抄録
Purpose: Our aim was to re-evaluate the usefulness of perfusion single-photon emission computed tomography (SPECT) for
the diagnosis of AD and DLB.
Methods: A total of 1,232 patients, including 180 cognitively normal subjects (NSs), 148 patients with amnestic mild cognitive
impairment (aMCI), 802 patients with AD and 102 patients with dementia with Lewy bodies (DLB) were enrolled. A voxel-based stereotactic extraction estimation analysis was used to assess hypoperfusion areas using the cut-off value of 20% for
the “extent %” and >2 hypoperfused lesions.
Results: The top 10 areas with hypoperfusion in AD were the angular gyrus, supramarginal gyrus, precuneus, superior
occipital gyrus, inferior parietal lobule, middle temporal gyrus, superior parietal lobule, inferior temporal gyrus, superior
temporal gyrus. and posterior cingulate, while DLB had 6 areas in common with AD. The respective sensitivity and specificity
for differentiating AD from NSs were 84.2% and 78.9% for 50-59 years old, and 91.8% and 70.0% for 60-69 years old, but
those values decreased from 70 years old. The respective sensitivity and specificity in DLB patients were 100% and 75.6%
for 50-69 years old, 84.6% and 70.1% for 70-79 years old and 81.4% and 77.1% for 80-89 years old. In aMCI patients, the
1-year change in the MMSE scores was -2.38±2.02 (n=16) in patients with abnormal SPECT images and -0.67±1.81 (n=18) in
those with normal SPECT images (P<0.02).
Conclusions: Our study confirms that perfusion SPECT is still a valid alternative for the diagnosis of AD and DLB when
fluorodeoxyglucose-PET is unavailable.
