Significance Statement: Patients with end stage CKD often develop cognitive decline, but whether this is related to the underlying disease or to hemodialysis remains unclear. We performed three-dimensional pseudocontinuous arterial spin labeling and quantitative susceptibility mapping prospectively in 40 patients with stage 1–4 CKD, 47 nondialysis patients with stage 5 CKD, and 44 healthy controls. Our magnetic resonance imaging data demonstrate that changes in cerebral blood flow–susceptibility coupling might underlie this cognitive decline, perhaps in the hippocampus and thalamus. These results suggest that magnetic resonance imaging parameters are potential biomarkers of cognitive decline in patients with CKD. Moreover, our findings may lead to discovery of novel therapeutic targets to prevent cognitive decline in patients with CKD. Background: Cerebral blood flow (CBF) and susceptibility values reflect vascular and iron metabolism, providing mechanistic insights into conditions of health and disease. Nondialysis patients with CKD show a cognitive decline, but the pathophysiological mechanisms underlying this remain unclear. Methods: Three-dimensional pseudocontinuous arterial spin labeling and quantitative susceptibility mapping were prospectively performed in 40 patients with stage 1–4 CKD (CKD 1–4), 47 nondialysis patients with stage 5 CKD (CKD 5ND), and 44 healthy controls (HCs). Voxel-based global and regional analyses of CBF, susceptibility values, and vascular-susceptibility coupling were performed. Furthermore, the association between clinical performance and cerebral perfusion and iron deposition was analyzed. Results: For CBF, patients with CKD 5ND had higher normalized CBF in the hippocampus and thalamus than HCs. Patients with CKD 5ND had higher normalized CBF in the hippocampus and thalamus than those with CKD 1–4. The susceptibility values in the hippocampus and thalamus were lower in patients with CKD 5ND than in HCs. Patients with CKD 5ND had higher susceptibility value in the caudate nucleus than those with CKD 1–4. More importantly, patients with CKD 5ND had lower CBF-susceptibility coupling than HCs. In addition, CBF and susceptibility values were significantly associated with clinical performance. Conclusions: Our findings demonstrate a new neuropathological mechanism in patients with CKD, which leads to regional changes in CBF-susceptibility coupling. These changes are related to cognitive decline, providing potential imaging markers for assessing clinical disability and cognitive decline in these patients.

Background: Neurovascular coupling (NVC) reflects the close connection between neural activity and cerebral blood flow (CBF) responses, providing new insights to explore the neuropathological mechanisms of various diseases. Non‐dialysis patients with chronic kidney disease (CKD) exhibit cognitive decline, but the underlying pathological mechanisms are unclear.Methods: The prospective study involved 53 patients with stage 1–3a CKD (CKD1–3a), 78 patients with stage 3b–5 CKD (CKD3b–5), and 52 healthy controls (HC). Our investigation involved voxel‐based assessments of both global and regional BOLD signal characteristics. Additionally, we explored the correlations between neuroimaging indices, Montreal Cognitive Assessment (MoCA) scores, and clinical laboratory findings.Results: Compared to HC, the CKD3b–5 and CKD1–3a groups exhibited lower ALLF and ReHo in the default mode network (DMN), higher CBF in bilateral hippocampus (HIP), higher susceptibility values in bilateral caudate nucleus (CAU) and putamen (PUT), and lower susceptibility values in bilateral HIP. At the global level, the coupling coefficients were lower in CKD1–3a and CKD3b–5 groups than in HC. At the ROI level, the CBF‐ALFF and CBF‐ReHo coupling in HIP and basal ganglia regions were lower in CKD3b–5 groups than in the CKD1–3a group. Most importantly, susceptibility‐ALFF in ANG.R may mediate the effects of phosphorus on cognitive decompensation in patients with CKD1‐3a.Conclusions: Non‐dialysis patients with CKD exhibit abnormal NCV, which is associated with the cognitive decline. Specifically, the susceptibility‐ALFF may serve as a valuable biomarker for early assessment of cognitive decline in CKD, offering insights into the pathogenesis of cognitive decline in CKD.

To investigate glymphatic function in nondialysis-dependent ESRD (ND-ESRD) patients with diabetic kidney disease (DKD) or non-DKD using the diffusion tensor image analysis along the perivascular space (DTI-ALPS) method and choroid plexus (CP) volume to explore the associations among DTI-ALPS index, CP volume, clinical characteristics, and cognitive function. 25 DKD patients, 25 non-DKD patients, and 25 healthy control (HC) were included in this study. All participants underwent cerebral MRI and cognitive assessments. Bilateral DTI-ALPS index and CP volume were calculated and were compared among three groups. Correlations among the DTI-ALPS index, CP volume, clinical characteristics, and cognitive function were analyzed. DKD patients had significantly lower left DTI-ALPS index (p = 0.001) and mean DTI-ALPS index (p = 0.003) than non-DKD patients. In ND-ESRD patients, there was a significant positive correlation between the left DTI-ALPS index and phosphorus (r = 0.377, p = 0.007) and attention score (r = 0.434, p = 0.002). In DKD group, the mean DTI-ALPS was positively correlated with total cholesterol (r = 0.631, p = 0.001). In non-DKD group, there was a significant positive correlation between the left DTI-ALPS index and phosphorus (r = 0.696, p < 0.001). Both DKD and non-DKD patients exhibited significant higher CP volume than HC group. In non-DKD group, CP volume was negatively correlated with mean DTI-ALPS (r = −0.535, p = 0.006) and attention score of MoCA (r = −0.538, p = 0.006). Glymphatic system dysfunction may contribute to the pathogenesis of clinical and cognitive impairment in ND-ESRD patients especially with DKD.

Background: Cognitive decline exists in the chronic kidney disease (CKD) population and is particularly severe in patients with stage 5 CKD, but the mechanisms underlying this relationship are unclear. Structural-functional coupling, an integrated measure that combines functional and structural networks, offers the possibility of exploring changes in network relationships in patients with stage 5 CKD. This study aimed to investigate the brain network topology and structural-functional coupling characteristics in patients with non-dialysis-dependent stage 5 CKD (CKD 5ND) and the correlation between network changes and cognitive scores. Methods: We prospectively performed diffusion tensor and resting-state functional magnetic resonance (rs-fMRI) imaging on 40 patients with CKD 5ND disease and 47 healthy controls (HCs). Graph theory analysis of functional and structural connectivity (SC) was performed. Small-world properties and network efficiency properties were calculated, including characteristic path length (Lp), clustering coefficient (Cp), normalized clustering coefficient (Gamma), normalized characteristic path length (Lambda), small-worldness (Sigma), global efficiency (Eglob), and local efficiency (Eloc). The SC-functional connectivity (FC) coupling characteristics and the association between Montreal Cognitive Assessment (MoCA) scores and graph-theoretical features were analyzed. Results: For SC, the Sigma (P=0.009), Cp (P=0.01), Eglob (P<0.001), and Eloc (P=0.01) were significantly lower in patients with CKD 5ND than in HCs, while Lp (P<0.001) and Lambda (P<0.001) were significantly higher in the patients than in the HCs. For FC, the Sigma (P=0.008), Gamma (P=0.009), Eglob (P=0.04), and Eloc (P<0.0001) were lower in patients with CKD 5ND than in HCs; however, the Lp (P=0.02) was higher in the patients than in the HCs. SC-SC coupling (P<0.001) was greater in patients with CKD 5ND than in HCs. The structural (Cp, Eloc, Eglob) and functional network parameters (Sigma, Gamma, Eglob) of the patients with CKD 5ND were positively correlated with MoCA scores; however, the Lp of both structural and functional networks was negatively correlated with MoCA scores. Conclusions: All patients with CKD 5ND included in the study exhibited changes in their structural and functional brain network topology closely related to mild cognitive impairment. SC-SC coupling was elevated in the patients compared with that in the controls. This may provide vital information for understanding and revealing the underlying mechanisms of cognitive impairment in patients with CKD 5ND.