A retrospective clinical review of 20 children seen during a 7-year period who had atlantoaxial rotatory subluxation.

To define the effectiveness of imaging and treatment measures and to identify risk factors for recurrence, the series was reviewed to analyze cause, management, and outcome.

Rotatory subluxation of the atlantoaxial complex remains a poorly understood entity. Despite many reports in the literature, there is no consensus about which imaging studies should be used for diagnosis and which patients benefit from collar immobilization, traction, or surgical fusion.

Between August 1990 and April 1997, 20 children with atlantoaxial rotatory subluxation were treated. Fourteen patients (70%) were girls and six (30%) were boys (mean age, 6.4 years). All patients had torticollis and neck pain with decreased cervical motion for a mean of 11.2 days before diagnosis. Seven patients (35%) had a history of pharyngitis or otitis media, four (20%) had recently undergone head or neck surgery, and four (20%) had sustained a traumatic injury; in five patients (25%), no clear cause was determined. All patients were neurologically intact and underwent plain cervical radiographs and dynamic cervical computed tomography to document atlantoaxial rotatory subluxation. Patients were then treated with a rigid cervical collar and anti-inflammatory agents (n = 5) or with cervical traction followed by immobilization (n = 15).

In four of the five patients in collars, reduction occurred spontaneously, whereas the fifth required cervical traction and eventual fusion for recurrence. In the 16 patients treated with traction (median, 1.8 kg), the normal atlantoaxial alignment was restored in 15 patients (94%) within a mean of 4 days. Of the 20 patients treated overall, conservative management failed in 6 (30%), and they required posterior fusion because of recurrence of the atlantoaxial rotatory subluxation or unsuccessful reduction. The major factor predicting the failure of conservative management was the duration of subluxation before initial reduction. Patients with long-standing subluxation were more likely to experience recurrence and require surgery. There were no complications noted. At follow-up, all patients who were treated conservatively remained neurologically intact with a normal atlantoaxial relation. All patients who underwent surgery remained neurologically intact and had radiographic documentation of fusion.

Optimal management of atlantoaxial rotatory subluxation entails early diagnosis with plain cervical radiographs and dynamic computed tomography. Closed reduction with cervical traction followed by rigid immobilization accomplished reduction in 15 of 16 patients (94%) and was curative in 10 of 16 patients (63%). Although reduction was achieved more rapidly and effectively with traction than with a collar, there may be a role for simple immobilization without reduction in patients with a short duration of symptoms. There does not appear to be a correlation between cause of atlantoaxial rotatory subluxation, age, or sex and the likelihood of recurrence.