal lobe metabolism related to Alzheimer's disease and dementia with lewy bodies

Abstract

Background: Association of medial temporal lobe (MTL) metabolism with Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) has not been evaluated considering their mixed disease (MD).

Methods: 131 patients with AD, 133 with DLB, 122 with MD, and 28 normal controls (NCs) underwent neuropsychological tests, assessments for parkinsonism, cognitive fluctuation (CF), and visual hallucinations (VH), and ¹⁸F-fluorodeoxyglucose PET to quantify MTL metabolism in the amygdala, hippocampus, and entorhinal cortex. The effects of AD and DLB on MTL metabolism were evaluated using general linear models (GLMs). Associations between MTL metabolism, cognition, and clinical features were evaluated using GLMs or logistic regression models separately performed for the AD spectrum (NC + AD + MD), DLB spectrum (NC + DLB + MD), and disease groups (AD + DLB + MD). Covariates included age, sex, and education.

Results: AD was associated with hippocampal/entorhinal hypometabolism, whereas DLB was associated with relative amygdalar/hippocampal hypermetabolism. Relative MTL hypermetabolism was associated with lower attention/visuospatial/executive scores and severe parkinsonism in both the AD and DLB spectra and disease groups. Left hippocampal/entorhinal hypometabolism was associated with lower verbal memory scores, whereas right hippocampal hypometabolism was associated with lower visual memory scores in both the AD spectrum and disease groups. Relative MTL hypermetabolism was associated with an increased risk of CF and VH in the disease group, and relative amygdalar hypermetabolism was associated with an increased risk of VH in the DLB spectrum.

Conclusions: Entorhinal-hippocampal hypometabolism and relative amygdala-hippocampal hypermetabolism could be characteristics of AD- and DLB-related neurodegeneration, respectively.

Keywords: Alzheimer disease; Lewy body disease; Metabolism; Mixed dementia; Positron-emission tomography.

Fig. 1

Group comparisons of metabolism in the MTL of the study groups. The data show the distribution of the z-transformed residuals in each group from the general linear models for MTL metabolism after controlling for age, sex, and education. P-values were obtained after correcting for multiple comparisons across 36 tests (6 MTL regions by 6 tests) with false discovery rate methods (* P < 0.05; ** P < 0.01; *** P < 0.001). Abbreviations: AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; MD, mixed disease; MTL, medial temporal lobe; NC, normal control

Fig. 2

Representative examples of MTL metabolic changes in each group. (A) An NC participant with preserved metabolism in the bilateral entorhinal cortex (black arrows). (B) patient with AD with progressive memory impairment exhibiting decreased metabolism in the bilateral entorhinal cortex (black arrows). (C) A patient with DLB with cognitive fluctuations and parkinsonism who exhibit bilateral relative hippocampal hypermetabolism (orange arrows). (D) MD patient with visual hallucinations and parkinsonism who exhibits bilateral relative amygdala hypermetabolism with an emphasis on the left side (red arrows) and left relative hippocampal hypermetabolism (orange arrow). Abbreviations: AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; MD, mixed disease; MTL, medial temporal lobe; MMSE, Mini-Mental State Examination; NC, normal control; UPDRS, Unified Parkinson’s Disease Rating Scale