Report-based arterial FDG uptake and the NAFLD–atherosclerosis overlap phenotype: a two-centre cross-sectional PET/CT study with an exploratory decomposition of the hepatic FDG signal
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background Qualitative arterial fluorodeoxyglucose (FDG) uptake descriptions in routine positron emission tomography/computed tomography (PET/CT) reports may contain information relevant to cardiometabolic disease. We examined their association with coexisting non-alcoholic fatty liver disease ( N AFLD) and atherosclerosis, and whether report-based hepatic FDG uptake statistically accounted for this association. Methods This retrospective cross-sectional study included 556 adults from two Chinese centres (Xiamen, n = 222; Xiangya, n = 334). Arterial FDG uptake was graded from 0 to 3 using routine reports. Four mutually exclusive phenotypes (neither condition, NAFLD only, atherosclerosis only and overlap) were ascertained independently of PET uptake. Centre-specific multinomial logistic regression assessed associations per one-grade increase. The clinically adjusted model was primary, and its stability was checked by bootstrap and penalised likelihood. Full adjustment in 471 participants served as a saturation test. Pairwise contrasts compared phenotype-specific associations. Exploratory cross-sectional decomposition evaluated hepatic FDG uptake as a statistical intermediate. Results Among participants with NAFLD, 99.4% met at least one cardiometabolic criterion for metabolic dysfunction-associated steatotic liver disease (MASLD). Higher arterial grades were associated with the overlap phenotype in both centres (adjusted odds ratio [OR] 2.23, 95% confidence interval [CI] 1.49–3.34 in Xiamen; OR 1.83, 95% CI 1.37–2.46 in Xiangya). Bootstrap, penalised and fully adjusted estimates were consistent. Associations were stronger for overlap than for either condition alone, but there was no evidence that the overlap association exceeded the combined single-condition associations on the log-odds scale. No exposure-by-centre interaction was detected ( P = 0.622). Hepatic uptake statistically accounted for 11.0% (Xiamen) and 21.1% (Xiangya) of the total association, and the indirect-component CI included the null in Xiamen. This component disappeared when the outcome excluded the hepatic condition. Alternative exposure definitions and sensitivity analyses, including selection and reporting-bias analyses, supported the primary association. Conclusions Report-based arterial FDG uptake grades were associated with coexisting NAFLD and atherosclerosis across two centres. The hepatic component largely reflected steatosis-related construct overlap, although a shared metabolic substrate such as insulin resistance cannot be excluded. These hypothesis-generating findings support further investigation of routine report text for opportunistic cardiometabolic phenotyping but do not establish causality or clinical utility.
Authors: Fu Su(Conceptualization, Writing – original draft, Writing – review & editing), Hong Zhang(Data curation, Funding acquisition, Writing – review & editing), Chao Ma(Data curation, Writing – review & editing), Liang Xu(Data curation, Writing – review & editing), Qun Fan(Data curation, Writing – review & editing)
Language: English
Subfield: Hepatology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7235747002
