Background: Aseptic loosening [AL] of cementless implants is causing approximately 60% of total knee and total hip arthroplasty revisions. AL is caused among other factors by micromotion. The result is that the bone around the prosthesis is replaced by a fibrous membrane. This fibrous membrane allows for more micromotion and we thus enter a vicious circle of loosening. In-vivo research for treatments of AL and research into preventing AL is rare. When in vivo research is conducted, this is commonly done in large animals such as sheep and dogs. To bring down costs and increase reproducibility it is desired to recreate the fibrous membrane in a smaller animal. Therefore, an unstable knee hemiprosthesis was designed for the femur of the Wistar rat. Methods: First a test was conducted on cadaverous adult male Wistar rat femurs to assess whether it was easier to access the medullary canal of the femur from the hip or the knee side. Then, a second test was conducted on cadaverous rat femurs (n=8) to assess the depth and width to which we could implant a cylindrical prosthesis. The third set of tests was to see if the conceptual unstable prosthesis allowed micromotion of 200 μm. Miniature silicone rubber and polyurethane resin [PUR(r)] springs were tested and a mock implantation in bone surrogate was conducted. Results: The results obtained from the first test show that it is easier to access the medullary canal of the femur from the knee side. The straight canal of the distal femur was also found to be more suitable to allow for micromotion than the curved canal of the proximal femur. The second test showed that we could implant a cylinder with a diameter of 2.3 mm and a length of 2.3 cm in the femoral canal. The third test showed that the conceptual prosthesis allowed for micromotion, but the spring’s material needs to be optimized. PUR(r)’s creep and compression set were too high. Silicone showed the most potential, because of its low compression set (permanent compressive deformation). However, the stiffness of the silicone needs to be increased and it showed high wear during fatigue testing with respect to the PUR(r). Conclusion: The conceptual prosthesis showed promise, but improvements on the spring material are necessary. The silicone can be made stiffer and altering the design can also increase the stiffness of the prosthesis. A flaw in the conceptual design is that micromotion is restricted in vivo, because the femoral prosthesis is abutted by the distal cortical bone. A change in the design of the prosthesis can remedy this problem. Another remaining challenge is to design a prosthesis which allows for localized application of wear debris particles and treatment solutions in the peri-prosthetic tissue, fully simulating the circumstances that occur during AL.