By Glenn R. Rechtine; Jonathan Landsman
The typical fracture patterns of the second cervical vertebrae involve fractures of the odontoid process, traumatic spondylolisthesis, and miscellaneous fractures. Management of odontoid fractures continues to evoke the most controversy.
The axis has four ossification centers: the odontoid process, the axis body, and the two neural arches. These ossification centers are joined to one another by cartilaginous plates or synchondroses rather than by epiphyseal plates and are fused by age 7. The transverse synchondrosis that joins the odontoid to the C2 body lies inferior to the junction of the odontoid and the body in the adult{1} and can predispose to a type III fracture before its fusion. The odontoid process is a well-vascularized structure and loss of blood supply due to injury is not a likely cause of nonunion. A rich arterial network, including branches from the vertebral internal and external carotid arteries, anastomose and supply the odontoid process from base to tip.{2,3}
The odontoid process and atlas with the associated ligamentous structures are the key to stability of the atlantoaxial complex.{4} The transverse ligament, which arises from the lateral masses of the atlas and passes directly behind the odontoid process, prevents anterior dislocation of the atlas on the axis. The anterior arch of C1 prevents posterior displacement of the atlas on the axis. Little intrinsic stability is gained from the facet joints because they lie in the horizontal plane and are designed to facilitate rotation. The paired accessory ligaments arise from the lateral masses of the atlas and attach to the base of the odontoid on each side. The paired alar ligaments attach the occipital condyles to the superior aspect of the odontoid process.{4,5} The apical ligament attaches the foramen magnum to the tip of the odontoid. The majority of odontoid fractures occur below the attachment of these ligaments to the proximal fragment, which explains the unstable nature of these injuries.{1,4,5}
The recent interest in the use of odontoid screw fixation has led to investigations into the bony morphology of the odontoid process. The dens is a conical structure that is thickest at its base. It projects an average of 15 mm from the body of the second vertebra. The size of the odontoid process does not correlate with the height or weight of the specimen.{6,7}
Odontoid fractures account for approximately 15% of all cervical spine fractures.{8-10} Over the last few decades, the most common mechanism of injury producing an odontoid fracture has been motor vehicle accidents.{5,8,11} The average age of persons sustaining an odontoid fracture is approximately 40 years of age,{5,8,11} with a wide range from the very young to the very old. In fact, the odontoid fracture is the most common individual cervical spine fracture in people younger than 8 years of age and older than 70 years of age.{12,13}
The typical high-energy injury pattern is not always seen in the very young and very old patient population. Apparently low-energy injuries, such as falls from short distances, are often responsible. Thus, a high index of suspicion is needed for a diagnosis of odontoid fractures, regardless of the mode of injury.{9,14} Odontoid fractures are among the most commonly missed spinal injuries.
Odontoid fractures are produced by flexion and extension moments with rotation probably also being a factor. Flexion causes anterior displacement of the odontoid process after it has fractured; extension produces posterior displacement. Knowledge of this may be helpful in fracture reduction. Anterior displacement is usually more common than posterior displacement, but posterior displacement may be more common in the elderly.{9,11,13,15}
Odontoid fractures are among the most commonly missed spinal injuries. In spite of this, the majority of odontoid fractures can be diagnosed from plain radiographs. In one study, 94% of odontoid fractures were diagnosed with combined lateral anteroposterior (AP) open mouth and oblique radiographs.{16} If these initial studies do not show the odontoid well or if a high index of suspicion persists despite normal radiographs, additional studies are warranted. Tomography or thin section computed tomography (CT) with sagittal and coronal refiguration may be helpful.{17}
The classification proposed by Anderson and D'Alonzo{5} is still universally accepted and useful in determining programs and directing treatment (Figure 1). The type I fracture is an oblique fracture through the apex of the dens, which probably represents an avulsion of the alar ligaments.{4,5} The type I fracture is extremely rare, but may be associated with atlanto-occipital dislocation.{11,18} The type II fracture, the most common fracture, occurs at the base of the odontoid at its junction with the body of the axis. The type III fracture extends downward through the body of the axis.{5}
Prognosis may be dictated by associated fractures, which are seen in up to 20% of patients. Type I fractures may be treated symptomatically in a collar or brace and should heal uneventfully.{5} Their association with atlanto-occipital dislocation has been established.{11,18} Type III fractures have also been typically thought of as being benign-type fractures that heal readily when properly diagnosed and immobilized. Clark and White's{11} multicenter study drew attention to a 13% nonunion rate with these fractures. The mode of treatment from the various centers ranged from no treatment at all to the use of halo immobilization. The authors found no statistically significant relation between union and displacement or angulation but noted a trend for nonunion with displacement greater than 5 mm and angulation greater than 10 degrees for these type III fractures. Thus, these authors recommended halo immobilization until healed (8 or 12 weeks) for type III fractures, with the understanding that surgery could be required for especially unstable fractures.
Significant controversy still exists as to the proper management of type II odontoid fractures. The literature varies with respect to the rate of nonunion with nonsurgical management of these fractures. Rates noted range from 10% to 67%.{5,8-11,13,19,20} There are several predictors of nonunion; amount of initial displacement (more than 5 to 6 mm), direction of displacement (nonunion more likely for posterior than for anterior), angulation (greater than 10 degrees), age (more than 60 years), smoking, time to diagnosis, and method of treatment.{5,8-11} {13,19,20} To further cast doubt on the validity of nonsurgical management of type II fractures, one series showed the union rate for primary posterior cervical fusion to be 96%, and thus the authors recommended considering primary fusion for patients with significant displacement, angulation, or both.{11} In this same study, however, all fractures that failed to unite by primary nonsurgical management united after delayed surgical fusions.{11} Other authors have documented similar results after delayed fusions.{8,21}
It seems wise then, as Anderson and D'Alonzo recommended 20 years ago,{5} to educate the patient as to the possibility of nonunion and the need for delayed fusion after several months of halo immobilization versus primary fusion with the inherent risks of surgery in this area. Should nonsurgical management fail, the results from secondary fusion seem to be as good as those from primary fusion.{8,11,21} The proper nonsurgical management includes reduction of displaced fractures and then serial monitoring with halo immobilization until healing occurs, usually in 12 weeks.{8,10,11} If traction fails to achieve stable alignment, or if displacement recurs with halo immobilization, surgical intervention is warranted.
The management of odontoid fractures in older patients deserves special mention. Dens fractures in persons older than 60 years of age are frequently delayed in diagnosis, are often the result of low-energy trauma, are posteriorly displaced, and have a high likelihood of nonunion.{9,11,13} The clinical outcome in patients with nonunion may not be unsatisfactory, however, and thus may not warrant aggressive treatment, such as surgery with its inherent risks. Several authors have reported that halo immobilization should always be considered, but is not necessarily needed for a successful outcome.{9,13} Late myelopathy or other neurologic deficit from nonunion has not been seen.{9,13}
The recent popularity of anterior screw fixation of odontoid fractures deserves mention. Proponents cite the advantage that it allows direct fracture control in unstable type II fractures and some "shallow" type III fractures, and that the loss of motion (rotation) seen with the typical C1-C2 posterior fusion is not found with anterior stabilization procedures.{22-25} Limited neck motion has been documented after C1-C2 posterior fusion, but it was rarely of clinical concern to the patients.{11} Anterior screw fixation can be considered with a type II odontoid fracture associated with a C1 ring fracture (which may occur in up to 16% of patients with type II odontoid fractures).{22-24} However, other authors state that it is appropriate to treat concomitant C1 ring fractures with type II odontoid fractures in halo fixation for a time sufficient to allow the atlas fracture to heal. If the odontoid fracture does not heal, a posterior C1-C2 fusion is indicated.{10,26}
If the decision to use anterior screw fixation is made, controversy exists as to the number of screws needed to stabilize the fractured odontoid. Two screws have been advocated by some authors{22,24} while single screw fixation has been successful for others.{23,27,28} It has been shown biomechanically that one screw may suffice.{29}
The rate of union for anterior fixation is similar to posterior C1-C2 fusion{22,25} but the complication rate may be higher.{22} Because of the potential for complications with this relatively new technique, it cannot yet be recommended in the routine management of odontoid fractures.
The relevant anatomy with respect to odontoid fractures is discussed above. Further expanding upon this, the second cervical vertebra may be thought of as a transitional unit (including the head, atlas, odontoid process, and body of the axis cephalad) with the typical subaxial vertebrae below. The critical area is in the region of the pars interarticularis of the axis, which is the focal point of the forces that are typically encountered in this injury.{1,30-32}
The mechanism most often cited as producing the typical injury is one of extension combined with an axial load.{1,26,30,33} These combined forces, typically seen in motor vehicle accidents when the face strikes the windshield or in falls, produce a bilateral fracture through the pars region. The amount of displacement, seen in the form of anterior displacement of the body of C2 on C3, and angular deformity relate to the degree of loading initially seen as well as secondary forces apparent in acceleration-deceleration injury patterns.{1,26,33} The association of this injury pattern to the historic hangman's fracture is mentioned here to eliminate confusion. The typical "hangman's fracture" is produced by an extension-distraction force{31,32,34-37} and is rarely seen today except when certain patterns of traumatic spondylolisthesis are treated with overzealous traction as will be discussed in the following section.
This classification was modified in 1985 by Levine and is currently the one most often quoted.{33} It consists of four types of fracture patterns (Figure 2). A biomechanical explanation of each injury pattern has been proposed and is useful in planning treatment and prognosis. The classification is based on the configuration of the fracture on a lateral radiograph. Type I injuries result from a hyperextension-axial load mechanism, as has been previously discussed. These injuries include all nondisplaced fractures and those with up to 3 mm of displacement with no angulation.{26,33} Type II injuries show C2-C3 displacement as well as angulation resulting from initial hyperextension-axial loading with the addition of "rebound flexion."{26-33} Type IIA injuries show moderate degrees of displacement combined with severe angulation. Levine{33} reported this in 5% of patients studied and recognized this pattern as including a different mechanism of injury: flexion-distraction. It is in these patients that excessive traction with attempted reduction may produce radiographs typically of the "hangman's fracture," representing, in fact, an iatrogenic hanging of sorts.{35,36} Type III injuries are also rare but are very difficult to treat. These represent unilateral or bilateral facet dislocations in addition to the bilateral posterior element fractures and result from a flexion-compression mechanism.{26,33}
Neurologic deficits are rarely produced by the typical injury patterns and can be explained by the fact that these fractures typically produce a widened canal in an area of the spine that already has ample room for the spinal cord.{30,32} Most neurologic deficits can be explained by the associated injuries to the head and spine, especially in the cervical region. Associated cervical injuries are seen in 14% to 30% of patients, and the great majority of these are in the upper three cervical segments.{31-33,37} Therefore, once the initial diagnosis is made, a search for other injuries is critical.{31,32,37,38}
Type I injuries are stable and may be treated with a cervical collar for 8 to 12 weeks.{26,30,33} Levine has shown in long-term follow up that type I injuries may be more problematic than type II fractures. Type I injuries rarely produce a spontaneous fusion anteriorly between C2 and C3, as seen with type II fractures. The lack of fusion allows mobility at the C2-C3 facet joint, which, combined with the original axial loading injury, can produce degenerative arthritis in 30% of patients with type I injuries.{39} Type II injuries with severe displacement are treated in traction initially with slight extension of the neck. The reduction that is gained is typically lost after application of the halo-vest because the traction cannot be maintained with halo immobilization. This is usually of no consequence, however, because of the rare association with neurologic injury and the high likelihood of achieving union with these fractures. If displacement is great enough (more than 6 mm), it may be necessary to treat in traction for several weeks to maintain alignment.{8,26,30,33} Halo immobilization typically lasts 10 to 12 weeks. The rare nonunion can be treated with anterior C2-C3 fusion. C1-C3 posterior fusion also can be attempted, but is less desirable because an additional intact level is added to the arthrodesis.{39}
Type IIA fractures should be treated initially with halo immobilization with some degree of compression and extension in order to counteract the injury-producing force.{26,33} This fracture pattern is likely to distract with the application of traction. The rare type III fracture pattern has an associated unilateral or bilateral facet dislocation at C2-C3 in addition to angulation and displacement anteriorly. This pattern is associated with neurologic deficits and is the only pattern in which surgical intervention is typically needed.{26,30,33} Attempted closed reduction of the facet dislocation is usually unsuccessful. Surgical stabilization of the facet joint(s) is often needed even if closed reduction is successful. The neural arch fracture is then managed by closed means with halo immobilization.{26,33}
In addition to the typical odontoid fracture and traumatic spondylolisthesis injury to the axis, other fractures are seen, albeit less commonly. These include C2 vertebral body fractures, lateral mass fractures, lamina fractures, and spinous process fractures. As one might suspect, there are no reports of large series with specific recommendations for managing these fractures. Traction has been recommended for displaced vertebral body fractures of C2. Residual translation is tolerated and usually will heal with halo immobilization. Angulation of the C2 vertebral body may not be as well tolerated because this may narrow the spinal cord via the odontoid process.{8,26} Less severe injuries can be treated with a rigid collar.
1. The Cervical Spine Research Society Editorial Committee: The Cervical Spine, ed 2. Philadelphia, PA, JB Lippincott, 1989.
2. Althoff B, Goldie IF: The arterial supply of the odontoid process of the axis. Acta Orthop Scand 1977;48:622-629.
3. Schiff DC, Parke WW: The arterial supply of the odontoid process. J Bone Joint Surg 1973;55A:1450-1456.
4. Schatzker J, Rorabeck CH, Waddell JP: Fractures of the dens (odontoid process): An analysis of thirty-seven cases. J Bone Joint Surg 1971;53B:392-405.
5. Anderson LD, D'Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg 1974;56A:1663-1674. This classification is useful in determining prognosis and directing treatment. It is universally accepted and is the classification system most often quoted in the literature to date.
6. Schaffler MB, Alson MD, Heller JG, et al: Morphology of the dens: A quantitative study. Spine 1992;17:738-743.
7. Heller JG, Alson MD, Schaffler MB, et al: Quantitative internal dens morphology. Spine 1992;17:861-866.
8. Hadley MN, Browner C, Sonntag VK: Axis fractures: A comprehensive review of management and treatment in 107 cases. Neurosurgery 1985;17:281-290.
9. Pepin JW, Bourne RB, Hawkins RJ: Odontoid fractures with special reference to the elderly patient. Clin Orthop 1985;193:178-183.
10. Rizzolo SJ, Cotler JM: Unstable cervical spine injuries: Specific treatment approaches. J Am Acad Orthop Surg 1993;1:57-66.
11. Clark CR, White AA: Fractures of the dens: A multicenter study. J Bone Joint Surg 1985;67A:1340-1348. This multicenter study included 144 patients. The results confirmed the validity of the classification by Anderson and D'Alonzo. The authors also emphasized the potential difficulty in treating type II fractures with the degree of angulation and amount of displacement being important criteria in guiding management.
12. Crawford AH: Operative treatment of spine fractures in children. Orthop Clin North Am 1990;21:325-339.
13. Ryan MD, Taylor TK: Odontoid fractures in the elderly. J Spinal Disord 1993;6:397-401.
14. Seimon LP: Fracture of the odontoid process in young children. J Bone Joint Surg 1977;59A:943-948.
15. Southwick WO: Management of fractures of the dens (odontoid process). J Bone Joint Surg 1980;62A:482-486.
16. Ehara S, el-Khoury GY, Clark CR: Radiologic evaluation of dens fractures: Role of plain radiography and tomography. Spine 1992;17:475-479.
17. Sutterlin CE, Gutentag I, Martinez CR, et al: False-positive diagnosis of an odontoid fracture by CT scan. J Orthop Trauma 1989;3:348-351. CT of acute cervical spine trauma has a role but not for the diagnosis of type II odontoid fractures. Conventional polytomography is superior to CT as an adjunct to plain films in the diagnosis of odontoid fractures, according to the authors.
18. Pedersen AK, Kostuik JP: Complete fracture-dislocation of the atlantoaxial complex: Case report and recommendations for a new classification of dens fractures. J Spinal Disord 1994;7:350-355.
19. Lind B, Nordwall A, Sihlbom H: Odontoid fractures treated with halo- vest. Spine 1987;12:173-177.
20. Schweigel JF: Management of the fractured odontoid with halo-thoracic bracing. Spine 1987;12:838-839.
21. Apuzzo ML, Heiden JS, Weiss MH, et al: Acute fractures of the odontoid process: An analysis of 45 cases. J Neurosurg 1978;48:85-91.
22. Aebi M, Etter C, Coscia M: Fractures of the odontoid process: Treatment with anterior screw fixation. Spine 1989;14:1065-1070.
23. Esses SI, Bednar DA: Screw fixation of odontoid fractures and nonunions. Spine 1991;16(suppl 10):S483-S485.
24. Etter C, Coscia M, Jaberg H, et al: Direct anterior fixation of dens fractures with a cannulated screw system. Spine 1991;16(suppl):S25-S32.
25. Montesano PX, Anderson PA, Schlehr F, et al: Odontoid fractures treated by anterior odontoid screw fixation. Spine 1991;16:S33-S37.
26. Levine AM, Edwards CC: Treatment of injuries in the C1-C2 complex. Orthop Clin North Am 1986;17:31-44.
27. Borne GM, Bedou GL, Pinaudeau M, et al: Odontoid process fracture osteosynthesis with a direct screw fixation technique in nine consecutive cases. J Neurosurg 1988;68:223-226.
28. Chang KW, Liu YW, Cheng PG, et al: One Herbert double-threaded compression screw fixation of displaced type II odontoid fractures. J Spinal Disord 1994;7:62-69.
29. Sasso R, Doherty BJ, Crawford MJ, et al: Biomechanics of odontoid fracture fixation: Comparison of one- and two-screw technique. Spine 1993;18:1950-1953.
30. White AA III, Panjabi MM (eds): Clinical Biomechanics of the Spine, ed 2. Philadelphia, PA, JB Lippincott, 1990.
31. Effendi B, Roy D, Cornish B, et al: Fractures of the ring of the axis: A classification based on the analysis of 131 cases. J Bone Joint Surg 1981;63B:319-327.
32. Francis WR, Fielding JW, Hawkins RJ, et al: Traumatic spondylolisthesis of the axis. J Bone Joint Surg 1981;63B:313-318.
33. Levine AM, Edwards CC: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg 1985;67A:217-226. Fifty-two patients were studied in this retrospective analysis and a classification relating fracture type to mechanism of injury was developed.
34. Roda JM, Castro A, Blazquez MG: Hangman's fracture with complete dislocation of C2 on C3: Case report. J Neurosurg 1984;60:633-635.
35. Schneider RC, Livingston KE, Cave AJE, et al: "Hangman's fracture" of the cervical spine. J Neurosurg 1965;22:141-154.
36. Sherk HH, Howard T: Clinical and pathologic correlations in traumatic spondylolisthesis of the axis. Clin Orthop 1983;174:122-126.
37. Fielding JW, Francis WR Jr, Hawkins RJ, et al: Traumatic spondylolisthesis of the axis. Clin Orthop 1989;239:47-52.
38. Garfin SR, Shackford SR, Marshall LF, et al: care of the multiply injured patient with cervical spine injury. Clin Orthop 1989;239:19-29.
39. Levine AM, Rhyne AL: Traumatic spondylolisthesis of the axis. Semin Spine Surg 1991;3:47-60.
Chapter Figure Legends
Figure 1: Anderson and D'Alonzo classification of odontoid fractures, types I, II, and III. (Reproduced with permission from Anderson LD, D'Alonzo RT: Fractures of the odontoid process of the axis. J Bone Joint Surg 1974;56A:1663-1674.)
Figure 2: Classification of traumatic spondylolisthesis of the axis, types I, II, and III. (Reproduced with permission from Levine AM, Edwards CC: The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg 1985;67A:217-226.)