Congenital reducible atlantoaxial dislocation: Classification and surgical considerations

Background. Reducible atlanto-axial dislocation (AAD) may cause severe motor and respiratory compromise due to recurrent spinal cord and/or brain stem impingement. To the best of the authors' knowledge, this is the first study concentrating on the classification, the protocol of the surgical management and the outcome of congenital, reducible AAD. Methods. 109 patients with congenital, reducible AAD underwent posterior stabilization. Their preoperative disability was graded as: I (n = 11, 10.09%) no functional disability (a history of minor trauma led to quadriparesis that subsequently improved); 11 (n = 3 1, 28.44%) independent for activities of daily living with minor disability; III (n = 42, 38.53%) partially dependent on others for their daily needs; and, IV (n = 25, 22.93%) totally dependent. They were classified into 4 groups depending upon their association with: a normal odontoid and posterior arch of atlas (n = 27); a dysplastic odontoid and normal posterior arch (n = 25); an assimilated posterior arch (n = 49); and, Arnold Chiari malformation type I (n = 8). Nine patients with a dysplastic, odontoid had a "hypermobile" AAD with an unrestricted backward and forward movement of the axis relative to the atlas in flexion as well as in extension of the neck, respectively. The surgical procedures included Brooks' (n = 12) or modified Brooks' C1-2 fusion (n 39); Goel's C1-2 fusion (3); Ransford's contoured rod fusion (n 7); Jain's occipitocervical fusion (n = 47); and, transoral decompression and Jain's occipitocervical fusion (n = 1). There were 6 peri-operative mortalities in the series. Findings. At follow-up (ranging from 3 months to 6 years; n = 86), 64 patients had shown improvement by one grade or more; 8 patients, who had a history of transient quadriparesis but were without neurological deficits at presentation, remained in grade 1; I I had achieved stabilization of neurological functions; while 3 had deteriorated despite adequate radiological reduction of AAD and fusion of the construct. A follow-up of 6 months or more was available in 79 of these 86 patients, in whom a dynamic intrathecal CT scan showed a good osseous union. Interpretation. The patients with congenital reducible AAD, depending on their surgical management, may be classified into four groups. Some patients with a dysplastic odontoid have a "hypermobile" AAD and require special care during intubation, positioning and stabilization. An assimilated posterior arch is often associated with asymmetrical lateral occipito-C1-C2 joint synostosis rendering transarticular screw placement difficult. The various causes of failure of constructs are discussed.