Posterior-only Hemivertebra Resection for Congenital Cervicothoracic Scoliosis

STUDY DESIGN: A retrospective study. OBJECTIVE: To evaluate radiographic and cosmetic outcomes after posterior-only cervicothoracic hemivertebra resection and instrumentation. SUMMARY OF BACKGROUND DATA: Cervicothoracic hemivertebra is a rare congenital deformity. It locates between the mobile cervical spine and the fixed thoracic spine, leading to rapid curve progression, shoulder imbalance, fixed torticollis, and facial asymmetry. METHODS: This study reviewed a consecutive series of 18 patients with cervicothoracic hemivertebra. All received posterior-only hemivertebra resection and instrumentation and had a minimum follow up of 2 years. Assessments on radiographic and cosmetic outcomes were based on changes in measurements of scoliosis, shoulder balance, neck tilt, head shift, and sagittal profiles. RESULTS: There were 24 hemivertebrae, consisting of 16 fully segmented and 8 semisegmented. Mean age at surgery was 9.5 ± 3.1 years, and mean follow up was 32 ± 10 months. Mean fusion segments were 4.8 ± 0.6 segments. Operation time averaged 208 ± 33 minutes, with an average blood loss of 384 ± 40 mL. Local scoliosis was corrected from 39 ± 5° to 16 ± 4° (P < 0.001). The distal curve had a spontaneous correction of 41 ± 18%. With regards to shoulder balance, both T1 tilt and clavicle angle were significantly improved, with a correction rate of 55 ± 22% and 47 ± 32%, respectively. Accordingly, neck tilt was improved from 20 ± 7° to 11 ± 7°, and head shift from 22 ± 9 mm to 13 ± 8 mm. A remarkable pedicle screw malpositioning rate (20%) was observed, but no neurovascular injuries. One case developed with Horner syndrome, and another one had transient radicular pain on the right arm after surgery. CONCLUSION: For patients with congenital cervicothoracic scoliosis, posterior-only hemivertebra resection with instrumentation allows for excellent scoliosis correction and cosmetic improvement. Great care should be taken to reduce the rate of pedicle screw malpositioning. LEVEL OF EVIDENCE: 4.