Background: Few radiographic guidelines are available to assist clinicians in deciding when to order magnetic resonance imaging in patients with a normal history and physical examination. Most of the recent reports on the radiographic characteristics of scoliosis are limited by a small number of patients and a shortage of large curves. The association between radiological features and the severity of scoliosis has little been elaborated. The purpose of this study is to further explore the radiological presentations in relation to curve severity in scoliosis associated with Chiari malformation and syringomyelia.
Methods: A total of 87 children and adolescents were divided into 3 groups: group 1 (10 degrees ≤ Cobb angle ≤ 30 degrees), group 2 (30 degrees < Cobb angle ≤ 60 degrees), and group 3 (Cobb angle > 60 degrees). Curves were classified into typical and atypical patterns in the coronal plane, and the sagittal profile was measured. Cerebellar tonsillar descent or syrinx patterns in relation to curve severity and the frequency of atypical curves were also investigated.
Results: The frequency of atypical curve patterns from groups 1 to 3 was 46.2%, 45.2%, and 40.7%, respectively. A total of 65.3% of patients with typical curve patterns had atypical features in all of the 3 groups. There was a significant difference of kyphotic angle among the 3 groups showing that the larger curves tended to have greater thoracic kyphosis. Both the degree of cerebellar tonsillar descent and syrinx patterns had no correlation with the curve severity or the frequency of atypical curves.
Conclusions: These results show that radiographic presentations including atypical curve patterns, atypical features in typical curve patterns, and a normal to hyperkyphotic thoracic spine may suggest the need for a preoperative magnetic resonance imaging. Kyphosis may be indicative of progressive scoliosis. There is no evidence to suggest that the degree of cerebellar tonsillar descent and syrinx patterns have an effect on the progress of scoliosis and the frequency of atypical curves.
Level of Evidence: Systematic review 2 of Level III studies.