OBJECTIVES: To compare Gadoxetic Acid-enhanced MRI (PV-MRI) with extracellular contrast media-enhanced MRI (ECCM-MRI) and 3-phase MD-CT (MD-CT) in the planning of surgery for patients with colorectal cancer liver metastases (CRCLM). METHODS: A decision-tree model was used. The probabilities were collected and validated by radiologists and liver surgeons in Germany, Italy and Sweden using a Delphi panel process. Cost was derived from published sources and the mean costs for the three countries were used in the model. The model focused on the initial imaging, the need for further or confirmatory imaging and unnecessary or modification of surgery. The model will be available as an iPad application during ISPOR. RESULTS: Using PV-MRI resulted in increased certainty in the decision-making, demonstrated by the reduced need for further imaging: 23.5% for MD-CT, 18.5% for ECCM-MRI and 8.6% for PV-MRI. There was a need for confirmatory imaging in high-risk surgery patients for 7.6% and 9% in the MD-CT and ECCM-MRI groups respectively. Unnecessary surgery was performed in 3.4% of PV-MRI and 5.4% of MD-CT and ECCM-MTI patients respectively. The need for intra-operative modification of the surgical plan was 7.7% for PV-MRI, 10.5% for MD-CT and 10.3% for ECCM-MRI. The cost of imaging, unnecessary and modified surgery was €892 for MD-CT, €997 for ECCM-MRI and €888 for PV-MRI. Additionally, it was, as expected more patients undergoing potential curative surgery in the PV-MRI group (49.4% versus 47.6% for MD-CT and 45.4% for ECCM-MRI). CONCLUSIONS: Using PV-MRI rather than ECCM-MRI and MD-CT for the initial imaging in patients with CRCLM leads to expected cost savings and reduction of unnecessary and modified surgery. A clinical trial, using a similar protocol as the model, is currently being analyzed. The results of the trial will be used to further validate the modeling approach and to demonstrate the impact on health outcomes.