Differences in the rate of distal junctional problems between L3 and L4 pedicle subtraction osteotomy

Pedicle subtraction osteotomy (PSO) is a powerful technique for sagittal plane deformity correction. The authors aimed to investigate the differences in radiographic outcomes and rates of distal junctional problems (DJPs) between L3 and L4 PSOs.

Patients who underwent L3 or L4 PSO at a quaternary care center between 2005 and 2021 were retrospectively identified. DJPs were defined as either hardware failure or pseudarthrosis distal to the PSO level.

In total, 116 patients were included: 86 (74.1%) underwent L3 PSO and 30 (25.9%) underwent L4 PSO. The mean imaging follow-up was 4.1 (range 1.0-10.9) years. There were no statistically significant differences in age, sex, BMI, operative time, and estimated blood loss. Preoperatively, there were no significant differences in mean sacral Hounsfield units and spinopelvic parameters, with the exception of pelvic incidence (PI; L3: 51.1° ± 11.2° vs L4: 57.9° ± 14.1°, p = 0.012) and the L1 pelvic angle (L3: 23.6° ± 10.1° vs L4: 34.8° ± 13.5°, p < 0.001). Postoperatively, there were no statistically significant differences in primary rod type, 2-rod versus multirod constructs, unilateral versus bilateral iliac fixation, number of levels fused, graft material, L5-S1 interbody fusion approach, and PI-lumbar lordosis mismatch. There were no significant differences between the cohorts in uni- versus bilateral pelvic fixation or type of fixation (iliac vs S2AI); however, patients who underwent L4 PSO had, on average, more pelvic screws placed (mean 1.9 ± 0.7 vs 1.5 ± 0.6, p = 0.002). L4 PSO resulted in larger postoperative L4-S1 segmental lordosis (37.2° ± 13.3° vs 21.4° ± 11.4°, p < 0.001) and reduced rates of postoperative low lordosis distribution index (20.0% vs 60.0%, p < 0.001). There were no significant differences in postoperative complication rates including CSF leak, iatrogenic dorsiflexion weakness, and 30- or 90-day readmissions. The L4 PSO cohort experienced lower DJP rates (6.7% vs 29.1%, p = 0.012), including hardware failure (3.3% vs 20.9%, p = 0.024) and pseudarthrosis (3.3% vs 25.6%, p = 0.008). Multivariate analysis found that multirod construct versus dual-rod configuration (OR 0.31, 95% CI 0.09-0.96) and L4 PSO (OR 0.18, 95% CI 0.02-0.80) were independently associated with decreased DJP rates. Age was also a risk factor for DJPs. The number of pelvic screws and pelvic screw fixation type did not predict DJPs.

In addition to multirod configurations, L4 PSO resulted in a lower rate of DJPs compared with L3 PSO. This result might be due to a more physiological distribution of lumbar lordosis with L4 PSO.