Synopsis

Keywords: Kidney, Kidney

Diabetic Kidney Disease (DKD) is the leading cause of end-stage renal disease worldwide. Morphological evaluation of renal compartment volume changes and functional evaluation of renal ADC changes provide comprehensive insight into DKD disease progression. This study compares renal compartment volume changes including: 1) parenchyma volume ,2) pelvis volume, 3) whole kidney volume and 4) parenchyma volume percentage and the functional parameter ADC for DKD groups with different CKD staging. Our results demonstrate that cortical/medullary ADC, parenchyma volume and whole kidney volume are correlated with eGFR and suggest that the cortical ADC is a stronger predictor of eGFR in DKD.

Introduction

With the prevalence of type 2 diabetes mellitus (DM), diabetic kidney disease (DKD) is the leading cause of end-stage renal disease worldwide with high morbidity and mortality(1-3). Mapping the apparent diffusion coefficient (ADC) and renal volume assessment are clinically meaningful tools for evaluating the renal micro- and macro- structure changes during DKD(4). Here the semi-automatic twelve-layer concentric objects (TLCO) technique is used for layer-specific assessment of ADC maps to provide a highly reproducible and reliable assessment of renal ADC in the cortex and medulla compared to the conventional region-of-interest method(5,6). This study compares morphological features of renal volume including the 1) parenchyma volume, 2) pelvis volume, 3) whole kidney volume, and 4) parenchyma volume ratio and functional renal MR parameter ADC that assessed with TLCO, among DKD groups with different CKD staging.

Method

This cross-sectional study was approved by the local ethics review board of Guangdong Provincial People’s Hospital. Written informed consent was obtained from all subjects. Subject characteristics are presented in Figure 1. The DKD subjects were stratified by eGFR(calculated according to CKD-EPI) into 5 groups according to Kidney Disease: Improved Global Outcome of chronic kidney disease criteria(7,8). Due to the limited sample size, DKD groups are merged into mild DKD group (CKD G1-2), with eGFR > 60 ml/min/1.73m², moderate DKD group (CKD G3), with 30 ml/min/1.73m² < eGFR < 59 ml/min/1.73m², and severe DKD group (CKD G4-5), with eGFR < 29 ml/min/1.73m².
MRI
MRI was conducted using a whole-body 3T scanner (SIGNA EXCITE, GE Healthcare) using the built-in body coil for signal excitation and an 8-channel RF torso array for signal reception. Respiratory triggering was used to avoid motion artifacts. For anatomical imaging, T2-weighted coronal fat-suppressed 2D multi-slice FSE imaging was implemented: TE=215ms, TR=15789ms, matrix=512×512, in-plane resolution=0.625×0.625mm², slice thickness=3mm, slice gap=0.6mm. For DWI, coronal fat-suppressed 2D multi-slice single-shot spin-echo EPI was implemented: TE=60ms, TR=3000ms, matrix=96×96，in-plane resolution=1.25×1.25mm²， slice thickness=3mm, slice gap=0.6mm, b value=0 and 600 s/mm².
Data analysis
Figure 2 illustrates the post-processing of the T2WI and ADC map that was analyzed by using the TLCO approach.
Morphological evaluation of the kidney includes the assessment of 1) the volume of renal parenchyma, 2) the volume of the renal pelvis, 3) the volume of the whole kidney, and 4) the volume percentage of renal parenchyma (i.e. the volume of renal parenchyma divided by whole kidney volume). To measure the volume of renal parenchyma and pelvis, renal parenchyma and pelvis were manually labeled by the experienced radiologist on T2WI by using ITK-SNAP(9). Figure 2A illustrates the labeling of the renal parenchyma and pelvis, and the reconstructed 3D surface mesh. Following this segmentation, the volume was corrected by the body surface area (BSA) which is calculated according to: BSA(m²)=0.61×height(m) +0.0128×body weight(kg) – 0.1529 (10).
The middle coronal slice DWI data was analyzed by using Matlab scripts (Matlab 2018b). ADC map was calculated first followed by the TLCO approach which segmented the renal parenchyma into 12 layers with equal thickness. Mean ADC values were calculated for each layer (Figure 2B). The mean ADC obtained for the outer 6 layers was assigned to the ADC of the renal cortex (ADC_cor), and the mean ADC value derived from the inner 6 layers was considered as the renal medulla ADC (ADC_med). ADC_cor and ADC_med were compared among DKD groups. Layer-wise comparison among DKD groups was conducted.
Statistical analysis
Statistical analysis was conducted using SPSS (SPSS 26, IBM). The mean value of volume and ADC of the left and right kidneys were calculated and considered as the MR biomarkers for the subject. MR biomarkers were compared among groups using one-way ANOVA with post-hoc LSD or Kruskal-Wallis one-way ANOVA with post-hoc Bonferroni depending on data normality. Pearson correlation between MR biomarkers and eGFR was analyzed. P<0.05 was considered significant.

Results

Figure 3 summarizes the MR biomarkers for the DKD group with different CKD staging. With the advanced CKD stages, ADC_cor and ADC_med are significantly decreased for the moderate DKD and severe DKD groups compared to the mild DKD cohorts. This observation is confirmed by the layer-wise ADC comparison (Figure 4). Significant parenchyma volume and whole kidney volume reduction were observed for the moderate DKD group and severe DKD group versus the mild DKD group, while the pelvis volume and parenchyma volume percentage didn’t show significant changes. Pearson correlation analysis showed that ADC_cor is moderately correlated with eGFR. ADC_med, parenchyma volume, and whole kidney volume are weakly to moderately correlated with eGFR (Figure 5).

Conclusion

This study evaluated renal volume as a macro-morphological feature and renal water diffusion as a functional and micro-morphological feature of the kidney assessed by using the TLCO approach among DKD cohorts with different CKD staging. Our results demonstrate that ADC_cor, ADC_med, parenchyma volume, and whole kidney volume are correlated with eGFR, and ADC_cor is a stronger predictor of eGFR in DKD.

Acknowledgements

no