PurposeUnicompartmental knee arthroplasty (UKAs) are available in the market as fixed‐ and mobile‐bearing (FB and MB) and can be characterised by a different set of design parameters in terms of geometries, materials and surgical approaches, with overall good clinical outcomes. However, clear biomechanical evidence concerning the consequences of variations of these features on knee biomechanics is still lacking; therefore, the present study aims to perform a sensitivity analysis to see which outcomes are affected by these variations.MethodsFor both MB‐UKA and FB‐UKA, five design and surgical parameters were defined (bearing insert thickness, tibial component material, implant components friction coefficient, antero‐posterior slope angle and level of tibial bone resection). Two control models were defined based on standard configurations for both implants. Finite element analysis was chosen to perform this study, and different parameter combinations (216 models in total) were implemented and tested at both 0° and 90° of flexion, using a previously validated finite element knee model. The results were then evaluated in terms of bone and polyethylene Von Mises stress and tibio‐femoral contact area.ResultsBearing thickness, tibial bone cut and slope angle were found to be the most sensitive parameters for both types of UKAs. Specifically, changes in these parameters in the FB‐UKA appeared to induce more significant variations in the polyethylene insert (both in terms of polyethylene stress and contact area), while in the MB‐UKA, these changes influenced bone stress distribution more.ConclusionsSurgical parameters returned to have a more significant influence than material and friction variations; furthermore, the outcomes most affected by parameter variations were the insert‐related ones for FB‐UKA while for the MB‐UKA were the ones regarding tibial bone stresses.Level of EvidenceNot Applicable.