Given the relatively short survival time of total hip replacement in young patients, a different concept of femoral implant was designed and evaluated biomechanically prior to clinical application. The femoral component has no intramedullary stem and fits into an angular resection of the femoral neck. A trans-trochanteric screw provides immediate strong fixation. A comparative biomechanical study with cadaver femora and a dynamic hip simulator was performed. The study had three objectives. (1) The strains generated under loading in the intact proximal femur were compared to those generated in the presence of the experimental implant and a classic cemented implant. (2) The short-term stability of the experimental prosthesis under loading conditions encountered during fast walking was evaluated. (3) Most importantly, the implant design was improved in the light of the experimental observations prior to any clinical trial. Indeed, the prototype underwent several important modifications in the course of the study. With the modified design, later used in a clinical trial, it was demonstrated that if accurate fit and position of the prosthesis were provided, fixation remained sound for over one million loading cycles with peak loads of three times the body weight. The measured interface micromotion was compatible with direct bone apposition and ingrowth. No statistical difference was found between the strain measurements obtained before and after implantation of the experimental prosthesis whereas significant stress shielding was observed with a stemmed, cemented prosthesis.