109 - Interpreting 2-Dimensional and 3-Dimensional Alignment in Adolescent I...

General Session: Biomechanics - Hall F

Presented by: N. Frangella

Author(s):

S. Ramchandran(1), A. Sure(1), J. Moon(1), P. Zhou(1), N. Frangella(1), L. Steinmetz(1), T. Errico(1), A. Buckland(1)

(1) New York University Langone Orthopedic Hospital, Division of Spinal Surgery, New York, NY, United States

Abstract

Introduction: Adolescent Idiopathic Scoliosis (AIS) is often associated with thoracic apical hypokyphosis. By nomenclature, thoracic kyphosis (TK) is measured as the difference in tilt between T1-T12, but does not consider the shape of the thoracic apex. Previous studies have described coronal and sagittal alignment using 2-Dimensional (2D) radiographs, which may not project orthogonal to the true plane of TK, nor account for the difference between apical- and overall TK. The advent of 3Dimensional (3D) imaging has redefined the concept of measurement of TK and by utilizing these 3D reconstructions, our study found that while T1-T12 tilt increased perioperatively, true kyphosis, as measured at the thoracic apex, decreased. The aim of this study was to analyze and compare 2D and 3D pre- and post-operative alignment in AIS.

Methods: Retrospective Radiographic Study at a Single Tertiary Academic Center. Radiographic analysis of AIS patients with Type 1 curves at baseline and at 6-month follow-up visit was performed. 2D and 3D radiographic measurements were performed on standing stereoradiographs using validated software. The thoracic ratio was created and defined as the ratio between the vertical distance from the superior endplate of the T1 to the inferior endplate of the T12 (Td), and the orthogonal transverse distance between the center of apical vertebral body and the line Td (Ta), as measured in 3D. Comparison was made between baseline 2D and 3D alignment, pre- and post-operatively.

Results: 22 AIS patients (mean age 15.9 years, 19 females) were included. Comparison of 2D vs. 3D alignment showed a significant difference in preoperative TK (32.6 vs 24.6°, p=0.01), but not post-operatively after curve de-rotation (31.5 vs 33.3°, p=0.47). Perioperative alignment significantly reduced thoracic Cobb angle (17.4 vs 42.2° p< 0.005) and Apical. Vertebral Rotation (4.5 vs 11.4°, p=0.002). 3D TK increased from baseline to post-op (24.6 vs 33.3°, p=0.041). However, while the length of the thoracic spine (Td) increased (24.5 vs 23.3cm, p< 0.001), transverse apical distance (Ta) decreased (2.7 vs 3.9cm, p< 0.001). This represents a decrease in true kyphosis as measured at the apex, as seen with an increase in Thoracic Ratio (Td/Ta, 11.1 vs 7.1, p< 0.001).

Conclusions: Our study shows that measurement of thoracic kyphosis as measured in 2D imaging differs to that in 3D due to the obliquity of the plane of kyphosis to the radiograph beam. Although post-operative increase in the TK was noted in 3D using the T1-T12 angle, the actual kyphosis, as determined by the thoracic ratio reduced. The noted decrease in thoracic kyphosis may imply that lung volumes are not necessarily improved by corrective AIS surgery.