Radiologists, MRI physicists and scientists.Dynamic contrast-enhanced MRI (DCE-MRI) is widely used for the diagnosis of tumors in clinical applications when fat and enhancing tissue co-exist in one voxel (1, 2). In these cases fat compartment does not get enhanced and its impact on the behavior of the contrast are not fully investigated. This work was to discuss the impact of fat on the behavior of Gd-based contrast (3, 4).Twenty-four 15 ml vials were filled with mixture of water based Gd-BOPTA (MultiHance, Bracco Diagnostic, NJ) solution, 3% emulsifying wax and canola oil (fat). Vials were divided into four groups; each had six vials with the same fat volume fraction. The fat volume fraction were 0, 20%, 33%, 50%. In each group, the concentration of Gd-BOPTA ([Gd-BOPTA]) in the water based solution in each vial was 0.2 mM, 0.4 mM, 0.5 mM, 0.6 mM, 0.8 mM and 1 mM respectively. The water based Gd solution was not heated and there were big visible fat droplets in the suspension (Figure 1). The phantom was scanned the next day on a clinical 3T scanner (Tim Trio, Siemens, Germany). First, the proton density fat fraction (PDFF) was determined using signal voxel spectroscopy (MRS STEAM). Then T₁ values were measured in both water-only and in-phase images acquired with VIBE-Dixon (5) (that is, 3D Spoiled Gradient Echo sequence) and Turbo Spin Echo sequence with 3 point Dixon. With Vibe-Dixon TR=10ms and flip angles =5°, 10°, 15°, 20°, and 25°. With TSE TR=100ms, 200ms, 400ms, 800ms, 1600ms, 3200ms, and 5000ms and echo train length=5. The in-phase and the water-only images were used to calculate T₁ values. Gd-BOPTA concentration were calculated both in water ([Gd]_water) and in the whole water-fat mixture ([Gd]_all). Regressions of R₁ (1/T₁) vs. [Gd]_water and [Gd]_all were performed, slopes of the regression should be equal to the relaxivity.The PDFF in four groups were 0, 25%, 31% and 45% respectively. The calculated T₁ using VIBE-Dixon were different with that using TSE, both with water only images and with in-phase images, for all vials with fat in the mixture (Figure 2 & 3). The regression slope of R1 vs. [Gd]_water in water image acquired with VIBE-Dixon were the closest (minimum standard deviation among all 4 figures) among all four PDFF groups. In images acquired with TSE, however, the regression slopes were the closest for R₁ vs. [Gd]_all in water only images.Given that the Gd-based contrast does not enter the fat component, and the fat-water mixtures were prepared the day before the experiment, the difference in measured T₁ and measured relaxivity using different sequence (we think it is mainly the difference in TR) is very interesting and may provide important information in the DCE-MRI applications. Previous studies from other group also demonstrated that the T₁ measured using variable flip angle method with SPGRE are different from that measured with TSE technique in tissue, indicating that the T₁ decay in tissue might not be mono-exponential (6). It was also demonstrated that even with fat suppression or spectroscopy technique, the T₁ measured in water-only signal or images will be affected by the fat component in the mixture (7). Our results were similar with these previous results but further indicating that there might be some exchange of proton between water and fat in a mixture, affecting the impact of the Gd in the mixture and making the apparent relaxivity to change with the increase of TR.