Surgical treatment of double major adolescent idiopathic scoliosis can lead to excessive lumbar lordosis in the instrumented segments

Purpose: The purpose of this study was to characterize the sagittal spine in AIS patients with double major curves fused into the lumbar spine to determine the effects of posterior spinal fusion and instrumentation (PSFI) on global and segmental lumbar sagittal parameters. Methods: A consecutive series of AIS patients undergoing a PSFI from 2012 to 2017 having Lenke 3, 4 or 6 curves were analyzed. Sagittal parameters included pelvic incidence (PI), lumbar lordosis (LL), and segmental lordosis were measured. The difference in segmental lumbar lordosis between the preoperative, 6-week, and 2-year radiographs was analyzed and correlated to outcomes using SRS-30 patient questionnaires. Results: Seventy-seven patients had improvement in their coronal Cobb from 67.3 ± 11.8° to 25.43 ± 10.7° (66.4%) at 2 years. There was no change in thoracic kyphosis (23.0 ± 13.4° to 20.3 ± 7.8°) and pelvic incidence (49.9 ± 13.4° to 51.1 ± 15.7°) from preoperative to 2 years (p > 0.05) while lumbar lordosis increased from 57.6 ± 12.4° to 61.4° ± 12.3° (p = 0.002). Segmental lumbar analysis showed increased (+) lordosis at each instrumented level when comparing the preoperative and 2-year films for: T12-L1 (+ 3.24°, p < 0.001), L1-L2 (+ 5.70°, p < 0.001), and L2-L3 (+ 1.70°, p < 0.001). Loss (–) of lordosis was noted at every level below the LIV: L3-L4 (− 1.70°, p < 0.001), L4-L5 (− 3.52°, p < 0.001), L5-S1 (− 1.98°, p = 0.02). Preoperative LL of L4–S1 comprised 70 ± 16% of the global LL compared to 56 ± 12%, at 2 years (p < 0.001). Changes in sagittal measurements did not correlate with SRS outcome scores at two-year follow-up. Conclusion: When performing PSFI for double major scoliosis, global SVA was maintained at 2 years, however, overall lumbar lordosis increased due to an increased lordosis in the instrumented segments and a smaller decrease in lordosis below the LIV. Surgeons should be wary of the tendency to create instrumented lumbar lordosis with a compensatory loss of lordosis below LIV which may be a set-up for poor long-term outcomes in adulthood.