Significant correlation between cerebral hypoperfusion and neuropsychological assessment scores of patients with mild cognitive impairment.

PURPOSE: The regions of significant correlation between cerebral hypoperfusion and neuropsychological assessment scores were evaluated using (99m)Tc-HMPAO SPECT in patients with mild cognitive impairment (MCI) on the basis of its three subtypes, namely, single-domain amnestic MCI (aMCI-s), multiple-domain amnestic MCI (aMCI-m), and nonamnestic MCI (naMCI), following which comparisons were made among the three subtypes of MCI.

METHODS: Regions of hypoperfusion were determined by comparing the three groups with the normal group. Neuropsychological assessment included tests to evaluate attention, language and related functions, visuospatial function, memory, frontal-executive function, and mini-mental state examination and depression scores. Regions of cerebral hypoperfusion were identified by comparing the three groups of MCI patients with the normal group (P<0.05, uncorrected). One-way analysis of variance was used to examine differences across groups, and post-hoc a-priori pairwise comparisons were used for between-group analyses. The regions of significant correlation, related to the neuropsychological assessment scores, were identified by simple regression of SPM8 within the masking image of the area of cerebral hypoperfusion (P<0.05, uncorrected).

RESULTS: The regions of cerebral hypoperfusion were identified by comparing members of the normal group with patients with aMCI-s, aMCI-m, and naMCI. The patients with aMCI-m showed significant correlation with all neuropsychological assessment scores, but the patients with aMCI-s correlated with four neuropsychological assessment scores of attention. The patients with naMCI revealed no significantly correlated regions (P<0.05, uncorrected). The regions that correlated with neuropsychological assessment scores in patients with aMCI-s were very small compared with those in patients with aMCI-m. The correlated regions in patients with aMCI-m were restricted to the left cerebrum and cerebellum. Brain areas showed significant correlation between neuropsychological assessment scores and hypoperfusion, which was evaluated by simple regression with the threshold being P less than 0.05, uncorrected. Rey complex figure test 20 min delayed, Korean-color word stroop test word reading, and Korean mini-mental state examination scores correlated more strongly with cerebral hypoperfusion compared with other assessment scores.

CONCLUSION: The specific pattern of significant correlation of cerebral hypoperfusion with neuropsychological assessment scores was classified into three subtypes (aMCI-s, aMCI-m, and naMCI) according to the patients' deficits in their cognitive domains.