Gadolinium Presence in the Brain After Administration of the Liver-Specific Gadolinium-Based Contrast Agent Gadoxetate: A Systematic Comparison to Multipurpose Agents in Rats.

Abstract

Clinical studies have reported different results regarding the signal intensity (SI) increase in the dentate nucleus on unenhanced T1-weighted magnetic resonance imaging (MRI) after repeated administrations of gadolinium-based contrast agents (GBCAs). The aim of this study was to evaluate MRI SI changes and gadolinium (Gd) brain concentrations in an animal model after repeated administration of liver-specific linear gadoxetate in comparison to multipurpose linear and macrocyclic GBCAs. Recently, it was demonstrated that small amounts of GBCAs are able to cross the blood-cerebrospinal fluid (CSF) barrier. Therefore, a secondary aim was to test if the administration of these GBCAs directly into the CSF results in a similar MRI pattern and brain Gd concentration than after systemic intravenous injection. Forty-eight Han-Wistar rats were equally divided into the following 4 groups: gadoxetate (liver-specific linear), gadodiamide (multipurpose linear), gadobutrol (multipurpose macrocyclic), and control (saline, artificial CSF). For systemic application, 6 animals per group received 8 intravenous injections on 4 consecutive days per week over 2 weeks using a dose of 0.15 mmol/kg for gadoxetate and 0.6 mmol/kg for multipurpose GBCAs per injection, which corresponds to the recommended clinical dose in humans. For CSF application, 6 animals per group received one intracisternal administration of 0.31 μmol Gd (gadoxetate) and 1.25 μmol Gd (multipurpose GBCAs) or an equal volume of artificial CSF. Brain MRI was performed after a period of 5 weeks to evaluate the SI in deep cerebellar nuclei (DCN) and brain stem. Subsequently, animals were euthanized and their brains were dissected for Gd quantification by inductively coupled plasma-mass spectrometry. Visually evident increased SIs in the DCN were observed in blinded image review only after administration of gadodiamide. The respective SI ratios between DCN and brain stem were significantly higher compared with the control groups (P = 0.009 and P = 0.002 for intravenous and intracisternal application, respectively), whereas no difference was found for gadoxetate and gadobutrol (P ≥ 0.9). Inductively coupled plasma-mass spectrometry revealed the lowest Gd content in the brain tissue after administration for gadoxetate. The mean Gd concentrations in the cerebellum were 0.08 nmol/g (gadoxetate), 2.66 nmol/g (gadodiamide), and 0.26 nmol/g (gadobutrol) after intravenous administration, and 0.28 nmol/g (gadoxetate), 3.23 nmol/g (gadodiamide), and 0.69 nmol/g (gadobutrol) after intracisternal application. This rat study demonstrates distinct differences in the presence of gadolinium in the brain between the liver-specific linear gadoxetate and the multipurpose linear GBCA gadodiamide. No MRI signal alterations were observed after 8 dose-adapted intravenous or a single intracisternal administrations of gadoxetate and multipurpose macrocyclic gadobutrol. The Gd concentrations in the brain 5 weeks after intravenous administration of gadoxetate were an order of magnitude lower compared with gadodiamide and slightly lower than for gadobutrol. Likely reasons for these differences are the 4-fold lower dose, the dual excretion pathway, and the higher complex stability of gadoxetate compared with multipurpose linear GBCAs. The similar findings for both routes of GBCA administration underlines the assumption that the very small amount of GBCAs that cross the blood-CSF barrier is further transported into the brain tissue.