Preoperative prediction of sagittal imbalance in kyphosis secondary to ankylosing spondylitis after one-level three-column osteotomy

Background This study aimed to determine preoperative predictors for sagittal imbalance in kyphosis secondary to ankylosing spondylitis (AS) after one-level three-column osteotomy. Methods A total of 55 patients with AS who underwent one-level three-column osteotomy were enrolled. The patients were divided into two groups according to sagittal vertical axis (SVA) value at the final follow-up (group A: SVA > 5 cm; group B: SVA ≤ 5 cm). The radiographic measures included global kyphosis, lumbar lordosis (LL), pelvic tilt (PT), pelvic incidence (PI), sacral slope, T1 pelvic angle (TPA), SVA, osteotomized vertebral angle and PI and LL mismatch (PI − LL). Postoperative clinical outcomes were evaluated using Scoliosis Research Society-22 questionnaire (SRS-22) and Oswestry Disability Index (ODI). Results Fifty-five AS patients had an average follow-up of 30.6 ± 10.2 months (range 24–84 months). Group A had larger preoperative and postoperative LL, PT, PI − LL, TPA and SVA values compared with group B ( P < 0.05), and no significant differences were found in ODI and SRS-22 scores between the two groups ( P > 0.05). Preoperative LL, PT, PI − LL, TPA, and SVA values were positively correlated with the follow-up SVA value ( P < 0.05). Among them, TPA > 40.9°, PI − LL > 32.5° and SVA > 13.7 cm were the top three predictors with the best accuracy to predict sagittal imbalance. Immediate postoperative SVA value of ≤ 7.4 cm was a key factor in reducing the risk of sagittal imbalance during follow-up. Conclusions Preoperative TPA > 40.9°, PI − LL > 32.5° and SVA > 13.7 cm could predict sagittal imbalance in AS kyphosis after one-level three-column osteotomy, and additional osteotomies were recommended for this condition. Immediate postoperative SVA ≤ 7.4 cm was an optimal indicator for preventing sagittal imbalance. Level of evidence IV.