Longitudinal Evolution of the Brain Microstructure in Cirrhotic Patients on Diffusion Kurtosis Imaging

Yuan‐Yuan Chen,Zi‐Ning Lu,Qi Zhang,Yi‐Ning Zhang,Wen‐Ting Ma,XiaoDi Zhang,Xiao‐Dong Zhang,Hong‐Yan Ni,Yue Cheng
DOI: https://doi.org/10.1002/jmri.29648
IF: 4.4
2024-11-03
Journal of Magnetic Resonance Imaging
Abstract:Background Although improvement of cognitive function after liver transplantation has been demonstrated in several neuropsychological studies, there is limited research on longitudinal changes in the cirrhotic patients' brain structure before and after transplantation. Purpose To investigate longitudinal changes of brain microstructure in cirrhotic patients using diffusion kurtosis imaging (DKI). Study Type Prospective. Subjects A total of 153 cirrhosis patients, comprising 60 hepatic encephalopathy (HE) patients (16 females/44 males) and 93 no‐HE patients (35 females/58 males), along with 93 healthy controls (HCs) (53 females/40 males) were enrolled. Subsequently, 58 recipients completed 1‐month postoperative follow‐up, 29 patients completed 1‐, 3‐months, and 17 patients completed 1‐, 3‐, 6‐month follow‐up. Sequence Spin‐echo single‐shot echo‐planar sequence using a 3.0 T scanner. Assessment Diffusion kurtosis estimator software was used to estimate the DKI parameter maps by a MR imaging physicist (Y.‐Y.C. with 12 years of experience). Statistical Tests The diffusion metrics (eg, radial kurtosis [RK], mean kurtosis, fractional anisotropy, mean diffusivity) of the patients before transplantation were compared with those of the HCs using voxel‐wise analysis of variance (ANOVA), along with t tests for post hoc analysis. Linear mixed‐effects models were applied to the longitudinal data. We imposed a cluster level Family Wise Error (FWE) correction rate of PFWE = 0.05 with voxel‐wise cutoff of P = 0.001 together with a cluster‐size cutoff of N ≥ 56, and generated smoothness estimates from the preprocessed data using the mixed‐model autocorrelation function. Results The RK metrics of the patients decreased significantly in the anterior cingulate cortex (HE/no‐HE
radiology, nuclear medicine & medical imaging
What problem does this paper attempt to address?