Introduction

With the use of each pedicle screw for surgical correction of adolescent idiopathic scoliosis (AIS), there is an increase in instrumentation-related costs, operative time, risk of neural injury, and overall health-care expenses. As such, alternate level screw strategy (ALSS) has been reported as a potential alternative to contiguous multilevel screw strategy (CMSS). Moreover, studies have shown the importance in accounting for the flexibility of the curve based on the fulcrum bending radiograph when assessing postoperative curve correction. Therefore, this study addressed a radiographic and cost analysis comparing CMSS with ALSS for the treatment of thoracic AIS with titanium screws and rod application.

Introduction: Anterior convex epiphysiodesis and posterior concave distraction has not been previously described in the literature for the treatment of thoracolumbar hemivertebrae. We describe our experience with long-term follow-up.

Method: Six consecutive patients with a mean age of 3.4 years were operated on with this technique. Levels of fusion extended two levels above and below the hemi-vertebra, while the instrumentation span the full length of the curve. Further concave distraction was carried out when there was evidence of loosening of the hooks.

Results: The average follow-up was 10.8 years (range 8 to 14). The mean Cobb angle before surgery was 49°, and at the latest follow-up was 26°. There was a mean 41% improvement in the scoliosis. In 5 of these cases, this correction was achieved immediately after surgery and did not significantly change despite repeated distraction.

Conclusion: The addition of concave distraction provided better correction than convex epiphysiodesis alone. It is technically easier and safer than hemivertebra excision in the correction of such deformities. This method of treatment is recommended for patients with single fully segmented hemivertebrae located at the thoracolumbar junction that has a significant deformity.

The FBCI has been shown to be a better method for describing scoliosis correction because it takes spinal flexibility into consideration. 1

Objective: To use FBCI prospectively to compare the efficacy of four different posterior instrumentations in the correction of thoracic scoliosis.

Method: 123 idiopathic scoliosis patients with thoracic curves were surgically treated prospectively using 4 different posterior instrumentations: TSRH (n=35); ISOLA (n=33); CD-Horizon (CD-H: n=32); and Moss-Miami (MM: n=23). All the operations were performed by the same team of surgeons using standard techniques. The curve was measured using the Cobb’s method on the pre-operative PA standing, fulcrum bending and 1-week post-operative PA standing radiographs. The conventional correction rate and the FBCI were calculated. One-way ANOVA and independent sample t-test were used for statistical analysis.

Results: (1) There were no significant differences between any of the 4 instrumentations when assessed using the FBCI, however, the correction rate was better in CD-H than in ISOLA and TSRH (Table 1). (2) Higher FBCIs were observed in the stiff curve group (fulcrum flexibility £ 50%) compared with those in the flexible group (fulcrum flexibility > 50%), while the correction rates were lower in the former than in the latter (Table 2).

Discussion: Better correction rate obtained in the CD-H group was attributed to the more flexible curves rather than the instrumentation itself. In the flexible curve group, the instrumentations have been able to take up all the flexibility revealed by the fulcrum-bending radiograph. Although the correction rate was less in the stiff curve group, the FBCI showed that the deformity correction was actually more than that indicated by the fulcrum bending radiographs. One possible explanation of this phenomenon may be that the fulcrum-bending radiograph is less effective in eliciting all the flexibility in the stiff curves.

Conclusion: All 4 instrumentations were EQUALLY effective in correction of thoracic scoliosis when the curve flexibility was taken into consideration.

In scoliosis, it is well known that lateral deformity is coupled with vertebral axial rotation. Coupled motion in the sagittal plane, however, has not been investigated.

Objective: To investigate the behavior of the sagittal alignment changes when coronal deformity was corrected in idiopathic thoracic scoliosis.

Method: 36 idiopathic scoliosis patients with thoracic curves were examined before surgery. Coronal deformity was corrected using the Fulcrum Bending technique1, and biplane radiographs were taken to monitor the correction of the deformity, as well as the coupled sagittal alignment changes. Sagittal alignment was measured from T4/T5 to T12 using Cobb’s method. Difference of less or equal to 3 degrees between two measurements was treated as no change. Results were compared with those measured from standing lateral radiographs prior to and at 1 week after surgery (Posterior correction and fusion with ISOLA: n=15; CD-Horizon: n=8; Moss-Miami: n=11, USS: n=2). Pearson correlation was used for statistical analysis.

Results: (A) When scoliosis was corrected under fulcrum bending, the coupled changes in the thoracic kyphosis were decreased if it was greater than 20 degrees (n= 18), increased if less than 20 degrees (n= 2), and kept no change if it was around 20 degrees (n= 16). These changes were not related to the amount of deformity or flexibility in the coronal plane (Table I). (B) There was strong relationship between the sagittal alignment measured on the pre-operative fulcrum bending and postoperative lateral radiographs (P< 0.01). However, the final sagittal alignment was neither correlated with the magnitude or flexibility of the coronal deformity, nor the instrumentation applied (P> 0.05)

Discussion: A coupling exists between the coronal lateral deformity and the sagittal alignment in thoracic scoliosis. It seems that the sagittal alignment in a scoliotic spine tends to “normalize” with correction of the deformity: a “hyper-kyphotic” spine tends to reduce, and a “hypo-kyphotic” one tends to increase the kyphosis. Post-operative sagittal alignment seems to be decided by the coupling motion and the amount of curvature of the pre-bent rod, as neither the nature (degree or flexibility or curve pattern) of the coronal deformity nor the choice of instrumentation were related to the post-operative sagittal alignment.