[Cortical morphological study in cerebral small vessel disease patients with subcortical ischemic depression]

Objective: To explore the characteristics of cortical morphology in cerebral small vessel disease (CSVD) patients with subcortical ischemic depression (SID) and its relationship with clinical symptoms. Methods: A total of 88 patients with CSVD in the First Affiliated Hospital of Anhui Medical University were enrolled from July 2017 to July 2020. The subjects were divided into CSVD-non depression group (CSVD-ND, n=58) and SID group (n=30) according to the geriatric depression scale (GDS). The 3D-T₁ MRI images were obtained from all subjects. The computed anatomy toolbox 12 (CAT 12) was used for image processing and cortical segmentation to obtain the cortical thickness (CTh) and surface metrics, including gyrification index (GI), sulcus depth (SD) and fractal dimension (FD). A comparison at the vertex- and region-of-interest (ROI)-wise levels were performed by the general linear model, and correlation analysis were conducted between cortical morphometric measurements and GDS scores. Finally, mean CTh (mCTh) was extracted for binary logistic regression analysis. Results: At the vertex-wise level, compared with the CSVD-ND group, the SID patients showed increased CTh in clusters mainly located in the posterior default mode network (pDMN), such as the precuneus(Pcu), the superior parietal gyrus (SPG) and the right postcentral gyrus (PoCG). As for the surface measurements, the GI value and the FD value were increased in clusters of Pcu and inferior temporal gyrus (ITG), respectively, in the SID group. ROIs analyses showed that apart from the Pcu, the SPG and the right PoCG, CTh alterations in the SID group were involved in a wider range of regions, extending to the right precentral gyrus ((2.27±0.20) cm³ vs (2.12±0.26) cm³, P=0.007), the left paracentral gyrus ((2.18±0.20) cm³ vs (2.05±0.23) cm³, P=0.008) and so on, than that in the CSVD-ND group.Compared with the CSVD-ND patients, the SID patients showed increased GI in the right PoCG ((25.31±1.11) vs (24.23±1.27), P<0.05). Correlation analysis showed that CTh in the right Pcu was positively correlated with the GDS scores (r=0.4, P<0.05). Further binary logistic regression analysis showed that in comparison with the subjects in the reference group (<2.367 cm³), the odds ratio(95%CI) for SID patients in the highest tertile of mCTh (>2.473 cm³) were 6.373 (1.254-32.389) after multivariable adjustment (sex, age, years of education, total intracranial volume, traditional imaging findings of CSVD, cognitive function (CAMCOG-C) and mCTh). Conclusion: Both CTh and cortical complexity were increased in CSVD patients with SID, especially in the clusters of pDMN, and CTh may be an important risk factor for SID.