Correlation Between Computer‐Assisted Ultrasonographic Parameters and Histological Findings in Glomerular Diseases: A Multivariate Ordinal Analysis

INTRODUCTION: This study aimed to correlate renal ultrasonographic and Doppler parameters with histological findings from kidney biopsies using computer-assisted image analysis. Objective measures included the cortical-to-liver (or spleen) echogenicity ratio, corticomedullary ratio (CMR), renal resistivity index (RRI), splenic resistivity index (SRI), and a derived index (DI-RISK). METHODS: The patients' age and CKD-EPI-derived estimated glomerular filtration rate (eGFR) were recorded. Histological lesions encompassed all components of the lupus nephritis activity and chronicity indices. Ultrasound images were analyzed to extract grayscale values from manually defined regions of interest. RRI was measured in interlobar arteries, and SRI at the intraparenchymal splenic artery. The DI-RISK index expressed the RRI difference from the SRI as a percentage of the RRI. Associations were evaluated using multivariate ordinal logistic regression. RESULTS: Increased cortical echogenicity was independently correlated with severe interstitial fibrosis/tubular atrophy (IF/TA) and lower eGFR. CMR was reduced in the presence of interstitial infiltrates. IF/TA was also associated with reduced kidney size. While resistivity indices correlated with vascular lesions, age, and eGFR in the univariate analysis, only age and eGFR remained significant in the multivariate models. CONCLUSION: Computer-assisted ultrasonography correlates with key histological changes. Cortical echogenicity and kidney size may help identify advanced chronic damage noninvasively.