Effect of lumbar pedicle subtraction osteotomy level on lordosis distribution and shape.

Purpose: Little is known about the qualitative results (postoperative upper/lower lumbar arches distribution and lumbar apex or inflection point positioning) of lumbar pedicle subtraction osteotomies (L-PSO) depending on the level of L-PSO.
Methods: We conducted a retrospective analysis of prospectively collected data of adult deformity patients undergoing single-level L-PSO. We analyzed several variables in preoperative and postoperative sagittal radiographs: L-PSO level, Roussouly classification (R-type), inflection point (InfP), lumbar apex (LApex), spinopelvic parameters, lordosis distribution index (LDI = L4-S1/L1-S1), and number of levels in the lordosis (NVL). Comparisons between PSO levels were performed to determine lordosis distribution and sagittal shape using ANOVA test and Chi-squared statistics.
Results: A total of 126 patients were included in this study. L5-PSO mainly increased the lower lumbar arch, thereby increasing LDI. L4 increased upper/lower arches similarly. PSOs at and above L3 increased the upper lumbar arch, thereby decreasing LDI (P < 0.001). L4-PSO added 1 vertebra into the lordosis (NVL =  + 1.2 ± 2.2). PSOs above L3 added 2 vertebrae into the lordosis (NVL =  + 2.3 ± 1.4). Overall P = 0.007. PSOs above L4 shifted the LApex cranially in 70% of the cases (mean 1.12 levels) and the InfP in 85% of the cases (mean 2.4 levels). L5-PSO shifted the LApex caudally in 70% of the cases (mean - 1.1 levels) and the InfP in 50% of the cases (mean - 1.6 levels). Overall P < 0.006. The L-PSO level was not associated with a specific Roussouly-type P > 0.05.
Conclusions: The level of L-PSO influenced upper/lower lumbar arches distribution, and lumbar apex and inflection point positioning. The correct level should be chosen based on the individual assessment of each patient.