Comparison of the lowest instrumented, stable, and lower end vertebrae in "single overhang" thoracic adolescent idiopathic scoliosis: anterior versus posterior spinal fusion.

Study Design: A retrospective multicenter study.

Objective: To investigate the relationship between the lowest instrumented, stable, and lower end vertebrae in patients with "single overhang" thoracic (main thoracic) curves treated with anterior or posterior spinal fusion.

Summary Of Background Data: Previous studies have shown "saving" fusion levels with anterior spinal fusion, as opposed to posterior spinal fusion; however, to our knowledge, none of these studies evaluated the relative position to the lower end vertebra to compare study groups accurately. For clarification, "single overhang" includes Lenke 1A and 1B curves. For these thoracic curves, the lumbar curve does not cross the midline.

Materials And Methods: A retrospective multicenter study of adolescent idiopathic scoliosis was performed to identify specifically patients with "single overhang" thoracic (Lenke 1A and 1B) curves with more than a 2-year follow-up. To analyze relative fusion levels, the differences were computed as follows: (1) the difference between the vertebra position for the stable vertebra of the main thoracic (MT) curve and the lowest instrumented vertebra, as noted on postoperative radiographs, or [equation: see text] (2) the difference between the vertebra position for the lower end vertebra of the main thoracic (MT) curve and the lowest instrumented vertebra, as noted on postoperative radiographs, or [equation: see text].

Results: A total of 298 "single overhang" thoracic curves (148 Lenke 1A, 150 Lenke 1B) were identified, of which 293 had either an anterior spinal fusion or posterior spinal fusion; 5 patients underwent a combined anterior-posterior spinal fusion. Anterior spinal fusion was performed in 70 patients (23.9%) and posterior spinal fusion in 223 (76.1%). While comparing the lowest instrumented vertebra to the stable vertebra with anterior spinal fusion, the lowest instrumented vertebra was identified either at the level of the stable vertebra or above in 97% of 1A/B curves (P < 0.001). Using posterior spinal fusion techniques, the lowest instrumented vertebra was identified either at the stable vertebra or above in 65% of the 1A/B curves (P < 0.05).

Conclusions: These data confirm that anterior spinal fusion techniques result in a mean shorter fusion of 1.5 vertebral segments/patient when compared to posterior spinal fusion techniques with respect to the position of the lowest instrumented and stable vertebrae for "single overhang" thoracic (Lenke 1A/B) curves. However, because this is a retrospective multicenter study over 10 years, it represents various posterior spinal fusion techniques that do not include all pedicle screw constructs.