Page 130 - Reconstructive Techniques in Musculoskeletal Tumor Surgery

Chapter 6
Discussion
In this nationwide retrospective survey, we evaluated (1) mechanical complications and infection, (2) oncological outcome, and (3) failures and allograft survival following hemicortical allograft reconstructions for the treatment of primary bone tumors. To the best of our knowledge, this study represents the largest series on hemicortical reconstructions to date.
Mechanical Complications and Infection
The most frequent complication was host bone fracture, the rate of which was 18%, which is in accordance with rates of 10% to 27% found in previous studies on hemicortical resection (table 5)3,11,15. Other authors reported no fractures, but they did not describe the extent of cortical resection, which was the most important risk factor in our patients12-14,24. The association between fractures and the extent of cortical resection may be explained by greater stresses acting on a smaller portion of remaining cortex25. Additional factors should, however, be considered. First, perfect tting of allografts may reduce fracture rates26. Three- dimensional CT scanning of allografts27 may aid in the selection of better- tting grafts. Second, osteotomies with sharp angles and screw xation perpendicular to the bone axis ( gures 1A, 1B, and 1C) act as stress-risers and should be avoided28,29. We advise surgeons to perform rounded osteotomies (“boat-shaped resections”) when possible and to insert screws in an oblique fashion29,30. Recommendations for when to use plate xation are proposed in gure 5.
Nonunion occurred in 7% of our patients, and resulted in failure in 2%. In previous reports, none of the patients required surgery to facilitate union (table 5). Autograft use may improve union rates, but it is not suitable for reconstruction of larger defects. Also, harvesting of autografts has been associated with substantial complication rates, especially prolonged pain at the donor site31-33. On the other hand, 24% to 47% of segmental allografts demonstrate nonunion so the rate in the current study may be considered encouraging7-10. Various factors may explain these di erences, including the fact that hemicortical reconstructions have a larger contact surface between allogeneic and host bone. The extent of soft-tissue dissection is generally limited in hemicortical resections; authors have hypothesized that this provides a superior environment for incorporation3,28. Moreover, the number of patients receiving adjuvant radiation or chemotherapy was limited in our study. Adjuvant therapies are known to delay bone-healing34.
128