STUDY DESIGN

Analysis of radiographic outcomes following surgical correction of scoliosis.

OBJECTIVES

To compare the curve correction and derotation following anterior spinal fusion (ASF) versus posterior spinal fusion (PSF) with thoracic pedicle screws.

SUMMARY OF BACKGROUND DATA

The benefits of ASF in adolescent idiopathic scoliosis include saving distal fusion levels and historically greater correction and derotation compared with PSF. However, comparative studies between ASF and PSF have generally consisted only of posterior hook instrumentation or hybrid constructs, with no direct comparisons between anterior fusion and thoracic pedicle screw (TPS) series.

METHODS

We performed a retrospective review of the radiographic and medical records of 40 patients (two curve-matched groups) with Lenke Type I main thoracic adolescent idiopathic scoliosis. There were 20 patients who underwent open ASF with single-rod instrumentation with a mean age at surgery of 15 years + 6 months (range, 12-20 years) and 20 patients who underwent PSF with TPS constructs with a mean age at surgery of 13 + 6 (range, 12-15). Radiographic follow-up averaged 44.1 month (24-80) for the ASF group and 55.1 month (25-83) for the PSF/TPS group. We evaluated the sagittal alignment, Cobb angles, rib hump deformity (RH), apical rib spread difference (ARSD), and apical vertebral body-rib ratio (AVB-R), measures of rotation and thoracic torsion, between both groups.

RESULTS

Before surgery, the main thoracic curve was 55.1 degrees (range, 47-66 degrees) for the ASF group and 52.5 degrees (range, 46-68 degrees ) for the PSF/TPS group (P = 0.16). Additionally, there was no difference in the pelvic tilt curves, thoracic kyphosis, lumbar lordosis, RH, or ARSD. However, there was a slightly greater preoperative thoracolumbar-lumbar (TL/L) curve (34.6 degrees versus 29.5 degrees , P = 0.04) and AVB-R (1.75 versus 1.5, P = 0.003) in the ASF group. After surgery, an average of 6.5 levels (range, 6-8) were fused in the ASF group, compared with 7.7 levels (range, 5-12) in the PSF/TPS group (P = 0.001) or 1.2 additional levels for PSF/TPS. At final postoperative follow-up, spontaneous pelvic tilt curve correction was greater in the ASF group (47% versus 35%), although this difference did not reach statistical significance (P = 0.07). For the main thoracic and TL/L curves, there was greater correction in the PSF/TPS group (62% versus 52%, P = 0.009; and 56% versus 41%, P = 0.03), respectively. Additionally, the PSF/TPS group demonstrated significantly greater RH correction (51% versus 26%, P = 0.005) and AVB-R ratio improvement (73% versus 32%, P < 0.0001). We also noted a trend towards increased correction of the ARSD in the PSF/TPS group (58% versus 32%, P = 0.07). Further, the postoperative thoracic kyphosis decreased 4.4 degrees in the PSF/TPS group (postop avg. 25.0 degrees ) and increased 5.7 degrees (average, 30.6 degrees ) in the ASF group (P = 0.04).

CONCLUSIONS

In this curve-matched cohort of Lenke Type I curves, PSF with TPS provided superior instrumented correction of main thoracic curves and spontaneous correction of TL/L curves. Perhaps more importantly, PSF/TPS demonstrated improved correction of thoracic torsion and rotation as compared with ASF in terms of RH (P = 0.005) and AVB-R ratio (P= 0.0001), with only one additional spinal segment fused on average.