At present, the recommended surgical intervention for osteochondroma is the marginal resection. However, a large bone defect will be left after complete resection of the giant osteochondroma, although the definition of the size of massive bone defect has not been well determined [10]. Derek F Amanatullah et al. suggests in the biomechanical study of long bone defect that the more increase of the area of cortical defect, the less hardness of defective bone[12]. Moreover, the average torsional stiffness has a strong linear correlation with the size of cortical defect, so the remaining bone after resection of bone defect is of great significance to its stability. These biomechanical analyses predicted a severe loss of torsional integrity when the cortical defect approaches 50% of the width of the femur. Therefore, it is necessary to actively carry out reconstruction surgery after the resection of osteochondroma in order to restore the biomechanical stability of defective bone [9, 11, 12].
In our surgical procedure, allograft was used to obtain the strength and mechanical stability of the reconstructed bone. In the process of bone union, the contact area of allograft and host bone needs enough matching and long-term absolute stability to complete the slow process of creeping substitution. Therefore, the allograft needs long-term protection to share the stress beyond its supporting capacity. At the same time, allogenic cancellous bone was applied in our investigation to fill in the medulla to preserve the bone mass after the defect as much as possible. The application of cancellous bone could fill the gap left after the bone graft, so that it can increase the bone contact area between allograft and host bone. The enlarged contact surface is capable of accelerating the process of creeping substitution between allograft and host bone, which could shorten the time of bone graft fusion. The usage of cancellous bone can also induce ideal osteoinduction and osteoconduction to achieve satisfactory bone union.
In our study, 3 patients had tumors in the humerus and 4 in the femur. Due to the femur and tibia are the main weight-bearing bones of the human body, the stress load after reconstruction is greater. Thus, for some patients with larger femoral defects, the allograft is fixed with steel plates and screws to strengthen the mechanical stability of the affected limb. Regarding the enhancement of the stability of allograft and reduction of the risk of fracture, Gupta S et al. reported that allograft augmented with intramedullary cement and plate fixation is a reliable solution [13, 14]. Being the non-weight-bearing bone, the humerus has a lower bearing requirement compared with femur or tibia. Therefore, we added absorbable screws for fixation instead of traditional metallic screws, and the implantation of steel plates were not necessary. Compared with the traditional plate and metallic screw fixation, this technique can reduce the use of metal materials, which can save the patients from the pain of removing the internal fixation again in the future. Besides, it has the advantages of reducing the opacity effect of metals in X-ray films and metal artifacts in CT and MRI, so we are able to make an early assessment of fusion process of cortical cancellous bone and local recurrence [15, 16]. Another benefit is that postoperative pain associated with elastic modulus mismatch may be reduced. Compared with metal plates and screws, cortical allogeneic bone scaffolds can better reconstruct the biomechanical elastic modulus of bone. In some published literature, this mismatch is considered to be one of the considerable causes of postoperative pain and eventual implant failure [17–19].
Because the texture of allograft is more brittle than that of normal bone, it is difficult for other tools to grind it into a suitable shape. As a consequence, piezosurgery was used in the process of operation to cut the interface between host bone and allograft directly into bevel in order to increase the contact area of biological bone graft and make it more matching, which could shorten the healing time and enable patients to recover and exercise early. At the same time, we ground the growth axis of the bone defect into an oval in the same direction as the length and diameter of the bone according to precise match orientation, so as to better adapt to the biomechanics of the reconstructed bone.
In spite of allograft has been used as a biological bone preservation technique, there still remains several potential problems such as allograft fracture [20, 21, 24], infection [20–23, 25], delayed union and nonunion [5, 25–27]. In our case, none of the patients had these complications above for the time being. Although it is generally believed that the incidence of allograft is high, we found that the incidence of postoperative infection, fracture and delayed bone union is low in our study. One possible reason is that our technique can shorten the operation time, reduce intraoperative bleeding, reduce the incidence of short-term and long-term postoperative complications. As a result, patients were able to carry out simple rehabilitation training early. Passive functional exercise was necessary and could be carried out as soon as possible after operation, which can prevent muscle atrophy and anchylosis. However, the active movement of the affected limb should be restricted in the short term to avoid fractures caused by excessive load or rotational violence.
There are several limitations although our method of operation was very effective. First of all, our study was a retrospective study, which lacked a direct comparison with other treatment techniques, especially plate and screw fixation. At the same time, we currently had an average follow-up period of 11.3 ± 3.0 months, which was still shorter than that reported in relevant studies. And the population was relatively small as a result of the rarity of the procedure, so late follow-up is needed to evaluate whether there would be new long-term complications. In addition, our research can be combined with the prevalent 3D printing technology, so that we could better plan the bone defect and its reconstruction, including the development of personalized guide plate and vascularized stent [28–30]. But it also exists some shortcomings, such as high cost, long learning cycle. And the long-term outcomes and the remedy after failure of 3D printing technology have not been certified clearly [31].