Severity of polycystic kidney disease revealed by multiparametric MRI
Feng Wang, Seo Yeon Lee, Fatemeh Adelnia, Keiko Takahashi, Kevin D. Harkins, Lilly He, Zhongliang Zu, Philipp Ellinger, Manuel Grundmann, Raymond C. Harris, Takamune Takahashi, John C. Gore
Abstract
Purpose
We aimed to compare multiple MRI parameters, including relaxation rates (R1, R2, and R1ρ), ADC from diffusion weighted imaging, pool size ratio (PSR) from quantitative magnetization transfer, and measures of exchange from spin-lock imaging (Sρ), for assessing and predicting the severity of polycystic kidney disease (PKD) over time.
Methods
Pcy/Pcy mice with CD1 strain, a mouse model of autosomal dominant PKD, were imaged at 5, 9, and 26 wk of age using a 7T MRI system. Twelve-week normal CD1 mice were used as controls. Post-mortem paraffin tissue sections were stained using hematoxylin and eosin and picrosirius red to identify histological changes.
Results
Histology detected segmental cyst formation in the early stage (week 5) and progression of PKD over time in Pcy kidneys. In T2-weighted images, small cysts appeared locally in cystic kidneys in week 5 and gradually extended to the whole cortex and outer stripe of outer medulla region from week 5 to week 26. Regional PSR, R1, R2, and R1ρ decreased consistently over time compared to normal kidneys, with significant changes detected in week 5. Among all the MRI measures, R2 and R1ρ allow highest detectability to PKD while PSR and R1 have highest correlation with pathological indices of PKD. Using optimum MRI parameters as regressors, multiple linear regression provides reliable prediction of PKD progression.
Conclusion
R2, R1 and PSR are sensitive indicators of the presence of PKD. Multiparametric MRI allows a comprehensive analysis of renal changes caused by cyst formation and expansion.