Résumé : Introduction: Total knee arthroplasty has proved to be a safe, effective and reproducible surgical treatment for patients with serious/advanced degenerative joint disease of the knee, but the optimal results after these implants can be achieved only if the joint kinematics and kinetics are carefully respected and not significantly altered after the replacement. In order to enhance the capability of matching the healthy configuration, therefore, different prosthesis models in terms of constraints and designs are available for the surgeons to choose among. As an example of this variety, mobile bearing models allow the surgeon to choose among different insert designs in terms of geometry and relative biomechanical approaches, with the relative performances depending on which one is adopted. This article aims to analyse the effect of different levels of congruency of mobile bearing inserts through an experimental activity performed on an ultra-congruent insert design, also referred as deep-dished. Methods: The experimental activity followed the same protocol used in a previous study focused on the effect of symmetricity/asymmetricity (in order to allow a comparison among the results obtained), i.e. it analysed passive squat on 5 cadaveric knee specimens; internal-external rotations of femur and tibial insert respect to the tibia tray were thus measured via a motion capture system in order to compare the insert-femur relative rotation among the different congruency configurations. Results: The overall knee kinematic (in terms of flexion-extension and internal-external rotation ranges of motion) with an ultra-congruent insert did not differ from the ones found for the other inserts; the insert-femur rotation values, instead, returned to be between the ones obtained with the other two inserts. Conclusion: In terms of adaptability to tibio-femoral rotation, therefore, this insert represents a valid option for the surgeon seeking for an intermediate solution among the opposed approaches of the other two.