Symptomatic atlantoaxial instability in Down syndrome

Transcription

1 J Neurosurg (Pediatrics 3) 103: , 2005 Symptomatic atlantoaxial instability in Down syndrome ALI NADER-SEPAHI, F.R.C.S. (SN), ADRIAN T. H. CASEY, F.R.C.S., RICHARD HAYWARD, F.R.C.S., H. ALAN CROCKARD, F.R.C.S., AND DOMINIC THOMPSON, F.R.C.S. (SN) Department of Neurosurgery, Great Ormond Street Hospital for Sick Children, London, United Kingdom Object. The aim of this study was to audit the treatment of symptomatic atlantoaxial instability in Down syndrome and to assess factors associated with fusion failure in this group of patients. Methods. The authors reviewed the cases of 12 children with Down syndrome presenting with symptomatic atlantoaxial instability who underwent surgery for internal fixation and fusion. A review of clinical histories, radiological investigations, and details of operative interventions was also performed. The mode of presentation was acute spinal cord injury (five cases), progressive myelopathy (four cases), and neck pain or stiffness (three cases). The atlantodental interval ranged from 5 to 13 mm. The posterior atlantodental interval at the C-1 level was 9.5 mm (range 6 11 mm). In 10 patients an os odontoideum was present. Translocation of the odontoid process occurred in one patient, and occipitoatlantal instability was also identified in two cases. Twenty-four operations were performed in the 12 patients. A transoral odontoidectomy was required in four children. Successful fusion was demonstrated in seven patients at the first operation. Three of the five patients with acute cord injury made significant functional recovery and were left with no or mild disability. Conclusions. Additional bone abnormalities at the CVJ are common in the Down syndrome population. Young age at the time of fusion and multiple osseous anomalies pose a higher risk factor in fusion failure. The authors recommend an aggressive surgical approach for management of symptomatic cases of CVJ instability. KEY WORDS Down syndrome atlantoaxial instability craniovertebral instability os odontoideum pediatric neurosurgery A TLANTOAXIAL subluxation associated with ligamentous laxity is a well-known entity in 15 to 20% of patients with Down syndrome, yet symptomatic atlantoaxial global subluxation is estimated to occur in only 1% of cases. Symptoms may include neck pain, cervical deformity, cervicomedullary compression leading to neurological deficit, and even sudden death. The indications for surgical intervention and the nature of the most appropriate surgical stabilization technique for these children remain controversial. Previous studies have highlighted significant neurological morbidity and particularly high failure rates for fusion in this group of patients. Ours is the largest series to investigate symptomatic atlantoaxial subluxation and its management exclusively in a population of pediatric patients with Down syndrome. We conducted a retrospective review of symptomatic atlantoaxial instability in patients with Down syndrome treated at the Great Ormond Street hospital to accomplish the following: 1) audit our own surgical experience in children with Down syndrome; 2) evaluate the underlying anatomical changes specific in this group; and 3) attempt to identify particular risk factors for failure of surgical stabilization. Abbreviations used in this paper: ADI = atlantodental interval; CT = computerized tomography; CVJ = craniovertebral junction; MR = magnetic resonance; PADI = posterior atlantodental interval. Clinical Material and Methods The medical records and radiographic studies of children with Down syndrome who had symptoms referable to atlantoaxial subluxation and underwent surgical stabilization formed the basis of this study. Demographic information and details of the presenting symptoms and clinical signs were obtained. These data covered a period between 1993 and The radiological studies reviewed included cervical spine x-ray films in flexion and extension, CT scans, and MR images. An ADI (distance between the posterior surface of the anterior arch of C-1 and the anterior surface of the dens) of more than 5 mm and a PADI (distance between the posterior surface of the dens and the anterior surface of the posterior arch of C-1) of less than 14 mm were considered to be abnormal. Two children had undergone previous attempts at fixation at other institutions that had failed. In all the cases in this series, the final operation was performed at our institution. The aim of surgery was stabilization of the unstable atlantoaxial or craniovertebral complex. In cases of irreducible atlantoaxial instability, transoral decompression of the cervicomedullary junction was performed in addition to posterior stabilization. Atlantoaxial subluxation was reduced and halo body vest stabilization achieved prior to the posterior arthrodesis. Posterior stabilization comprised internal fixation by using suboccipital 231

2 A. Nader-Sepahi, et al. TABLE 1 Summary of data on study patients* No. of Age (yrs), ADI PADI Ops Sex (mm) (mm) Fusion Clinical Manifestation Neurological Outcome Group A 1 6, M 7 8 yes acute cord compression following general remained ventilator dependent (IIIB) anesthetic for ear, nose, & throat procedure, ventilator dependent (IIIB) 7, F 11 7 yes neck pain (I) resolution of neck pain (I) 4, F 5 11 yes progressive myelopathy (II) full recovery (I) 6, M 8 11 yes acute cord compression (IIIA) full recovery (I) 14, F 7 11 yes progressive myelopathy (II) no further deterioration of myelopathy (II) 5, F yes neck pain & stiffness (I) resolution of neck pain (I) 15, M 13 9 yes acute cord compression (IIIA) full recovery (I) Group B 2 3, F 5 8 yes neck pain & torticollis (I) resolution of neck pain & correction of torticollis (I) 3 2, M 11 6 yes acute cord compression (II) no further deterioration of myelopathy (II) 6, F 5 8 yes progressive myelopathy (II) gradual improvement, normal gait, & resolution of upper motor neuron signs (I) 4 5, F 6 10 yes acute cord compression, wheelchair mobile w/ a frame, arrest of myelopathy bound (IIIB) (IIIA) 7, F 5 11 yes progressive myelopathy (II) improved neurologically; walking unaided (II) * Ranawat classification for spinal cord deficit. and sublaminar wires, C1 2 transarticular screws, or Cervifix (Synthes, Stratec Medical, United Kingdom) augmented with split calvarial bone, rib, or iliac crest cancellous bone as bone graft material. Postoperative immobilization was performed for a minimum of 3 months in every patient. Every patient underwent fluoroscopic examination to determine the state of fusion before removal of the halo body jacket. The only exception was the case of a 7-year-old child who had transarticular fixation and did not need halovest immobilization. Results There were eight girls and four boys with a median age of 6 years (range 2 15 years). The mode of presentation was variable, ranging from neck pain and stiffness to acute spinal cord compression. Three patients presented with neck pain, stiffness, and torticollis. Headache and pain in the upper and lower extremities were additional presenting features in another patient. Four others had progressive deterioration in gait, spastic paraparesis, unsteadiness, and hemiparesis suggestive of progressive myelopathy. In five patients the onset of symptoms was acute in the form of hypotonia, transient tetraparesis, or tetraplegia; one patient was unable to breathe and was dependent on a ventilator following an iatrogenic injury. This group either had a history of minor trauma or had no trauma at all (Table 1). Radiological Findings The following radiological features were identified on reviewing the plain cervical spine radiographs in flexion and extension, the finecut, bone-windowed CT scans of the CVJ with two- and three-dimensional reconstructions, and the MR images of the CVJ. Ten patients had an os odontoideum (Fig. 1). In three patients the os odontoideum moved with the arch of C-1 as an entity in relation to C-2. Abnormalities of ossification of the arch of C-1 were noted in three patients and comprised an incomplete anterior arch in one, incomplete posterior arch in another, and incomplete anterior and posterior arches (hence, a bipartite atlantal ring) in the third (Fig. 2). Atlantooccipital instability (posterior subluxation) was present in two cases leading to additional compromise of the neuraxis; atlantoaxial subluxation was present in all patients and resulted in ADI ranging from 5 to 13 mm (median 7.5 mm). The median PADI measurement was 9.5 mm (range 6 11 mm). In four cases a fixed deformity was present, with no evidence of reduction on extension. These cases required transoral removal of the odontoid process and excision of the interposing transverse ligament (Fig. 3). Rotatory subluxation resulting in torticollis was present in three patients and was reducible at surgery in all instances. Finally, translocation of the odontoid process into the foramen magnum was a feature in one patient, resulting in cervicomedullary junction compression. On MR imaging, high signal intensity in the neuraxis was present in three cases. The best means of identifying the interposing transverse ligament in cases of irreducible atlantoaxial subluxation was MR imaging (Fig. 4). Surgical Procedure Table 2 summarizes the 24 operations performed in these 12 patients; this figure takes into account four previous failed attempts at posterior arthrodesis in two children prior to referral to our unit. Four transoral procedures to remove an irreducible ventrally compressing odontoid process and intervening transverse ligaments were performed. Posterior fixation took the form of occipitocervical in 10 patients and atlantoaxial in another two. These figures depict only the final operations that led to successful fusion. A bone graft from an autologous calvarial source as described by Casey, et al., 5,6 a rib graft, or cancellous bone from the iliac crest were used as the source of bone grafts. In Table 1, we divided the patients in our series into two groups. Group A con- 232

3 Atlantoaxial instability in Down syndrome FIG. 1. Axial (left) and sagittal reconstruction (right) CT scans of the CVJ (bone windows) demonstrating atlantoaxial instability and the presence of an os odontoideum. tained seven patients in whom fusion was successful after the first operation and Group B included five patients in whom more than one attempt at posterior fixation was required. The mean age of the patients was 8.1 years in Group A compared with 4.6 years in Group B (Table 1). Surgical Outcome We ultimately achieved osseous fixation as judged on flexion extension radiographs in all of our patients (100%). The deformity was corrected in all three patients with torticollis. Three of five patients with acute cord compression recovered and were left with minor neurological deficits or none at all. In the other two patients, we achieved arrest of neurological deterioration and neurological improvement from Ranawat IIIB to IIIA. After induction of a general anesthetic for an ear, nose, and throat procedure, one patient remained quadriparetic and dependent on a ventilator (Table 1). Postoperative Complications We had no postoperative infections. Neurological deterioration occurred in one patient, a 6-year-old girl who presented with headaches and pain in her upper and lower limbs and abnormal gait. Her examination revealed a spastic gait with exaggerated reflexes in her lower limbs. Imaging demonstrated apparently reducible atlantoaxial subluxation, os odontoideum, bifid posterior arch of C-1, and cervicomedullary compression. She underwent atlantoaxial fixation in which an autologous calvarial bone graft was placed. Her neurological condition deteriorated gradually during the postoperative period. Repeated imaging demonstrated persisting ventral compression from the os odontoideum. A transoral odontoidectomy and removal of the transverse ligament were performed. The child later made a full recovery. Her posterior fixation ultimately failed to fuse and she had a successful instrumented posterior fixation with Cervifix and a bone graft. After occipitocervical fusion one patient was found to have a small cerebellar hematoma of no clinical significance. Discussion The incidence of radiological atlantoaxial instability in patients with Down syndrome ranges from 14 to 24%. Dzenitis 10 first described a symptomatic case of atlantoaxial subluxation in Down syndrome. According to published reports, the incidence of symptomatic atlantoaxial subluxation is thought to be less than 1%. 15,27 Only children with clear symptoms referable to atlantoaxial subluxation have been included in this series. Although this issue has received the most attention, our experience and that of others 20,37 indicates that concomitant instability at the atlantooccipital segment is not uncommon. For this reason, we prefer to use the term craniovertebral instability in cases with mixed instability both at the atlantoaxial and the atlantooccipital joints and to emphasize that, particularly in Down syndrome, the mechanical problem can extend beyond the atlantoaxial joint, reflecting a more generalized ligamentum laxity at the CVJ. In addition to ligamentous laxity, patients with Down syndrome have osseous abnormalities at the CVJ such as os FIG. 2. Three-dimensional reconstruction CT scans of the CVJ (bone windows) demonstrating anterior (left) and posterior (right) defects of the arch of atlas. 233

4 A. Nader-Sepahi, et al. FIG. 3. Three-dimensional reconstruction CT scans of the CVJ (bone windows). Preoperative image (left) demonstrating an odontoid peg (a), an os odontoideum (b), and an incomplete anterior arch of C-1 in a case of irreducible atlantoaxial subluxation (c). A postoperative image (right) demonstrating the decompression of the CVJ following a transoral odontoidectomy (d). odontoideum, hypoplastic odontoid process, and abnormal ossification of the arch of C-1. Of the 12 patients in our series, 10 had os odontoideum. Braakhekke, et al., 1 reviewed the literature on 20 cases of Down syndrome with myelopathy in 1985 and found that nine of 20 patients with atlantoaxial instability had odontoid hypoplasia or os odontoideum. Giacomini 13 first described os odontoideum in In a postmortem examination, he found an independent bone cranial to the abnormal dens. The incidence of os odontoideum is higher in patients with conditions characterized by instability at the CVJ, such as Down syndrome, Morquio syndrome, and spondyloepiphysial dysplasia. Doubt has arisen in the past as to whether this was cause or effect. An increasing body of opinion now holds that the os odontoideum is a consequence of the hypermobility at the atlantoaxial joint rather than being a cause of it. 35,36 A traumatic origin for os odontoideum is supported by reported cases in which a normal odontoid process was later replaced by an os odontoideum. Fielding, et al., 12 Menezes, 19 and Sherk and Nicholson 33 have all reported cases in which an intact odontoid process had been visualized radiologically prior to the later appearance of an os odontoideum. In his study of 32 individuals, Burke, et al., 11 reported on seven individuals with atlantoaxial instability, in three of whom odontoid abnormalities developed. Stevens and Crockard and their colleagues 8,35 introduced a new concept of abnormal ossification pattern caused by instability. In a series of patients with Morquio Brailsford disease, they observed regeneration of normal or near-normal odontoid processes and closure of the neural arch defects after occipitocervical arthrodesis. 36 They proposed that the underlying instability and hypermobility from ligamentum laxity creates abnormal shearing stresses at the cartilaginous stage of development. This in turn interferes with the normal ossification process, which extends from the centrum of C-1 into the portion projecting above the articular surfaces of C-2. Eventually, the ossification centers separate and the interposition of the transverse ligament and hypermobility results in established nonunion. Hypermobility at the atlantoaxial complex causes repeated minor trauma, fracturing the odontoid process. The apical portion of the dens is pulled up by the alar ligaments toward the basion and survives on its apical blood supply from the carotid artery through the apical arcade. Because the os is not attached to the base of the odontoid peg, the transverse ligament is incompetent. Interposition of this ligament can lead to an irreducible atlantoaxial subluxation and in established nonunion (that is, an os odontoideum). 35 An os odontoideum is synonymous with instability of the atlantoaxial complex. Our series of patients strongly supports this analysis; 10 of 12 cases of symptomatic atlantoaxial subluxation involved an os odontoideum. The prevalence and type of atlantal anomalies seen in this series may similarly be a reflection of the regional instability at the CVJ, with excessive movement delaying or preventing the ossification of the C-1 ring. Operative management of atlantoaxial instability in Down syndrome is complex and controversial, although most authors would recommend surgery for symptomatic cases. 17,27,29 The atlantoaxial motion segment is the most mobile region of the vertebral column. Various techniques have been used for posterior arthrodesis. The primary aim of these surgical treatments is to provide translational and rotatory stability. Hitherto, the most commonly used method has been sublaminar wiring holding the bone graft in place until fusion is achieved. This technique provides stability in the anteroposterior direction but little rotational stability, which has to be provided by postoperative halo immobilization. This technique is not without risk and significant morbidity rates, and deaths associated with this technique have been reported. 21,24,34 In several studies, patients with Down syndrome had the highest rate of neurological complications or failure of fusion. 7,31,34 Although onlay bone grafts supplemented by sublaminar wiring still have a place in the management of the very young child with atlantoaxial instability, recent developments in spinal instrumentation have led to encouraging results in older children. 3 The risk of neurological complications is especially high where reduction has not been achieved preoperatively, leading to the passage of sublaminar wires in an already compromised canal; therefore, it is essential to establish reduction prior to a posterior fusion procedure. Nordt and Stauffer 24 reported on two patients who became tetraparetic after intraoperative manipulation. Neither patient had preoperative reduction of atlantoaxial subluxation. In our 234

5 Atlantoaxial instability in Down syndrome FIG. 4. A T 2 -weighted sagittal MR image of the CVJ in a case of atlantoaxial instability depicting the anterior arch of C-1 (a), the os odontoideum (b), and the interposing transverse ligament (c). series, incompletely reduced atlantoaxial subluxation resulted in postoperative neurological deterioration, which was corrected in one child after a transoral odontoidectomy. The failure to appreciate irreducibility and ventral compression were deemed to be the foremost reasons for fusion failure in the two patients in this series who had undergone prior surgery at other institutions. Both had undergone two previous attempts at posterior fixation and both needed a transoral odontoidectomy before the successful stabilization procedure. Failure to appreciate coexisting atlantooccipital subluxation can result in an inadequate stabilization procedure so that it becomes necessary to include the occiput in the fusion procedure. In our experience, incorporation of the occiput in the fixation construct is also indicated in a young child, in whom the atlas may be both structurally immature and incomplete and in whom there may have been previous, unsuccessful attempts at atlantoaxial fixation. 20 The choice of posterior fixation depends, therefore, on a variety of factors. We perform atlantoaxial fixation with soft wires and bone grafts when isolated atlantoaxial subluxation exists. We recommend incorporation of the occiput into the fixation construct routinely in cases of craniovertebral instability (that is, concomitant atlantooccipital and atlantoaxial subluxation) and also where it has been necessary to undertake transoral resection of the anterior arch of C-1, the odontoid peg, and the transverse ligament. We recommend that incorporation of the occiput also be considered in cases with preexisting cranial settling and/or basilar impression, where there is a congenital osseous anomaly of the C-1 ring, and in cases with repeated failure of atlantoaxial fusion. When possible (the child s age and therefore bone stock permitting) we prefer instrumented occipitocervical fixation and bone grafting. In both situations, we prefer using a split calvarial bone graft or an iliac crest bone graft for the latter if the child s bone stock allows. Three-point fixation with maximal stability is achieved through addition of internal rotational stabilization to the above techniques. This is achieved using C1 2 transarticular screws, which at the same time obviates the need for halo-vest immobilization. 14,16,23 Originally described by Magerl and Seemann, this technique is reported to achieve fusion rates of 95 to 100%. 9,14 Superior knowledge of the anatomy of this region is required, however, along with a thorough preoperative workup to determine anatomical suitability for placement of transarticular screws. In their study, Paramore, et al., 26 concluded that 18 to 23% of candidates may not be suitable for insertion of transarticular screws on at least one side. In his article, Brockmeyer 2 advocates the use of this technique for pediatric patients. An additional risk of neurological deficits from vertebral artery injury exists as well. 18 Other potential concerns that should not be ignored include limitation of future growth, development of lesions adjacent to the fused level, and determination of the youngest age at which screws can be placed safely. Notwithstanding, Brockmeyer, et al., 3 has shown that in the majority of cases ( 89%) children are anatomically suitable for transarticular screw placement and recommend this procedure for patients as young as 4 or 5 years of age. Only one patient in our series had pure atlantoaxial subluxation and was anatomically suitable for successful fusion by using this technique. Rates of fusion after posterior arthrodesis in patients with Down syndrome vary between studies, ranging from to 100%. 20 In one study, three of four patients with an os odontoideum experienced a failure to fuse. 7 Why surgeries in these patients are so complex is thought to be multifactorial. Segal, et al., 31 postulated that resorption of bone graft in six of their 10 cases arose from an inherent collagen defect in Down syndrome. A deficient immunological response in the initial inflammatory stage after the posterior fusion technique has also been implicated. 1 In young children, the relatively large head and small spine as well as the immaturity of the bones themselves are likely additional factors that contribute to the difficulties in achieving long-term stability. Although not statistically significant, the observation that the children in whom fusion was achieved after a single procedure (Group A) were generally older than those who required multiple procedures supports this contention. In our opinion the osseous abnormalities that appear so prevalent in this group of patients, represent additional risk factors for fusion failure. Two patients had an incomplete posterior atlantal arch, which we believe may have compromised the mechanical stability of the construct, resulting in persistent movement and fusion failure. Also, the presence of an os odontoideum is both indicative of significant instability and may also be a cause of atlantoaxial subluxation irreducibility. We therefore recommend that the potential significance of these features be incorporated into the operative decision-making process as we described previously. The natural history of atlantoaxial instability in Down 235

6 A. Nader-Sepahi, et al. TABLE 2 Stabilization and fusion operations performed for craniovertebral instability in patients with Down syndrome* Op No. Age Rationale for Choice (yrs) 1st 2nd 3rd 4th of Posterior Fixation 6 occipitocervical fixation w/ NA NA NA severe craniovertebral soft wires & SCBG instability 7 C1 2 transarticular screw fix- NA NA NA well-defined lat ation & Gallie fusion: soft masses of C-1 & C-2 wire & ICBG & isolated AA instability 4 occipitocervical fixation w/ NA NA NA severe craniovertebral soft wires & SCBG instability 6 occipitocervical fixation w/ NA NA NA occipitoatlantal instabilsoft wires & SCBG ity 14 occiput C4 Ransford loop, NA NA NA occipitoatlantal instabilwires, & ICBG ity & translocation of the dens 5 Cervifix occipitocervical fix- NA NA NA incomplete ring of C-1 ation & rib graft & severe craniovertebral instability 15 transoral decompression Cervifix occipitocervical NA NA transoral decompression fixation & ICBG 3 AA fixation w/ soft wires AA fixation w/ soft NA NA isolated AA subluxation & SCBG wires & rib graft 2 occipitocervical fixation w/ occipitocervical fixation Cervifix occipitocervical NA severe craniovertebral soft wires & SCBG revised fixation & ICBG instability & failure to fuse on initial 2 attempts 6 AA fixation w/ soft wires & transoral decompression Cervifix occipitocervical NA transoral decompression SCBG fixation & ICBG 7 C1 2 fixation w/ soft wires & C3 4 added to fixation transoral decompression C-1 lamin & oc- transoral decompression bone graft construct ciput C3 soft wire & SCBG 5 C1 2 wire fixation & bone C-1 lamin & Ransford transoral decompression occiput C5 soft transoral decompression graft loop, occipitocervical wire & SCBG fixation * AA = atlantoaxial; ICBG = iliac crest bone graft; lamin = laminectomy; NA = not applicable; SCBG = split calvarial bone graft. Operation performed in another institution prior to referral to Great Ormond Street Hospital. syndrome is debated. Long-term follow up of patients with Down syndrome has revealed very little change in the ADI over many years. Ohsawa, 25 Morton, 22 and colleagues have even reported a reduction in the ADI over a 5-year period. Ohsawa, et al., are the only source, however, to separate the patients with os odontoideum from those with ligamentous laxity. Roy, 30 Selby, 32 and associates found no correlation between the ADI and myelopathic findings. Furthermore, Ferguson, et al., 11 reported no statistical difference between the incidence of symptomatic myelopathy in the subluxator and nonsubluxator groups. At the same time, Ferguson acknowledges that the short duration of their study explained why Pueschel, et al., 28 and Burke, et al., 4 noted a progression of the ADI in some patients after longer follow up. Clements, et al., 6 reported on a patient with documented os odontoideum without instability who by the 5-year follow-up examination had suffered symptomatic atlantoaxial instability requiring surgical stabilization and fusion. Some controversy exists regarding the treatment of the asymptomatic child with Down syndrome found to have atlantoaxial subluxation. Some neurosurgeons use the ADI as a criterion on which to base the decision to operate, but the correlation between the ADI and the risk of developing a neurological deficit is unclear; furthermore the measurements of ADI have been shown to be inconsistent and can vary significantly on repeated radiographs on the same day. 32 Current recommendations in the United Kingdom are that routine screening is not indicated and that intervention is reserved for symptomatic cases. Symptomatic cases present a more straightforward surgical decision. Although this approach has been challenged by some, Ferguson, et al., 11 suggest that the presence of atlantoaxial subluxation does not automatically represent a cause-and-effect relationship. They therefore recommended that not all symptomatic patients should undergo surgery. In their article, Menezes and Ryken 20 highlighted the significance of recognizing atlantooccipital subluxation in decision making for treating these children. From our series, it is clear that presence of other osseous abnormalities such as os odontoideum is a major contributing factor in symptomatic presentation of these children. Ten of 12 patients in our series have os odontoideum as an osseous abnormality. Ferguson, et al., 11 clearly stated in their paper that no osseous abnormality of the odontoid or atlas was found in either group of patients their group studied. Conclusions We recommend treatment for all children with Down 236

7 Atlantoaxial instability in Down syndrome syndrome who have clinical symptoms and radiological evidence of instability at the CVJ. Because ours is a small, retrospective series bearing numerous patient and treatment variables, it is not possible to apply any useful statistical analysis; however, our experience has allowed us to devise some general principles to help in the management of these complex pathological entities. For the older child, instrumented fixation is our preferred treatment of choice. In younger children ( 5 years of age), autologous bone graft fusion with soft wires is warranted. In all cases, it is essential to establish reducibility prior to posterior fixation. Incorporation of the occiput in the fixation is advisable in the following instances: when there is evidence of atlantooccipital instability, when there is a congenital osseous anomaly of the C-1 ring, when atlantoaxial fusion has repeatedly failed, and after transoral odontoidectomy. References 1. Braakhekke JP, Gabreels FJ, Renier WO, van Rens TJ, Thijssen HO, Begeer JH: Cranio-vertebral pathology in Down syndrome. Clin Neurol Neurosurg 87: , Brockmeyer D: Down syndrome and craniovertebral instability. Pediatr Neurosurg 31:71 77, Brockmeyer DL, York JE, Apfelbaum RI: Anatomical suitability of C1 2 transarticular screw placement in pediatric patients. J Neurosurg 92:7 11, Burke SW, French HG, Roberts JM, Johnston CE II, Whitecloud TS III, Edmunds JO Jr: Chronic atlanto-axial instability in Down syndrome. J Bone Joint Surg Am 67: , Casey AT, Hayward RD, Harkness WF, Crockard HA: The use of autologous skull bone grafts for posterior fusion of the upper cervical spine in children. Spine 20: , Clements WD, Mezue W, Mathew B: Os odontoideum congenital or acquired? that s not the question. Injury 26: , Coyne TJ, Fehlings MG, Wallace MC, Bernstein M, Tator CH: C1-C2 posterior cervical fusion: long-term evaluation of results and efficacy. Neurosurgery 37: , Crockard HA, Stevens JM: Craniovertebral junction anomalies in inherited disorders: part of the syndrome or caused by the disorder? Eur J Pediatr 154: , Dickman CA, Sonntag VK: Posterior C1-C2 transarticular screw fixation for atlantoaxial arthrodesis. Neurosurgery 43: ; Dzenitis AJ: Spontaneous atlanto-axial dislocation in a mongoloid child with spinal cord compression. Case report. J Neurosurg 25: , Ferguson RL, Putney ME, Allen BL Jr: Comparison of neurologic deficits with atlanto-dens intervals in patients with Down syndrome. J Spinal Disord 10: , Fielding JW, Hensinger RN, Hawkins RJ: Os odontoideum. J Bone Joint Surg Am 62: , Giacomini C: Sull esistenza dell os odontoideum nell uomo. Gior Accad Med Torino 49:24 28, Grob D, Crisco JJ III, Panjabi MM, Wang P, Dvorak J: Biomechanical evaluation of four different posterior atlantoaxial fixation techniques. Spine 17: , Grobovschek M, Strohecker J: Congenital atlanto-axial subluxation in Down s syndrome. Neuroradiology 27:186, Hanson PB, Montesano PX, Sharkey NA, Rauschning W: Anatomic and biomechanical assessment of transarticular screw fixation for atlantoaxial instability. Spine 16: , Kobori M, Takahashi H, Mikawa Y: Atlanto-axial dislocation in Down s syndrome. Report of two cases requiring surgical correction. Spine 11: , Madawi AA, Casey AT, Solanki GA, Tuite G, Veres R, Crockard HA: Radiological and anatomical evaluation of the atlantoaxial transarticular screw fixation technique. J Neurosurg 86: , Menezes AH: Congenital and acquired abnormalities of the craniovertebral junction (children and adults), in Youmans J (ed): Neurological Surgery, ed 4. Philadelphia: WB Saunders, 1995, Vol 2, pp Menezes AH, Ryken TC: Craniovertebral abnormalities in Down s syndrome. Pediatr Neurosurg 18:24 33, Minderhoud JM, Braakman R, Penning L: Os odontoideum, clinical, radiological and therapeutic aspects. J Neurol Sci 8: , Morton RE, Khan MA, Murray-Leslie C, Elliott S: Atlantoaxial instability in Down s syndrome: a five year follow up study. Arch Dis Child 72: , Naderi S, Crawford NR, Song GS, Sonntag VK, Dickman CA: Biomechanical comparison of C1-C2 posterior fixations. Cable, graft, and screw combinations. Spine 23: , Nordt JC, Stauffer ES: Sequelae of atlantoaxial stabilization in two patients with Down s syndrome. Spine 6: , Ohsawa T, Izawa T, Kuroki Y, Ohnari K: Follow-up study of atlanto-axial instability in Down s syndrome without separate odontoid process. Spine 14: , Paramore CG, Dickman CA, Sonntag VK: The anatomical suitability of the C1 2 complex for transarticular screw fixation. J Neurosurg 85: , Pueschel SM, Scola FH: Atlantoaxial instability in individuals with Down syndrome: epidemiologic, radiographic, and clinical studies. Pediatrics 80: , Pueschel SM, Scola FH, Pezzullo JC: A longitudinal study of atlanto-dens relationships in asymptomatic individuals with Down syndrome. Pediatrics 89: , Rizzolo S, Lemos MJ, Mason DE: Posterior spinal arthrodesis for atlantoaxial instability in Down syndrome. J Pediatr Orthop 15: , Roy M, Baxter M, Roy A: Atlantoaxial instability in Down syndrome guidelines for screening and detection. J R Soc Med 83: , Segal LS, Drummond DS, Zanotti RM, Ecker ML, Mubarak SJ: Complications of posterior arthrodesis of the cervical spine in patients who have Down syndrome. J Bone Joint Surg Am 73: , Selby KA, Newton RW, Gupta S, Hunt L: Clinical predictors and radiological reliability in atlantoaxial subluxation in Down s syndrome. Arch Dis Child 66: , Sherk HH, Nicholson JT: Rotatory atlantoaxial dislocation associated with ossiculum terminale and mongolism. J Bone Joint Surg Am 51: , Smith MD, Phillips WA, Hensinger RN: Complications of fusion to the upper cervical spine. Spine 16: , Stevens JM, Chong WK, Barber C, Kendall BE, Crockard HA: A new appraisal of abnormalities of the odontoid process associated with atlanto-axial subluxation and neurological disability. Brain 117: , Stevens JM, Kendall BE, Crockard HA, Ransford A: The odontoid process in Morquio-Brailsford s disease. The effects of occipitocervical fusion. J Bone Joint Surg Br 73: , Taggard DA, Menezes AH, Ryken TC: Treatment of Down syndrome-associated craniovertebral junction abnormalities. J Neurosurg (Spine 2) 93: , 2000 Manuscript received December 5, Accepted in final form May 17, Address reprint requests to: Dominic Thompson, F.R.C.S., Great Ormond Street Hospital, Great Ormond Street, London WC1N 3JH, United Kingdom. thompd@gosh.nhs.uk. 237