The Hangman’s fracture is a bilateral fracture of the pars interarticularis of the C2 vertebra that causes traumatic spondylolisthesis of C2, the axis. The term “hangman’s fracture” was first used by Schneider in 1965 to describe an avulsion fracture of the lamina of C2 with traumatic dislocation and listhesis of the axis upon C3.1 This injury pattern was correlated to the fracture pattern described from judicial hangings that utilized a submental knot. Thus, the injury mechanism was first thought to be the result of an extension-distraction injury, and that still remains true with the exception of the more severe subtypes of Hangman’s fractures and some of the atypical patterns. Due to the inherently unstable flexion type Hangman’s fractures, great caution and diligence is required when treating these fractures and properly diagnosing these injuries allows for the distinction between stable and unstable fracture patterns and hence nonoperative versus operative treatments.
INCIDENCE AND DEMOGRAPHICS
Fractures of axis are the most common cervical spine fractures in patients older than 70 years in the United States. Hangman’s fractures are the second most common fracture of the C2 vertebrae. In an initial study of 229 consecutive cases of axis fractures performed in 1989, Hadley and colleagues calculated that 46 of 229 cases or 20% had Hangman’s fractures with males outnumbering females 2:1. The most common C2 fracture is an odontoid fracture accounting for 60% of C2 fractures. They additionally report that 65% of axis fractures were caused by motor vehicle accidents, 15% by falls, and 6% by diving injuries.2 Another study by the Swedish national health care registry demonstrated that the overall incidence of C2 fractures has doubled from 3 to 6 per 100,000 inhabitants between 1997 and 2014.3 This registry study utilized ICD10 codes to determine the incidence of C2 fractures and was not able to determine the subtypes of C2 fractures. Fortunately, a follow-up study examined the radiographs of all patients treated at 2 university hospitals in Sweden during this same time period and included 233 consecutive patients with C2 fractures and found 183 odontoid fractures (79%), 26 hangman’s fractures (11%), and 24 atypical fractures (10%).4 Interestingly, they also found dramatic increases of axis fractures with age, with 17.8 cases per 100,000 person-years in geriatric patients (age above 70 y) and 1.4 cases per 100,000 person-years in patients under the age of 70. Hangman’s fractures show less dependence upon age; patients over the age of 70 had an incidence of 0.9 while patients under the age of 70 had an incidence of 0.3 cases per 100,000 person-years.
ANATOMY
The axis of the cervical spine possesses unique/distinct anatomy when compared with other vertebra of the spine. Perhaps the most unique aspect of the axis is the dens, an anterior structure around which the atlas rotates and provides much of the axial rotational motion of the cervical spine and relies heavily upon its ligamentous integrity.5 The dens measures ~15 mm in height and 10 mm in width.6 Laterally, the superior facets of the axis measure ~17 mm in width and 18 mm in length and angle about 23 degrees from horizontal toward the sagittal plane. Finally, the pedicles—where bony failure occurs in Hangman’s fractures—are oriented approximately 33 degrees from the sagittal axis and measure about 8 mm in diameter and 25 mm long. The upper cervical spine relies heavily upon its ligamentous integrity, as such injuries related to this region of the spine deserve intense scrutiny as not to miss unstable patterns of injuries.
FRACTURE CLASSIFICATION
The first Hangman’s fracture classification was proposed in 1981 by Effendi et al.7 This classification system is based upon the degree and type of displacement of the anterior and posterior fragments of the C2 fracture. Type I lesions involve minimal displacement of the C2 body and maintain an intact C2–C3 disk space. Type II lesions show displacement of the anterior fragment and have an abnormal C2–C3 disk space. Finally, type III lesions involve displacement of the anterior fragment, with the anterior fragment in flexion and dislocation of the C2–C3 facet joints.
Levine and Edwards
The classification scheme proposed by Effendi and colleagues in 1981 was modified by Levine and Edwards8 in 1985 with the addition of a subclassification of type II fractures. It is the most commonly used fracture classification system for typical Hangman’s fractures and it is important to note that these are canal widening fractures so associated neurological injury is quite rare in typical Hangman’s fractures.
Type I injuries were defined as all nonangulated fractures with displacement <3 mm. These injuries likely result from extension-axial load mechanisms that fracture the axis, but do not disrupt the C2–C3 disk or the anterior or posterior ligaments.
Type II injuries were again defined by the presence of angulation or displacement of the anterior fragment, with an average angulation of 11 degrees and an average displacement of 5 millimeters. Type II injuries are thought to be caused by a combined mechanism with extension and axial loading followed by a flexion and compression load. Patients with type II injuries were initially treated with halo traction. Three of the 32 patients with this injury in the initial study showed immediate widening of the posterior disk space with application of traction. These patients with disk space widening were defined as having type IIa injuries. The mechanism of injury of type IIa injuries was proposed to be predominantly flexion-distraction. This force results in an oblique fracture anterior to the facet joints. The fracture line tends to be more horizontally oriented or in the axial plane and the disk space has angular deformity and lacks translation as compared with the typical type II. It is this subtype of fracture caused predominantly by a flexion force that is imparted on the C2 vertebrae that leads to a more unstable injury pattern.
Finally, flexion-compression was proposed as the mechanism of type III injuries. In addition to a fracture of the neural arch, type III injuries result in dislocation of bilateral facets and an unstable injury pattern.
Typical Versus Atypical
The classification of Hangman’s fractures was further modified to define typical and atypical fracture types by Starr and Eismont in 1993.9 This classification depends upon the symmetry of the fracture, with symmetric patterns included in the typical fractures, whereas asymmetrical patterns were included with atypical fracture patterns. Importantly, Starr and Eismont noted that, in contrast to typical fracture patterns, atypical fracture patterns were associated with significant canal narrowing with potential for neurological injury. Included among these atypical fractures are those patterns with a coronal fracture line through the posterior vertebral body of C2, thus the boney ring that forms the spinal canal remains intact to behave more like facet dislocations creating canal narrowing with displacement/translation. It is this unique fracture pattern in atypical Hangman’s fractures that allows for canal narrowing to occur with injury/listhesis through the C2–C3 disk space.10
DIAGNOSIS
Plain Radiographs
Because the majority of trauma patients seen in the United States receive a computed tomographic (CT) scan of the cervical spine before plain radiographs, the use of plain radiographs for the diagnosis of Hangman’s fractures is largely mentioned for historical relevance. Nonetheless, low-energy falls represent a fraction of Hangman’s fractures and recognition of this injury on plain radiographs is important for rapid and reliable diagnosis.2 After the diagnosis is made the plain radiograph aids in characterization of the fracture, especially how it will behave with the patient upright, and provides a baseline for future comparison if nonoperative treatment is chosen.
CT
CT scan of the cervical spine is the most common method of diagnosis of Hangman’s fractures. The defining aspect of this fracture is a discontinuity of the bilateral pars of the C2 vertebra. As described above, classification of the fracture depends upon the angulation and displacement of the anterior fracture fragment and the relationship of the C2–C3 facet joints. The regional angulation of the fracture is measured as the angle formed by the comparison of lines along the inferior aspects of the C2 and C3 vertebrae. Displacement is measured as the maximal distance between the anterior and posterior fragments when measured in a plane perpendicular to the posterior aspect of the C3 vertebral body and at the level of the disk space of the second and third vertebrae.8
Magnetic Resonance Imaging
Magnetic resonance imaging plays less of a role in diagnosis but may be used in conjunction with CT to further assess the neurological status and provide information regarding the integrity of the disk and posterior ligaments.
TREATMENT
Historically, treatment of Hangman’s fractures has centered on nonoperative treatment with immobilization either with an orthosis or a halo vest. However, poor outcomes, difficulties with prolonged external immobilization, and advancements in cervical spine implants has prompted a shift toward operative intervention for the unstable subtypes. While surgery has become more commonplace in Hangman’s fracture treatment, the ultimate treatment determinant remains fracture stability. As described previously, the Levine and Edwards classification system provides both a framework to discuss these fractures as well as generalized groupings for treatment recommendations. Type I and many type II fractures are stable and thus are treated nonoperatively. In contrast, type IIA and type III fracture patterns are inherently unstable and typically require operative intervention due to the predominant flexion force imparted in these injuries and the subsequent injury to the C2–C3 disk space.8 In addition, the atypical Hangman’s fracture, with a fracture passing through the vertebral body, may require operative intervention due to the greater potential for neurological injury and must be closely monitored if nonoperative intervention is selected.9 What these surgical fractures share is a concomitant injury or equivalent injury to the secondary stabilizers of the cervical spine, most notably the C2–C3 disk space.
Treatment of these fracture patterns begins in the emergency department. A standard evaluation should be undertaken to rule out concomitant injury including other spinal fractures. Secondary evaluation is critical to recognize inherently unstable fracture patters which place patients at risk for neurological injury including type IIa and III injuries. While traction is a common temporizing treatment for other cervical spine injuries with neurological or pending neurological injury, it should never be utilized in the predominant flexion subtypes of injuries, type IIa and III or atypical patterns. As Levine and Edwards astutely recognized, traction creates widening of the C2–C3 disk space with the potential to create or worsen neurological injury. If distraction of the C2–C3 disk space is seen, consideration for collar removal and immobilization in sandbags should be considered to avoid catastrophic neurological injury.
Nonoperative Management
All Levine and Edwards type I fractures are stable and should be managed nonoperatively with immobilization. Options for immobilization include hard collar, Minerva Jacket, Sternal Occipital Mandible Immobilizer (SOMI) brace, or HALO vest.11 A systematic review of the literature from 1966 to 2002 revealed a healing rate of 100% for type I fractures treated nonoperatively with immobilization.12 Further, a recent observational study, 21 patients (20 type I and 1 type II), were treated nonoperatively with a Minerva brace. Only 2 patients failed nonoperative treatment and required surgery. It is important to note that only 2 of the 21 patients showed bony union on CT scan at 3 months; however, all patients went on to heal their fracture when evaluated again at 1 year.13 Thus, CT follow-up is not necessary for the decision to complete the nonoperative immobilization which can be made based on plain radiographs and time.
To date, no prospective studies have directly compared the effectiveness of the various immobilization methods available for treating these injuries. However, a recent systematic review compared the use of a rigid orthosis (hard collar, Minverva jacket, or SOMI brace) to Halo vest.11 The study identified 131 Hangman’s fractures treated nonoperatively, 45 with a rigid orthosis and 86 with Halo. The rigid orthosis group had 3 (6%) nonunions but zero patients required operative intervention while the Halo group had 11 (13%) nonunions and 5 of these required surgical management. These differences were not statistically significant, suggesting that a rigid orthosis is a viable immobilization method for Hangman’s fractures. This is especially important given the inconvenience, morbidity, and mortality associated with Halo vests but should be interpreted with caution as there was likely a selection bias of increasing instability in the halo group at the outset of these studies.14,15 Also, final alignment can vary.
As described earlier, atypical Hangman’s fractures occur when the fracture passes through the posterior cortex of the vertebral body as opposed to through the neural arch.9 Prior case series advocate that treatment strategies should be the same for these atypical fracture patterns, though some controversy does exist.7,9,16 Some feel that the atypical fracture pattern may be more prone to instability.17 However, in a case series of 28 patients with atypical fracture patterns only 1 required surgical management, suggesting that these fractures warrant a trial of nonoperative management.10 Further, in a retrospective study of 63 patients with variant Hangman’s fractures and rigid external immobilization, all progressed to union without complication.18
Operative
Unstable type II Hangman’s fractures, type IIa, and type III patterns are indications for operative intervention. Historically, unstable patterns treated nonoperatively led to poor healing rates, 60% of type II, 43% of type IIA, and <40% for type III fractures. These poor healing rates, as well as the advent of modern cervical instrumentation, has prompted a shift to treat these fracture patterns surgically.11
Unstable Hangman’s fractures are amenable to multiple reduction and fixation methods. Judet originally advocated for direct reduction and fixation of the fracture with transpedicular lag screws but this has shown to have high failure rates and the fracture anatomy amenable to this type of fixation is uncommon. Both anterior and posterior approaches have been published as viable methods for treating these fractures. Anterior-based treatment relies on a C2–C3 anterior cervical discectomy and fusion (ACDF) while posterior approaches rely predominantly on a C2–C3 posterior cervical decompression and fusion (PCDF) with C2 pedicle and C3 lateral mass screws. Additional levels may be included if additional injuries to the cervical spine prompt further intervention or if the C2 fracture does not allow for C2 pedicle screws, fracture anatomy and/or aberrant vertebral artery anatomy, then extension to C1 is often necessary at the outset.
A retrospective review directly compared ACDF to PCDF in patients with Hangman’s fractures.19 In this single-center study, 24 patients underwent ACDF while 14 had a PCDF. All 38 patients achieved bony fusion at 6 months, but the ACDF group had significantly lower mean operative time and estimated blood loss. For these reasons, the authors advocate for an anterior approach. However, limited conclusions should be drawn from a small, single-center case series. Though there are no randomized, prospective studies directly comparing ACDF to PCDF for Hangman’s fractures 1 systematic review analyzed the combined results of 25 case series.11 Of the 417 patients who underwent surgical intervention, 200 were treated with an ACDF, 193 were treated with a PCDF, and 24 underwent a combined ACDF and PCDF. They found no significant differences in rates of fusion, treatment failure, complications, or mortality between surgical approaches. These authors advocate for an isolated C2 pedicle construct or a posterior C2–C3 fusion if the fracture can be lagged together. If a transpedicular lag screw is not possible, they then advocate for an ACDF in a young patient and a C1–C3 posterior fusion in an elderly patient.11
The use of a combined approach, with a C2–C3 ACDF and posterior cervical fusion, has also been described in the literature.20 This approach has been applied to both type II, IIA, and III fractures but is mainly reserved for fractures that have significant displacement and the C2 vertebral body dislocated anterior to C3.20 One case series of 11 patients with this fracture pattern showed that a combined approach resulted a 100% fusion rate for these unusual fracture patterns.
CONCLUSIONS
Fractures of the bilateral pars of the axis are the second most common injury to the C2 vertebra after odontoid fractures. These fractures occur most often as the result of motor vehicle accidents and falls, and the incidence of these injuries increase with patient age. With an aging population, we should expect that the incidence of these injuries will continue to increase.
The classification of these injuries has evolved over time, with the Levine and Edwards classification system as the most commonly used system at this time. In addition to providing a descriptive system that may be utilized to study these fractures, this system has also proven to reliably predict outcomes and determine optimal treatment. All type I and most type II injuries should be managed nonoperatively with a rigid orthosis or halo immobilization. In contrast, type IIa and type III fracture patterns should be managed with surgical fixation. If a C2 pedicle screw is possible, this is a good option for reduction and stabilization of the fracture. If this is not possible, either ACDF or posterior spinal fusion also provide robust treatment options.
There are many options available for immobilization of these fractures, including hard collar, Minerva Jacket, SOMI brace, and HALO vest. There are no studies that have found any difference between use of a rigid collar and the Halo vest for immobilization of these fractures.
Atypical Hangman’s fractures should be rapidly diagnosed because of the possibility of neurological injury. However, these fractures are often stable and most can be treated nonoperatively with close observation.
Hangman’s fractures can be classified as stable (type I and most type II) or unstable (type IIa and III) and the optimal treatment depends upon this distinction. Stable injuries do well with rigid immobilization and rarely require operative intervention. In contrast, unstable injuries do poorly if treated nonoperatively and warrant surgical intervention. Properly identifying and treating these injuries represents an opportunity for the spine surgeon to optimize patient outcomes with these complicated fractures.
Authors Preferred Treatment
Type I
Minerva jacket or SOMI brace. Upright films in brace after placement with close clinical follow-up (upright x rays again at 1 wk intervals for first month followed by x rays at 8 wk, 12 wk, 6 mo, and 12 mo). Collar is typically removed at 12 weeks with physical therapy referral.
Type II
Minerva jacket or SOMI brace. A halo can still be a viable option in a type II fracture that requires reduction, especially in younger patients. Upright films in brace after placement with close clinical follow-up. If traumatic spondylolisthesis (Figs. 1–3) is noted then patient will require surgical stabilization. Immobilization is typically removed at 12 weeks with physical therapy referral.
;)
FIGURE 1: Computed tomographic scans of 56-year-old male driver involved in a high-speed motor vehicle collision demonstrate fractures of the bilateral pars interarticularis of C2 without significant displacement or angulation of C2 relative to C3. Left (A), midline (B), and right (C) sagittal slices are provided in addition to an axial (D) and coronal (E) images.
FIGURE 2: Upright radiographs were obtained after the injury with the patient in a hard cervical collar (A) demonstrating no spondylolisthesis or angulation of the C2–C3 disk. However, at follow-up, upright radiographs in the cervical collar (B) show new spondylolisthesis of C2 on C3.
FIGURE 3: The patient was treated with a C2–C3 anterior cervical discectomy and fusion and postoperative radiographs (A) demonstrate correction of the patient’s spondylolisthesis, and maintenance of his alignment on radiographs obtained 6 months postoperatively (B).
Type IIa. These often require surgical stabilization and are axially unstable so traction is contraindicated. If treated nonoperatively then simple extension is utilized and can be maintained in a halo. C2–C3 ACDF or posterior C2–C3 fusion is preferred. If fracture anatomy or vertebral artery anatomy does not allow for a C2 pedicle screw placement then C2–C3 ACDF is recommended in younger patients and C1–C3 posterior cervical fusion in older patients (Figs. 4, 5). Postoperatively, a standard cervical collar is worn with close clinical follow-up and removed at 6 weeks with resumption of gentle range of motion exercises.
FIGURE 4: Computed tomographic scans of a 53-year-old female involved in a high-speed motor vehicle collision demonstrate fractures of the bilateral pars of C2 (A–D). There is anterolisthesis of C2 on C3 without associated facet dislocation.
FIGURE 5: Because the patient demonstrated clinically significant anterolisthesis of C2 on C3 and demonstrated instability of the C2–C3 disk complex with the application of traction, the patient was diagnosed with a type IIa injury. She was treated with a C1–C3 posterior cervical fusion. Postoperative radiographs (A) and 4-year follow-up radiographs (B) demonstrate intact posterior fusion hardware with a healed fusion mass.
Type III
These require surgical stabilization. Posterior C2–C3 fusion is preferred to obtain an open reduction of the facet dislocations. If fracture anatomy or vertebral artery anatomy does not allow for a C2 pedicle screw placement then C2–C3 ACDF is recommended in younger patients and C1–C3 posterior cervical fusion in older patients. A combined anterior/posterior approach is rare with the exception of significant displacement, spondyloptosis, in a young patient (Figs. 6–10). A standard cervical collar is worn postoperatively with close clinical follow-up and removed at 6 weeks with resumption of gentle range of motion exercises.
FIGURE 6: A 49-year-old male was struck by an automobile while riding his bicycle and computed tomographic scans were obtained (A–C) demonstrating fractures of bilateral pars of C2 with dislocation of the C2–C3 facets. This patient was diagnosed with a type III Hangman’s fracture.
FIGURE 7: The patient was placed in traction and reduction of the injury was obtained. Postreduction computed tomography redemonstrates fractures of the bilateral pars of C2 and shows an extension injury at C6–C7 (A–C).
FIGURE 8: This patient was treated with anterior fusion of C2–C3 and posterior cervical fusion extending from occiput to C7. The previous fracture-dislocations show appropriate reduction and good alignment.
FIGURE 9: Two years after his injury, the patient complained of pain and lack of range of motion in his neck. A computed tomographic scan at this time demonstrates good maintenance of alignment and formation of a fusion mass from occiput to C7 (A–C).
FIGURE 10: Due to his pain and impaired of range of motion, the patient was treated with removal of hardware from occiput to C3, occiput-C1 osteotomy, and posterior fusion from C1–C3 with wiring. Postoperative (A) and 1-year postoperative (B) radiographs demonstrate continued maintenance of alignment and no evidence of complication. The patient was very satisfied with his improved neck range of motion postoperatively.
Atypical Hangman’s Fractures
Minerva jacket or SOMI brace is typically initiated. Upright films in brace after placement with close clinical follow-up (Figs. 11, 12). If traumatic spondylolisthesis is noted then patient will require surgical stabilization and is more commonly seen in this population than the traditional type II fractures. Therefore, there needs to be a high index of suspicion of converting to surgical intervention with appropriate patient counseling and close neurological monitoring.
FIGURE 11: A 55-year-old male sustained an injury at work when a large piece of equipment fell onto him. A computed tomographic scan obtained at the time reveals a fracture of the left pars interarticularis and right lamina of C2 (A–E). The asymmetric fracture pattern is representative of an atypical Hangman’s fracture.
FIGURE 12: The patient was managed nonoperatively in a hard cervical collar. Upright radiographs obtained in the hospital (A) and at his 1-year follow-up visit (B) demonstrate maintained alignment and interval healing of fractures.
REFERENCES
1. Schneider RC, Livingston KE, Cave AJE, et al. Hangman’s Fracture of the cervical spine. J Neurosurg. 1965;22:141–154.
2. Hadley MN, Dickman CA, Browner CM, et al. Acute axis fractures: a review of 229 cases. J Neurosurg. 1989;71:642–647.
3. Robinson A-L, Olerud C, Robinson Y. Epidemiology of C2 fractures in the 21st century: a national registry cohort study of 6370 patients from 1997 to 2014. Adv Orthop. 2017;2017:6516893.
4. Robinson A-L, Möller A, Robinson Y, et al. C2 fracture subtypes, incidence, and treatment allocation change with age: a retrospective cohort study of 233 consecutive cases. BioMed Res Int. 2017;2017:8321680.
5. Ishii T, Mukai Y, Hosono N, et al. Kinematics of the upper cervical spine in rotation: in vivo three-dimensional analysis. Spine. 2004;29:E139–E144.
6. Xu R, Nadaud MC, Ebraheim NA, et al. Morphology of the second cervical vertebra and the posterior projection of the C2 pedicle axis. Spine. 1995;20:259–263.
7. 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 Br. 1981;63-B:319–327.
8. Levine AM, Edwards CC. The management of traumatic spondylolisthesis of the axis. J Bone Joint Surg Am. 1985;67:217–226.
9. Starr JK, Eismont FJ. Atypical Hangman’s fractures. Spine. 1993;18:1954–1957.
10. Al-Mahfoudh R, Beagrie C, Woolley E, et al. Management of typical and atypical Hangman’s fractures. Global Spine J. 2016;6:248–256.
11. Murphy H, Schroeder GD, Shi WJ, et al. Management of Hangman’s fractures: a systematic review. J Orthop Trauma. 2017;31:S90–S95.
12. Li X-F, Dai L-Y, Lu H, et al. A systematic review of the management of Hangman’s fractures. Eur Spine J. 2006;15:257–269.
13. Prost S, Barrey C, Blondel B, et al. Hangman’s fracture: management strategy and healing rate in a prospective multi-centre observational study of 34 patients. Orthop Traumatol Surg Res. 2019;105:703–707.
14. Majercik S, Tashjian RZ, Biffl WL, et al. Halo vest immobilization in the elderly: a death sentence? J Trauma. 2005;59:350–357.
15. Tashjian RZ, Majercik S, Biffl WL, et al. Halo-vest immobilization increases early morbidity and mortality in elderly odontoid fractures. J Trauma. 2006;60:199–203.
16. Samaha C, Lazennec JY, Laporte C, et al. Hangman’s fracture: the relationship between asymmetry and instability. J Bone Joint Surg Br. 2000;82:1046–1052.
17. Burke JT, Harris JH. Acute injuries of the axis vertebra. Skeletal Radiol. 1989;18:335–346.
18. Niemeier TE, Manoharan SR, Mukherjee A, et al. Conservative treatment of Hangman variant fractures. Clin Spine Surg. 2018;31:E286–E290.
19. Ge C, Hao D, He B, et al. Anterior cervical discectomy and fusion versus posterior fixation and fusion of C2–3 for unstable Hangman’s fracture. Clin Spine Surg. 2015;28:E61–E66.
20. Liu J, Li Y, Wu Y. One-stage posterior C2 and C3 pedicle screw fixation or combined anterior C2-C3 fusion for the treatment of unstable hangman's fracture. Exp Ther Med. 2013;5:667–672.
