OBJECTIVE: The objective of this study was to measure the protein-specific response of r1 and r2 relaxivities of commercially available gadolinium-based magnetic resonance imaging contrast agents to variation of plasma-protein concentrations. MATERIALS AND METHODS: In this in vitro study, contrast agent (gadofosveset trisodium, gadoxetate disodium, gadobutrol, and gadoterate meglumine) dilution series (0-2.5 mmol Gd/L) were prepared with plasma-protein (human serum albumin [HSA] and immunoglobulin G [IgG]) concentrations at physiological (42 and 10 g/L HSA and IgG, respectively, Normal) and at 3 pathological levels with HSA/IgG concentrations of 10/10 (solution Alb low), 42/50 (IgG mild), and 42/70 (IgG severe) g/L. Contrast-agent molar relaxivities and relaxivity-enhancing proteincontrast-agent interaction coefficients were determined on the basis of inversion-recovery and spin-echo data acquired at 1.5 and 3.0 T at 37°C. Protein-induced magnetic resonance imaging signal changes were calculated. RESULTS: The effective r1 and r2 molar relaxivities consistently increased with albumin and IgG concentrations. At 1.5 T, the r1 values increased by 10.2 (gadofosveset), 4.3 (gadoxetate), 1.3 (gadobutrol), and 1.1 L s mmol (gadoterate), respectively, from the Alb low to the IgG severe solution. At 3.0 T, the r1 values increased by 2.9 (gadofosveset), 2.3 (gadoxetate), 0.7 (gadobutrol), and 0.9 (gadoterate) L s mmol, respectively. An excess of IgG most strongly increased the r1 of gadoxetate (+40 and +19% at 1.5 and 3.0 T, respectively, from Normal to IgG severe). An albumin deficiency most strongly decreased the r1 of gadofosveset (-44% and -20% at 1.5 and 3.0 T, respectively, from Normal to Alb low). The modeling confirmed a strong gadofosveset r1 enhancement by albumin and suggested stronger IgG than albumin effects on the apparent molar relaxivity of the other agents per protein mass concentration at 1.5 T. CONCLUSIONS: Pathological deviations from normal plasma-protein concentrations in aqueous solutions result in changes of effective r1 and r2 contrast-agent relaxivities and projected signal enhancements that depend on the contrast agent, the blood-serum protein profile, and the field strength.

No abstract