Chapter 63 Trauma Surgery
Occipitocervical Junction
Occipitocervical junction injuries can be classified in several ways (Box 63-1). One useful system describes occipitocervical junction injuries as isolated ligamentous injuries, isolated fractures, or mixed ligamentous and bony injuries. Occipitocervical junction trauma can also be described by the site and/or level(s) of injury. At most sites, classification systems have been developed for specific injury patterns (e.g., C2 odontoid fractures). Finally, occipitocervical junction injuries can be described on the basis of their stability. Stability is generally determined with clinical and radiographic assessment, sometimes using dynamic flexion/extension radiographs. A stable injury does not demonstrate significant radiographic deformity, pain, or neurologic dysfunction with normal physiologic loads and movement. An example of a stable injury would be an isolated C2 spinous process fracture that meets the preceding criteria. Some injuries are clearly unstable, such as occipitocervical dislocations. Other injuries may initially appear stable but have a reasonable chance of developing delayed instability with time, gravity, movement, and/or relaxation of paraspinal muscle spasm. This category reflects the reality that clinical and radiographic assessment of long-term stability may be indeterminate.
BOX 63-1 Occipitocervical Junction Injury Classification Systems
The preceding classification systems are helpful in injury assessment and planning management. However, the management of a patient with occipitocervical junction trauma is best determined by considering the nature of the injury (including associated injuries), patient characteristics (e.g., age, medical risk factors, bone quality, desire and ability to tolerate use of a halo orthosis), and the physician’s experience. Although much less common, penetrating trauma to the occipitocervical junction presents unique issues that relate to the specific location and trauma modality (e.g., bullet, knife). This class of injuries is not specifically discussed in this chapter but is addressed in Chapter 73. Although most of the principles from blunt trauma are applicable to penetrating trauma, it is important to point out some important differences. Compared with blunt trauma, penetrating trauma typically results in less ligamentous injury and, therefore, for a similar fracture, may be more stable. However, penetrating trauma more commonly results in trauma to vascular or other important regional structures.
General Principles
The initial management of occipitocervical junction injuries is focused on basic trauma management principles, including establishment and maintenance of airway, breathing, and circulation; careful immobilization and transportation; and recognition and management of any associated injuries. These principles have evolved over time and have been published in numerous settings.1,2
Occipitocervical junction injuries are frequently recognized on routine cervical spine imaging. However, these injuries may be difficult to detect on initial diagnostic studies. Clinical suspicion based on history and physical examination can aid recognition. Routine radiographs and clinical assessment are often inadequate to fully characterize the injury, and more specialized imaging is usually indicated. Coronal, curved coronal, sagittal, and/or three-dimensional CT reconstruction views can be extremely helpful in characterizing the presence and nature of injury. MRI may be difficult or impossible to obtain acutely but can often provide essential information on spinal canal compromise and may suggest the presence and degree of ligamentous injury. Dynamic imaging with plain radiographs, CT, and/or MRI can be valuable in assessing stability but should be performed carefully. Occasionally, stability is checked with real-time fluoroscopy during careful flexion and extension controlled by a qualified examiner. For example, fluoroscopic flexion/extension imaging may be helpful when there is urgent need to assess the stability of the cervical spine in an unresponsive patient but there is still controversy about its interpretation.3
Once occipitocervical junction injuries are diagnosed, management decisions are based on several factors, including the extent and stability of injury, the presence or progression of neurologic deficits, and patient-specific factors that influence the risks with different treatments. Nonoperative management typically includes some type of rigid (halo) or semirigid (collar) orthosis. Operative management is generally indicated for injuries that are unstable, have significant potential for delayed instability, have progressive neurologic deficits, and/or cause significant deficits or symptoms that are not controlled with nonoperative measures. Operative planning may include obtaining additional imaging (e.g., dedicated studies for image guidance), ensuring the availability of appropriate instrumentation, and arranging neurophysiologic monitoring where appropriate.
Diagnosis and Management
Occipitocervical Dislocations
Occipitocervical dislocations are relatively uncommon ligamentous injuries that usually result from hyperflexion and distraction during high-impact blunt trauma.4,5 These injuries are highly unstable, frequently fatal, and usually result in significant neurologic injury from stretching, compression, and/or distortion of the spinal cord, brainstem, and cranial nerves.6 In addition, significant morbidity and mortality can result from associated cerebrovascular injury, which varies among trauma series, diagnosis test used (CT, conventional angiogram), and severity of injuries.7,8 Recognition and rapid management of these injuries may limit further injury, but even with appropriate care, neurologic deficits can progress. Although these were initially felt to be rare, several series of trauma fatalities have revealed an incidence between 8% and 19%.4
Lateral cervical spine radiographs may recognize occipitocervical dislocations (sensitivity, 0.57), especially in severe injuries. However, these injuries can be difficult to diagnose with plain radiographs alone, especially with less-severe dislocations. In addition, the frequent presence of coexisting significant head trauma can delay recognition of spinal injury. Diagnostic clues include prevertebral soft tissue swelling, an increase in the dens-basion distance, and separation of the occipital condyles and C1 lateral masses (Fig. 63-1). CT imaging with reconstruction views (sensitivity 0.84) usually provides a better assessment of fractures and alignment than plain radiographs do. The presence of subarachnoid hemorrhage supports but does not confirm the diagnosis. MRI imaging can be helpful for diagnosis (sensitivity 0.86), to assess the extent of spinal cord compression and injury, and to demonstrate compressive hematoma lesions.2
On the basis of the injury pattern, occipitoatlantal dislocations have been classified by Traynelis et al.9 into four types: type I (anterior), type II (longitudinal), type III (posterior), and “other” (complex). A number of diagnostic radiographic criteria have been described that assess the relationship between the skull base and cervical spine (Fig. 63-2). Although developed for lateral plain radiographs, these criteria can also be used on sagittal reconstruction CT views, provided that there are no significant artifactual distortions. The Wackenheim clival line extends along the dorsal surface of the clivus and should be tangential to the tip of the dens.10 Ventral or dorsal translation of the skull in relation to the dens will shift the clival line to either intersect or run dorsal to the dens, respectively. The Powers ratio is based on the relationship of the B–C line (from the basion to the C1 dorsal arch) and the O–A line (between the opisthion and the C1 ventral arch).11 Normal B–C/O–A ratios average 0.77, while pathologic ratios (>1) typically represent occipitocervical dislocations. However, false negatives can occur with longitudinal or dorsal dislocations.12 The Wholey dens-basion technique assesses the distance from the basion to the dens tip.13 Although variability is common, the average distance in adults is about 9 mm, and pathologic distances are greater than 15 mm.14 The Dublin method, the least reliable method, measures the distance from the mandible (posterior ramus) to the ventral part of C1 (normally 2–5 mm) and C2 (normally 9–12 mm).15
Initial management of these injuries focuses on immobilization, almost always with a halo orthosis. Cervical collars are potentially dangerous because they may produce distraction and thereby promote further injury. Similarly, traction can cause neurologic worsening (2 of 21 patients) and should be avoided or used with extreme caution.1,2 Nonoperative management does not provide definitive treatment of these injuries because of the significant ligamentous disruption that cannot be expected to heal even with prolonged rigid (halo) external immobilization (11 of 40 patients had a nonunion and/or neurologic deterioration).2 Operative stabilization consists of an occipitocervical arthrodesis with rigid internal fixation (discussed later and in Chapter 143). Decompression and restoration of alignment may also be necessary to maximize neurologic recovery.
Transverse Ligament Injuries
Transverse ligament injuries are suggested or diagnosed with radiographic imaging. A widened atlantodental interval on flexion lateral cervical radiographs (>3 mm in adults, >5 mm in children) suggests transverse ligament insufficiency. Thin-cut CT imaging with reconstruction views may suggest the diagnosis by demonstrating a C1 lateral mass avulsion fracture at the ligamentous insertion. Thin-cut MRI with attention to the transverse ligament when using gradient echo sequences can directly demonstrate a transverse ligament injury.16 If the diagnosis is uncertain, dynamic (flexion/extension) imaging is appropriate for cooperative patients. On the basis of CT and MRI, traumatic transverse ligament injuries can be classified into two categories (Fig. 63-3). Type I injuries involve disruptions of the midportion (IA) or periosteal insertion laterally (IB). Type II injuries involve fractures that disconnect the C1 lateral mass tubercle for insertion of the transverse ligament via a comminuted fracture (IIA) or an avulsion fracture (IIB).17
The management of transverse ligament injuries is based on the type of injury. Type I injuries are pure ligamentous injuries that cannot be expected to heal with nonoperative external fixation. Therefore, operative stabilization with a dorsal C1-2 arthrodesis and fixation is indicated. The surgical options include C1-2 dorsal wiring, C1-2 Halifax clamps, C1-2 transarticular screws, and/or C1-2 segmental screw fixation (see later section and Chapter 143). Type II injuries have a much higher chance of healing with halo immobilization (up to 74%).17 If a nonunion is still present after a prolonged period of immobilization (>3 months), then operative stabilization is generally appropriate.
Rotatory C1-2 Subluxations
Rotatory C1-2 subluxations are ligamentous injuries that are more common in children and adolescents. These injuries typically present with neck pain and a fixed, rotated “cock-robin” head position. Open-mouth radiographs may demonstrate an asymmetry of the C1 and C2 lateral masses. CT imaging can confirm the rotatory subluxation diagnosis and demonstrate coexisting fractures. C1-2 axial rotation greater than 47 degrees confirms the diagnosis. Three-view CT imaging (15 degrees to the left, neutral, and 15 degrees to the right) can also be helpful in establishing the diagnosis.18,19 MRI may detect a coexistent transverse ligament injury.
Occipital Condyle Fractures
Occipital condyle fractures generally occur with axial trauma and are almost always unilateral (>90%). These injuries are classified into three types according to Anderson and Montesano.20 Type I injuries are comminuted fractures that result from axial trauma. Type II fractures are extensions of linear basilar skull fractures. Type III injuries, the most common, are avulsion fractures of the condyle that can result from a variety of mechanisms. The incidence of occipital condyle fractures has been estimated to be between 1% and 3% of blunt craniocervical trauma cases.21 Although plain radiographs (usually open-mouth radiographs) may occasionally identify the injury, they have an unacceptably low sensitivity (estimated at 3.2%) and should not be relied on when the diagnosis is suspected. CT imaging with reconstruction views provides the best assessment of fracture pattern and alignment.21,22
Occipital condyle fractures are generally stable and therefore are typically managed with an external nonrigid orthosis (collar) until the fracture heals (often 12 weeks). Type III fractures are felt to be more prone to instability, and when significant displacement or clinical concern exists, halo immobilization may be appropriate.23 Operative stabilization with an occipitocervical fusion is generally reserved for situations in which there are associated cervical fractures or significant ligamentous injuries.
C1 Fractures
Isolated C1 fractures account for approximately 5% of cervical spine fractures. These injuries occur with axial trauma with or without lateral bending.24 Open-mouth radiographs may suggest the injury, but CT imaging with reconstruction views provides the best assessment of fracture pattern and alignment. Fractures can include almost any part of the ring or lateral masses of C1. Aside from unilateral lateral mass fractures, the fractures usually occur at multiple sites (Fig. 63-4). Jefferson fractures are four-part fractures with bilateral ventral and dorsal ring fractures. The assessment of these injuries is focused on evaluating the integrity of the transverse ligament and on recognizing any additional fractures.
The management of C1 fractures is based on the integrity of the transverse ligament that can be assessed indirectly with several radiographic criteria such as a widened atlantodental interval (>3 mm) and increased spread of the lateral masses of C1 over C2 (>6.9 mm, rule of Spence)25 or directly through high-resolution MRI (Fig. 63-5). If the transverse ligament is intact, isolated C1 fractures are generally stable and can be treated with an external orthosis (e.g., SOMI) primarily for symptom control until the fracture heals. With transverse ligament insufficiency, operative stabilization is indicated by using a C1-2 fusion technique such as dorsal C1-2 wiring techniques, C1-2 transarticular screws, C1 lateral mass-to-C2 pars/pedicle screws, or ventral C1-2 screw fixation (see Chapter 143). The surgical choice is based primarily on patient anatomy and fracture pattern as well as the surgeon’s experience and preference. Postoperatively, most operations employing rigid internal fixation can be managed with a nonrigid external orthosis (e.g., a collar, SOMI), but C1-2 dorsal wiring without additional instrumentation generally warrants the use of a halo.26
