Concern has grown about the finding of gadolinium deposits in the brain after administering gadolinium-based contrast agents (GBCAs). The mechanism is unclear, and related questions remain unanswered, including the stability over time. Therefore, we conducted a three-time-point study to explore T1-weighted (W) signal changes in the dentate nucleus (DN) and globus pallidus (GP), after the first, fifth, and tenth injections of either a macrocyclic agent (gadoterate meglumine) or a linear agent (gadobenate dimeglumine).

For this retrospective, multicenter, longitudinal study, two groups of 18 (gadoterate meglumine) and 19 (gadobenate dimeglumine) patients were identified. The evolution of the signal over time was analyzed using DN/pons (DN/P) and GP/thalamus (GP/T) ratios.

DN/P and GP/T ratios tended to increase after the fifth administration of gadobenate dimeglumine, following by a downward trend. A trend in a decrease in DN/P and GP/T ratios were found after the fifth and tenth administrations of gadoterate meglumine.

After exposure to gadobenate dimeglumine, the signal intensity (SI) tended to increase after the fifth injection owing to gadolinium accumulation, however, a SI increase was not found after the tenth administration supporting the hypothesis of a slow elimination of the previously retained gadolinium (wash-out effect) from the brain or of a change in form (by dechelation), causing the signal to fade. No increasing SI was found in the DN and GP after exclusive exposure to gadoterate meglumine, thus confirming its stability. We found, instead, a trend for a significative gadolinium elimination over time.