Forty-one Wagner revision stems were implanted at the Orthopedic Department of the University of Tübingen between July 1990 and January 1993. We report the results of 37 patients at an average follow-up of 27 months (13-48 months) postoperatively. The main indication was stem loosening with considerable loss of bone. In addition, we used the implant 4 times in primary arthroplasty. At follow-up examination 33 patients (89%) were satisfied with the postoperative outcome. According to the Merle D'Aubigné score (12-point scale), 32 patients showed a poor functional result of less than 6 points preoperatively. Postoperatively, the results of 36 patients could be classified as very good to good. To categorise the radiological destruction of the implant bed, we used the femoral shaft defect classification of the DGOT (Deutsche Gesellschaft für Orthopädie und Traumatologie) in conjunction with the classification of Pak and Paproski [5, 11]. Twenty patients presented with trochanteric and calcar defects, and 11 patients with a combination of a calcar and shaft defect. We found a circular shaft defect in 2 patients. In 7 cases we assessed the bone remodelling postoperatively as very good, with strong newly formed bone structures, and in 25 cases as good, with remodelling of the old stem bed and bony structuring of the osteolyses. A secondary sinking in of the Wagner stem was seen in 7 cases. Only one stem had to be revised because of pain symptoms and loosening; in all other cases a secondary stabilisation of the revision-stem took place. With the Wagner revision stem, there is the possibility of achieving mechanical stability even in situations with massive bone loss. The evacuation of bone cement and granulation tissues is facilitated by the transfemoral approach, bony remodelling is accelerated, and bone grafting is often not necessary. As our short-term results show, the concept is a promising one. Nevertheless, we will be very careful in following these patients in the long term, as we have noticed stem sinkage in a small percentage of cases.