Abstract
BACKGROUND: Adolescent idiopathic scoliosis (AIS) surgery typically involves posterior spinal fusion (PSF) using rods contoured by the surgeon, which may be time-consuming and may not reliably restore optimal sagittal alignment. However, pre-contoured patient-specific rods may more optimally restore sagittal spinal alignment. This study evaluates the radiographic outcomes of AIS patients who underwent PSF utilizing surgeon contoured vs. pre-contoured rods. METHODS: This is a retrospective cohort study of AIS patients who underwent PSF with either surgeon contoured or pre-contoured rods. Demographics, Lenke classification, fused levels, osteotomies, estimated blood loss (EBL), and surgical time were also obtained via chart review. Coronal curve magnitude, T5-T12 thoracic kyphosis (TK), lumbar lordosis (LL), pelvic incidence (PI), PI-LL mismatch, and T1 pelvic angle (TPA) were obtained pre-operatively, postoperatively and at last follow up. Outcome measures included rate of achievement of postoperative radiographic alignment goals (TK between 20 and 40 degrees, PI-LL mismatch within 10 degrees, and TPA <14 degrees). Predicted post-operative sagittal alignment was also compared with observed measurements. Student's and paired t-tests were performed to determine significant mean differences for continuous variables, and chi-square for categorical variables. RESULTS: No differences were found in demographics, Lenke classification, preop radiographic measurements, fused levels, osteotomies, EBL, and surgical time in the surgeon contoured cohort (n=36; average follow up 11.3 months) and pre-contoured cohort (n=22; average follow up 9.7 months). At last follow up, 95.5% of patients with pre-contoured rods vs. 61.1% of patients with surgeon contoured rods (P=0.004) met TK goal. During assessment of first standing postoperative X-ray, 72.7% of patients with pre-contoured rods vs. 33.3% of patients with surgeon contoured rods met PI-LL mismatch goal (P=0.004). Other radiographic measurements were similar. Artificial intelligence (AI) predicted and observed differences for the pre-contoured group were 3.7 for TK (P=0.005), -7.6 for PI-LL mismatch (P=0.002), and -2.6 for TPA (P=0.11). CONCLUSIONS: AI and pre-contoured rods help achieve global sagittal balance with high accuracy and improved kyphosis restoration and PI-LL mismatch than surgeon contoured rods in AIS patients.