Is Proximal Femur Reconstruction With a Vascularized Fibula and Allograft Successful at Reconstructing a Tumor Resection in Children 6 Years of Age or Younger?

Abstract

Background: Treating bone sarcomas in young children, particularly in the proximal femur, is challenging because of the need to preserve growth potential, maintain joint function, and conserve bone for future revisions. In 1997, we introduced a new technique combining a vascularized fibula that preserved the proximal epiphysis to substitute for the femoral head and physis with a massive bone allograft aimed at providing a supportive scaffold for the autograft during growth. This approach initially showed promising results in both anatomic and functional restoration, but because it was a complex operation, we believed that longer term follow-up of this procedure would be helpful to judge its value as a potential reconstructive option.

Questions/purposes: (1) What proportion of patients achieved allograft union; what proportion exhibited radiologic signs of vitality in the vascularized fibula, demonstrating bone growth and remodeling; what proportion avoided revision surgery; and what complications were observed with this technique? (2) What were the observed outcome scores using the Musculoskeletal Tumor Society (MSTS) score? (3) What was the status of the hip after treatment?

Methods: Between 1997 and 2010, a total of 14 patients who were 6 years old or younger underwent resection of the proximal femur for a bone sarcoma in two sarcoma centers. Six patients were treated with a vascularized fibular autograft plus bulk allograft reconstruction. The median (range) length of the proximal femur resection was 13 (11 to 14) cm. During the period in question, we generally performed the vascularized fibular autograft plus bulk allograft when the acetabular diameter was < 36 mm. None of those who had the fibula autograft plus bulk allograft reconstruction were lost to follow-up before 2 years without meeting a study endpoint (revision, reoperation) or died prior to 2 years with implants intact. All six patients in the original group had either a follow-up of at least 14 years or had met a study endpoint (revision, reoperation) before that minimum surveillance duration; these patients were the focus of the current study. We noted that two patients had follow-up periods of 20 and 27 years but have not been seen in the last 5 years; they were included for survivorship analysis purposes, yet we emphasize that we cannot ascertain their current status. The median (range) follow-up was 17 (14 to 27) years.

Results: Six of six patients achieved allograft union, while five of six underwent reoperations, and four of six had the graft removed as part of a revision procedure at a median (range) of 19 (7 to 40) months. Only one patient maintained the original reconstruction after 27 years, despite an epiphysiolysis that occurred 56 months after surgery, which was corrected surgically. The other five patients experienced a complication of the reconstruction at a median (range) onset at 19 (7 to 40) months after surgery. Two of the six patients had a common peroneal nerve palsy at the harvested site. Allograft union was observed a median of 8 months from the index procedure. All patients presented residual limb length discrepancy ranging from 1 to 7 cm. The median (range) MSTS score before any revision was 38% (33% to 93%). Four of the six patients underwent removal of the biological reconstruction and were reconstructed by other methods. They displayed a median (range) MSTS functional score of 88% (73% to 93%) at most recent follow-up. One of the six patients showed progressive diaphyseal hypertrophy, growth, and remodeling of the epiphysis, along with integration with the allograft, and had a final limb length discrepancy of -1 cm. The vascularized fibular epiphysis was enlarged and remodeled to provide a new femoral head and grew similar to the physis of the native femoral head in this one patient.

Conclusion: Biological reconstruction using a vascularized fibula and bone allograft in young patients with proximal femur sarcoma was not a durable solution for most patients, and it resulted in frequent unplanned reoperations and revisions. Only one of six patients in this small series had a reconstruction that lasted into adulthood. Even though this was a small group of patients, the poor survival of the construct, the frequent complications, and the unpredictable results associated with it led us to discontinue its use in favor of prosthetic replacement.

Level of evidence: Level IV, therapeutic study.