Aims: The purpose of this study was to analyze treatment of postoperative femoral periprosthetic fractures at the Institute for orthopedic surgery ÒBanjicaÒ. Methods: There were 26 patients included in this study. Twenty-one case was of periprosthetic fracture after total hip replacement and in 5 patients there was a femoral fracture after total knee replacement. Fractures were classiþed as: type A Ð prosthesis and fracture stable; type B Ð prosthesis stable and fracture unstable; type C Ð prosthesis and fracture unstable, adequate bone stock; and type D Ð prosthesis and fracture unstable, inadequate bone stock. Functional assessment was made using the modiþed Harris score. Results: Follow up period ranged from 2 to 9 years with a mean of 3.2 years. Time spent from premier procedure till fracture averaged 2.5 years with range from 3 months to 6 years. In 14 cases treatment consisted of prostethic revision with longer cemented stem and in 10 fracture þxation was done using plate and screw and/or wire þxation. 2 patients went on for nonoperative treatment. Cortical allografts were used in 4 cases. 18 out of 26 patients returned to preinjury level of function as assessed by Harris hip score. Conclusions: Analysis conþrmed two basic principles in the treatment of periprosthetic femoral fractures: 1) Unstable and loose stems should be replaced with longer stems that would provide implant and fracture stability and 2) Displaced fractures should be stabilized.