The 3D Sagittal Profile of Thoracic Versus Lumbar Major Curves in Adolescent Idiopathic Scoliosis

Abstract

Study DesignRetrospective. ObjectiveTo compare the 3D sagittal profile of patients with main thoracic or thoracolumbar/lumbar adolescent idiopathic scoliosis (AIS) to a normal cohort. Summary of Background InformationThoracic AIS is often associated with a loss of kyphosis. Classically, this measure has been made in 2D, which may underestimate the true sagittal deformity. MethodsBiplanar upright radiographs were obtained on 152 primary thoracic (TH: Lenke 1–4), 50 primary thoracolumbar/lumbar (TL/L: Lenke 5–6) curves, and 89 normal controls (NC). 3D spinal reconstructions were created using sterEOS software. MATLAB code was used to create segmental measurements of kyphosis/lordosis for each vertebral and disc segment from T1 to S1 in the local coordinate system of each motion segment. Comparisons were made between groups for the 3D summed segmental measures (T1–T5, T5–T12, T12–S1), pelvic incidence, sacral slope, and pelvic tilt. ResultsMean 2D Cobb was 57°±12° (range 40°–115°) for TH curves and 52°±9° (range 37°–75°) for TL/L curves. Significant differences in 3D sagittal measures were found between the 3 groups. Post hoc tests revealed significant differences at T1–T5, TH<NC, and TL/L<NC. All groups differed from each other from T5–T12, with the least kyphosis in TH curves. T12–S1 lordosis was significantly greater in TH and TL/L curves compared with NC. Lumbar lordosis extended proximally an average of one segment in AIS compared to normal spines (T11 vs T12). Pelvic incidence, sacral slope, and pelvic tilt were significantly greater for TH curves compared to NC. ConclusionsThere is a substantial average loss of thoracic kyphosis (∼15°–25°) for both primary thoracic and primary thoracolumbar/lumbar AIS curves compared to normal adolescents. Three-dimensional assessment of scoliosis allows the “true” deformity to be measured by correcting for error due to out-of-plane measurement associated with conventional 2D measurements. Level of EvidenceLevel II, prognostic.