Recent studies reported an increase in the dentate nucleus (DN)-to-pons signal intensity (SI) ratio (DN-pons SI ratio) on unenhanced T1-weighted images in patients who received consecutive serial injections of linear gadolinium-based contrast agents (GBCAs). In contrast, most studies found no increase in the DN-pons SI ratio when patients were treated with consecutive serial injections of macrocyclic GBCAs. However, the potential difference between macrocyclic and linear GBCAs has never been assessed in individuals who received subsequent applications of both contrast agents. In this retrospective study, we assessed the evolution of the DN-pons SI ratio change in patients that were treated with a comparable number of serial consecutive injections of the linear GBCA gadopentetate dimeglumine and subsequent serial injections of the macrocyclic GBCAs gadobutrol and gadoterate meglumine. Data of 36 patients was analyzed. All patients underwent at least 5 consecutive administrations of the linear GBCA gadopentetate dimeglumine followed by an equal number of consecutive administrations of the macrocyclic GBCA gadobutrol. In 12 of the 36 patients, 5 or more final consecutive injections of the macrocyclic GBCA gadoterate meglumine were analyzed additionally. The difference of DN-pons SI ratios on unenhanced T1-weighted images was calculated by subtracting the ratio at the first examination from the ratio at the last examination in each of the 3 periods. The mean DN-pons SI ratio difference in the gadopentetate dimeglumine period was significantly greater than 0 (mean ± SD, 0.0448 ± 0.0345; P < 0.001), whereas the mean DN-pons SI ratio difference in the subsequent gadobutrol and gadoterate meglumine period was significantly smaller than 0 (gadobutrol: -0.0178 ± 0.0459, P = 0.026; gadoterate meglumine: -0.0250 ± 0.0284, P = 0.011). In this observational study, the application of the linear GBCA gadopentetate dimeglumine was associated with a DN-pons SI ratio increase, whereas subsequent applications of the macrocyclic GBCAs gadobutrol or gadoterate meglumine in the same patients were not. Rather, the current data tentatively suggest a decrease in preexisting hyperintensities over time when linear GBCAs are changed to macrocyclic GBCAs, potentially indicating a washout effect or precipitation of gadolinium. Future patient studies need to include control groups to replicate the present results, and additional animal studies should be conducted to clarify the underlying mechanism of the proposed SI decrease.