Objective To indicate whether ex vivo magnetic resonance imaging (MRI)-proton density fat fraction (PDFF) could be advantageous for selection of liver grafts and to achieve information about pre-transplant indicators of the quality of liver grafts.Materials and methods MRI-PDFF was performed using a 3.0 T scanner. Totally, 35 liver grafts were pathologically confirmed and 19 liver grafts were used for liver transplantation. Surgeon established a grading system for cadaveric livers based on visual inspection of liver color. The grades ranged from normal (score 3) to mild/moderate (score 2) to severe (score 1). One region of interest (ROI) was selected in each of the liver I-VIII segments. The Pearson or Spearman correlation coefficient was used to analyze the correlation according to the distribution of data. MRI-PDFF and histological findings of grafts obtained before implantation ex vivo, together with the graft survival rate and incidence of early allograft dysfunction (EAD), were analyzed. The Kaplan-Meier method was used to calculate graft survival rates, and log-rank test was utilized to compare them.Results The average of nine-ROI PDFF in the cohort was 4.1 +/- 4.4%. The macrovesicular steatosis group had higher MRI-PDFF value than the normal group, with statistical significance (P < 0.05). The visual score in the macrovesicular steatosis group was higher than that in the normal group (P = 0.004). The 1, 2, and 3-year graft survival rates in the macrovesicular steatosis group were 100%, 66.7%, and 66.7%, respectively, which were lower than those in the normal group (100%, 78.6%, and 78.6%, respectively). The incidence of early allograft dysfunction (EAD) was 50% in the macrovesicular steatosis group and 33.3% in the normal group (P = 0.60).Conclusion It was confirmed that ex vivo MRI-PDFF is a safe, accurate, and non-invasive tool for selection of liver grafts and it provides information about pre-transplant indicators of the quality of liver grafts.