Graft Infections in Biologic Reconstructions in the Oncologic Setting: A Systematic Review of the Literature.

Biologic graft infection (BGI) is one of the main complications in graft reconstructions. However, very little evidence exists regarding the epidemiology of BGI, as most of the data come from sparse reports. Moreover, most of the series did not detail the treatment and outcome of graft infections. The aim of this systematic review of the literature is to provide a comprehensive data collection on BGI after oncologic resections. Three different databases (PubMed, Scopus, and Web of Science) were searched for relevant articles, and further references were obtained by cross-referencing. 139 studies met the inclusion criteria. A total of 9824 grafts were retrieved. Among these, 684 (6.9%) were in the humerus, 365 (3.7%) in the pelvis, 2041 (20.7%) in the femur and 1660 (16.8%) in the tibia. Most grafts were osteoarticular (2481, 26.7%) and intercalary 2112 (22.7%) allografts. In 461 (5.0%), vascularized fibula grafts (VFGs) were used in combination with recycled autografts. Recycled grafts were reported in 1573 (16.9%) of the cases, and allograft-prosthetic composites in 1673 (18.0%). The pelvis and the tibia had the highest incidence of BGI (20.4% and 11.0%, respectively). The most reported first treatment was debridement and implant retention (DAIR) in 187 (42.8%) cases and two-stage revision with graft removal in 152 (34.8%). Very little data are reported on the final outcome specified by site or type of graft. This systematic review of the literature confirms a high incidence of infections in biologic reconstructions after resections of primary bone tumors. Despite DAIR being a viable attempt, in most cases, a two-stage approach with graft removal and reconstruction with endoprosthesis presented the highest chance to overcome infection, guaranteeing a reconstruction. We emphasize the need for future multicentric studies to focus on the management of infections after biological reconstructions in bone sarcomas.