Introduction. How translation of different parts of spine responds to selective thoracic fusion has not been well investigated. Furthermore, how posterior pedicle-screw-only constructs affect spontaneous lumbar curve correction (SLCC) remains unknown. In a retrospective study, we aimed to investigate the balance change after selective thoracic fusion in Lenke 1C type adolescent idiopathic scoliosis (AIS) treated with posterior pedicle-screw-only constructs. Methods. All AIS cases, surgically treated between 2002 and 2008 in our institute, were reviewed. Inclusion criteria were: patients with Lenke 1C scoliosis treated with posterior pedicle-screw-only constructs; the lowest instrumented vertebra (LIV) ended at L1 level or above; and a minimum 2-year radiographic follow-up. Standing anteroposterior (AP) and lateral digital radiographs from different timepoints (preoperative, immediately postoperative, 3 months postoperative, and final follow-up) were reviewed. In each standing AP radiograph, centre sacral vertical line (CSVL) was drawn first, followed by measurement of the translation (deviation from the CSVL) of some key vertebrae, such as the LIV, LIV+1 (the first vertebra below LIV), LIV+2 (the second vertebra below LIV), LIV+3 (the third vertebra below LIV), lumbar apical vertebra, thoracic apical vertebra, and T1. Additionally, the Cobb angles of major thoracic and lumbar curve were measured at different timepoints, and the correction rate was calculated. Furthermore, clinical photos of patients' back appearance were taken preoperatively and postoperatively. Results. Of the 278 patients reviewed, 29 met the inclusion criteria. The continuous follow-up of our study revealed an interesting occurrence. Criteria were established to define this finding, and the validity of the criteria was verified by a series of tests. The results of our study showed that the selective thoracic fusion was prone to cause left imbalance in patients with Lenke 1C AIS. Even though some patients regained spinal balance by postoperative spinal alignment remodelling, there were still about a third who remained imbalanced at 2 year follow-up. Smaller thoracic curve before surgery did not show higher risk of causing postoperative imbalance. Conclusions. Selective thoracic fusion is prone to cause left spinal imbalance in patients with Lenke 1C scoliosis. Postoperative spinal alignment remodelling can make some patients regain spinal balance. The postoperative spinal balance condition in patients with this disorder could be improved by selecting LIV at stable vertebrae or above, or by checking balance condition during surgery to prevent overcorrection. Additionally, to establish the presence of postoperative spinal alignment remodelling, the criteria we recommend is that thoracic apical vertebra shifts towards the right side by more than 5 mm 2 years after surgery
