With extended life expectancy, the number of primary joint arthroplasties has also increased. Revision surgery is also more often necessary, due to aseptic or septic loosening of the prosthesis or periprosthetic fracture. Large bone defects often occur in these patients and several non-modular, conventional implants are available to handle this difficult situation. Custom made implants offer an individually designed and defect-adapted shape with perfect covering of the lesion, but may delay the operation. The Modular Universal Tumor And Revision System (MUTARS®) offers the possibility to vary the length and angle of the femoral neck and the antetorsion angle. Thus, it permits intraoperative adaption to the individual patient's defect. The aim of our study is to present clinical, functional results and the rate of complications in a cohort of patients undergoing revision surgery for failed endoprosthetic replacement or failed trauma surgery, using the Modular Universal Tumor And Revision System with short-to midterm follow-up. Between August 2005 and September 2014, 25 patients (17 female, 8 male) with an average age of 72 years (min. 56, max. 92 years) were included in a retrospective study using the MUTARS® system. The patients underwent surgical revision of osseous defects that were not susceptible to conventional care. The mean follow-up was 16 months (min. 12, max. 72 months). The indications which led to revision surgery were as follows: loosened metal-on-metal prosthesis with an acetabular defect caused by a metal-induced granuloma (n = 1), pseudoarthrosis after implantation of femur nail (n = 2), cut-out of a femur nail (n = 1), aseptic loosening of an implanted endoprosthesis (n = 4), septic loosening of an implanted endoprosthesis (n = 1), periprosthetic fracture (n = 6) and periprosthetic infection after two stage revision surgery (infection consolidated Girdlestone situation) (n = 10). All patients were followed up at regular intervals, both clinically and radiologically. Apart from comorbidities, clinical and functional parameters were measured, including the Harris Hip Score. An outcome analysis was also performed with respect to peri- and postoperative complications. Serial plain X-rays were followed-up. In all patients, stable reconstruction without the use of an allograft was possible. In total, 23 patients underwent surgery by using a proximal femur prosthesis; three patients also received a custom made acetabular component. One patient was treated by using a distal femur and one patient was treated by implantation of a total femur prosthesis. The average length of hospital stay was 25 days (min.11, max. 47 days). The Harris Hip Score improved from 28 points preoperatively to 81 points after surgery.In total we recorded 24 % of complications after surgery. Two patients suffered recurrent periprosthetic infection and the prosthesis had to be revised in a further two stage exchange; one patient had a tractus gap and was revised by surgery. Two patients suffered periprosthetic fracture due to trauma...