a b c Figure 10. Halo a, b Correct positioning of the skull pins, c halo vest. aspect of the head perimeter. Vest size is determined by measurement of chest circumference with a tape measure. The halo vest ( Fig. 10c)seemstobethefirst choice for conservative treatment of unstable injuries of the upper cervical spine, although pin track problems, accurate fitting of the vest, and a lack of patient compliance lead to clinical failures [165]. Because of these drawbacks, the authors’ preference is a Minerva cast. Spinal Cord Injuries Spinal cord injury frequently results from cervical fracture/dislocation Spinal cord injuries are frequently associated with traumatic cervical spine frac- tures and cervical facet dislocation injuries due to a displacement of fracture fragments or subluxation of one vertebra over another. Reduction of the defor- mity helps to restore the diameter of the bony canal and eliminates bony com- pression of the spinal cord. Theoretically, early decompression of the spinal cord after injury may lead to improved neurological outcome. However, indication and timing of surgical interventions in patients with complete and incomplete spinal cord injuries has been debated in the literature [6]. Yablon et al. [211] found that patients who underwent operative stabilization more frequently improved regarding neurological level than patients who were treated conserva- tively. In tetraplegic patients, such improvement can be essential for quality of life. 848 Section Fractures Role of Steroids in Acute Spinal Cord Injury High-dose methylpredniso- lone is highly controversial in acute SCI The role of steroids in acute spinal cord injury is very controversial [35, 122]. Although the use of corticosteroids can usually be considered safe in surgical patients [166, 168, 190], the potential side effects of high dose methylpredniso- lone such as infections [84, 86], pancreatitis [100], myopathies [157], psychosis [194], and lactate acidosis in combination with intravenous adrenaline treatment [98] are important arguments against this treatment. After the release of the NASCIS (National Acute Spinal Cord Injury Study) II study [36], the use of high- dose methylprednisolone in spinal cord injury became the standard of care. However, many researchers found the study methodology and statistics ques- tionable. Short [180] revisited this concern within the evidence-based frame- work of a critical appraisal of the accumulation of clinical studies and concluded that high-dose methylprednisolone cannot be justified as a standard treatment in acute spinal cord injury within current medical practice. On the other hand, the fact that there may be some hope of benefit and that adverse medicolegal implica- tions are feared has led many centers to adhere to the NASCIS II guidelines. Nev- ertheless, many centers are currently revising these guidelines to limit or discon- tinuetheuseofmethylprednisolone[131].Weonlyconsiderhigh-dosemethyl- prednisolone treatment for young patients with a monotrauma of the spine, i.e., without significant additional injuries. Role and Timing of Spinal Cord Decompression Secondary SCI due to additional fracture/ dislocation must be avoided Particularly in unstable fractures, further mechanical injury to the spinal cord by secondary dislocations must be avoided. The severity of the injury is related to the force and duration of compression, the displacement and the kinetic energy. Many animal models, including those of primates, have demonstrated that neu- rological recovery is enhanced by early decompression [72]. However, this experimental evidence has not been translated to patients with acute spinal cord injury. This may in part be due to: heterogeneous injury patterns absence of well-designed RCTs Even delayed decompression may improve neurology While one randomized controlled trial (RCT) showed no benefit of early (<72 h) decompression [197], several recent prospective series suggest that early decom- pression (<12 h) can be performed safely and may improve neurological outcomes [72]. Aebi et al. [12] demonstrated in 100 retrospectively examined patients that reduction within the first 6 h revealed the best neurological results. Lee et al. [124] found that 26% of patients who were reduced within 12 h improved the Frankel scale two or more grades, whereas only 8% improved if reduction was performed after 12 h. Immediate closed reduction is the most rapid and effective procedure for decompression in patients presenting with significant motor deficits [90]. However, pre-reduction MRI performed in patients with cervical fracture disloca- tion injury will demonstrate disrupted or herniated intervertebral discs in one- third to one-half of patients with facet subluxation [3, 90]. These findings do not seem to significantly influence outcome after closed reduction in awake patients and the usefulness of pre-reduction MRI can be questioned in this setting. A num- ber of studies have documented recovery of neurological function even after delayed decompression of the spinal cord (months to years) after the injury [21, 33, 34, 123, 193]. The improvement in neurological function with delayed decompres- sion in patients with cervical or thoracolumbar spinal cord injury who have pla- teaued in their recovery is noteworthy and suggests that compression of the cord is an important contributing cause of neurological dysfunction [3]. Cervical Spine Injuries Chapter 30 849 Urgent decompression is indicated for an incomplete SCI There are currently no standards regarding the role and timing of decompression in acute spinal cord injury. An immediate operative intervention is recom- mended in patients with incomplete spinal cord injury or progressive neurologi- cal deterioration, and in whom there is a persistent mechanical compression of thespinalcordbyfracturefragmentsordiscmaterial[6,72]. Specific Treatment of Upper Cervical Spine Injuries For the vast majority of cervical injuries, there is insufficient scientific evidence to support diagnostic and treatment standards or guidelines. At best it is possible to indicate options which are evidence enhanced but not evidence based [2]. We acknowledge that the anecdotal experience of the authors has been used to attempt to fill in the gap in those areas where scientific evidence is lacking. We therefore ask the reader to cr itically evaluate any t reatment recommendation before adaptation. Fractures of the Occipital Condyle Occipital condyle fractures are rare and require CT/MRI assessment Traumatic occipital condyle fracture (OCF) was first described by Bell in 1817 [28]. Occipital condyle fractures are rare injuries. Clinical suspicion should be raised by the presence of one or more of the following criteria: blunt trauma patients sustain- ing high-energy craniocervical injuries, altered consciousness, occipital pain or tenderness, impaired cervical motion, lower cranial nerve paresis, or retropharyn- geal soft tissue swelling. Computed tomographic imaging allows the establishment of the diagnosis of OCF and for a precise assessment of fracture displacement. MRI is recommended to assess the integrity of the craniocervical ligaments [8]. Classification Occipital condyle fracture can be distinguished into three types (Fig. 11): Figure 11. Classification of occipital condyle fractures Type I: fractures may occur with axial loading. Type II: fractures are extensions of a cranial basilar fracture. Type III: frac- tures result from an avulsion of the condyle during rotation, lateral bending, or a combination of these mechanisms. Treatment Occipital condyle fractures are usually treated by exter- nal immobilization The choice of treatment depends on the extent of fracture displacement (as seen in CT) and ligamentous injury. Depending on the severity of injury, the treat- ment ranges from collar immobilization to more rigid halo jacket or cast immo- 850 Section Fractures bilization [8]. Patients with untreated OCF may develop lower cranial nerve defi- cits which then require rigid immobilization [8]. However, OCFs are rarely asso- ciated with neurological deficits and can usually be treated conservatively [212]. In 2002, a review of the literature of OCF revealed 47 articles including a total of 91 patients. Based on this review, treatment with external cervical immobiliza- tion is recommended [8]. Although Type III OCFs are considered unstable, not all patients will develop neurological deficits and require surgery [8]. Atlanto-occipital Dislocation Atlanto-occipital dislocation is a rare and often fatal condition Atlanto-occipital dislocation (AOD) is a rare and often fatal traumatic injury that is difficult to diagnose. Immediate death may result from injuries to the brain, spinalcord,andlesionstothevascularstructures,particularlythevertebral arteries [1]. In individuals who have survived the initial injury, the diagnosis is often overlooked because AOD is frequently combined with traumatic brain injury or multiple organ trauma. Patients who survive often have neurological impairment, such as unilateral or bilateral weakness, lower cranial neuropathies, or tetraplegia. The diagnosis is frequently missed on initial lateral cervical X-rays [1]. Interestingly, nearly 20% of patients with acute traumatic AOD will have a normal neurological examination on presentation [1]. Prevertebral soft tissue swelling on a lateral cervical radiograph or craniocervical subarachnoid hemor- rhage on axial CT has been associated with AOD and should increase the suspi- cion of this lesion. CT with 3D image reformation, MRI and angiography are the imaging modalities that will allow the diagnosis of AOD and to exclude addi- tional concomitant injuries [121]. Avulsion fractures of the occipital condyles, apical dens fractures, and a retropharyngeal hematoma may lead to the diagnosis of an AOD [63]. The presence of upper cervical prevertebral soft tissue swelling on an otherwise non-diagnostic plain X-ray should prompt additional imaging [1]. If there is clinical suspicion of AOD, and plain X-rays do not suffice, CT and/ or MRI is necessary [1]. Classification A lateral cervical radiograph is recommended for the diagnosis of AOD to calcu- late the ratio of basion/posterior arch of C1 to anterior arch of C1/opisthion according to Kricun [120] ( Fig. 5c). Three types of AOD can be classified accord- ing to Traynelis [196] ( Fig. 12). A systematic review of the literature published between 1966 and 2001 revealed 48 articles including a total of 79 patients with AOD (29 Type I, 32 Type II, 4 Type III). However, 14 cases were unclassifiable because these fractures were lateral, rotational, and multidirectional dislocations not fitting the three types of Traynelis [196]. Treatment Rule out AOD before applying traction All patients with AOD should be treated [1]. Without treatment, nearly all patients develop neurological deterioration and recovery is unlikely. In the pres- ence of AOD, traction may result in devastating neurological deficits [1]. There- fore, AOD must be ruled out before applying traction. Internal fixation and fusion is indicated in all patients with AOD Therapeutic options aim to stabilize the cervico-occipital junction and to avoid secondary neurological deterioration [185]. Consequently, craniocervical fusion with internal fixation (using a Y-plate or newer generation occipital plate- rod systems) is recommended for the treatment of patients with acute traumatic AOD to allow for early mobilization [1]. Cervical Spine Injuries Chapter 30 851 Figure 12. Atlanto-occipital dislocations Type I: anterior dislocation. Type II: vertical dislocation. Type III: posterior dislocation. Fractures of the Atlas Fractures of the atlas account for approximately 1–2% of all fractures and for 2–13% of all acute cervical spine fractures [94, 129, 179]. Cooper was the first to demonstrate a fracture of the atlas in 1822 at autopsy. In 1920, Jefferson [114] reviewed 42 previously described cases of atlas fractures adding 4 of his own cases. Although his article documentsa variety of atlas fracture patterns, it is best known for the characterization of the “Jefferson fracture,” i .e. , a b ur st f r ac tu re injury of the atlas ring [99]. Acute atlas fractures comprise a large variety of frac- ture types. These fractures are frequently associated with other cervical fractures or ligamentous traumatic injuries [95, 150]. Classification Burst fractures of the atlas are caused by massive axial loads and often occur at the sulcus vertebralis, the weakest site of the arch. These fractures are very fre- quently associated with other fractures of the craniocervical junctions. Accord- ing to Jefferson [114], five types can be differentiated ( Fig. 13). Treatment The extent of lateral mass displacement is decisive for the treatment The treatment of atlas fractures in combination with other cervical fracture inju- ries is most commonly linked to the treatment of the associated injury [95]. The decision for the treatment of atlas fracture depends on the stability of the frac- ture. The main criteria to determine C1–C2 instability due to transverse atlantal ligament injury include the sum of displacement of the lateral masses of C1 com- 852 Section Fractures Figure 13. Classification of atlas fractures pared to C2 of more than 8 mm on plain X-rays (rule of Spence [183] corrected for magnification [102]), a predental space of more than 4 mm in adults [206], and MRI evidence of ligamentous disruption or avulsion [4]. Unstable burst fractures should be treated with rigid external fixation or instrumented fusion The literature does not allow treatment recommendations to be given on solid scientific evidence. So far, treatment options are based on the specific atlas frac- ture type [4]. It is recommended to treat isolated fractures of the atlas with intact transverse alar ligaments (implying C1–C2 stability) with cervical immobiliza- tion alone (rigid collar, halo vest, or Minerva cast) for a duration of 10–12 weeks [4]. It is recommended to treat isolated fractures of the atlas with disruption of the transverse ligament with rigid external fixation (halo vest or Minerva cast) or with atlantoaxial screw fixation and fusion [4]. Atlantoaxial Instabilities Atlantoaxial instabilities are rare after trauma Atlantoaxial instability results from either a purely ligamentous injury or avulsion fractures. While atlantoaxial dislocation and subluxation is relatively common in patients with rheumatoid arthritis [40], a traumatic origin due to a rupture of the transverse ligament is rare [62]. Atlantoaxial dislocations occur more frequently in elderly patients when compared to other traumatic cervical injuries [112]. These injuries are significant, because complete bilateral dislocation of the articular pro- cesses can occur at approximately 65° of atlantoaxial rotation. When the transverse ligament is intact, a significant narrowing of the spinal canal and subsequent potential spinal cord damage is possible [54]. With a deficient transverse ligament, complete unilateral dislocation can occur at approximately 45° with similar conse- quences. In addition, the vertebral arteries can be compromised by excessive rota- tion which may result in brain stem or cerebellar infarction and death [173, 202]. A special form of atlantoaxial instability is referred to as atlantoaxial rotatory subluxations, which may occur with or without an initiating trauma. Non-trau- matic etiologies include juvenile, rheumatoid arthritis, surgical interventions such as tonsillectomy or mastoidectomy, and infections of the upper respiratory tract (“Grisel syndrome”). Cervical Spine Injuries Chapter 30 853 Classification Atlantoaxial instabilities can be classified according to the direction of the dislo- cation as [20]: anterior (transverse ligament disruption, dens or Jefferson fracture) posterior (dens fracture, see Fielding Type IV) lateral (lateral mass fracture of C1, C2, or unilateral alar ligament ruptures) rotatory (see Fielding Types I–III) vertical (rupture of the alar ligaments and tectorial membrane) Rotatory Atlantoaxial Instability Only Types I and II occur as a result of trauma Rotatory injuries of the atlantoaxial joint are a spectrum of rare lesions that range from rotatory fixation within the normal range of C1–C2 motion to frank rota- tory atlantoaxial dislocation [51, 74, 75, 128]. Atlantoaxial rotatory dislocations frequently occur in children but rarely in adults. According to Fielding et al. [74, 75], four types can be differentiated ( Fig. 14): Figure 14. Atlantoaxial rotatory subluxation Type I: rotatory fixation with no anterior displacement (transverse ligament intact) and the dens working as pivot. Type II: rotatory fixation with anterior displacement of 3–5 mm and one lateral articular process acting as the pivot. Type III: rota- tory fixation with anterior displacement of more than 5 mm. Type IV: rotatory fixation with posterior displacement. Type III and IV were only observed in non-traumatic conditions. Treatment Reduction and instrumented fusion is the treatment of choice Anterior dislocations of more than 3 mm are regarded as unstable and usually fail to heal conservatively. Therefore, reduction and atlantoaxial fusion is recom- mended as the treatment of choice [101]. The internal fixation should reduce and prevent further translation of C1 on C2. In both cases, the transarticular screw technique or the C1–C2 fusion technique described by Harms [96] is a good sur- gical option. A Gallie or Brooks fusion should be added to obtain long-term sta- bility. The treatment of posterior and lateral instabilities depends largely on the concomitant injury (e.g., dens fracture). Ver tical instability is treated by an occi- pitocervical fusion [20]. Type I rotatory instabilities are often stable and can be treated by reduction, and rigid external fixation for 4–6 weeks. In recurrent Type I rotatory instabilities as well as in unstable Type II instabilities, an atlantoaxial fusion is indicated [20]. Dens Fractures Themostcommonaxisinjuryisafracturethroughtheodontoidprocess.Atlan- toaxial motion is primarily rotational, accounting for about one-half of the axial 854 Section Fractures rotation of the head on the neck [203]. Translational motion of C1 on C2 is restricted by the transverse atlantal ligaments that center the odontoid process to the anterior arch of C1. With a fracture of the odontoid process, restriction of translational atlantoaxial movement is lost [205]. Classification According to the classification of Anderson and D’Alonzo [19], three types can be differentiated ( Fig. 15): Figure 15. Odontoid fractures Type I: oblique fractures through the upper portion of the odontoid process. Type II: across the base of the odontoid pro- cess at the junction with the axis body. Type III: through the odontoid that extends into the C2 body. Comminuted (Type IIA) fractures are associated with severe instability In 1988, Hadley et al. [94] added a comminuted fracture involving the base of the odontoid as a Subtype IIA. The incidence of a Type IIA fracture was 5% of all Type II fractures. Importantly, Type IIA fractures were associated with severe insta- bility and inability to obtain and maintain fracture reduction and realignment. Treatment A variety of non-operative and operative treatment alternatives have been pro- posed for odontoid fractures based on [5]: fracture type degree of (initial) dens displacement extent of angulation patient’s age Non-operative Treatment The non-operative treatment options consist of: cervical collar traction Cervical Spine Injuries Chapter 30 855 Minerva cast halo jacket Cervical collar is an option for Type I fractures Several authors proposed treatment of odontoid fractures with cervical collars. In a series by Polin et al. [156], 36 Type II fractures were treated either with a Phil- adelphia collar or with halo vest immobilization. The fusion rate was lower in the patients treated with collars compared with patients managed in halos (53% vs. 74%, respectively). The infrequent Type I odontoid fracture seems to have an acceptable rate of fusion with rigid cervical collar immobilization, approaching 100% in one study [19, 47, 49]. Type III odontoid fractures have been treated with cervical collars as well, but the fusion rates are in the range 50–65% in small series. Traction and cervical collars are inappropriate for Type II fractures Reviews by Traynelis [195] and Julien et al. [118] address the treatment of odontoid fractures with traction and subsequent immobilization in a cervical collar. The authors concluded that the non-union rate of Type II dens fractures is almost 50% indicating that traction and cervical collar immobilization is not appropriate for Type II fracture patients. Halo immobilization is an option for Type I and III odontoid fractures Greene et al. [91] reviewed 199 odontoid fractures and reported that successful fusion was obtained with halo vest immobilization in the Type I (100%) and Type III fractures (98.5%). Non-union resulted in 28% of Type II fractures treated with external immobilization for a median of 13 weeks. A displacement of the dens of 6 mm or more was associated with a high non-union rate (86% failure rate), irrespective of patient age, direction of displacement, or neurological defi- cit. Julien et al. [118] reviewed nine articles that dealt with treatment of odontoid fractures (total of 269 patients) using halo/Minerva fixation for 8–12 weeks. The non-union rate for Type I, II and III odontoid fractures was 0%, 35% and 16%, respectively. The high non-union rate of Type II dens fractures is due to inadequate fracture immobilization White and Panjabi [205] have outlined that it is unlikely that the high non- unionrateofTypeIIfracturesisduetoalimitedbloodsupplytothefracture fragments but rather due to the inadequate immobilization of the fracture. Operative Treatment Surgical techniques to stabilize the atlantoaxial joint complex are technically demanding. Proper understanding of the fracture, careful preoperative planning (e.g., CT studies of the anatomical landmarks), adequate knowledge of the surgi- cal anatomy, good intraoperative fluoroscopic control, and precise surgical tech- nique will yield the best results. Based on recent literature reviews [5, 118, 195], TypeIIandTypeIIIodontoidfracturesshouldbeconsideredforsurgicalfixation in cases of: dens displacement of 5 mm or more dens fracture (Type IIA) inability to achieve fracture reduction inability to achieve main fracture reduction with external immobilization Greene et al. [91] have found that patients with dens displacement of 6 mm or more had a non-union rate of 86%, compared with a non-union rate of 18% for patients with displacement of less than 6 mm. The surgical armamentarium consists of: anterior dens screw fixation ( Fig. 16a–d) anterior atlantoaxial screw fixation and fusion ( Fig. 16e, f) posterior atlantoaxial fusion (Gallie or Brooks) ( Fig. 17a–d) posterior atlantoaxial screw fixation and fusion ( Fig. 17e, f) posterior atlas and axis screw-rod fixation and fusion ( Fig. 17g, h) 856 Section Fractures ab cd ef Figure 16. Anterior surgical stabilization of dens fractures Anterior dens screw fixation: a The dens fracture is reduced prior to surgery by traction and patient positioning. Two Kir- schner wires are inserted in an anterior-caudal to posterior-cranial direction. b The Kirschner wires should be convergent but must allow for enough interspace for the insertion of the cannulated screws. c, d Cannulated screws are inserted over the Kirschner wires. When inserting the screw care must be taken that the screw is not angulated to the guide wire in order not to cause breakage or proximal advancement of the guide wire. After screw insertion the wires are removed. e, f Anterior transarticular screw fixation: As an augmentation of the anterior dens screw or in cases of a salvage proce- dure, screws can be inserted over Kirschner wires from a medial-anterior-caudal to a lateral-posterior-cranial direction crossing the atlantoaxial joint. Anterior screw fixation is indicated in Type II fractures Anterior odontoid screw fixation is indicated in Type II fractures with either a horizontal or anterior cranial to posterior caudal direction of the fracture line. In cases in which the fracture line is running in the anterior caudal to posterior cra- nial direction, fracture displacement is likely and therefore a contraindication. This direct osteosynthesis technique aims to maintain rotational motion at the atlantoaxial joint. Transverse alar ligament disruption is a contraindication for anterior screw fixation because of persistent transverse instability. In the review by Julien et al. [118], the fusion rate of Type II fractures treated with anterior screw fixation was 89%. Cervical Spine Injuries Chapter 30 857 . evidence-based frame- work of a critical appraisal of the accumulation of clinical studies and concluded that high-dose methylprednisolone cannot be justified as a standard treatment in acute spinal cord injury. canal and eliminates bony com- pression of the spinal cord. Theoretically, early decompression of the spinal cord after injury may lead to improved neurological outcome. However, indication and. 10c)seemstobethefirst choice for conservative treatment of unstable injuries of the upper cervical spine, although pin track problems, accurate fitting of the vest, and a lack of patient compliance lead to clinical