Review Article

Odontoid Fractures: Update on Management

Abstract
Wellington K. Hsu, MD Paul A. Anderson, MD

From the Department of Orthopaedic Surgery and the Department of Neurological Surgery, Feinberg School of Medicine, Northwestern University, Chicago, IL (Dr. Hsu), and the Department of Orthopaedic Surgery and Rehabilitation, University of Wisconsin Hospitals and Clinics, Madison, WI (Dr. Anderson). Dr. Hsu or an immediate family member is a member of a speakers bureau or has made paid presentations on behalf of Stryker and has received research or institutional support from Pioneer Surgical, Medtronic, and Baxter. Dr. Anderson or an immediate family member has received royalties from Pioneer Surgical and Stryker; serves as a paid consultant to or is an employee of Aesculap/B. Braun, Expanding Orthopedics, Medtronic Sofamor Danek, Spartec, and Titan Surgical; and has stock or stock options held in Titan Surgical, Pioneer Surgical, Spartec, and Titan. J Am Acad Orthop Surg 2010;18: 383-394 Copyright 2010 by the American Academy of Orthopaedic Surgeons.

Recognition of the incidence of odontoid fractures as well as the associated morbidity and unexpectedly high mortality rates has prompted signicant changes in the management of these fractures in the past decade. Nonsurgical management of type II odontoid fracture has historically been associated with a high nonunion rate. Thus, new classication systems have been devised to identify patients who might benet from early surgical treatment. The decision-making process is particularly difficult when treating elderly patients. Increased familiarity with anterior and posterior surgical techniques has led to more aggressive treatment of odontoid fracture, with the intent of hastening functional rehabilitation. However, these clinical decisions have been associated with a signicant rate of complications. The treatment algorithm for odontoid fractures continues to evolve based on the improved understanding of, and evidence-based literature on, anterior screw xation, posterior spinal fusion, and halo-vest immobilization.

dontoid fractures make up 9% to 15% of cervical spine fractures in the adult population.1,2 These injuries are associated with a high rate of morbidity and mortality regardless of treatment method. Several factors account for this, including nonunion, comorbidities in elderly patients, and potentially fatal complications from the use of restrictive bracing, such as a halo vest. Treatment guidelines based on the best available medical evidence are required to optimize clinical outcomes. Definitive management of odontoid fractures remains controversial. These injuries typically occur in a bimodal distribution, and treatment priorities should be individualized based on patient age, fracture pattern, associated neurologic deficits, and overall medical condition. An ar-

ray of nonsurgical and surgical recommendations has been made.3,4 Certain patient characteristics and risk factors for nonunion are important to consider in the management of odontoid fractures. Long-term studies analyzing the clinical outcomes of nonsurgical therapy and surgical fixation from anterior and posterior approaches have advanced our understanding of treatment outcomes.

Classication
In 1974, Anderson and DAlonzo5 proposed a classification system for odontoid fractures that remains widely used (Figure 1). Distinctions were made between fractures that occur at the very tip (type I), the base of the dens (type II), and the body of the axis (type III). This system has

July 2010, Vol 18, No 7

383

Odontoid Fractures: Update on Management

Figure 1

Figure 2

The Grauer modication of the odontoid fracture classication further delineated the difference between type II and type III fractures. Those without involvement of the C2 superior articular facet were classied as type II (A), and those with facet involvement were classied as type III (B). (Redrawn with permission from Grauer JN, Sha B, Hilibrand AS, et al: Proposal of a modied, treatment-oriented classication of odontoid fractures. Spine J 2005;5[2]:123-129.)

The Anderson and DAlonzo classication of odontoid fracture. Type I fractures involve avulsion near the tip of the dens. Type II fractures occur at the base of the odontoid process. Type III fracture lines extend into the body of the axis. (Redrawn with permission from Anderson LD, DAlonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg Am 1974;56:1663-1674.)

proved to be useful in defining the natural history of odontoid fracture and identifying appropriate treatment options. One significant contribution of this system is the distinction between type II and type III fractures, which led to a different treatment approach for each. Much has been learned of the natural history of type II fracture, which has a demonstrated nonunion rate of 26% to 85%.6-8 Compared with a type III injury, fracture patterns across the odontoid base involve considerably less trabecular bone and are under distractive forces from the apical ligament, which can contribute to an unfavorable healing potential.9 The dens is surrounded by

synovial cavities, which results in diminished periosteal blood supply to type II fractures. Consequently, it has been suggested that type II injuries must be managed surgically to avoid late complications.4 Conversely, nonsurgical bracing has remained the mainstay of management of stable type III odontoid fractures, which have demonstrated high healing rates.6 The Anderson and DAlonzo classification has limitations, however. Barker et al10 reported only fair interrater and intrarater reliability between spine surgeons and neuroradiologists. With regard to the evaluation of plain radiographs, the kappa coefficients for interobserver and intraobserver agreement were reported to be 0.30 and 0.25, respectively. Although agreement was significantly improved with the use of reformatted CT (interobserver, 0.46; intraobserver, 0.56), the authors concluded that there is substantial variability in the use of the Anderson and DAlonzo classification even with the most experienced practitioners.

Grauer et al11 proposed a modification to the classification system to address some of these limitations. To aid in distinguishing type II from type III fractures in this treatmentoriented classification system, the authors narrowed the definition of a type II fracture to one that does not involve the C2 superior articular facets (Figure 2). By this definition, a type II fracture could still extend into the body of the axis; however, it would presumably still be under the same unfavorable forces typical of other type II fractures. In addition, three subtypes were added to further stratify the type II fracture pattern (Figure 3). Type IIA odontoid fractures have a transverse pattern and demonstrate <1 mm of displacement. Type IIB fractures have an oblique pattern extending from the anterosuperior to the posteroinferior portion of the dens. Type IIIC fractures begin anteroinferior and extend posterosuperior; these may be associated with significant anterior comminution. Using the associated fracture subclassification, the authors suggested treating type IIA fractures with exter-

384

Journal of the American Academy of Orthopaedic Surgeons

Wellington K. Hsu, MD, and Paul A. Anderson, MD

Figure 3

Figure 4

The Grauer subclassication of type II odontoid fractures. Type IIA injuries demonstrate a transverse fracture pattern and displacement of <1 mm. Type IIB injuries have an oblique fracture pattern extending from the anterosuperior to the posteroinferior portion of the dens. Type IIC fractures begin anteroinferior and extend posterosuperior. These may be associated with signicant anterior comminution. (Redrawn with permission from Grauer JN, Sha B, Hilibrand AS, et al: Proposal of a modied, treatment-oriented classication of odontoid fractures. Spine J 2005;5[2]:123-129.)

A, Standard measurement technique for odontoid fracture displacement. A tangent line is drawn along the anterior aspect of the dens and another is drawn along the anterior aspect of the body of C2. At the level of the fracture, a line is drawn connecting these two lines; this measurement indicates the amount of sagittal fracture displacement. B, Standard measurement technique for odontoid fracture angulation. The degree of fracture angulation is represented by the angle subtended by a tangent line along the posterior aspect of the odontoid and along the posterior aspect of the body of C2. The location of the fracture apex angulationanterior or posterioris used as the descriptor. (Reproduced with permission from Bono CM, Vaccaro AR, Fehlings M, et al: Measurement techniques for upper cervical spine injuries: Consensus statement of the Spine Trauma Study Group. Spine [Phila Pa 1976] 2007;32[5]:593-600.)

nal immobilization, type IIB fractures with anterior screw fixation, and type IIC fractures with posterior atlantoaxial spinal fusion.11

Imaging
The initial evaluation of a suspected cervical spine injury should include three plain radiographic views (ie, AP, cross-table lateral, open-mouth odontoid) or CT. In patients with a high index of suspicion for upper
July 2010, Vol 18, No 7

cervical spine fracture, CT provides efficient evaluation of the cervical spine in the absence of plain radiographs. Sagittal, coronal, and axial CT reconstructions allow for characterization of fracture patterns, displacement, and associated soft-tissue swelling. Measurement of fracture displacement and angulation is critical in the treatment algorithm for odontoid fractures (Figure 4). These measurement techniques have recently been standardized.12 The use of CT has also been associated with higher interrater and intrarater agreement in the classification of odontoid fractures.10 CT is also critical in assessing associated bony and ligamentous injuries of the cervical spine. Up to 34% of patients with odontoid fracture present with additional spinal injuries,13 including subaxial fracture, occipitoatlantal dissociation, transverse ligament rupture, and posterior

interspinous ligament disruption. An accurate way to evaluate the spatial relationship between the occiput, atlas, and axis is to measure the basion-dental interval and basion posterior axial line interval.12 This method was originally described by Harris et al,14 and it is also known as Harris rule of twelve. Intervals >12 mm are suggestive of occipitoatlantal instability. MRI is indicated in the presence of a neurologic deficit as well as with suspected ligamentous instability. MRI allows for the assessment of myelomalacia, congenital stenosis, canal stenosis, and ligamentous injury. To fully evaluate the upper cervical spine using MRI, sagittal and axial images should be obtained.

Management
Odontoid fractures present in several different patterns in patients with a

385

Odontoid Fractures: Update on Management

Table 1 Odontoid Fracture Classication, Subtypes, and Treatment Recommendations Type I II Subtype Nondisplaced (occiput-C1) Displaced (occiput-C1) Low risk High risk Elderly patient III
a

Displacement None <5 mm >5 mm

Transverse Ligament Intact Intact Intact Intact

Treatment Recommendation Cervical collara Occiput-C2 fusiona Cervical collar, halo vest,a odontoid screw, posterior C1-2 fusion Cervical collar, halo vest, odontoid screw,a posterior C1-2 fusiona Cervical collar,a halo vest, odontoid screw, posterior C1-2 fusiona Cervical collar,a cervicothoracic brace,a halo vest

Body fracture

Authors preferred treatment = with or without

wide array of characteristics; thus, treatment must be individualized. Patients should be informed that both nonsurgical and surgical treatment are associated with significant morbidity. General guidelines gleaned from evidence-based literature may be helpful in navigating this complicated treatment algorithm (Table 1). Several recent systematic reviews and meta-analyses of odontoid fractures have been published, which have aided in the treatment decisionmaking process.6,15

stable type I fracture is immobilization with a cervical collar, unless other spine fractures are present. Most spine surgeons recommend posterior occiput-C2 fusion for occipitoatlantal instability. Halo vests are not typically recommended for type I injuries because of the possible detrimental effects of introducing distraction to the fracture site. In an evidence-based review, Julien et al6 demonstrated healing in all type I fractures, regardless of treatment.

Type II Type I
Type I fractures are avulsion fractures involving the alar ligament, which is responsible for craniocervical stability. Thus, type I injuries may be associated with occipitoatlantal instability.16 Radiographic evaluation typically demonstrates a displaced fragment involving a small portion of the tip of the dens. These fractures are deemed to be stable when at least one alar ligament and the transverse atlantal ligament are intact. Soft-tissue swelling surrounding this injury can be a harbinger of current or previous subluxation.17 The recommended management of The management of type II odontoid fractures is controversial, with no consensus. Halothoracic vests have traditionally been used; however, the morbidity associated with these braces has been well-documented. Because of the reported high rates of pseudarthrosis with this fracture pattern, in the past decade, type II odontoid fractures have been managed surgically more often than not. However, retrospective studies have demonstrated high rates of complications resulting from these procedures, as well. The end points for successful treatment include preservation of neurologic function, a sta-

ble atlantoaxial interval, reduction of neck pain, and return to preinjury functional status. Mller et al8 reported the results of immobilization with a rigid cervical orthosis in 19 patients with type II odontoid fracture. Fracture gap was <2 mm, AP displacement was <5 mm, and angulation was <11; 74% of fractures achieved radiographic union. Delayed internal fixation was performed in two patients with >2 mm of motion on flexion/extension radiographs, without further sequelae. These authors concluded that in stable type II odontoid fractures, rigid cervical collar treatment can contribute to successful clinical outcomes.8 Other authors have confirmed the observations of Mller et al.8 Ekong et al18 demonstrated that risk factors for nonunion are present in patients aged >40 years who present with odontoid fracture involving posterior displacement >5 mm, angulation >11, and concomitant neurologic deficits, as well as in patients with significant comminution. Chronic nonunion can lead to clinical symptoms such as persistent pain, upper extremity paresthesia, and late-onset myelopathy.5 Koivikko et al19 re-

386

Journal of the American Academy of Orthopaedic Surgeons

Wellington K. Hsu, MD, and Paul A. Anderson, MD

viewed 69 patients with type II fractures treated nonsurgically, of whom only 32 (46%) healed. Nonunion was associated with posterior displacement >5 mm, fracture gap >1 mm, 4-day delay to the start of treatment, and loss of position >2 mm. Anterior displacement, sex, and age were not risk factors.

Table 2 Outcomes of Treatment of Type II Odontoid Fractures Treatment Non-halo orthosis Andersson et al22 Clark and White4 Polin et al3 Hanigan et al24 Hanssen and Cabanela25 Total Halo-vest orthosis Clark and White4 Koivikko et al19 Lind et al26 Polin et al3 Seljeskog27 Stoney et al28 Total Odontoid screw Andersson et al22 Apfelbaum et al20 Fountas et al29 Henry et al30 Harrop et al23 Jenkins et al21 Montesano et al31 Total Posterior fusion Andersson et al22 Campanelli et al32 Clark and White4 Frangen et al7 Total No. of Patients No. Healed Healing Rate (%)

Nonsurgical Management Traction and prolonged best rest are outmoded modalities and are of historical significance only. Currently, nonsurgical management of type II fracture involves early immobilization using a halo vest or a nonrigid orthosis (eg, collar, cervicothoracic orthosis). A systematic review of the clinical evidence of healing of type II fractures using these orthoses demonstrates wide variability (Table 2). Specific fracture patterns are associated with high rates of healing. In a recent systematic, evidencebased review of the treatment of type II odontoid fractures, the use of traction for 6 weeks followed by a 6-week period of cervical collar immobilization resulted in nonunion in 49% of patients (19 of 39).6 Nonunion rates were lower when more rigid bracing (halo/Minerva cast) was used (53 of 168 patients [32%]). Surgical Management High risk of nonunion, proven instability, persistent cord compression in patients with neurologic deficits, and failure of nonsurgical treatment are accepted indications for surgical management. Some authors recommend surgery for all acute type II fractures regardless of risk of nonunion.20 Surgical choices include odontoid screw fixation and posterior atlantoaxial arthrodesis. In the case of arthrodesis, autogenous bone graft has been most commonly added. The use of allograft and other biologic substitutes/supplements has not been adequately studied.
July 2010, Vol 18, No 7

10 3 16 5 5 39

4 0 9 3 4 20

40 0 56 60 80 51

38 69 9 20 15 22 173

25 32 8 15 14 18 112

66 46 89 75 93 82 65

11 117 31 61 9 42 14 285

3 99 26 56 7 30 12 233

27 85 84 92 78 71 86 82

7 7 26 21 61

7 6 24 20 57

100 86 92 95 93

Posterior Atlantoaxial Arthrodesis Posterior atlantoaxial arthrodesis was historically performed using Gallie and Brooks wire fixation. Although these techniques are efficacious in most patients, they have largely been abandoned for the management of odontoid fractures because of the inability to prevent anteroposterior shear and because

postoperative halo-vest immobilization is required.33,34 Biomechanical experiments confirm this clinical observation, showing that stiffness in flexion and extension as well as anteroposterior shear are significantly better with screw fixation.35 Wire fixation remains an alternative in young children and in patients with contraindications resulting from vas-

387

Odontoid Fractures: Update on Management

cular or bony anomalies. Atlantoaxial screw fixation can be obtained in one of two ways: with screws placed across C1 and C2, or with screws placed into each vertebra separately and connected with a rod. With the C1-2 transarticular screw technique, a screw is placed starting just above the C2-3 facet joint; the screw is positioned to enter the lateral mass of C2 and then to perforate the C1-2 joint, ending in the midlateral mass of C1.35 The steep trajectory angle of this screw requires percutaneous starting points as low as T1, with open guidance and posterior bone grafting. Transarticular screws have also been classically used in conjunction with interspinous wiring techniques. Although this method is technically demanding, it is cost-effective and requires only a standard set of fracture repair screws. Alternatively, atlas fixation can be obtained with direct placement of lateral mass screws or sublaminar wires. Axial fixation is performed with either pedicle or translaminar screws. In these cases, polyaxial screws, which are connected by rods, are advantageous. Although biomechanical differences with the various screw fixation techniques have been identified, they appear to be of little clinical significance. Odontoid screw fixation is an osteosynthesis procedure; it preserves atlantoaxial rotation by avoiding arthrodesis. However, this advantage is merely theoretical; C1-2 range of motion has been shown to be reduced by 50% following this technique.36 The definitive indications for use of an odontoid screw rather than other treatment options remain controversial. Grauer type IIB injuries, which have a fracture orientation from anterosuperior to posteroinferior, are optimal in achieving interfragmentary compression with anterior screw fixation. In the authors

opinion, odontoid screw fixation is indicated in younger patients with favorable fracture patterns who are at risk for nonunion or who have multiple spine fractures. The most important risk factors are initial displacement of >5 mm and the inability to maintain alignment nonsurgically. Either one or two screws can be used for fixation. Biomechanical and clinical studies have demonstrated no significant difference in relative strength or union rates.21 Although there have been reports of the use of variable pitch screws,37 most surgeons currently recommend partially threaded lag screws to increase the amount of compression across the fracture. An essential step in odontoid fixation is anatomic fracture reduction before screw placement. Adequate biplanar imaging and body habitus that allows for a flat screw trajectory almost parallel to the spine are critical to a successful operation.

Occipitocervical Fusion With C1 Laminectomy Occipitocervical fusion with C1 laminectomy is occasionally required for patients who are irreducible and who have posterior cord compression behind the atlas. This condition is a contraindication for sublaminar wire passage. Occipitocervical fusion is performed using occipital screws connected to C1 and C2 screw fixation, as described above. Reported clinical outcomes of surgical treatment of type II odontoid fractures are almost exclusively based on fracture or arthrodesis healing (Table 2). Patient-derived outcomes and general health assessment are lacking. In an evidence-based review of 95 studies containing American Medical Association class III data, Julien et al6 reported successful fusion in 74% of patients treated with posterior cervical fixation; most patients underwent wire fixation. The choice

of postoperative immobilization in these studies varied; however, a halothoracic vest was used in three studies. The most common complications were loss of fracture reduction and the development of new neurologic deficits. An evidence-based review of the recent literature demonstrates a healing rate of 93% with posterior atlantoaxial fusion4,7,22,32 (Table 2). Posterior atlantoaxial fusion can also be performed as a salvage operation when anterior screw fixation fails. Reported complications following posterior fixation include loss of reduction, increased neurologic deficit, and vertebral artery injury.38 Review of seven articles reporting on the use of anterior fixation to treat type II odontoid fractures shows that 233 of 285 patients achieved successful fracture healing (82%)20-23,29-31 (Table 2). In addition to nonunion, anterior screw fixation is associated with significant rotatory motion loss,36 respiratory problems, hardware issues22 (eg, screw cutout), and mortality.39 Mortality has been reported to be as high as 9% following anterior screw fixation for odontoid fractures in elderly persons.40

Authors Treatment Recommendations Many treatment options are acceptable for patients with type II odontoid fracture. It is important to use a shared decision-making protocol with patients after obtaining proper informed consent following an explanation of the associated risks and benefits. Patients at low risk for nonunion (eg, nondisplaced and stable fracture after initial immobilization) can be treated with a halo-vest orthosis. The authors recommend surgery for patients at high risk for nonunion. Fractures with favorable patterns (ie, Grauer IIB) and body habitus are treated with an odontoid screw. Other patients are treated using posterior at-

388

Journal of the American Academy of Orthopaedic Surgeons

Wellington K. Hsu, MD, and Paul A. Anderson, MD

Figure 5

Preoperative sagittal CT scans of the midspinal canal (A), the right midlateral mass (B), and the left midlateral mass (C), as well as an axial CT scan of C1 (D) in a 55-year-old man with a Grauer IIB odontoid fracture pattern associated with a C1 anterior arch fracture. Lateral (E) and AP odontoid (F) radiographs of the cervical spine on postoperative day 2. The patient was treated with posterior C1-2 fusion with a transarticular screw technique and posterior bone grafting with demineralized bone matrix. Lateral (G) and odontoid (H) plain radiographs of the cervical spine taken 3 months postoperatively. The patient demonstrated a stable atlantoaxial interval with fracture healing 1 year postoperatively.

lantoaxial arthrodesis with screw fixation (Figure 5). Posterior C1-2 fusion with screw fixation is also indicated in type II odontoid fractures that are not amenable to anterior fixation, that are associated with significant osteoporosis,23 and that are the result of nonunion.

Type III
Type III fractures make up less than one third of all odontoid fractures; they are defined by a fracture line through the cancellous body of C2 (Figure 1). In general, these fractures have a better prognosis than do type II injuries because the fracture occurs in a large bony contact area with an
July 2010, Vol 18, No 7

adequate vascular supply. Historically, these injuries have been treated with a variety of surgical and nonsurgical approaches;6 however, much of the recent analysis has demonstrated acceptable healing with cervical collar immobilization. In seven publications containing data for type III odontoid fractures treated with cervical collar immobilization, only 8% demonstrated nonunion after treatment (5 of 60)3,4,8,41-44 (Table 3). Halo-vest immobilization and cervical orthosis offer similar rates of healing (95% versus 92%, respectively)3-5,8,18,41-48 (Table 3). However, characteristics of certain type III patterns may lead to

a poorer prognosis. For example, the incidence of nonunion may be greater in high type III fractures that involve the waist of the dens and in those that exhibit significant anterior or posterior displacement, such as in a type II odontoid fracture. In addition, in a certain subset of type III fractures that demonstrate vertical instability with >5 mm of distraction, surgical treatment should be considered to prevent the development of a progressive neurologic deficit.49 For stable type III odontoid fracture without significant distraction, immobilization with a cervical orthosis has yielded union rates of 86% to 100% (also see Table 3).

389

Odontoid Fractures: Update on Management

Table 3 Outcomes of Halo-vest and Nonrigid Immobilization of Type III Odontoid Fractures Treatment Nonrigid immobilization Chiba et al41 Clark and White4 Mller et al8 Pepin et al42 Polin et al3 Ryan and Taylor43 Wang et al44 Total Halo Anderson and DAlonzo5 Bucholz and Cheung45 Clark and White4 Ekong et al18 Fujii et al46 Greene et al47 Polin et al3 Seybold and Bayley48 Wang et al44 Total No. of Patients No. healed Healing Rate (%)

17 10 7 13 5 6 2 60

16 9 6 11 5 6 2 55

94 90 86 85 100 100 100 92

12 9 15 6 14 67 13 20 12 168

11 9 13 5 13 66 13 19 10 159

92 100 87 83 93 99 100 95 83 95

Considerations in the Elderly

Odontoid fractures in elderly persons deserve special consideration because of the high rate of associated morbidity and mortality. The exact age cutoff is unclear, but several published studies have reported early mortality rates as high as 40%.50,51 Mller et al50 reported a 35% inhospital mortality rate following odontoid fractures in patients aged >70 years. Several reasons have been postulated for this finding, including the mechanism of trauma, the poor rehabilitative potential of patients in this age group, the presence of comorbidities, and the consequences of fracture management. Odontoid fractures are seen commonly in this

age group, and many factors must be considered in selecting the best treatment option for acceptable healing while minimizing complications. In elderly patients, halo-vest immobilization is associated with a high rate of morbidity and mortality; thus, its use should be avoided. In a retrospective analysis, Tashjian et al52 reported that patients aged >65 years with a type II or III odontoid fracture managed with a halo vest had a 42% mortality rate, compared with a 20% rate in the nonhalo group (P = 0.03). In this and other published studies, the high mortality rate was attributed to a significant increase in cardiopulmonary complications. In fact, aspiration pneumonia and cardiac arrest occurred in 34% and 26%, respectively, of patients treated with a halo vest.52 In an elderly person

with poor pulmonary and rehabilitative reserve, rigid immobilization can lead to devastating complications. Conversely, however, nonrigid immobilization may not adequately stabilize the fracture to allow for healing. Kuntz et al53 reported failure with nonsurgical bracing in 6 of 12 patients aged 65 years. Consequently, many surgeons have advocated for early surgical treatment in elderly patients to avoid the complications resulting from pseudarthrosis and rigid bracing. Both anterior fixation and posterior atlantoaxial fusions have also led to acceptable healing rates when used in elderly patients.7,23 When surgery can be performed expeditiously with minimal trauma and postoperative complications, there is great potential to improve patients ability to rehabilitate early and to decrease the need for postoperative halo vest immobilization. This has led many experts to recommend surgical stabilization for type II odontoid fractures in patients aged >50 years.15 However, significant postoperative complications have been reported. In one study of odontoid fracture in patients aged >70 years, mortality rates were reported to be 40% with anterior screw fixation, 13% with hard collar immobilization, and 33% with a halo vest.50 Although this study had limited statistical power, the data suggest that elderly persons may have a lower mortality rate when treated with a cervical collar alone. Furthermore, a published case series has suggested that clinical outcomes involving chronic, unstable, dens nonunions are acceptable provided that the patient does not have myelopathic symptoms.54 These studies suggest that type II odontoid fracture with a stable pattern can be treated with a cervical orthosis in the geriatric population. Platzer et al40 reported the clinical results of two patient cohorts, one

390

Journal of the American Academy of Orthopaedic Surgeons

Wellington K. Hsu, MD, and Paul A. Anderson, MD

with patients aged 65 years and one with patients aged >65 years. Although patients in both cohorts returned to their preinjury activity level, the elderly group had a significantly higher rate of nonunion (12% versus 4%, respectively), morbidity (22% versus 8%), and, most important, mortality (9% versus 1%) (P < 0.05 for each). Frangen et al7 reported high union rates in the geriatric population after dorsal C1-C2 fusion. However, the mortality rate was 22% (6 of 27 patients). The authors concluded that although the risk of mortality is still very high, the rates are considerably lower than with halo-vest immobilization. No comparison was made directly to a group treated in a cervical orthosis, such as a Miami-Jackson brace. For all type I and type III as well as stable type II odontoid fractures in elderly persons, we recommend using a hard cervical collar or cervicothoracic brace. For unstable type II patterns, posterior C1-2 arthrodesis should be performed.

Complications
Odontoid fractures are associated with an alarming rate of complications, such as respiratory problems, nonunion, pain, and mortality. The goal of treatment for odontoid injury is fracture healing with cervical spine stabilization. However, the avoidance of life-threatening complications is more important. Odontoid fractures are not commonly associated with neurologic deficits; however, a more recent review of the literature has identified important consequences related to the diagnosis and treatment of neurologic deficits. Przybylski et al55 reported the results of 13 patients with a posteriorly displaced odontoid fracture who developed severe respiratory compromise after closed reJuly 2010, Vol 18, No 7

duction with flexion. One possible explanation for this outcome is that axial skeletal traction and neck flexion may result in displacement of retropharyngeal edema or hematomas from the acute soft-tissue injury. These authors recommended that surgeons consider securing the airway before manipulating such fractures. Harrop et al56 reported that odontoid fractures associated with concomitant spinal cord injury are more likely to occur in males; in persons with cervical spondylotic myelopathy, resulting in stenosis; and in conjunction with high-velocity injuries. Although a halo-vest orthosis is a valid treatment option for many patients with odontoid fracture, significant complications have been reported with its use. Cardiopulmonary complications are well-established in elderly patients treated with a halo for odontoid fractures.50,52 Other well-established complications include pin site loosening, infection, pressure sores, and limitation of respiratory function. Morishima et al57 reported significant impairment in swallowing in healthy test subjects wearing a halo vest placed in extension. Using manometric evaluation of swallowing, the authors demonstrated that progressive cervical extension from 0 to 45 causes increased narrowing of the pharyngoesophageal junction, leading to delay in pharyngeal swallow. The authors concluded that these radiographic findings can explain the increased risk of aspiration in this patient population. Although anterior fixation remains the treatment of choice for odontoid fractures in certain spine centers, significant complications have been reported with its use. In our experience, elderly patients incur a higher rate of respiratory complications from a high anterior approach (Figure 6). Osteoporosis and inadequate screw purchase have led to several

negative outcomes, as well.22 Hardware complications such as screw cutout require surgical revision (Figure 7). Anterior fixation also has limitations because only those patients with the ideal body habitus (ie, without a short neck or barrel-chest deformity), satisfactory bone quality, appropriate anesthetic risk, and oblique fracture pattern are candidates for this treatment. Posterior C1-2 fusion is used to treat odontoid fractures that cannot be treated either nonsurgically or with anterior fixation. Taking away motion at the atlantoaxial articulation results in a significant loss of neck rotation and involves surgical expertise to avoid vascular injury. Furthermore, posterior muscle-splitting approaches are associated with postoperative pain.38 Pseudarthrosis after posterior C1-2 fusion with wire fixation may occur in up to 26% of cases.6

Summary
Management of type II odontoid fractures remains a source of substantial controversy. Treatment algorithms have evolved over the past decade to accommodate differences in fracture pattern as well as in patient age, medical condition, and body habitus. Furthermore, the number of elderly persons is increasing at a significant rate compared with the general population, and surgeons are often presented with several challenges when managing odontoid fractures in octogenarians. The recent development of novel surgical techniques in the anterior and posterior upper cervical spine has led to the use of more aggressive fracture management techniques to facilitate postoperative rehabilitation and avoid rigid bracing. Although many of the issues associated with halothoracic vests are prevented with this approach, recent retrospective data evaluating

391

Odontoid Fractures: Update on Management

Figure 6

Preoperative AP (A) and lateral (B) CT scans demonstrating distraction and posterior displacement at the site of type II odontoid fracture in an 81-year-old man. Because of the patients favorable medical condition, the decision was made to perform osteosynthesis of the fracture with an anterior screw. However, on postoperative day 2, the patient experienced airway compromise and required reintubation. AP odontoid (C) and lateral (D) radiographs of the cervical spine obtained on postoperative day 1, demonstrating signicant anterior retropharyngeal swelling adjacent to the surgical site.

Figure 7

Preoperative sagittal (A) and axial (B) CT scans demonstrating a posteriorly displaced type II odontoid fracture in a 78-year-old man. C, Intraoperative lateral plain radiograph demonstrating successful reduction of the odontoid fracture. Postoperative lateral (D) and AP odontoid (E) uoroscopic views of anterior odontoid screw xation. F, Lateral plain radiograph on postoperative day 7 demonstrating screw cutout, despite the use of a hard collar. Intraoperative lateral (G) and AP odontoid (H) uoroscopic views demonstrating posterior transarticular screw xation with C1-2 fusion using a transarticular screw and interspinous wiring.

392

Journal of the American Academy of Orthopaedic Surgeons

Wellington K. Hsu, MD, and Paul A. Anderson, MD

surgical treatment of odontoid fractures have demonstrated unexpectedly high morbidity rates, especially in the elderly. As with many degenerative and traumatic conditions in the spine, treatment decisions must be based on individual patient characteristics and fracture patterns to optimize outcomes. Because significant complications are associated with this injury regardless of treatment, surgeons must move toward a treatment plan that attempts to limit these complications as well as possible while achieving stability in the upper cervical spine. Recent and ongoing studies will continue to significantly aid in efforts to formulate a comprehensive algorithm that balances the risks, benefits, and outcomes for odontoid fractures in both young and elderly patients.

5.

Anderson LD, DAlonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg Am 1974;56(8):16631674. Julien TD, Frankel B, Traynelis VC, Ryken TC: Evidence-based analysis of odontoid fracture management. Neurosurg Focus 2000;8(6):e1. Frangen TM, Zilkens C, Muhr G, Schinkel C: Odontoid fractures in the elderly: Dorsal C1/C2 fusion is superior to halo-vest immobilization. J Trauma 2007;63(1):83-89. Mller EJ, Schwinnen I, Fischer K, Wick M, Muhr G: Non-rigid immobilisation of odontoid fractures. Eur Spine J 2003; 12(5):522-525. Amling M, Hahn M, Wening VJ, Grote HJ, Delling G: The microarchitecture of the axis as the predisposing factor for fracture of the base of the odontoid process: A histomorphometric analysis of twenty-two autopsy specimens. J Bone Joint Surg Am 1994;76(12):1840-1846. Barker L, Anderson J, Chesnut R, Nesbit G, Tjauw T, Hart R: Reliability and reproducibility of dens fracture classification with use of plain radiography and reformatted computeraided tomography. J Bone Joint Surg Am 2006;88(1):106-112. Grauer JN, Shafi B, Hilibrand AS, et al: Proposal of a modified, treatmentoriented classification of odontoid fractures. Spine J 2005;5(2):123-129. Bono CM, Vaccaro AR, Fehlings M, et al: Measurement techniques for upper cervical spine injuries: Consensus statement of the Spine Trauma Study Group. Spine (Phila Pa 1976) 2007; 32(5):593-600. Green RA, Saifuddin A: Whole spine MRI in the assessment of acute vertebral body trauma. Skeletal Radiol 2004; 33(3):129-135. Harris JH Jr, Carson GC, Wagner LK, Kerr N: Radiologic diagnosis of traumatic occipitovertebral dissociation: 2. Comparison of three methods of detecting occipitovertebral relationships on lateral radiographs of supine subjects. AJR Am J Roentgenol 1994;162(4):887892. Isolated fractures of the axis in adults. Neurosurgery 2002;50(3 suppl):S125S139. Bellabarba C, Mirza SK, West GA, et al: Diagnosis and treatment of craniocervical dislocation in a series of 17 consecutive survivors during an 8-year period. J Neurosurg Spine 2006; 4(6):429-440. Eismont FJ, Bohlman HH: Posterior atlanto-occipital dislocation with fractures of the atlas and odontoid

process. J Bone Joint Surg Am 1978; 60(3):397-399. 18. Ekong CE, Schwartz ML, Tator CH, Rowed DW, Edmonds VE: Odontoid fracture: Management with early mobilization using the halo device. Neurosurgery 1981;9(6):631-637. Koivikko MP, Kiuru MJ, Koskinen SK, Myllynen P, Santavirta S, Kivisaari L: Factors associated with nonunion in conservatively-treated type-II fractures of the odontoid process. J Bone Joint Surg Br 2004;86(8):1146-1151. Apfelbaum RI, Lonser RR, Veres R, Casey A: Direct anterior screw fixation for recent and remote odontoid fractures. J Neurosurg 2000;93(2 suppl): 227-236. Jenkins JD, Coric D, Branch CL Jr: A clinical comparison of one- and twoscrew odontoid fixation. J Neurosurg 1998;89(3):366-370. Andersson S, Rodrigues M, Olerud C: Odontoid fractures: High complication rate associated with anterior screw fixation in the elderly. Eur Spine J 2000; 9(1):56-59. Harrop JS, Przybylski GJ, Vaccaro AR, Yalamanchili K: Efficacy of anterior odontoid screw fixation in elderly patients with type II odontoid fractures. Neurosurg Focus 2000;8(6):e6. Hanigan WC, Powell FC, Elwood PW, Henderson JP: Odontoid fractures in elderly patients. J Neurosurg 1993;78(1): 32-35. Hanssen AD, Cabanela ME: Fractures of the dens in adult patients. J Trauma 1987;27(8):928-934. Lind B, Nordwall A, Sihlbom H: Odontoid fractures treated with halovest. Spine (Phila Pa 1976) 1987;12(2): 173-177. Seljeskog EL: Non-operative management of acute upper cervical injuries. Acta Neurochir (Wien) 1978; 41(1-3):87-100. Stoney J, OBrien J, Wilde P: Treatment of type-two odontoid fractures in halothoracic vests. J Bone Joint Surg Br 1998;80(3):452-455. Fountas KN, Kapsalaki EZ, Karampelas I, et al: Results of long-term follow-up in patients undergoing anterior screw fixation for type II and rostral type III odontoid fractures. Spine (Phila Pa 1976) 2005;30(6):661-669. Henry AD, Bohly J, Grosse A: Fixation of odontoid fractures by an anterior screw. J Bone Joint Surg Br 1999;81(3): 472-477. Montesano PX, Anderson PA, Schlehr F, Thalgott JS, Lowrey G: Odontoid

6.

7.

19.

8.

20.

9.

21.

22.

10.

23.

References
Evidence-based Medicine: Levels of evidence are described in the table of contents. In this article, references 3-8, 10, 11, 14, 18-21, 33, 39, 40, 42, 51, 52, and 57 are level III studies. References 2, 16, 22, 23, 28, 31, 32, 34, 49, 50, and 53-56 are level IV studies. Citation numbers printed in bold type indicate references published within the past 5 years.
1. Vaccaro AR, Madigan L, Ehrler DM: Contemporary management of adult cervical odontoid fractures. Orthopedics 2000;23(10):1109-1115. Lee PC, Chun SY, Leong JC: Experience of posterior surgery in atlanto-axial instability. Spine (Phila Pa 1976) 1984; 9(3):231-239. Polin RS, Szabo T, Bogaev CA, Replogle RE, Jane JA: Nonoperative management of types II and III odontoid fractures: The Philadelphia collar versus the halo vest. Neurosurgery 1996;38(3):450-457. Clark CR, White AA III: Fractures of the dens: A multicenter study. J Bone Joint Surg Am 1985;67(9):1340-1348.

11.

24.

12.

25.

13.

26.

14.

27.

28.

2.

15.

29.

16.

3.

30.

4.

17.

31.

July 2010, Vol 18, No 7

393

Odontoid Fractures: Update on Management

fractures treated by anterior odontoid screw fixation. Spine (Phila Pa 1976) 1991;16(3 suppl):S33-37. 32. Campanelli M, Kattner KA, Stroink A, Gupta K, West S: Posterior C1-C2 transarticular screw fixation in the treatment of displaced type II odontoid fractures in the geriatric population: Review of seven cases. Surg Neurol 1999;51(6):596-600. Maiman DJ, Larson SJ: Management of odontoid fractures. Neurosurgery 1982; 11(6):820. Coyne TJ, Fehlings MG, Wallace MC, Bernstein M, Tator CH: C1-C2 posterior cervical fusion: Long-term evaluation of results and efficacy. Neurosurgery 1995; 37(4):688-693. Lapsiwala SB, Anderson PA, Oza A, Resnick DK: Biomechanical comparison of four C1 to C2 rigid fixative techniques: Anterior transarticular, posterior transarticular, C1 to C2 pedicle, and C1 to C2 intralaminar screws. Neurosurgery 2006;58(3):516521. Jeanneret B, Vernet O, Frei S, Magerl F: Atlantoaxial mobility after screw fixation of the odontoid: A computed tomographic study. J Spinal Disord 1991;4(2):203-211. 39. Platzer P, Thalhammer G, Oberleitner G, Schuster R, Vcsei V, Gaebler C: Surgical treatment of dens fractures in elderly patients. J Bone Joint Surg Am 2007; 89(8):1716-1722. Platzer P, Thalhammer G, Ostermann R, Wieland T, Vcsei V, Gaebler C: Anterior screw fixation of odontoid fractures comparing younger and elderly patients. Spine (Phila Pa 1976) 2007; 32(16):1714-1720. Chiba K, Fujimura Y, Toyama Y, Fujii E, Nakanishi T, Hirabayashi K: Treatment protocol for fractures of the odontoid process. J Spinal Disord 1996;9(4):267276. Pepin JW, Bourne RB, Hawkins RJ: Odontoid fractures, with special reference to the elderly patient. Clin Orthop Relat Res 1985;193:178-183. Ryan MD, Taylor TK: Odontoid fractures in the elderly. J Spinal Disord 1993;6(5):397-401. Wang GJ, Mabie KN, Whitehill R, Stamp WG: The nonsurgical management of odontoid fractures in adults. Spine (Phila Pa 1976) 1984;9(3): 229-230. Bucholz RD, Cheung KC: Halo vest versus spinal fusion for cervical injury: Evidence from an outcome study. J Neurosurg 1989;70(6):884-892. Fujii E, Kobayashi K, Hirabayashi K: Treatment in fractures of the odontoid process. Spine (Phila Pa 1976) 1988; 13(6):604-609. Greene KA, Dickman CA, Marciano FF, Drabier JB, Hadley MN, Sonntag VK: Acute axis fractures: Analysis of management and outcome in 340 consecutive cases. Spine (Phila Pa 1976) 1997;22(16):1843-1852. Seybold EA, Bayley JC: Functional outcome of surgically and conservatively 54. managed dens fractures. Spine (Phila Pa 1976) 1998;23(17):1837-1845. 49. Jea A, Tatsui C, Farhat H, Vanni S, Levi AD: Vertically unstable type III odontoid fractures: Case report. Neurosurgery 2006;58(4):E797. Mller EJ, Wick M, Russe O, Muhr G: Management of odontoid fractures in the elderly. Eur Spine J 1999;8(5):360-365. Majercik S, Tashjian RZ, Biffl WL, Harrington DT, Cioffi WG: Halo vest immobilization in the elderly: A death sentence? J Trauma 2005;59(2):350-358. Tashjian RZ, Majercik S, Biffl WL, Palumbo MA, Cioffi WG: Halo-vest immobilization increases early morbidity and mortality in elderly odontoid fractures. J Trauma 2006;60(1):199-203. Kuntz C IV, Mirza SK, Jarell AD, Chapman JR, Shaffrey CI, Newell DW: Type II odontoid fractures in the elderly: Early failure of nonsurgical treatment. Neurosurg Focus 2000;8(6):e7. Hart R, Saterbak A, Rapp T, Clark C: Nonoperative management of dens fracture nonunion in elderly patients without myelopathy. Spine (Phila Pa 1976) 2000;25(11):1339-1343. Przybylski GJ, Harrop JS, Vaccaro AR: Closed management of displaced type II odontoid fractures: More frequent respiratory compromise with posteriorly displaced fractures. Neurosurg Focus 2000;8(6):e5. Harrop JS, Sharan AD, Przybylski GJ: Epidemiology of spinal cord injury after acute odontoid fractures. Neurosurg Focus 2000;8(6):e4. Morishima N, Ohota K, Miura Y: The influences of halo-vest fixation and cervical hyperextension on swallowing in healthy volunteers. Spine (Phila Pa 1976) 2005;30(7):E179-E182.

40.

50.

51.

33.

41.

34.

52.

42.

35.

53.

43.

44.

36.

45.

55.

37. Magee W, Hettwer W, Badra M, Bay B, Hart R: Biomechanical comparison of a fully threaded, variable pitch screw and a partially threaded lag screw for internal fixation of type II dens fractures. Spine (Phila Pa 1976) 2007;32(17):E475E479. 38. Shilpakar S, McLaughlin MR, Haid RW Jr, Rodts GE Jr, Subach BR: Management of acute odontoid fractures: Operative techniques and complication avoidance. Neurosurg Focus 2000;8(6):e3.

46.

56.

47.

57.

48.

394

Journal of the American Academy of Orthopaedic Surgeons