With the collaboration of multidisciplinary team, the life expectancy has been increased significantly in patients suffer from malignant tumor. Therefore, to improve the quality of life of patients, various reconstruction methods after bone tumors excision have been developed, including megaprosthesis, allografting, and tumor-bearing bone grafts (irradiated bone, pasteurized bone, and frozen bone). Given that biological reconstructions can achieve acceptable long-term functional outcome, biological reconstructions have received increasing attention [10].
Tumor-bearing bone graft is one of the biological reconstructions. In the past 20 years, tumor-bearing autografts frozen with liquid nitrogen have been reported as safe and effective methods for treating osteoblast tumors of various types and locations in basic experimental studies and clinical practices[8, 11-15]. The beneficial effects include a shorter union period, restoration of bone stock, lower cost, osteoinduction, osteoconduction, perfect fit, ease of soft-tissue attachment, activation of antitumor immune response and decreased disease transmission[8, 12]. In fact, FAPC by pedicle method shows significant advantage in proximal femoral tumors. First, it does not require femoral osteotomy or wait for the junction healing. Second, it is easy to reattach the ligaments and soft tissue around the proximal femur to the original anatomic site to increased hip joint stability. Third, it can potentially preserve maximal bone matrix to avoid further retreatment resulting from insufficient bone mass. In theory, all biological reconstructions have similar advantages and disadvantage. However, due to the loss of osteoinductive and osteogenic properties after thermal or radiation treatment, the allograft might have potential risk factors which require further surgery, such as nonunion with the host bone, graft fracture, bone resorption and immunological reactions [5-7]. Takata et al. also reported that tumor-bearing frozen bones maintains their microstructure and osteoinductive ability compared to pasteurization, autoclaving and allograft [11].
Biomechanical stability is of great concern to biological reconstructions. Lee et al. reported that the pasteurization decreases the biologic and mechanical properties and reduces strength to less than that of an allograft[16]. Interestingly, Yamamoto et al. had reported that the frozen bone has sufficient biomechanical strength for limb reconstruction that is comparable to pasteurized autografts and allografts[13].
Previous studies reported that an APC reconstruction exhibited satisfactory 5-year survival rate (72–90%) and MSTS score (77–90%), respectively. APC is a better reconstruction option if easily available. Remarkably, graft–host junction union is a major problem, and the nonunion rate is reportedly 5%–19%[5, 6, 17-19]. On the other hand, Eid et al.[20] reported the outcomes of the application of the pasteurized APC in 18 patients; MSTS functional score was 80%, 5- and 10-year graft survival rates of 86%, respectively; mean graft-host junction union time of 13 months, and 1 case of non-union. Another study using extracorporeal irradiated APC by Chen et al.[21] reported an MSTS score of 72%, mean graft-host junction union time of 20 months, and 5-year graft survival rate of 85%. It seems that all biological reconstruction methods feature acceptable functional outcomes and implant survival rates. A few cases of nonunion or delayed union have been reported, although many studies reported that using the step-cut osteotomy, autogenous or allogenous bone graft into host bone, and non-cemented prostheses increased graft-host junction healing and stability in all biological reconstructions[5, 19, 22] (Table 2).
Greater trochanter stability in biological reconstruction is also a concern. Instability of abductors reconstruction prone to potentially severe complications, such as greater hip abductors avulsion, resorption, hip abductor avulsion, Trendelenburg gait or dislocation, which lead to poor function and a protracted postoperative rehabilitation period[19, 20]. Although abductors reconstruction methods are controversial, the hip abductor strength and gait which APC seems to be superior to megaprothesis [5, 19, 23]. A few studies emphasize the importance of preserving the proximal capsule and re-sutured onto the allograft to prevent hip dislocations [17]. In the current study, similar procedures were also applied to peeled gluteus medius from greater trochanter and reattached the gluteus medius to the original site. Bone to tendon reconstruction which might easy to restore abduction strength due to integrity of the gluteus medius. In this study, only one patient occurred joint dislocation, which is comparable to other methods of biological reconstruction [7, 23, 24].
In addition, bone cement or graft plus long-stem-prosthesis has been used in some patients with metastases combined lytic to make up for osteolytic destruction in the proximal femur. Such treatment will provide sufficient mechanical strength and stabilization of bones. Given that pedicle freezing required more a longer incision and muscle dissection, the authors encouraged isometric exercise to avoid complication in early postoperative period. The bone cement and prosthesis provide rigid fixation and avoid bone union which make it possible for full weight-bearing at 6 weeks post operation. Meanwhile, all the patients underwent physical exercise to learn prevention hip dislocation, and postoperative bracing provided sufficient support to minimize dislocations if necessary. In case of large resection of muscles, the mesh can be used to attach surrounding soft tissues, and abduction brace is used to prevent hip dislocation. Of note, if patients showed signs of severe destruction of cortex (>2/3) in the proximal femur, megaprosthesis was recommended [12].
Megaprosthesis has satisfactory short- to medium-term outcome, early mobilization and weight-bearing, and short operative time; however, abductor muscles reattachment remains an issue need to be concerned. Accumulating evidence indicates using an artificial ligament to affix the megaprosthesis can promote soft-tissue reconstruction and achieve better joint stability and functional outcomes[25]. In fact, artificial ligament use is probably unable to reduce the occurrence of prosthetic complications such as aseptic loosening, prosthesis breakage, infection, and stress shielding[26]. Moreover, long-term prosthetic failure rate is between 6–33%, mean MSTS score is between 63–83%, and the major complications were infection (5–13%) and dislocation (0–20%) [4, 18, 26-28] (Table 2). The application of silver– or iodine– coated implants had been reported to reduce overall infection rates[29, 30]. However, a higher cost burden, unavailability in some countries and the limited bone mass complicate revision surgeries.
The various reconstructive alternatives have acceptable oncological and functional outcomes. However, each method has its own limitations, and thus it is crucial for choose the proper method to maximize the benefits for each patient. Surgeons must carefully consider the patient’s age, general condition, response to chemotherapy, and expectations when individualizing a treatment plan.
In the current study, the functional outcome was similar to those of other reconstruction methods. The mean MSTS functional score was 88%, and the 5- and 10-year graft survival rates were 100% and 50%, respectively. At the last follow-up, no prosthesis loosening or obvious lucent shadows around the autograft bone was observed on radiological examination. Only 1 patient had local recurrence around the residual soft tissue for which hip disarticulation was performed after 2 re-excisions. Wear of the acetabulum occurred in 2 patients despite the use of bipolar hemiarthroplasty, but this might be an inevitable long-term complication of joint replacement.
This study had several limitations. First, due to its retrospective design and single center, a relatively small number of patients were enrolled, a follow-up >10 years was available for only 4 of the 19 patients. Second, no control group was available for comparison of functional outcomes; thus, our results could be compared to only those of prior studies. Similarly, the accuracy of our results was lower than those of randomized study. Third, the patients had various diagnoses and were treated with various chemotherapy regimens, which might affect survival rates and functional outcomes. Therefore, to assess the efficacy and safety of this procedure, a prospective study which compares the functional outcomes and survival rates of several reconstruction methods over a long-term follow-up period needs to be performed in the future.