STUDY DESIGN: Retrospective cohort study.

OBJECTIVES: To investigate spinopelvic alignment and spine shape in patients surgically treated for adolescent idiopathic scoliosis (AIS) and to assess the distribution and clinical applicability of the Roussouly classification.

SUMMARY OF BACKGROUND DATA: How spinopelvic alignment is affected in AIS patients is not well established. Roussouly et al. proposed a classification based on the sagittal spinal profile and spinopelvic alignment that may have clinical utility in these patients.

METHODS: A consecutive cohort of 134 surgically treated AIS patients were included. Whole-spine standing lateral radiographs were analyzed preoperatively, one-week postoperatively and at two-year follow-up. Patients were categorized using the modified Roussouly classification and analyzed for sagittal alignment.

RESULTS: Postoperatively, global thoracic kyphosis (TK) decreased by 2.6° and lumbar lordosis (LL) decreased by 6.2°(p ≤ .012) while Pelvic tilt (PT) increased 1.4° (p = .024). At two-year follow-up, TK and LL had returned to preoperative values (p ≥ .346) while PT had decreased from preoperative 9.7 ± 7.6° to 7.0 ± 7.5° (p > .001). Proximal junctional angle increased from 8.4 ± 5.0° preoperatively to 12.8 ± 8.9 (p < .001). Preoperatively, Roussouly curve types were distributed equally apart from a lower rate of type 1 (12%). At final follow-up, 30% were categorised as type 3 with pelvic anteversion which is considerably higher than the normal adolescent population. Only three patients were type 1 at the final follow-up. Overall, we found a high rate of proximal junctional kyphosis (16%), PI-LL mismatch (60%) and pelvic anteversion (38%). In preoperative type 1 patients, the rate was 50%, 82% and 64%, respectively.

CONCLUSION: We found that immediate postoperative changes in lordosis and kyphosis were reversed at final follow-up and found evidence of proximal junctional kyphosation and pelvic anteversion as the main compensatory mechanisms. Poor sagittal alignment was frequent in type 1 curves, and surgical treatment may need to be individualized according to the sagittal profile.

LEVEL OF EVIDENCE: III.