Platt, Andrew

OBJECTIVE The objective of this study was to compare a multiple pelvic screw fixation strategy (dual bilateral 4 pelvic screw fixation [4PvS]) with the use of single bilateral 2 pelvic screw fixation (2PvS), with the aim of addressing lumbosacral junction stability. METHODS This analysis is a single-center, retrospective review of ASD patients treated between 2015 and 2021. All patients had a minimum 2-year follow-up and spinal fusion to the sacrum without sacroiliac fusion and met at least one radiographic and procedural criterion: pelvic incidence–lumbar lordosis ≥ 20°, T1 pelvic angle ≥ 20°, sagittal vertical axis ≥ 7.5 cm, scoliosis ≥ 50°, three-column osteotomy, or spinal fusion of ≥ 8 levels. Two sacropelvic fixation methods were compared: 4PvS versus 2PvS. Primary outcomes included spinal implant-related reoperation and screw breakage, while secondary outcomes included reoperation for symptomatic pelvic screws, screw loosening or bending, L5–S1 pseudarthrosis, and patient-reported outcomes. Propensity score matching and inverse probability of treatment weighting (IPTW) were used to minimize selection bias and estimate causal treatment effects. Clinical outcomes were assessed using conditional multivariable logistic regression. RESULTS In this study of 406 patients (67.98% female, mean age 64.48 years), 349 patients (85.96%) received 2PvS and 57 (14.04%) received 4PvS. Age (OR 1.081, 95% CI 1.027–1.145) and total number of osteotomies (OR 1.180, 95% CI 1.048–1.355) emerged as independent predictors of receiving the 4PvS technique. In unmatched cohorts (n = 406), 2-year implant-related reoperation rates (p = 0.1896) and pelvic screw breakage rates (p = 0.2498) were not significantly different between groups. However, in the 4:1 propensity score–matched cohort, the 2-year reoperation rate (10.53% for 2PvS vs 3.51% for 4PvS; OR 3.27, 95% CI 1.10–9.74 [p = 0.0312]) and the pelvic screw breakage rate (9.21% for 2PvS vs 3.51% for 4PvS; OR 2.87, 95% CI 1.08–7.63 [p = 0.0349]) were significantly higher among the 2PvS groups. The IPTW analysis confirmed these findings, with reoperation rates of 10.45% for 2PvS and 1.18% for 4PvS (p = 0.0244) and pelvic screw breakage rates of 8.72% and 1.18%, respectively (p = 0.0477). A safety assessment revealed comparable operative times and intra- and perioperative complications between the two techniques. CONCLUSIONS Patients who underwent 4PvS demonstrated significantly lower 2-year implant-related reoperation and pelvic screw breakage rates compared with 2PvS, with no differences in intraoperative or perioperative complications.

BACKGROUND AND OBJECTIVES: Adult spinal deformity (ASD) with fixed sagittal malalignment (FSM) may require a pedicle subtraction osteotomy (PSO) for greater focal lordosis and restoration of global alignment. Despite growing trends in minimizing PSOs given their associated high risks, a considerable portion of patients with ASD still require a lumbar PSO most commonly because of iatrogenic flat back deformity. The purpose of this article is to describe a modified extended PSO technique with additional anterior column support coined the “sandwich” extended PSO (SE-PSO) to promote arthrodesis and report the outcomes in a consecutive case series. METHODS: Patients with ASD treated with a lumbar SE-PSO at a single institution from 2015 to 2020 were analyzed. Complications, radiographic data, and patient-reported outcomes were compared preoperatively, at immediate postoperative follow-up, and at a 2-year postoperative follow-up (FU). RESULTS: Fourteen patients who underwent revision operations for FSM were included. Improvements in segmental lordosis across the PSO site (14.8 ± 6.8 vs 39.9 ± 7.1, P < .0001), overall lumbar lordosis (14.6 ± 15.4 vs 44.6 ± 12.1, P < .0001), sacral slope (21.0 ± 10.5 31.1 ± 10.7, P = .0150), C7 sagittal vertical axis (140.1 ± 59.0 mm vs 35.9 ± 28.5, P < .0001), and spinopelvic mismatch (52.5 ± 21.3 vs 18.6 ± 14.1, P = .0001) were obtained in all patients. Eight patients experienced perioperative complications, with intraoperative durotomy being the most common (n = 7). Eight patients had a 2-year FU and demonstrated improvements in their segmental lordosis across the PSO site (14.3 ± 7.0 vs 41.3 ± 7.3, P = .0003), overall lumbar lordosis (8.7 ± 17.8 vs 46.1 ± 14.2, P = .0014), sacral slope (19.1 ± 12.8 vs 32.3 ± 12.5, P = .0479), C7 sagittal vertical axis (173.6 ± 54.4 mm vs 35.8 ± 30.0, P < .0001), and spinopelvic mismatch (63.0 ± 19.7 vs 21.1 ± 18.3, P < .0001), all of which were maintained at final FU ( P > .05). At 2 years, a significant increase in Scoliosis Research Society-22r total score (2.5 ± 0.8 vs 3.6 ± 0.7, P = .0023 was reported. There were no reports of symptomatic pseudarthrosis or mechanical complications. CONCLUSION: SE-PSO is an effective technique to correct FSM and is associated with low complications, improved patient-reported outcomes, and spinopelvic parameters that are maintained at 2 years.