Background
Bone defects, which are commonly induced by bone infection, removal of bone tumors, and severe trauma, account for 15% of fractures of limbs and cause a decline in the quality of life of patients. Bone defects longer than 2 cm will not heal by themselves and need to be repaired. A free bone graft is the “gold standard” for the repair of bone defects [
1‐
4]. Traditionally, a free cancellous bone graft was considered suitable only for segmental bone defects less than 4–6 cm and unsuitable for defects more than 4–6 cm due to frequent bone absorption, which resulted in a high rate of nonunion [
1‐
4]. Although multiple methods, such as the Ilizarov technique, vascularized bone graft, blood donor bone grafting, and large allograft bone grafts, are emerging that appear to be suitable for segmental long bone defects greater than 4–6 cm, all these methods have various deficiencies [
5,
6]. For example, the Ilizarov technique has shortcomings of reduced self-healing rates, long fixation times, and a high incidence of complications, such as joint stiffness, infection, and loosening of the nail path. Vascularized bone grafting requires microsurgical techniques, which result in difficult to be popularized. Besides, the surgical trauma is too large. The shortcomings of large allogeneic bone grafts include slow healing time, high rates of refractures and infections, and foreign body reactions, all of which restrict its widespread application. Tissue engineering techniques are an option but are only in the experimental research stage [
5,
6].
In the last two decades, many improved free bone graft methods have been proposed to treat segmental long bone defects. For example, in 2000, Masquelet [
5] reported a wrapped cancellous bone graft using the induced membrane in the treatment of 35 cases with long segmental bone defects of 4–25 cm. All the patients obtained clinical healing after an average of 4 months. In the same year, Cobb et al. [
6] reported a wrapped bone graft using titanium mesh in the treatment of large segmental bone defects. Later, Tian Wen et al. [
7] adopted a line mesh and line suture or line-binding cancellous bone graft to treat segmental long bone defects and obtained satisfactory effects. All the aforementioned methods employ cancellous bone wrapped in a mechanical device, thereby overcoming the shortcomings of traditional cancellous bone grafts [
7]. These improved free bone graft methods change the traditional view of the unsuitability of free bone graft as a treatment for large segmental one defects [
5,
6,
8‐
10]. Thus, modified free bone graft is becoming an effective method for the treatment of segmental bone defects. We named this method the “wrapped bone grafting technique”. Recently, the development of a new wrapped device-absorbed mesh with holes and progress in the harvesting of large volumes of nonstructural autogenous bone have advanced the development of wrapped bone grafting techniques [
8,
9].
Few literatures reported the influence factors of wrapped bone grafting and the comparison of therapeutic effects of different wrapped bone grafting methods. From 2007, we started to adopt various methods of wrapped cancellous bone grafting to treat segmental bone defects. In this study, we investigate the principle, therapeutic effect and influencing factors of wrapped cancellous bone graft for bone defects.
Discussion
The present study demonstrated that, as a modified free bone graft method, the wrapped cancellous bone grafting avoids or obviously decreases loosening of grafted bone and bone absorption, promotes the rapid healing and high healing rates and is suitable for both small and large segmental bone defects.
Cancellous bone has characteristics of bone induction, and osteogenesis and is regarded as the best bone graft material. The main reason of traditional free cancellous bone graft was considered unsuitable for large segmental bone defects is due to a high rate of failure caused by loosening and absorption resorption of the grafted bone as lack of wrapped device, the grafted bone is vulnerable to vibration from the surrounding tendons or muscles and body vibration, resulting in loosening or sliding, even bone absorption [
6]. For this reason, some authors suggest the use of a plaster cast or brace for a period of time postoperatively. However, immobilization cannot prevent bone absorption. Furthermore, it hampers functional recovery and bone healing.
As compared with the traditional method, the wrapped cancellous bone grafting improves both the volume and harvesting method of grafted cancellous bone [
6], which overcomes the shortcomings of traditional free bone graft method. First, the technique utilizes mechanical packing device, which fixes the bone graft material and avoids undesirable stimulation of surrounding tendons or muscles and vibrations of the bone graft material [
3,
13]. Second, the wrapped bone graft technique requires abundant autologous cancellous bone [
6]. In the technique using titanium mesh, the amount of grafted bone is 1.5–2.0 times that of the bone defect that has to be filled. The amount of grafted bone is 1.5 times that of the bone defect for line mesh, line suturing or binding, and induced membranes. If the amount of autogenous cancellous bone is insufficient (e.g., in cases of large bone defects), cancellous bone can be mixed with cortical bone or artificial bone, but no more than one-quarter of cortical/artificial bone can be used [
14,
15].
Harvesting methods of autologous cancellous bone have also made great progress. Studies have also shown that more cancellous bone can be obtained from the posterior iliac crest than from the anterior iliac crest, therefore the former is the first choice [
7,
16]. The proximal tibia is also considered a rich source of cancellous bone, with few complications [
5,
6,
14,
16]. Based on our experiences, autologous cancellous bone can be harvested from multiple sites to meet the need for large amounts of cancellous bone graft. For example, the amount of autogenous cancellous bone from the bilateral iliac crest mixed with attached cortical bone can fill a 1.5 measuring cup, which can meet the needs of a 6.0-cm-long tibial shaft defect. Autogenous cancellous bone from the bilateral proximal tibia can fill one cup. Using autogenous cancellous bone from the aforementioned four sites, sufficient graft material to meet the needs of a 9.0–10.0 cm-long tibial shaft defect can be obtained. By using a reamer-irrigator-aspirator (RIA), a large amount of granular autologous bone graft material can be collected from the femur or tibial medullary cavity [
8,
9]. The amount of granular autologous bone grafts collected from a unilateral femur using an RIA can be up to 40–90 cm
3 (average of 67 cm
3), which is more than the average volume of 26 cm
3 in the anterior iliac crest and 36 cm
3 in the posterior iliac crest. Bone graft material from a bilateral femur obtained by an RIA can meet the needs of a tibial bone defect longer than 10.0 cm [
8]. Furthermore, the bone graft material collected by an RIA contains a large amount of cancellous bone and cortical bone granules. These are rich in osteoblasts, marrow stromal stem cells, fibroblast growth factors, platelet-derived growth factors, insulin-like growth factors, bone morphogenetic protein-2, and transforming growth factor beta 1 and have the same osteogenic effect as ilium [
8,
9]. As a result, RIA provides an effective method of bone harvesting for large segmental bone defects treated using the wrapped bone graft technique.
An additional advantage of the wrapped bone graft technique is that patients can take early postoperative rehabilitation exercise due to the effect of wrapping on the bone graft material and reliable internal or external fixation [
8,
17‐
21]. Early rehabilitation is helpful to stimulate bone healing and functional recovery of the joint. In addition, the wrapped device plays an important role in vascularization and osteogenesis of the grafted bone because it has holes and good biological properties. The holes provide a pathway for new blood vessels and osteogenic factors. Therefore, the grafted bone is nourished by blood vessels and osteogenic factors from the surrounding area and forms new bone. In terms of the different wrapping materials, titanium mesh has good biocompatibility and some degree of a bone inductive effect [
6]. The induced membrane has good osteogenic properties and a blood supply, especially in the early stage when it has a rich vascular system and can secrete osteogenic growth factors (e.g., transforming growth factor beta 1 and bone morphogenetic protein-2) and angiogenic growth factors (e.g., CD31+ endothelial cells, vascular endothelial growth factors, and osteogenic precursor cells) [
16,
17,
19,
22]. The availability of RIAs overcomes previous difficulties in harvesting of rich autologous bone and makes it possible to obtain a large amount of autologous cancellous bone. Recently, an absorbable mesh for wrapped cancellous bone grafts to treat bone defects has become available [
23]. The use of this absorbable mesh can shorten the operation time and further promote the popularity of wrapped bone graft technique.
Cobos et al. [
6] reported two cases of 8.5–9.5 cm defects of tibia bone. Attias et al. [
13] reported three cases of tibial bone defects, where the average defect length was 12.2 cm. Ostermann [
20] reported one case of a segmental tibial bone defect. Attias et al. [
21] reported one case of an 8-cm humerus bone defect. In these cases [
6,
13,
21,
22], the defects were treated with titanium mesh wrapped cancellous bone graft, and bone healing was achieved 1 year postsurgery. Karger et al. [
14] reported 84 patients with long bone defects, with the longest being 23 cm, that were treated with an induced membrane wrapped autologous cancellous bone graft. They reported a healing rate was 90%. McCall et al. [
9] reported 21 cases of bone defects of the lower limb (an average of 6.6 cm) treated with an induced membrane wrapped bone graft by RIA. In their study, 20 cases obtained bone healing, and one case was lost to follow-up. Apard et al. [
17] reported 12 patients with tibial bone defects (average length of 8.7 cm) that were treated with an induced membrane wrapped bone graft and achieved a bone healing rate of 91.6%. Whately et al. [
8] reported one case of a 10-cm-long tibial defect treated using an absorbable polymer mesh wrapped bone graft by RIA plus intramedullary nail fixation. They reported clinical healing of bone 6 months postoperatively. Liu Yao-xi [
7] reported congenital tibial pseudarthrosis in 12 pediatric patients with bone defects (4–11 cm) treated with a line suturing cortical bone wrapped cancellous bone graft and achieved 100% healing. In the present study, the average bone defect length was 5.9 cm, the longest tibia defect was 9 cm, and the longest humerus defect was 7 cm. The average healing time in the patients treated with the four different materials using the cancellous bone graft method was 6.1 months, and the healing rate was 98%. The incidence of complications in the grafted area was 11.8, and 21.6% in the harvesting site, which are not higher than other methods [
1,
5,
6,
9]. Our data indicate that all four materials are effective for the treatment of segmental bone defects. They were associated with accelerated healing, high rates of healing, and few complications, irrespective of the length of the bone defect. Notably, the titanium mesh had the shortest healing time (average of 5.44 months), pointing to its superiority over the other wrapped materials. In theory, the healing effect of the induced membrane wrapped bone graft should have been the best, as the induced membrane has a mechanical package and fixation effect and osteogenic induction. As reported in the literature [
3,
5,
22], the shortest healing time using the induced membrane technique for the treatment of bone defects was 3 to 4 months, and the longest healing time was 6–10 months after grafting. In the present study, 89.5% of the patients who were treated with the induced membrane wrapped bone graft were infected defects. In all the patients, the infection was controlled at least more than 3 months before bone grafting. The time from bone cement filling to bone grafting averaged 5.1 months in our study. Previous studies showed that the best osteogenic activity of an induced membrane after bone cement filling occurred 4–6 wk. after filling and that the osteogenic activity declined gradually from then on [
16,
17,
19,
22]. In the present study, in the induced membrane group, the osteogenic activity was reduced and the blood supply was poor in 5 months after bone cement filling, and the induced membrane exhibited only mechanical and fixation effects. In addition, in this study, one patient had nonunion and required a secondary bone graft, in which the clinical healing time was 15 months. Thus, the induced membrane group was characterized by a relatively long healing time and high rate of complications.
As stability is the main factor affecting bone healing, reliable internal fixation must be chosen. According to our experience, intramedullary nails should be the preferred method of internal fixation for long bone shaft defects. This is because they have good biomechanical stability and save bone graft materials by occupying the position of the medullary cavity. If one side of the medullary cavity is large, the blocking nail technique should be used to prevent the end instability (Fig.
3c). Plate fixation is preferred for epiphyseal fixation. If the bone defect is close to the joint surface, internal fixation is difficult. In such cases, external fixation is a good choice [
16].
Li Lin et al. [
24] reported four cases of large segment bone defects of the tibia treated with cortical bone graft wrapped by an induced membrane. The nonunion rate was 50%, and bone healing of nonunion was achieved after the cortical bone graft was replaced with cancellous bone graft. In the present study, one case of bone nonunion occurred because the volume of bone cement filling at the ends of bone defect was too small and the induced membrane volume that formed was also small, which resulted in less bone graft and insufficient bone connection. The reasons for slower healing of the middle and lower segments of the tibia are a weak blood supply and coarser bone. Therefore, the quality of bone grafting, fracture stability, wrapped material properties, and peripheral blood supply are the main factors influencing the efficacy of bone healing by cancellous bone grafting.
The selection of different wrapped bone graft methods depends on the specific situation of the bone defects. For bone defects in location of non weight-load such as upper limbs, line mesh is the preferred method. If there are multiple free cortical bone blocks, the preferred method is line suturing or binding. For large segmental defects of long lower limbs, titanium mesh is preferred because of its reliable fixation and rich bone graft material can be filled [
10]. For open and infected bone defects, induced membrane technique is indicated [
16]. Irrespective of which wrapped method is chosen, fixation using an intramedullary nail is the first choice. For large segmental bone defects, RIA is the preferred way to harvest bone. If RIA is not available, the preferred site to harvest bone is the posterior iliac. If the quantity harvested from the posterior iliac cannot meet the need, then the proximal tibia can be selected. If the quantity of harvested aotogenous bone is insufficient and cannot meet the need, then allogeneic or artificial bone can be addited.
There are some deficiencies in performing wrapped cancellous bone grafting. First, it requires good conditions of soft tissue [
5,
16]. In bone defects accompanied by soft tissue defects, the wound repair with skin flap must have no tension, the skin scar near the bone defect should be removed and replaced with a skin flap. In our study, the incision in one case was disrupted because of high tension of the skin flap and required another operation Therefore, in cases of poor skin condition, the wrapped cancellous bone graft technique is not suitable. A more suitable choice in such cases is Ilizarov technique. Second, the infection must be controlled and keep a normal state in erythrocyte sedimentation rate and C-reactive protein level for more than 3 months before the bone graft [
6,
7]. Third, complications of harvesting large amount of graft materials from the iliac crest remain an unsolved problem. A previous study reported that the incidence of complications (pain and numbness) in the anterior iliac crest varied from 6 to 36% [
9]. In the present study, the incidence was 21.6%. Research also reported that average intraoperative blood loss when using RIA to harvest bone graft materials was 674 ml and that iatrogenic fractures may occur when harvesting a large amount of bone graft material [
25]. In addition, another study reported that although callus formation was fast, corticalization of the callus was very slow and usually took 2–3 y [
14]. Thus, stress fractures can occur before corticalization [
14]. Finally, the removal of titanium mesh and plates after a bone graft is difficult [
13].
In this paper, only four kinds of wrapped bone grafting methods were used in the earlier stage. This study did not include fascia, including femoral fascia, and absorbable mesh that have begun to be used recently in clinical applications. This paper was a retrospective clinical study, and only a few patients were treated with each of the wrapped materials. Thus, there is sampling error. Moreover, there were large baseline differences in the preoperative general data. Furthermore, there was bias, which was not suitable for the statistical analysis to compare the differences in the various wrapped methods. Therefore, more clinical data from multicenter studies, large samples, and experimental research are needed compare differences in bone healing and functional recovery using the various wrapped graft methods and confirm the therapeutic effect of these methods.