Serial casting as one of the applications to treat early-onset scoliosis has been reported efficiently to improve deformity, but no report has focused on the efficacy of braces in the treatment of congenital early-onset scoliosis and comparison with progressive idiopathic early-onset scoliosis. Patients with progressive EOS treated with braces in our institution with a minimum of 4 years follow-up were reviewed. Two groups according to the etiological diagnosis were analyzed and compared: the congenital scoliosis (CS) group and idiopathic scoliosis (IS) group. The success cases and the failure cases were also compared. 27 patients with an average main Cobb angle of 38.19° (20-55) underwent initial bracing at an average age of 55.7 months (24-108), the average follow-up time was 76.19 months (49-117). In IS group the main Cobb angle was corrected to 18.69 ± 12.06° (48.61%) following the first bracing; the final Cobb angle was 23.08 ± 22.15°(38.76%) after brace removal. In CS group the main Cobb angle was corrected to 33.93 ± 10.31°(17.1%) following the first bracing and 37.93 ± 14.74°(3.53%) after brace removal. Both coronal chest width and T1-T12 height increased dramatically from pre-bracing to the last follow-up. Patients diagnosed as IS tended to have a better result in main Cobb angle correction than that of CS (P = 0.049). By the time of last follow-up, 8 patients had undergone surgery, and the operation time was postponed by 68.88 ± 26.43 months. For patients with progressive early-onset scoliosis, bracing is an efficient nonsurgical alternative to casting, and some of them can be cured; if not, eventual surgical intervention can be delayed for a period of time without restrictions on the thoracic cavity.

METHODS: In vitro animal model.
OBJECTIVE: To compare the strength of 4 different anchor constructs commonly used as foundations in growing spine surgery.
BACKGROUND: Children with progressive early-onset scoliosis often require surgical intervention to control the deformity and allow continued growth. The foundation sites of growing spine constructs take a significant load and can fail. This study compares the strength of 4 commonly used constructs applying the same load in a porcine model.
METHODS: Forty immature porcine specimens including soft tissues (10 per group) were instrumented with 1 of 4 bilateral proximal anchors at T5-T6. The four groups were: screw-screw (SS), lamina hook-hook (HH), rib hook-hook (RR), and transverse process to lamina hook-hook (TPL). The entire specimen was kept intact except for surgical site exposure. A unique fixture was designed to brace the specimen and provide a counterforce. The ultimate load was identified as the greatest load recorded for a construct and analyzed by a set of 1-way analysis of variance using the SPSS 12.0 statistical package.
RESULTS: All specimens eventually failed at the bone-anchor interface. No failures were observed in the instrumentation used. The means and standard deviations of ultimate loads were measured as RR (429 ± 133 N), SS (349 ± 89 N), HH (283 ± 48 N), and TPL (236 ± 60 N). There was no statistically significant difference between the following construct pairs: RR/SS, SS/HH, and HH/TPL. Young\'s modulus was calculated for each construct type and no statistically significant difference was determined.
CONCLUSIONS: This study showed that RR and SS constructs had the greatest ultimate strength but also the greatest variability among the foundations tested. However, the HH and TPL constructs had lower ultimate strength but were less variable. Rib-based anchors may be considered as an alternative in upper foundation constructs in growing rod techniques.