Cerebral white matter integrity in type 2 diabetes mellitus: underlying mechanisms and associations with anhedonia, melancholia, and cognition

Introduction

Type-2 Diabetes Mellitus (T₂DM) is commonly associated with peripheral neural injuries, known as diabetic polyneuropathy. However, intracerebral White Matter (WM) integrity has been rarely studied in patients with T₂DM. Our previous studies have highlighted different mechanisms involved in the association between T₂DM and the brain, mainly low-grade inflammation and neurodegeneration. However, it is still unclear which factors might be involved in the impairment of WM integrity in T₂DM. It is also unclear whether T₂DM-specific WM lesions are correlated with the neuropsychological state of affected persons.

Aims

Compare different diffusion parameters between T₂DM and non-T₂DM participants, understand the underlying mechanisms, and evaluate their correlation with cognitive and psychological symptoms.

Methods

The multiethnic study participants, aged 50 years and older, were recruited in North Texas (HABS-HD). These underwent diffusion-weighted scans, which were preprocessed and analyzed; tensors were fitted, and diffusion parameters were averaged in WM tracts defined by the JHU-atlas. Effect sizes (Cohen’s d) were calculated after adjusting for relevant confounders and multiple testing. Mediation analysis was performed using 1000-fold non-parametric bootstrapping of 95% confidence intervals (CI). Spearman’s correlation was used to assess associations between relevant variables and neuropsychological outcomes.

Results

Significant impairment of WM integrity was observed in T₂DM participants, particularly in the Posterior thalamic radiation (PTR), Anterior corona radiata (ACR), Cingulum cingulate gyrus (CCG), Inferior fronto-occipital fasciculus (IFOF), Genu of corpus callosum (GCC), and Body of corpus callosum (BCC). In addition to the fractional anisotropy (FA), the mean diffusivity (MD) and radial diffusivity (RD) showed significant results, except in the PTR, ACC, and GCC for MD, and ACC for RD. These results are statistically robust as they were adjusted for relevant confounders and multiple testing. Low-grade inflammation mediated up to 15% of these associations. WM impairment explained up to 5% of systemic Neurofilament light Chain (NfL) levels. T₂DM-specific impairment of WM integrity correlated with a worse cognitive status (ACR: r_Spearman= 0.09 [0.05, 0.14], p<0.001; CCG: r_Spearman= 0.07 [0.03, 0.11], p<0.001; IFOF: r_Spearman= 0.14 [0.10, 0.18], p<0.001), executive function (PTR: r_Spearman= -0.14 [-0.18, -0.09], p<0.001; ACR: r_Spearman= -0.26 [-0.30, -0.22], p<0.001; CCG: r_Spearman= -0.15 [-0.19, -0.11], p<0.001; IFOF: r_Spearman= -0.16 [-0.21, -0.12], p<0.001; GCC: r_Spearman= -0.17 [-0.21, -0.12], p<0.001; BCC: r_Spearman= -0.14 [-0.18, -0.09], p<0.001), and dementia severity (PTR: r_Spearman= -0.12 [-0.16, -0.08], p<0.001; ACR: r_Spearman= -0.14 [-0.18, -0.10], p<0.001; CCG: r_Spearman= -0.14 [-0.18, -0.09], p<0.001; IFOF: r_Spearman= -0.11 [-0.16, -0.07]; p<0.001, GCC: r_Spearman= -0.15 [-0.20, -0.11]; p<0.001, BCC: r_Spearman= -0.13 [-0.17, -0.09], p<0.001). It was also correlated with higher anhedonia (PTR:r_Spearman= -0.11 [-0.15, -0.06], p<0.001; ACR: r_Spearman= -0.08 [-0.13, -0.04], p<0.001; CCG: r_Spearman= -0.07 [-0.11, -0.03], p=0.001; GCC: r_Spearman= -0.06 [-0.10, -0.01], p=0.007) and melancholia scores (PTR: r_Spearman= -0.06 [-0.10, -0.01], p=0.01; ACR: r_Spearman= -0.05 [-0.10, -0.01], p=0.01; IFOF: r_Spearman= -0.06 [-0.10, -0.01], p=0.01).

Conclusions

T₂DM-associated impairment of cerebral WM integrity is regionally defined, affecting specific WM tracts, and is partly mediated by inflammation. Findings suggested both axonal and demyelinating injuries. Impaired WM integrity correlated with worse cognitive performance and higher anhedonia and melancholia scores.

References

No references