Aims: The outcome analysis of different revision techniques should provide guidelines for future stem revisions. Methods: Between 1989 and 2000 revisions of 478 femoral stems were performed. 252 of them with cemented components and 226 without cement. In 86 cases a Wagner revision stem was used. The mean observation time was 6.4 years. Success and failures were analysed and classified. Results: Four types of defects and remaining bone stock patterns could be distinguished: Type I with normal diaphyseal bone and minor lesions at themetaphyseal side, type II with a regular isthmus but defects, mostly contained ones, of the proximal femur, type III a wide isthmus with a thin and cylindrical cortical wall, type IV with destructions of the proximal femur which need structural allografting to provide sufficient initial stability. Type I revisions can be performed with regular components as used in primary cases, type II revisions need a slightly longer revision stem with either conical or cylindrical diaphyseal part, type III should be treated with components, that allow for axial load transmission proximally and can take rotational loads distally (the “load sharing concept”). Type IV needs a long revision stem in combination with an appropriate choice of structural allografts. Conclusions: If the type and length of the revision stem is chosen according to this classification, the failure rate of stem revisions can be reduced and the us of expensive allografts can be limited.

Introduction: The operation method selected for acetabular revisions depends on the type of deþciency; Type 1: contained cavitary; the acetabular rim is preserved and thus supportive. Type 2: non-contained deþciencies; the acetabular rim and the peripheral zones are non- supportive defects. In non-contained, rim non-supportive defects acetabular reinforcement rings are used as a rule. Material and methods: Between July 1988 and December 1997 465 acetabular revisions with 229 Reinforcement Rings and 236 Press- Fit Cups (4 bilaterally) were performed. A 7.5 (5.0–10.8) year clinical and radiological follow-up of the Press-Fit Cup inserted in 130 women and 102 men was effected. Mean age at revisions surgery was 70 in women and 73 in men. 39 patients (17%) died without needing a re-revision, and 17 (7%) could only be interviewed by phone. One patient was lost to follow-up. Of the remaining 175 patients (75%) with 179 hips (76%) were available for complete evaluation. Results: The clinical results (according to Merle dñAubignŽ) was excellent and good in 82% 143 pts), moderate in 17% (30 pts) and poor in 1% (2 pts). 22 (9%) reinterventions were noted during the observation period: 12 for dislocations, 5 for aseptic stem loosening, 1 for septic loosening, 2 for secondary periprosthetic fractures and 2 for removal of heterotopic ossiþcations. Conclusions: No Press-Fit Cup had to be re-revised for aseptic loosening. Reconstruction for contained, rim supportive defects with the Press-Fit Cup provides excellent results, comparable to those of primary arthroplasties.