Glucose metabolic reduction of elderly subjects with subjective cognitive impairment using 3-dementional SPM approach with MRI-mask correction for brain atrophy.

Objectives Subjective cognitive impairment (SCI) is receiving increasing attention as a pre-mild cognitive impairment condition in the course of the clinical manifestation of Alzheimer disease (AD). However, brains of elderly people become atrophic with age, and brain atrophy makes false positive artifact for diagnosis of demented brain using FDG-PET, resulting in decreasing diagnostic accuracy. The purpose of the study was to propose new voxel-based 3-dementinal Z-score mapping software using SPM correcting brain atrophy with MRI, and to evaluate glucose metabolism for elderly subjects with SCI and to compare with patients with mild cognitive impairment (MCI) and AD.

Methods Thirteen subject with SCI (aged 65-80 years) and 7 patients with MCI and probable AD (aged 65-79 years) were examined using FDG-PET and MRI and neuropsychological tests such as MMSE, MoCA, CASIS, GDS, every day memory checklist based on Wilson’s list and the test of awareness of memory deficits. The awareness of memory deficits was defined as the difference of score between subject own rating and caregiver’s rating. FDG-PET images were analyzed using new voxel-by-voxel analyzing software, statistical parametric Z-score mapping based on SPM (Zsmap). In brief, images were spatially normalized, smoothed and compared with the mean and standard deviation of PET images of age-matched control subjects with correcting brain atrophy using their own MRI. Mean values of Z-scores were superimposed on the standardized MRI template with a color scale and measured with Talairach atlas that was equipped with hippocampal slices. Zsmap worked semi-automatically and provided parametric images within a couple of minutes. Statistical significance was determined as Z-score ≧ 2.0.

Results Glucose metabolism was decreased in the anterior cingulate (71%), posterior cingulate and the precuneus (100%), the superior parietal cortex (29%) and parahippocampus (57%) for patients with MCI and AD. On the other hand, glucose metabolic reduction was found in the anterior cingulate (77%) and posterior cingulate and the precuneus (23%) for subjects with SCI, and the reduction in the anterior cingulate was dominant. Depression was found in 43% of patients with MCI and AD, however, no depression in subjects with SCI. Scores of subjective awareness of memory deficits for the subjects with SCI were higher than those of objective awareness observed by caregivers.

Conclusions We found that glucose metabolism is reduced in the cingulate cortex, especially anterior cingulate, for the subjects with SCI, the finding of which is not related to depressed conditions but to subjective cognitive decline. We suppose that glucose metabolic reduction in the cingulate cortex is ahead of reduction in other regions, and some of those subjects with SCI may convert to MCI later. The proposed software, Zsmap, is useful to evaluate glucose metabolism for elderly subjects with SCI who have brain atrophy.