Purpose: To measure the systematic error in perfusion and filtration parameters derived from magnetic resonance (MR) renography caused by protein binding of MR contrast agents.
Materials and methods: Eight healthy Danish Landrace pigs were examined with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). In four pigs a bolus of gadopentetate-dimeglumine (Gd-DTPA; no protein binding) was injected, followed by gadobenate-dimeglumine (Gd-BOPTA; 10% protein binding). The order was reversed in the other four pigs. A two-compartment filtration model was generalized to allow for protein binding and fitted to whole-cortex region of interest (ROI) curves. Single-kidney plasma flow and volume, tubular flow (or GFR), and tubular transit time of both agents were compared.
Results: The data show a strong systematic underestimation (P < 0.001) in GFR by Gd-BOPTA (33 ± 7.2%), and no significant differences (P > 0.05) in plasma flow (2.2 ± 18%), plasma volume (-1.7 ± 7.8%) and tubular transit time (3.1 ± 7.2%). The order of injection had no significant effect.
Conclusion: Theory and experiments agree that perfusion parameters of both agents are comparable, whereas GFR is underestimated with Gd-BOPTA due to the dependence of relaxivity on protein content. Hence, GFR cannot be measured with protein-bound contrast agents, but the proposed dual-agent protocol may produce new functional indices measuring protein filtration.
Copyright © 2011 Wiley-Liss, Inc.