AI Summary & Key Takeaways
AI InsightThese hypothesis-generating findings support prospective investigation using standardized neuropsychological phenotyping, continuous quantitative PET measures, and independent etiological characterization.
📌 3 Key Takeaways
Abstract
Aims: Brain 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) provides functional information on regional cerebral glucose metabolism and may support evaluation of cognitive impairment. This study investigated associations between presenting clinical features and the topographic distribution of CortexID-derived regional hypometabolism in patients undergoing dementia-related brain 18F-FDG PET/CT. Methods: This retrospective exploratory study included 53 patients scanned at a tertiary referral center between January 2017 and January 2021. Presenting clinical features were coded as present or absent, and images were assessed visually and by CortexID semiquantitative analysis. Regions with a Z-score ≤-2.5 were classified as significantly hypometabolic. Associations were evaluated using Fisher’s exact test, odds ratios with 95% confidence intervals, Benjamini–Hochberg false discovery rate (FDR) correction, Spearman rank correlation, and supportive age- and gender-adjusted logistic regression. Results: The mean age was 66.1±4.7 years, and 46 patients (86.8%) were male. Memory complaint was most frequent (69.8%) and showed nominal associations with parietal, temporal, and combined parietotemporal patterns, whereas speech disturbance showed nominal associations with frontal hypometabolism, particularly the left frontal region. These associations remained directionally consistent in age- and gender-adjusted models but none survived FDR correction across the complete regional or composite-pattern families (lowest adjusted q=0.103). Symptom burden was uncorrelated with the number of threshold positive hypometabolic regions (Spearman ρ=0.035, p=0.805). Conclusion: Presenting memory and speech-related symptoms showed anatomically coherent but multiplicity-sensitive associations with regional hypometabolic topography on brain 18F-FDG PET/CT. These hypothesis-generating findings support prospective investigation using standardized neuropsychological phenotyping, continuous quantitative PET measures, and independent etiological characterization.