Purpose：To investigate the functional connectivity change of the frontal-parietal network (FPN) in end-stage renal disease (ESRD) dialysis patients using independent component analysis (ICA) of rest-state functional magnetic resonance imaging and to correlate the changes of the FPN connectivity with neuropsychological tests results.

Materials and methods: The study was approved by the local Ethics Review Board. Twenty-five volunteers (8 males and 17 females, with the mean age of 51.9 ± 10.4 years old, range 32 to 67) and forty ESRD patients undergoing standard dialysis (20 males and 20 females, average 51.6 ± 8.6 years old, range 24 to 70) were recruited. Patients with history of psychiatric diseases and/or brain stroke as well as other severe diseases were excluded All enrolled subjects gave their informed consent prior to the study. All MR data (FLAIR, BOLD and 3D T1 imaging) were collected with a clinical 3T MR (Verio, Siemens Healthcare, Erlangen, Germany) and a series of neuropsychological tests were finished after the MR examination. BOLD data were prepared by using DEPARSF software and ICA were performed with the GIFT box to isolate the FPN. Due to the z scores can indirectly provide a measurement of functional connectivity in the FPN, the intensity values were converted to z scores in each spatial map to determine which voxels contributed most strongly to an independent component. Maps of the FPN were compared between the two groups. Pearson correlation analysis was performed to correlate FPN functional connectivity changes with neuropsychological test results. Statistics analyses were performed with SPSS version 18.0 for windows (SPSS Inc., Chicago, Illinois).

Results: All the MR data and 52 neuropsychological tests (20 normal subjects and 32 ESRD patients) data were successfully collected. Patients with ESRD demonstrated significantly less functional connectivity in the bilateral inferior parietal lobule (IPL) (P < 0.01 AlphaSim) than did normal subjects (Figure 1 and Figure 2). The functional connectivity of the right IPL was negatively correlated with trail making test-A (r = - 0.285, P = 0.041) and trail making test–B (r = - 0.426, P = 0.002) (Figure 3). Age (P = 0.0896), sex (P = 0.201) and education years (P = 0.223) between the two groups had no significant difference.

Conclusion: Functional connectivity of the bilateral FPN were impaired in patients with ESRD, primarily in bilateral IPL. The right IPL functional connectivity correlated with trail making test–A and B scores, which means cognitive decline of ESRD patients might associated with FPN functional connectivity reduction.

Discussion: Many previously published studies (1, 2) focused on different imaging techniques, such as diffusion-tensor imaging, voxel-based morphometry, have demonstrated that multiple brain regions, especially the frontal and parietal lobes, had structural and metabolic abnormalities with cognitive decline in ESRD patients who had no clinical symptoms. Ni et al (3) also manifest the default-mode network of ESRD patients is changed and associated with reduced cognition. Our study shows the similar results in FPN and we observed the reduction of functional connectivity in bilateral IPL. FPN contributes to working memory, attention and executive functions (4), so impaired FPN functional connectivity may contribute to the neurocognitive dysfunctions in ESRD patients. Our findings were partially supported by the negative correlation between the average z score of the IPL and the scores of trail making test-A and B, which domains in attention and executive function. We only observed the correlation between right IPL and neuropsychological test results, but none left IPL. These may because of the relatively small sample size or the intrinsic differences between two cerebral hemisphere. Functional connectivity of FPN may serve as a noninvasive biomarker to monitor cognitive impairment, especially attention and executive function impairment, in ESRD patients.