In severe conditions of osteoarthritis, total replacement of the knee joint is essential and it is a well-established surgical procedure performed to relieve patient’s discomfort and to restore function in knee osteoarthritis. Surface roughness of knee prostheses components has been suggested as contributing to the mechanism for premature prosthesis failure and is very important to improve tribological performance. The objective of this research was to characterize the roughness of tibial ultra-high-molecular-weight-polyethylene of retrieved knee implants through an optical profiler, discussing the results in the context of the acting loads obtained from a multibody musculoskeletal model. The authors looked for roughness differences between medial and lateral condyles considering the different load exposure of the two condyles. As prove of the different stress to which the two areas are subjected, the authors simulated the loads acting on a knee joint by developing a model in AnyBody Modelling System™.11 total knee prostheses, of the same design and all posterior stabilised, were considered in this study. Roughness and topographic surface were acquired on them. These knee prostheses were explanted from 11 patients (mean age of 68years) after a mean of 3.2years. Statistical tests outlined a difference in surface roughness when different sizes of implants are considered, whereas the authors did not find a statistically significant difference between medial and lateral condyles in term of their surface roughness.