Clinical Application of the Roussouly Classification in the Sagittal Balance Reconstruction of 101 Adolescent Idiopathic Scoliosis Patients

PURPOSE: Although Roussouly classification has been widely used in spinal surgery, it was mainly applied to degenerative scoliosis patients and correlational studies concerning adolescent idiopathic scoliosis (AIS) are still insufficient. This retrospective study explored the clinical application of Roussouly classification in surgeries and prognosis prediction for AIS. METHODS: This clinical research selected 101 AIS patients who received surgeries between August 2005 and November 2019. Whole spine standing radiographs were obtained for each patient preoperatively, postoperatively, and at the last follow‐up (>24 months). All patients were classified into “theoretical types” and “current types.” Patients were further divided into mismatch or match groups based on the consistency of their current type and theoretical type. The main parameters include: proximal junctional angle (PJA), pelvic incidence (PI), sacral slope (SS), pelvic tilt (PT), fixed thoracic kyphosis (TK), global TK, fixed lumbar lordosis (LL), global LL, thoracic tilt, proximal thoracic alignment (PTA), lumbar tilt, spino‐sacral angle (SSA), and spinal tilt (ST). RESULTS: A total of 47.5% of AIS patients were subject to a preoperative mismatch of Roussouly classification. There was a significant difference in PI‐LL between the preoperative mismatch and match groups (p = 0.008). There was a significant difference in the rate of PI‐LL deformity between the match and mismatch groups with a preoperative mismatch (p = 0.037). A significant difference in thoracic tilt was observed between the postoperative mismatch and match groups (p = 0.019). The preoperative mismatch group has a higher risk of postoperative PI‐LL malformation than match group (OR = 2.303, 95% CI: 1.026, 5.165). When mismatch occurred postoperatively, there were significant differences between groups in the rate of pelvic deformity (p = 0.002) and PI‐LL deformity (p = 0.025) at the last follow‐up. Compared with the postoperative match group, mismatch group had an increased risk of pelvic deformity (OR = 5.029, 95% CI: 1.618, 15.629) and PJK deformity (OR = 3.017, 95% CI: 1.709, 11.375) at the last follow‐up. Short Form‐36 and Scoliosis Research Society 22 score of the match group was significantly higher than that of the mismatch group at the last follow‐up. CONCLUSION: The Roussouly classification mismatch before or after operation leads to increased risks of PI‐LL deformity and pelvis deformity postoperatively or at the follow‐up, which seriously worsens the clinical symptoms and prognosis of patients. Therefore, recovering to the theoretical type in Roussouly classification may effectively improve patients' prognosis.