The Role of Iron Homeostasis Imbalance in T2DM‐Associated Cognitive Dysfunction: A Prospective Cohort Study Utilizing Quantitative Susceptibility Mapping

Figures

• 

  FIGURE 1. Patient selection and study design. Flow chart of the cohort patient selection process and inclusion and exclusion criteria. AVLT, auditory verbal learning test; FBG, fasting blood glucose; HbA1c, glycated hemoglobin; HDL, high‐density lipoprotein cholesterol; LDL, low‐density lipoprotein cholesterol; MEDI+0, morphology‐enabled dipole Inversion with an automatic uniform cerebrospinal fluid zero reference algorithm; MEDI, morphology‐enabled dipole Inversion; SCWT, Stroop Color and Word Test; SDMT, symbol‐digit modalities test; TG, triglycerides; TMT, trail making test.

• 

  FIGURE 2. Flowchart of QSM image spatial registration, normalization, and ROI segmentation. By registering QSM images to T1‐weighted images, spatial consistency across different imaging modalities is achieved, enabling precise localization of subcortical regions in the QSM images. Subsequently, using the subcortex atlas derived from the redefinition of functional connectivity gradients and regional boundaries, QSM values for 32 subcortical regions are calculated. 3D‐T1, 3D T1‐weighted imaging; GM, gray matter; MNI, Montreal Neurological Institute; QSM, quantitative susceptibility mapping; WM, white matter.

• 

  FIGURE 3. Correlations between subcortical nuclei magnetic susceptibility values and cognitive test scores. p‐values from Spearman correlations were Bonferroni‐corrected; the significance threshold was set at p < 0.05/16 = 0.003. Scatterplots with overlaid Spearman partial correlation lines; marginal density plots show the distribution of each variable. Spearman correlations are presented. The included cognitive tests are SCWT (Stroop Color‐Word Test, evaluating the ability to resolve color‐word interference), AVLT (Auditory Verbal Learning Test, measuring memory performance in auditory verbal tasks), TMT (Trail Making Test, highlighting motivational characteristics), and SDMT (Symbol Digit Modalities Test, assessing cognitive processing speed and symbol‐digit association skills). aPUT, anterior putamen; lh, left hemisphere; pPUT, posterior putamen; pCAU, posterior caudate; rh, right hemisphere; THA‐DA, dorsoanterior thalamus.

• 

  FIGURE 4. Non‐linear relationship curve between magnetic susceptibility values of subcortical nuclei and HbA1c. aPUT, anterior putamen; lh, left hemisphere; pCAU, posterior caudate; rh, right hemisphere; THA‐DA, dorsoanterior thalamus.

• 

  FIGURE 5. Structural equation model of cognitive, metabolic, and magnetic susceptibility values interactions. The relationships among metabolic changes, magnetic susceptibility values of subcortical nuclei, and cognitive function are depicted in the SEM. The term “HbA1c squared” represents the squared value of HbA1c (HbA1c * HbA1c). Significant paths are marked as follows: *p < 0.05, **p < 0.01, and ***p < 0.001, indicating the strength of the statistical associations.