Implant Survival and Complication Profiles of Endoprostheses for Treating Tumor Around the Knee in Adults: A Systematic Review of the Literature Over the Past 30 Years

Abstract Background Endoprosthetic replacement has become the mainstream method of reconstruction after tumor resection around the knee for decades, but there is a lack of comprehensive review evaluating the implant outcomes. We performed a systematic review to analyze the implant survival and complication profiles of distal femoral replacement (DFR) and proximal tibial replacement (PTR) in adults, and to evaluate the effects of different fixation methods and hinge mechanisms. Methods A systematic review of 40 studies with 4748 DFR cases and 1713 PTR cases was performed after searching the PubMed and EMBASE databases. Results of the implant longevity, complications, and other relevant data were extracted, recategorized, and analyzed. An additional review of 227 cases of Compress DFR from 6 studies was also performed. Results The mean 5-, 10-, 15-, and 20-year implant survival rates of DFR were 78.3%, 70.1%, 61.6%, and 38.3%, slightly higher than those of PTR (75%, 60%, 55.3%, and 25.1%). Aseptic loosening (8.8%) and infection (8.5%) were the most devastating complications in DFR, while in PTR it was infection (16.8%). Cemented or cementless fixation did not significantly affect implant survival or aseptic loosening rate. Rotating-hinge mechanism might improve long-term implant survival and reduce bushing wear, but not necessarily prevented aseptic loosening. The series Kotz modular femur and tibia replacement system/Howmedica modular replacement system/global modular replacement system was one of the most durable implants. The Compress DFR showed no superiority in implant survival and complication profiles over other brands of endoprostheses. Conclusion Short-term to mid-term implant survival of adult tumor endoprostheses around the knee is acceptable but long-term outcome remains unsatisfactory. Efforts should be made in reducing loosening and infection.