Apical vertebral rotation in adolescent idiopathic scoliosis: comparison of uniplanar and polyaxial pedicle screws.

STUDY DESIGN: Retrospective radiographic outcome analysis.

OBJECTIVE: To compare residual postoperative apical vertebral rotation between uniplanar versus polyaxial bilateral pedicle screw constructs in thoracic adolescent idiopathic scoliosis (AIS).

SUMMARY OF BACKGROUND DATA: Vertebral rotation is a component of the 3-dimensional deformity of AIS. The correction of vertebral rotation is an important goal of surgery. A comparison of uniplanar versus polyaxial screws has yet to be studied.

METHODS: A review of a multicenter database of Lenke type 1 to 3 AIS patients who underwent surgical correction of thoracic AIS by posterior segmental pedicle screw instrumentation and fusion with either uniplanar or polyaxial pedicle screws and 5.5-mm steel rod constructs was performed. Curves greater than 100 degrees were excluded to control for the propensity to use polyaxial screws to correct larger curves. Postoperative apical vertebral rotation of thoracic curves was graded as 0, 1, or 2 based on a computed tomography scan-validated radiographic method that uses the relative position of the screw tips to grade apical vertebral rotation at the 6-week and 1-year postoperative visits. By this grading method, higher grades correspond to greater residual apical vertebral rotation.

RESULTS: Two hundred and ten patients met the inclusion criteria. The uniplanar screw group included 95 patients whereas the polyaxial screw group had 115 patients. The respective mean preoperative thoracic Cobb angle of 58±12 and 60±13 degrees (P=0.1), first erect postoperative coronal correction of 72% and 74% (P=0.38), and 1-year correction of 70% and 76% (P=0.07) were not significantly different between the uniplanar and polyaxial groups. At 6 weeks postoperation, the uniplanar group had 34% of patients with grade 0 rotation, 52% with grade 1, and only 14% with grade 2 thoracic apical vertebral rotation. In the polyaxial group, only 14% of patients were grade 0, 35% were grade 1, and 51% were the most rotated grade 2. This was a significant difference in the distribution of the axial rotation grades (P<0.001), with less residual apical vertebral rotation with the use of uniplanar screws. The same pattern of results was found at 1-year postoperative evaluation (P<0.001).

CONCLUSIONS: There was little difference in the coronal plane correction of thoracic curves between the 2 types of screws. However, the uniplanar pedicle screw group had a larger proportion of patients with greater thoracic apical vertebral derotation (less residual apical vertebral rotation) compared with the polyaxial screw group. This can be attributed to the increase in rotational leverage afforded by uniplanar screws during intraoperative bilateral direct apical vertebral derotation maneuvers.