Is Ipsilateral Femoral Head Autograft Reconstruction Durable in Patients Undergoing Enneking II/II+III Tumor Resections at a Minimum 5-year Follow-up?

Background The reconstruction options for Enneking II/II+III pelvic tumors—defined as periacetabular tumor with or without involving the pubis or ischium—remain controversial. Short-term outcomes of ipsilateral femoral head autograft for the reconstruction of acetabular defects have been reported in limited case series studies. However, the long-term stability and complications of this method remain unknown. Questions/purposes At a minimum of 5 years of follow-up: (1) What was the graft survival rate in patients who underwent pelvic Enneking II/II+III tumor resections using an ipsilateral autologous femoral head graft for reconstruction? (2) What were the radiologic outcomes? (3) What were the functional outcomes, and what factors were independently associated with poor functional outcomes (Musculoskeletal Tumor Society Score [MSTS]-93 scores < 80%)? (4) What percentage of patients experienced complications, and what were they? Methods Between January 2007 and June 2020, we surgically treated 583 patients with Enneking II/II+III pelvic tumors. Ten percent (61 of 583) of patients underwent amputation, 57% (335 of 583) underwent curettage, and the remaining 32% (187 of 583) underwent en bloc resection. During the period in question, we generally used en bloc resection in patients with primary pelvic tumors, invasive benign pelvic tumors, and metastatic pelvic tumors. Patients who underwent en bloc resection were considered potentially eligible for inclusion. This study only included patients who underwent Enneking II/II+III pelvic tumor resections with partial preservation of the acetabulum and ipsilateral femoral head autograft reconstruction and who had 5 years or more of follow-up. Twenty-one percent (120 of 583) of patients were excluded because they received other reconstruction methods. Although 2% (10 of 583) of patients had < 5 years of follow-up or were lost to follow-up, some patients may have experienced failure or complications prior to 5 years or at the time of loss to follow-up. Consequently, these patients should still be included, leaving 10% (57 of 583) of patients for analysis. The mean ± SD age at diagnosis was 41 ± 13 years, and the mean follow-up time was 11 ± 3 years. Fifty-one percent (29 of 57) of patients were men. The most common pathologic diagnoses were giant cell tumor of bone (42% [24 of 57]) and chondrosarcoma (35% [20 of 57]). At the last follow-up, 84% (48 of 57) of patients had no evidence of disease, 9% (5 of 57) were alive with disease, and 7% (4 of 57) died of disease. The graft cumulative removal rate and revision rate were estimated using the competing risk estimator. We used preoperative and postoperative thin-section CT scans to assess the displacement of the hip rotation center. We classified complications into major and minor complications according to whether patients underwent unplanned reoperation. The MSTS-93 score was used to assess functional outcomes. Preoperative and postoperative CT scans were used to assess the displacement distance of the hip rotation center and the femoral neck. Cox regression analysis was used to evaluate risk factors associated with poor postoperative functional scores (MSTS-93 < 80%). Results The cumulative implant removal rates at 2 years and 5 years were 2% (95% confidence interval [CI] 0.3% to 8%) and 5% (95% CI 1% to 13%), respectively. The revision rates at 2 years and 5 years were 10% (95% CI 4% to 18%) and 15% (95% CI 7% to 25%), respectively. The median (range) MSTS-93 score was 90% (57% to 100%). Sixteen percent (9 of 57) of patients developed a total of 11 complications, including five major complications and six minor complications. The most common complications were infection (5% [3 of 57]), delayed wound healing (5% [3 of 57]), and hip dislocation (4% [2 of 57]). After controlling for potential confounding variables such as gender, age, and previous surgery, we found that abductor muscle resection (cause-specific HR 2.9 [95% CI 1.3 to 6.1]; p = 0.012) was an independent risk factor associated with poor function. Conclusion Among patients with Enneking II/II+III pelvic tumors with partial preservation of the acetabulum, ipsilateral autologous femoral head autografting combined with hip arthroplasty is a viable reconstructive approach. For patients who cannot undergo tumor prosthesis or allograft reconstruction because of difficulties in obtaining grafts, costs, or social and/or religious reasons, this reconstruction technique may be considered. Additionally, we do not recommend this surgery for patients with complete acetabular resection, as autologous femoral head transplantation is insufficient to completely compensate for bone defects. Future studies with larger sample sizes should compare the advantages and disadvantages of this reconstruction technique with prosthetic or biological reconstruction and focus on clarifying the patient indications for existing reconstruction techniques. Level of Evidence Level IV, therapeutic study.