Choroid plexus volume in metabolic syndrome: a cross-sectional neuroimaging study
Related Experiment Video
Updated: Nov 2, 2026
Manual Segmentation of the Human Choroid Plexus Using Brain MRI
Published on: December 15, 2023
Xinxin Huo1, Tao Chen2, Zhenyu Cheng3
1Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, China.
Objectives:
Our investigation aimed to explore the associations between choroid plexus (CP) volume and metabolic syndrome (MetS) components and cognitive-motor performance in a cross-sectional sample.
Methods:
A total of 211 participants (73 with MetS and 138 non-MetS controls) were included in this study. CP volume was automatically segmented from 3D T1-weighted images using a 3D nnU-Net. CP volume was normalized to estimated total intracranial volume (eTIV) and expressed as the CP/eTIV ratio. Multivariable linear regression analyses and Spearman correlation analyses were used to examine associations among the CP/eTIV ratio, MetS components and cognitive-motor domains. The Boruta algorithm was applied for exploratory feature selection, and receiver operating characteristic (ROC) analysis was used to evaluate the discriminative capacity of the selected variables for identifying participants with lower global cognitive-motor performance (lowest 25th percentile of global cognitive-motor composite).
Results:
Participants with MetS exhibited a higher CP/eTIV ratio than non-MetS controls, and higher CP/eTIV was independently associated with lower high-density lipoprotein cholesterol (HDL-C) levels (FDR q = 0.024) after full adjustment and FDR correction. In unadjusted analyses, the CP/eTIV ratio was correlated with multiple cognitive-motor domains. However, these associations were no longer statistically significant after adjustment for confounders. Boruta analysis identified CP/eTIV, white matter hyperintensity (WMH) volume/eTIV, and BMI as key variables, and the combined model yielded an AUC of 0.787 (95% CI: 0.718 to 0.854) for discriminating participants with lower global cognitive-motor composite. DeLong's test showed that the combined model did not significantly outperform CP/eTIV alone (0.742 vs. 0.787; FDR q = 0.084) or the model of WMH/eTIV and BMI (0.729 vs. 0.787; FDR q = 0.084).
Conclusion:
A higher CP/eTIV ratio was associated with MetS and lower HDL-C levels, but was not independently associated with cognitive-motor performance after multivariable adjustment. CP structural variation may represent an imaging correlate of metabolic disturbance, and its relevance to cognitive vulnerability requires prospective validation.
