Pediatric congenital atlantoaxial dislocation: differences between the irreducible and reducible varieties

Abstract

Clinicoradiological differences and outcome following surgery in pediatric patients (< or = 16 years of age) with congenital irreducible atlantoaxial dislocation (IAAD) and reducible atlantoaxial dislocation (RAAD) were analyzed. Ninety-six patients (57 with IAAD and 39 with RAAD) were categorized as follows: Grade I, no deficits except hyperreflexia or neck pain (six patients); Grade II, minor deficits but independent for activities of daily living (25); Grade III, partially dependent (30); and Grade IV, totally dependent (35). Patients with RAAD underwent direct posterior fusion, and those with IAAD were treated with transoral decompression and posterior fusion. Patients with good outcomes included those who could walk unaided, with improvement in spasticity and weakness, and those who maintained Grade I status. The category of poor outcome included patients with the following conditions: postoperative deterioration or lack of improvement; inability to ambulate regardless of neurological improvement at a minimum follow-up duration of 3 months; or perioperative death. A significantly higher incidence of C-1 assimilation, C2-3 fusion, asymmetrical occiput-C2 facet joints, and basilar invagination were seen in patients with IAAD, and os odontoideum was noted in those with RAAD (p < 0.05). A good outcome was recorded in 35 patients with IAAD and 22 with RAAD, whereas 14 with IAAD and nine with RAAD had a poor outcome (eight patients in each category were lost to follow up). Radiological differences in the anatomy of patients with IAAD and those with RAAD may be due to improper segmentation of the occipital and upper cervical sclerotomes in the former and dysfunction of the transverse ligament in the latter. A significantly better outcome was noted in completely dependent patients with IAAD compared with those with RAAD. Respiratory compromise was an important prognostic factor.