BACKGROUND: Understanding the hemodynamics of the total cavopulmonary connection (TCPC) may lead to further optimization of the connection design and surgical planning, which in turn may lead to improved surgical outcome. While most experimental and numerical investigations have mainly focused on somewhat simplified geometries, the investigation of the flow field of true TCPC configurations is necessary for a true understanding. METHODS: This study details a manufacturing methodology yielding more accurate in vitro models that would provide a better understanding of the TCPC hemodynamics and adequate data for the validation of anatomical CFD simulations. This approach is illustrated on two different TCPC templates: an intra-atrial TCPC with a single superior vena cava (SVC) and a bilateral SVC with an extra-cardiac conduit. Power loss, flow visualization, digital particle image velocimetry (DPIV) flow measurements as well as computational fluid dynamics simulations are performed to characterize the anatomic flow structure. Additional parametric glass models of the TCPC were manufactured to help understand the fluid dynamics of the anatomical models and support the computational model validation effort. RESULTS/CONCLUSIONS: Both anatomic configurations revealed very different fluid dynamics underlining once again the need for at least one comprehensive experimental campaign per TCPC template for a good understanding of the flow phenomena. The absence of caval offset in the anatomical intra-atrial model resulted in important flow turbulence, which was enhanced by the large connection area and yielded high pressure drops and power losses. On the other hand, the bilateral SVC, which featured a smooth extra-cardiac conduit and wider vessels, led to power losses that were one order of magnitude lower than those of the anatomic intra-atrial model and a smooth flow field with lower levels of instability. The simplified glass models demonstrated that the diameter of the connecting vessels and of the pulmonary arteries in particular, was a parameter of prime importance. Finally, this study also reports on a combined experimental and numerical validation methodology, suggesting a cautious approach for the straightforward use of available CFD tools and pointing out the need for developing high resolution CFD techniques specifically tailored to tackle the complexities of cardiovascular flows.