Study Design: Retrospective cohort study. Objective: To evaluate radiographic predictors of success for L3 LIV selection in posterior spinal fusion (PSF) for idiopathic scoliosis (IS). Summary of background data: Determining the optimal lowest instrumented vertebra (LIV) in adolescent idiopathic scoliosis (AIS) surgery remains a challenge. Stopping at L3 ( vs . L4) preserves motion, but may increase distal decompensation risk and secondary pain. Methods: A single institution, multisurgeon database of idiopathic scoliosis (IS) was queried to identify PSF with LIV of L3. Exclusion criteria were insufficient radiographic follow-up (<2 yr), growth-friendly constructs, and revisions. A total of 101 patients were identified for study inclusion. Preoperative scoring criteria included: (1) central sacral vertical line (CSVL intersecting or falling medial to L3 pedicle), (2) direction of L3–L4 disc opening, (3) ≤1 Nash-Moe grade difference between L3 and L4, and (4) L3 stacking on L4-L5-S1 in nonweight-bearing radiographs. Success was defined as coronal balance (C7 plumb line ≤20 mm from CSVL) and L3-L4 disc wedging ≤10°. Results: Among 101 patients (mean age 14.8 yr, 87% female, mean follow-up 2.5 yr), 77.2% achieved radiographic success; no reoperations. The main thoracic deformity preoperatively was 51.2°, which improved postoperatively to 22° (57% improvement). The average lumbar curve of 43° preoperatively improved to an average of 15.5° (64% improvement). Individual criteria were met in 43% (n=43) for CVSL, 39.6% (n=40) for opening, 93.1% (n=94) for rotation, and 54.5% (n=55) for stacking. When the CSVL criteria and one other criteria was met, the success rate increased to 91.2% ( P =0.0744). Conclusion: The radiographic success with a LIV of L3 in PSF for IS was greatest in patients who met the CSVL criteria (86.1%). In addition, radiographic success increased with patients who met CSVL criteria and one additional criterion (92.1%). The data suggests using the CSVL as a key measure to optimize radiographic outcome when determining LIV of L3 versus L4. Evidence level: Level III.
Braithwaite et al. (Mon,) studied this question.