To test the hypothesis of washout from the anterior pituitary (AP) gland after serial injections of gadodiamide. We included 59 patients with history of at least 5 injections of gadodiamide. Values of mean signal intensity of the AP and of the central pons were measured on unenhanced sagittal T1-weighted images. AP-to-pons signal intensity ratios were calculated dividing the values of the AP by those of the pons. The measurements were performed using MR images acquired at four different time points including baseline (prior to any gadodiamide injection), minimum post-injection time delay, maximum post-injection time delay, and last available MR scans. Normalized ratios (i.e. ratios divided total volume of injected gadodiamide) were also calculated. To assess the difference between ratios, non-parametric Wilcoxon signed-rank test was applied. The correlations were tested with non-parametric Spearman correlation coefficient. A p-value < 0.05 was considered as statistically significant. A statistically significant increase of AP signal intensity was found by comparing the baseline scans with both the minimum time delay (p = 0.003) and maximum time delay scans (p = 0.005). We found significant higher normalized ratios for minimum post-injection time delay with respect to maximum post-injection time delay (p < 0.001). The normalized ratios demonstrated a statistically significant negative correlation with the post-injection time delay (r = − 0.31; p = 0.006). The findings of this study suggest that washout phenomena of retained/deposited gadolinium from the AP are influenced by the total injected volume and post-injection time delay.