RESUMO
INTRODUCTION: Vascularized fibular autografts (VFA) are used in the oncologic skeletal reconstructions of long bones, alone or combined with massive bone allografts (MBA). Data regarding the role of imaging in assessing these complex skeletal reconstructions are lacking, and have mainly focused on Computed Tomography (CT). Our aim was to evaluate if early conventional radiography (CR) findings are correlated with the outcome of these skeletal reconstructions. MATERIALS AND METHODS: All consecutive patients who underwent oncologic resection of lower limbs long bones followed by VFA reconstruction were included in this single-center retrospective study. We compared the CR obtained immediately after surgery with the CR at the 6-month control, as well as the CR at 6 months with the CT at 6 months when available. The following scores were assigned to the VFA: 0 (unchanged), 1 (osteopenia-cortical bone thinning), 2 (increase in bone density-cortical thickening). We then investigated whether this score correlated with the implant outcome within 12 months (optimal integration, suboptimal integration, integration requiring further surgery or lack of integration) using Kaplan-Meier and Cox regression analyses, considering the occurrence of integration and the duration time before the surgical removal of the whole bone reconstruction. RESULTS: Forty-five patients were included (32 men [71.1%], mean age 14.6 years), 26 affected by osteosarcoma, 14 by Ewing sarcoma, 3 by adamantinoma and 2 operated for the failure of previous reconstructions for bone sarcoma. VFA changes on 6-month CR were significantly associated with optimal integration of the implants (log-rank P = 0.0137, multivariate Hazard ratio = 7.62, 95% confidence interval = 1.13-51.25). None of the other clinical and surgical features were associated with the implant outcome. The findings on 6-month CR and CT follow-up were not significantly different. CT at 6 months was available in 36 patients (80.0%). CONCLUSION: The assessment of VFA morphological changes on CR performed at 6 months can predict the outcome of the skeletal implant. This data should be considered for clinical decision-making, selecting patients requiring additional images (CT), and possible subsequent revision surgical procedures.
