Assessment of Orthotopic Liver Transplant Graft Survival on Gadoxetic Acid–Enhanced Magnetic Resonance Imaging Using Qualitative and Quantitative Parameters

Objective The aim of this study was to evaluate the prognostic potential of a 3-parameter visual scoring (qualitative score [QS]) system for hepatobiliary phase gadoxetic acid–enhanced magnetic resonance imaging (MRI) in orthotopic liver transplant grafts. Materials and Methods This retrospective study of 128 patients was approved by our institutional review board. Two readers independently assigned 3 QSs to T1-weighted MRI scans, 20 minutes after the administration of gadoxetic acid (hepatobiliary phase), based upon the following: (1) liver parenchymal enhancement (EnQS, 0–2); (2) biliary contrast excretion (ExQS, 0–2); and (3) signal intensity of the portal vein relative to the liver parenchyma, that is, the portal vein sign (PVsQS, 0–2). The functional liver imaging score (FLIS) was calculated as the sum score of these 3 parameters. The relative liver enhancement (RLE) was measured as well. Demographic, clinical, laboratory parameters, and imaging findings were included in univariate and multivariate statistical analyses. The primary end point was graft failure, that is, retransplantation or death from liver failure. The probability of graft survival was calculated by Kaplan-Meier estimates and Cox proportional hazards regression. Results In the univariate analysis, EnQS, ExQS, PVsQS, and FLIS scores, as well as RLE, were significantly associated with the 1- to 3-year probability of graft survival (P < 0.001). For a FLIS of (0), the 3-year probability of graft survival was 6.5%, whereas it was 51.3% for a FLIS of (1–3) and 100% for a FLIS of (4–6) (P < 0.001). In the multivariate survival models, EnQS, ExQS, and PVsQS, each independently outperformed the majority of clinical and laboratory parameters, and the FLIS did even better regarding the prediction of 1- to 3-year graft survival. Conclusions In liver transplant recipients, gadoxetic acid–enhanced MRI-derived QSs (ie, EnQS, ExQS, and PVsQS), as well as the FLIS and RLE, can predict graft survival probability.