Synthetic bone models such as polyurethane (PU) foam are a well-established substitute to cadaveric bone for screw pull-out testing; however, little attention has been given to the effect of PU foam anisotropy on orthopaedic implant testing. Compressive and screw pull-out performance in three PU foam densities; 0.16 g/cm3 (PCF 10), 0.32 g/cm3 (PCF 20) and 0.64 g/cm3 (PCF 40) were performed in each of the X, Y or Z orientations. The maximum compressive force, stiffness in the linear region, maximum stress and modulus were determined for all compression tests. Pedicle screws were inserted and pulled out axially to determine maximum pull-out force, energy to failure and stiffness. One-way ANOVA and post hoc tests were used to compare outcome variables between PU foam densities and orientations, respectively. Compression tests demonstrated the maximum force was significantly different between all orientations for PCF 20 (X, Y and Z) while stiffness and maximum stress were different between X versus Y and X versus Z. Maximum pull-out force was significantly different between all orientations for PCF 10 foam. No significant differences were noted for other foam densities. There is potential for screw pull-out testing results to be significantly affected by orientation in lower density PU foams. It is recommended that a single, known orientation of the PU foam block be used for experimental testing.