Introduction Charcot arthropathy is a well recognised complication in denervated synovial joints. This is a late complication of traumatic spinal cord injury that is rarely reported in the literature. Early recognition is important and can be difficult as the clinical presentation can vary from pain, deformity, autonomic dysreflexia and audible noises with motion.

Methods We present 5 cases of Charcot’s arthropathy of the spine in patients with in patients with traumatic paraplegia. All patients had spinal surgery to stabilise the spine shortly after the acute injury.

Results The average time from initial injury to presentation with Charcot’s arthropathy was 27 years (range 10–41). A combination of localised and neuropathic pain was the dominant symptom (4 patients) causing re-presentation, but other symptoms included deformity (1 patient). The level of the initial spinal cord injury was at the thoracolumbar junction patients. The Charcot joint level was usually 1 to 2 segments caudal to the spinal fusion. The features noted on plain radiology were destructive changes of the endplate in 4 patients and deformity in 1. With one exception, all patients went on to have MRI to exclude infection and subsequently all were surgically stabilised. All patients were treated surgically. One had an anterior approach, one had posterior approach and one had staged anterior and posterior approaches. The remaining two had anterior and posterior stabilisation through a lateral extra-cavitary approach.

At an average follow-up of 36 months all patients reported good relief of their symptoms, and had returned to their best function post-injury.

Discussion Surgical stabilisation in this series yielded very good results. We observed a wide variation in presenting symptoms and therefore would indicate that a high index of suspicion is required. We believe that MRI is mandatory to exclude infection and would advocate early stabilisation. The lateral extra-cavitary approach allows posterior and anterior stabilisation in a single procedure and in now the preferred method in our institution. As patients with spinal cord injuries live increasingly active lives, this problem will be seen more frequently.

Introduction Both congenital and acquired aetiologies for Os Odontoideum have been proposed leading to confusion in a medico-legal setting. No large series with CT and MRI features has been reported.

Methods Clinical, radiological, CT and MRI data from 25 cases of cranio-cervical anomalies (including 18 with Os Odontoideum) was collected prospectively. The hypothesis that Os Odontoideum is associated with signs of dysplasia such as arch hypertrophy (as quantified by the arch-peg ratio) and a ‘jigsaw’ like atlanto-dens joint configuration was tested.

Results Only one of 18 cases with Os Odontoideum gave a history of remote trauma. None of the cases with other post traumatic conditions showed arch hypertrophy or dysplasia of the joint. Os Odontoideum was associated with two distinctive features on mid-sagittal CT reconstruction: arch hypertrophy and the presence of a ‘jigsaw sign’. A quantitative definition of arch hypertrophy (an arch/ peg ratio greater than 0.5) provided a sensitive and specific test for Os Odontoideum. This was further enhanced in combination with the jigsaw sign to give a test accuracy of 97%. Motor changes with MRI signs of cord damage were observed in one of 12 cases associated with trauma.

Conclusions The lack of a history of remote trauma and the dyplastic atlanto-dens joint seen on CT are more suggestive of a congenital than a post-traumatic aetiology for Os Odontoideum. Patients with Os odontoideum are able to tolerate moderate to severe levels of injury without sustaining significant acute cord damage.

Introduction: This is a prospective study of a series of consecutive cases of Os Odontoideum focussing on CT and MRI data. Both congenital and post-traumatic aetiologies have been proposed in the literature. This can lead to confusion in a medico-legal and clinical setting.

Methods: Radiological, CT and MRI data from 26 consecutive cases of cranio-cervical anomalies were collected prospectively. Demographic details, the presence of any recent or remote traumatic aetiology and the clinical presentation were obtained from the medical record. A reconstructed mid-sagittal CT was examined for the thickness of the arch of C1, the size and location of the Os and the morphology of the atlantodens joint. The presence of any cord impression or signal change was obtained from the MRI.

Results: 18 cases of Os Odontoideum were identified. Only one had a history of significant trauma remote from presentation. All adults had an abnormal archdens joint configuration (the ‘Jigsaw’ sign) with one exception. The atlantodens ratio was significantly greater in all cases of Os odontoideum indicating a relatively thickened anterior arch of the atlas. One case of non-union of a dens fracture presenting five or more years after the injury was identified in this series. Neither in this case nor two cases of transverse ligament rupture and two cases of Ossiculum terminale, was a thickened arch or an abnormal atlantodens joint observed. 12 of the cases presented after traumatic injury to the neck. In only three of these was there any abnormal neurological signs.

Conclusions: Os Odontoideum has a characteristic appearance of the anterior C1 arch and the atlantodens joint as viewed on CT. These radiological signs are not observed after dens fractures. They may be taken to indicate a congenital aetiology for the condition. Patients with Os odontoideum are able to tolerate moderate to severe levels of injury without sustaining significant acute cord damage.