RESUMEN
Chronic kidney disease (CKD) is characterized by inflammation and fibrosis in the kidney. Renal biopsies and estimated glomerular filtration rate (eGFR) remain the standard of care, but these endpoints have limitations in detecting the stage, progression, and spatial distribution of fibrotic pathology in the kidney. MRI diffusion tensor imaging (DTI) has emerged as a promising noninvasive technology to evaluate renal fibrosis in vivo both in clinical and preclinical studies. However, these imaging studies have not systematically identified fibrosis particularly deeper in the kidney where biopsy sampling is limited, or completed an extensive analysis of whole organ histology, blood biomarkers, and gene expression to evaluate the relative strengths and weaknesses of MRI for evaluating renal fibrosis. In this study, we performed DTI in the sodium oxalate mouse model of CKD. The DTI parameters fractional anisotropy, apparent diffusion coefficient, and axial diffusivity were compared between the control and oxalate groups with region of interest (ROI) analysis to determine changes in the cortex and medulla. In addition, voxel-based analysis (VBA) was implemented to systematically identify local regions of injury over the whole kidney. DTI parameters were found to be significantly different in the medulla by both ROI analysis and VBA, which also spatially matched with collagen III immunohistochemistry (IHC). The DTI parameters in this medullary region exhibited moderate to strong correlations with histology, blood biomarkers, hydroxyproline, and gene expression. Our results thus highlight the sensitivity of DTI to the heterogeneity of renal fibrosis and importance of whole kidney noninvasive imaging.NEW & NOTEWORTHY Chronic kidney disease (CKD) can be characterized by inflammation and fibrosis of the kidney. Although standard of care methods have been limited in scope, safety, and spatial distribution, MRI diffusion tensor imaging (DTI) has emerged as a promising noninvasive technology to evaluate renal fibrosis in vivo. In this study, we performed DTI in an oxalate mouse model of CKD to systematically identify local kidney injury. DTI parameters strongly correlated with histology, blood biomarkers, hydroxyproline, and gene expression.