Direct skeletal attachments for transfemoral amputees have been the subject of clinical trials since the early nineties. This method of attachment allows the amputee an unrestricted range of motion around the hip joint, better sitting comfort, improved sensory feedback through osseoperception, improved limb control and reduced soft tissue problems. However, the length of the rehabilitation period is perceived as a shortcoming by the amputees and the clinicians. The aim of the present study is to estimate the risk of failure during gait, for a patient with direct skeletal attachment of a femoral prosthesis, using finite element analysis (FEA). Material properties and loads were derived from subject-specific data and implant stability assumed secured by bone ingrowth into a porous implant surface. A simplified FEA was used to optimize the implant geometry with respect to load bearing capacity. The resulting geometry was then implemented in a subject-specific FE study. The results indicate that the risk of failure for the implant system is approximately three times greater than what can be expected for an intact femur. The main conclusion, based on the risk of failure factors calculated, is that it is likely that a porous-coated implant could be beneficial for osseointegrated fixation. It is also suggested that the proposed methodology can be used in future studies exploring the mechanical stability of osseointegrated fixation in the view of improving direct skeletal attachments for lower limb amputees.