BACKGROUND: Successful liver resection in oncologic surgery depends on safety, precision, and efficacy, all of which require a thorough understanding of liver anatomy. Contrast-enhanced computed tomography (CT)-generated three-dimensional (3D) models have been proposed as a valuable tool to enhance this understanding. However, a systematic comparison of different display modalities across professional groups has not yet been performed.

METHODS: In this prospective, monocentric randomized trial, we compared high-resolution two-dimensional (2D) CT images of liver malignancies with their corresponding standardized, non-colored 3D virtual and printed models in facilitating anatomical and spatial understanding as well as surgical decision-making. A total of 91 participants, including 40 surgeons, 10 radiologists, and 41 students, evaluated six clinical cases (three centrally and three peripherally located liver malignancies). Each participant assessed one central and one peripheral case per display modality, presented in a random order.

RESULTS: Compared to 2D CT images, both 3D virtual and printed models significantly improved the identification of tumor location (P < 0.001), enhanced the comprehension of spatial relationships with adjacent liver and portal veins (P < 0.001 and P = 0.019, respectively), and facilitated clinical decision-making (P < 0.001). No significant difference was observed between virtual and printed models in terms of effectiveness. Within the different groups, surgeons and students, but not radiologists, more accurately identified tumor location and spatial relationships with adjacent liver and portal veins using 3D models. Subjectively, most surgeons and students preferred 3D printed models over virtual models and 2D CT images.

CONCLUSIONS: This study demonstrated that standardized, non-colored 3D virtual and printed models equally help preoperative anatomical understanding and decision-making, particularly for surgeons and students. By isolating the influence of display modality, our findings clarify prior inconsistent results and support the integration of cost-effective 3D visualization by applying virtual models into surgical planning and education. Preference for printed models despite comparable efficacy highlights the importance of user-centered implementation strategies.