Letter: Reconsidering Bone Morphogenetic Protein in the Cervical Spine: Selective Use for Managing Type II Odontoid Fractures in the Elderly

To the Editor:

Type II odontoid fractures are common and result in significant morbidity and mortality in the elderly. The optimal treatment strategy for these fractures is controversial. Both nonoperative immobilization and operative fixation can be effective, but each is associated with unique complications. Failure of fracture healing is common and is likely due in part to poor bone quality among the elderly. Recombinant human bone morphogenetic protein (rhBMP) is an osteogenic peptide that has been used to facilitate fracture healing. However, reports of significant adverse events associated with the use of rhBMP in the cervical spine led to a Food and Drug Administration (FDA) warning in 2008. Given the limitations of currently available treatment options and the significant morbidity of type II odontoid fractures in the elderly, we propose considering the use of local, low-dose rhBMP as an adjunctive treatment for these fractures. To assess the technical feasibility of targeted rhBMP administration to the odontoid, we created a type II odontoid fracture in a cadaveric cervical spine specimen. An rhBMP-soaked collagen sponge homogenate was then delivered via an 18G spinal needle through a transoral approach to the fracture site under fluoroscopic guidance. The rhBMP solution was delivered without difficulty. The bony defect was noted to be filled completely without spillover of the injection material. While rigorous studies are necessary to assess both the clinical efficacy and potential complications associated with this therapy, we believe local delivery of rhBMP to the odontoid is feasible and may constitute a promising adjunctive treatment option for type II odontoid fractures in the elderly.

Odontoid fractures of the second cervical vertebra occur following hyperflexion of the spine and are classified as type I, II, or III.¹ Type I and III fractures are considered stable and are generally managed nonoperatively, healing well with immobilization. Type II fractures through the odontoid base are the most common, are unstable, and have a high rate of nonunion.¹ Odontoid fractures are overrepresented among the elderly.^2,3 Most are caused by low-energy trauma, such as fall from standing height.⁴ This population's increased susceptibility to odontoid fractures is likely due to degenerative changes in the cervical spine.⁵ The majority of these fractures are unstable type II fractures^6-8 for which high nonunion rates are further exacerbated by poor bone quality.

Type II fractures can be managed nonoperatively or operatively.⁹ Current evidence has not identified one strategy as superior to the rest, particularly in the elderly.^10-12 Nonoperative immobilization employs a rigid cervical spine collar or the halo vest^13,14 and is variably successful, resulting in fracture healing in 0%¹⁵ to nearly 100% of patients.^16-20 These less invasive options may be selected to minimize operative morbidity;²¹ however, immobilization can worsen quality of life²² and is not without complications, which include treatment failure²³, and for the halo vest, pin infection, pressure sores, nerve injury, and increased mortality.^18,24-27 Most studies demonstrate that operative management produces higher rates of bony fusion,⁹ and some additionally suggest that it improves functional outcomes²⁸ and overall survival.^15,29-31 However, these benefits may be diminished in older patients³² due to increased rates of complications and comorbidities.^28,33 It is this patient population that requires innovative treatment options.

A significant barrier to fracture healing is poor bone quality,^34-37 which worsens with age and thus becomes important in considering fracture management in the elderly. One strategy that may hold promise is the use of bone morphogenetic protein (BMP), an osteogenic material that induces bone formation and is important for achieving maximal bone mass in healthy individuals.^38,39 Evidence suggests that in osteoporotic bone, BMP efficacy is decreased^40-42 while supplementation of BMP augments bone healing and strength,^43-49 leading osteoporosis to be described by some as a “bone morphogenetic disorder.”⁵⁰ As such, increasing BMP back to physiological levels may be critical for facilitating healing when native bone quality is poor.⁵¹

Given the osteogenic properties of BMP, rhBMP peptides have been exploited to facilitate bony fusion in clinical populations. Based on small clinical trials,^52,53 the FDA approved rhBMP-2 in 2002 as a bone graft substitute for single-level anterior lumbar interbody fusion (ALIF). Exemption approval was then granted for rhBMP-7 for posterolateral fusion (PLF) revisions. Current FDA-approved indications for rhBMP in the spine remain limited to ALIFs and PLFs. However, the clinical use of rhBMP in the spine has expanded,⁵⁴ and up to 85% of the use of rhBMP may actually be off-label.⁵⁵ Initial studies of rhBMP-2 in cervical fusions demonstrated good outcomes,^56,57 but trial data revealed significant complications in 10% to 50% of cases.⁵⁸ Hoarseness, dysphagia, and, rarely, risk of respiratory compromise^59-61 were related to swelling,⁶² which was attributed to a local inflammatory response and seemed to be dose-related. The recognition of these adverse events led to a 2008 FDA safety alert for the use of rhBMP in cervical fusions.⁶³

Current strategies for managing type II odontoid fractures in the elderly remain suboptimal. These patients’ poorer health precludes surgical intervention under general anesthesia, and their poorer bone quality compromises fracture healing. The use of agents to stimulate local osteogenesis has the potential to improve outcomes. We therefore propose transoral delivery of rhBMP to the fracture site to facilitate fusion while avoiding operative management. Though the use of rhBMP in cervical spine fusions has been associated with significant adverse events, these complications seem to be dose-related and could plausibly be avoided by a focal application of a smaller peptide dose. Thus, we believe it prudent to consider the judicious use of rhBMP to improve management of type II odontoid fractures in the elderly. Specifically, we propose that local application of rhBMP directly to the fracture site would be an effective adjunct for nonoperative management.

To examine the feasibility of local delivery, we performed transoral administration of an rhBMP solution to the odontoid of a human cadaveric cervical spine specimen. Institutional Review Board approval or patient consent was not required. While maintaining the integrity of the spinal ligaments, joints, and discs, a type II odontoid fracture was created using an osteotome and mallet and was confirmed with fluoroscopy (Figure 1). One mg of rhBMP-2 (Medtronic, Dublin, Ireland) was combined with 10 mL of radiopaque gadodiamide medium (Amersham Health, Princeton, New Jersey) to produce a 0.1 mg/mL rhBMP solution. A collagen sponge was saturated with the rhBMP solution and homogenized for injection. Under fluoroscopic guidance, an 18G spinal needle was used to enter the fracture anteriorly to simulate a transoral approach (Figure 2A). The rhBMP solution was then slowly injected into the fracture (Figure 2B) until the defect was completely filled. Fluoroscopic imaging was used to confirm placement and volume (Figure 2C). The needle was easily guided into the fracture site, and the rhBMP solution was delivered without difficulty. No spillover of the injection material was noted.

FIGURE 1.

AP (anterior-posterior) fluoroscopic image demonstrating a type II fracture of the odontoid in a cadaveric cervical spine specimen.

FIGURE 2.

Fluoroscopic images demonstrating lateral view of the odontoid during administration of rhBMP solution to the fracture site. A, The tip of an 18G spinal needle was used to enter the fracture site. B, A homogenized collagen sponge soaked in rhBMP was then slowly injected into the defect until, C, complete filling of the defect was noted. Care was taken to avoid spilling of the rhBMP solution outside of the fracture site.

Given the high morbidity and mortality of type II odontoid fractures in the elderly in concert with the lack of adequate treatment options, we propose considering the judicious use of rhBMP as an adjunct to nonoperative management of these fractures. We demonstrate the feasibility of local administration of this osteogenic peptide via direct injection into the fracture site. Though concerns have been raised regarding the complications associated with the use of rhBMP in the high cervical spine, the focal application of low-dose rhBMP to treat type II odontoid fractures constitutes an alternative approach to facilitate fracture healing and merits further study. While assessing the efficacy of this treatment in a large clinical population is required before definitive conclusions can be drawn, this approach is promising and may have a significant impact on improving the morbidity associated with type II odontoid fractures in the elderly.

Disclosure

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.