A quantitative CT analysis of fibula inlayed in a massive allograft for femoral diaphysis reconstruction.

Introduction: In diaphyseal reconstructions for bone tumor resection, massive bone allografts (MBA) are historically regarded as the gold standard. However, these are not without complications, and they present an elevated risk of infection, nonunion and structural failure that increases over time as the graft remains largely avascular. To counteract this disadvantage, a combination of allograft with a vascularized fibula has been proposed. The aim of our study was to objectively review the results of combined vascularized fibula-allograft constructs compared to plain allograft reconstruction for bone defects in tumor patients and to assess fibular vitality predictive factors from imaging studies. Materials and methods: Our data was retrospectively reviewed for patients with femoral diaphysis reconstructions in the past ten years. Ten patients (six males and four females) with a mean average follow-up time of 43.80 months (range 20-83, SD 18.17) with combined graft (Group A) were included in the study. As a control group 11 patients (six males and five females) with a mean average follow-up of 56.91 months (range 7-118, SD 41.33) with a simple allograft reconstruction were analyzed (Group B). Demographic and surgical data, adjuvant therapy as well as complications were analyzed in both groups. Both groups were assessed with plain radiographs for bony fusion at the osteotomy sites. Patients in "Group A" had consecutive CT scans at 6 months and then annually to check for potential bone stock and bone density changes. We analyzed total bone density as well as incremental changes in three different areas of the reconstruction. This was done at two defined levels for each patient. Only patients with at least two consecutive CT scans were included in the study. Results: There were no statistical differences between the groups in terms of demographics, diagnosis or adjuvant therapy (p = 1.0). The mean average surgical time (599.44 vs 229.09) and mean average blood loss (1855.56 ml vs. 804.55 ml) were significantly higher in the combined graft group A (p < 0.001 and p = 0.01, respectively). The mean average length of resection (19.95 cm vs. 15.50 cm) was higher in the combined graft group (p = 0.04). The risk for non-union and infectious complication was higher in the allograft group, however, the difference was not significant (p = 0.09 and p = 0.66, respectively). The mean average time to union at junction sites was 4.71 months (range 2.5-6.0, SD 1.19) for cases of successful fibula transfer, 19.50 months (range 5.5-29.5, SD 12.49) for the three cases where we presumed the fibula was not viable and 18.85 months (range 9-60, SD 11.99) for the allograft group. The difference in healing time was statistically significant (p = 0.009). There were four cases of non-union in the allograft group.Seven out of ten patients in Group A exhibited incremental changes in all CT scan measured values. This difference was statistically significant already at 18 months from the index surgery (p = 0.008). The patients with a non-viable fibula had a smaller increase in the percentage of total bone density area measured in the CT scan compared to those patients with a successful fibula transfer (4.33, SD 2.52 vs. 52.29, SD 22.74, p = 0.008). The average bone density incremental increase in-between the fibula and allograft was different among patients with an unsuccessful fibula transfer (32.22, SD 10.41) and the ones with a viable fibula (288.00, SD123.74, p = 0.009). Bony bridges were observed in six cases of viable fibula and in none of the tree presumably dead fibulas (p = 0.03). The mean average MSTS score was higher for the subgroup of successful fibular transfer (26.7/30, SD 2.87) when compared to the group of non-viable fibular graft (17.00/30, SD 6.08) and this was also statistically significant (p = 0-007). Conclusion: A viable fibula enhances incorporation of the allograft and decreases the risk for both structural failure as well as infectious complications. Viable fibula also contributes to better functional status of the recipient. Consecutive CT scans proved to be a reliable method for assessing fibular vitality. When no measurable changes are present at 18-month follow-up, we can declare the transfer unsuccessful with a good amount of certainty. These reconstructions behave as simple allograft reconstructions with analogue risk factors. The presence of either axial bridges between the fibula and allograft or newly formed bone on the inner surface of the allograft is indicative of a successful fibular transfer. The success rate of fibular transfer in our study was only 70% and skeletally mature and taller patients seem to be at increased risk for failure. The longer surgical times and donor site morbidity therefore warrant stricter indications for this procedure.