STUDY DESIGN: A prospective study. OBJECTIVE: To compare the derotational effect of the two methods and to determine the effect of the position of axis of the rotation on derotation of the apical vertebrae. SUMMARY OF BACKGROUND DATA: Vertebral derotation about z-axis following a posterior instrumentation and its relation to the position of the rotational axis is still controversial. Rod derotation(RD) method rotates the vertebrae about the axis of the rod curvature located relatively anterior position whereas the vertebrae to rod(VTR) method, reducing the vertebrae to the contoured rod, rotates the vertebrae about the posteriorly located axis. MATERIALS AND METHODS: Eleven consecutive thoracic idiopathic scoliosis subjected to segmental pedicle screw instrumentation were analysed. Six were treated by RD and five by VTR. Average preoperative curve was 46.6 delta in RD and 51 delta in VTR with flexibility of 69% and 71% respectively (p>0.05). Mean preoperative relative apical vertebral rotation(RAVR) measured by computerized tomography were 11.2 delta in RD and 13.8 delta in VTR(p>0.05). RESULTS: Average postoperative curve magnitudes were 11.5 delta in RD and 12 delta in VTR with correction rates of 77% and 74% respectively (p>0.05). Postoperative relative apical vertebral rotation(RAVR) were 3.6 delta in RD and 6.1delta in VTR with correction rates of 68% and 56% respectively (p>0.05). The mean instrumentation time per vertebral segment instrumented was 4.7 minutes in RD and 8.5 minutes in VTR (p<0.05). Screw pullout during operative procedure in 8/51 screws(15%) in VTR and none(0/60) in RD. CONCLUSION: RD and VTR methods were not significantly different, both enabling a significant apical z-axis derotation and frontal curve correction. However, RD was more efficient than VTR with less operative time and intraoperative screw loosening. The position of the axis of rotation did not significantly influence the apical derotation effect of segmental pedicle screw instrumentation.