Keywordsbone graft, autograft, allograft, BMP, complication
Complications from bone grafting in the spine include poor bony fusion, pain at harvest sites, in addition to infection.
Grafting in spine surgery includes autograft, allograft, bone morphogenetic protein, ceramics, and novel materials. Each graft type possesses benefits and drawbacks.
New materials are intended to promote bony fusion while reducing further the possibility of complications.
Neurosurgeons and orthopedic spine surgeons often utilize various forms of bone grafting to achieve successful spinal fusion. The process of achieving fusion with autograft, allograft, ceramic materials, or bone morphogenetic proteins (BMP) is not completely benign for the patient and presents the possibility for a variety of complications. This chapter details the complications with grafting for spinal fusion with various available materials, ranging from donor site pain, to infection, and to pseudarthrosis.
Spinal operations involving fusion are increasingly performed in the United States. To facilitate fusion, grafting has become requisite. Numerous factors influence whether a bone graft will achieve successful fusion or not. They include quality of preparation of the recipient site, the patient’s history of radiation, the biomechanical stability of the graft complex, and the presence or absence of graft loading. Additionally, systemic patient factors such as general nutrition, history of smoking, osteoporosis, and presence of infection also affect the success of the bone graft and fusion. If a patient has an unsuccessful fusion or pseudarthrosis, there is a greater likelihood of poor clinical outcome. Poor clinical outcome often leads to increased financial expenditure on the medical care of the patient, thus placing further burden on the health care system.
Bone grafting is associated with three requirements for successful spinal fusion: osteogenesis, osteoinduction, and osteoconduction. Osteogenesis refers to the provision of cells that can directly form bone. Osteoinduction describes the ability to induce differentiation of progenitor stem cells into osteoblasts for bone formation. Finally, osteoconduction is the provision of a sufficient scaffold to support bone formation. Autograft remains the most successful form of bone graft for spinal fusion since it involves all three requirements. Unfortunately, utilizing autograft is associated with its own complications.
Autograft is a very effective means for achieving successful fusion in anterior cervical discectomy and fusions. In 1952, Bailey and Badgley first performed an anterior cervical fusion, and in 1960 described a technique where a patient with instability due to neoplasm required onlay strut grafts harvested from the patient, thereby demonstrating the benefit of autograft. Autografts have a reported fusion rate of 83% to 99%. The fusion rate decreases as the number of levels increases. After discectomy, an interbody potentially filled with structural bone can be utilized to avoid structural changes and maintain disc height.
Autograft, usually in the form of corticocancellous bone, is harvested locally ( Fig. 48.1 ) or with iliac crest bone graft (ICBG). This form of graft has been the most common and successful graft in spinal fusion, often referenced as the “gold standard.” In spine surgery, the bone harvested from a local site such as lamina or spinous process can be used later in the surgery for fusion. This cancellous autograft bone lacks possibility of disease transmission and has no risk of immunogenicity. Additionally, autograft is easily revascularized after removal from the donor site.
ICBG, though effective at ensuring successful fusion, possesses inherent procedural potential for complications. Complicating factors include inadequate bone quantity, creation of another surgical incision, increased operative duration for harvesting, possibility of increased blood loss, and the potential need for transfusion. Harvesting bone from the iliac crest can lead to pelvic fractures, vascular injuries, deep infection, and difficulty with ambulation due to pain. Minor complications include superficial infection and variable chronic donor site pain. Donor site pain can present as hyperesthesia or diminished sensitivity in the territory of the lateral femoral cutaneous nerve. The incision should be made to avoid injury to the superior cluneal nerves and away from the sciatic notch to avoid injury to superior gluteal artery and nerve, the sciatic nerve, and the ureter. Robertson and Wray performed a prospective analysis examining bone graft site donor morbidity for ICBG. Significant complications were low at 1.9% with donor site morbidity in 35% of patients. Patients reported donor site pain lasting up to 6 months that would significantly decrease by 12 months postoperatively.
One case report describes a patient with non-Hodgkin’s lymphoma on chemotherapy who experienced infection from a previous ICBG donor site 22 years prior for spinal arthrodesis. She was found to be infected with methicillin-resistant Staphylococcus aureus (MRSA) and underwent surgical debridement. This case underscores the possibility of complications at the donor site, even decades after the initial surgery. Alternative sites of autograft also include the rib, fibula, and vertebral body when ICBG is deemed unsuitable. The potential for complications associated with ICBG autograft has spurred spine surgeons to consider other options for bone grafting to achieve fusion.
Serving as the most common substitute for autograft, cadaver allograft ( Fig. 48.2 ), provides the benefits of fusion and spares the patient of a second surgical site. Three types of allograft exist in the form of fresh frozen allograft, freeze-dried allograft, and demineralized bone matrix (DBM). Unfortunately, allografts possess minimal osteogenicity because cells do not survive the processing. Allograft advantages include immediate availability, storage ability, reduced blood loss, and decreased duration of surgery. Allograft is a reliable substitute in that the biomechanical stability is higher in allograft when compared with autograft.
Fresh frozen allograft is the simplest of the allograft types. Typically, it is treated with antibiotic solution and stored at frozen temperature after harvesting. These allografts possess the greatest strength. Freeze-dried allograft is also treated with antibiotic and frozen, but the water is subsequently removed. This process can potentially decrease the mechanical stability and resistance to fracture.
There are concerns that preservation via freezing or freeze-drying may negatively affect the osteoinductive and osteoconductive properties of the allograft, while also reducing its immunogenic properties. Utilizing allograft also creates concern for donor to recipient disease transmission. This necessitates rigorous donor screening and sterilization to prevent bacterial and viral transmission. Though preservation effectively reduces immunogenicity, further sterilization may be needed with high-dose gamma irradiation or ethylene oxide gas. Unfortunately, these methods may further reduce osteoinductivity. Animal studies have shown that despite these sterilization procedures, viruses can survive, as in the case of feline leukemia virus, a retrovirus similar to the human immunodeficiency virus (HIV). Additionally, there are long-term financial benefits associated with allogeneic bone graft, likely due to the increased costs of harvest site of morbidity.
The third form of allograft is DBM. DBM has been acid treated for removal of the mineralized bone while keeping the organic scaffolding and growth factors. DBM allograft has osteoconductive and variable osteoinductive properties without the structural strength of allograft. It has been shown that DBM products can act as a bone graft extender for posterior fusion in adults and adolescents with scoliosis. One caveat noted in DBM animal studies is that it can be nephrotoxic. DBM can also be used as a bone substitute for anterior lumbar body interbody fusion. Unfortunately, one study found that when allograft is combined with DBM, there is a higher rate of pseudarthrosis when compared with that of autograft.
The properties of the allograft are often variable due to lack of standardization among manufacturers and the heterogeneity of the donor population. Human allograft has historically been available as mineralized or demineralized. Mineralized allograft is considered nonosteogenic, osteoinductive, and osteoconductive. DBM is osteoconductive and somewhat osteoinductive.
In addition to the risk of disease transmission, allografts can be associated with complications regarding malplacement and pseudarthrosis. Animal studies have shown that when compared with autograft in anterior and posterior spinal fusions, allograft is associated with a slower fusion rate, greater graft resorption, and increased infection rate. Stand-alone impacted femoral ring allografts have been associated with pseudarthrosis and graft extrusion. Historically, use of autograft was more effective in achieving fusion when compared with allograft. Godzik et al. describe the case where they performed occipito-cervical fusion with allograft. They found the fusion in their series to be comparable to autograft when providing compressive forces. To increase fusion rates, threaded allograft bone dowels were utilized in combination with recombinant human bone morphogenetic protein-2 (rhBMP-2). These threaded dowels were better able to resist expulsion and stabilize the implant.
In 2000, a metaanalysis was performed on four studies comparing allograft and autograft in anterior cervical fusions. There was a significantly higher rate of bony union with a lower incidence of collapse with autograft versus allograft for one- and two-level fusions based on radiographic findings. The authors did not report whether autograft was clinically superior to allograft. Ultimately, the decision to use autograft or allograft should be made by the surgeon to ensure the best clinical outcome. Fig. 48.3 shows intraoperative combination of allograft and autograft to achieve fusion in the posterior cervical spine.