Maintenance of Pelvic Tilt Normalization following Adult Spinal Deformity Corrective Surgery: Analysis of Prevalence, Timing, and Predictors influencing Occurrence
Maintenance of Pelvic Tilt Normalization Following Adult Spinal Deformity Corrective Surgery: Analysis of Prevalence, Timing, and Predictors Influencing Occurrence
Spine Surgeon NYU School of Medicine New Canaan, CT, US
Introduction: Increasing pelvic tilt (PT) is a primary compensatory mechanism in adult spinal deformity (ASD). The driving forces behind this lack of PT-response are not well defined. This study aims to determine the surgical and radiographic parameters associated with and predict maintaining PT normalization after ASD corrective surgery
Methods: ASD patients fused to S1/pelvis included. Patients stratified based on PT: Early Normalization 6 weeks (PTNorm6w); Delayed Normalization 2 years (PTNorm2y); Non-normalized 2 years (Non); Reversal Normalization (RevNorm). Univariate analyses compare normalized (PTNorm6w and PTNorm2Y) and non-normalized (Non and RevNorm) in postoperative alignment (PI-LL, SVA, and GAP score) and clinical outcomes (complication rates, HRQL). Multivariable logistic regression and ROC curve develop a model consisting of significant predictors.
Results: 253 met inclusion criteria (64.6±9.06 years, 79% female, BMI 27.9±5.57 kg/m2, CCI: 1.9±1.6, levels fused 11.8±4.4, EBL: 1841.6mL, op time: 449.6min). mean PT: BL: 30.5º, 6W: 22.5º, 2Y: 25°. By SRS-Schwab, 56.5% (n=143) had a moderate pelvic tilt at baseline and 43.5% (n=110) had severe pelvic tilt at baseline. 37.9% (n=96) were normalized 6W and 6.3% (n=16) normalized 6W and 2Y. PT6WNorm and PT2YNorm were more likely to be overcorrected at 6W (p < .05). GAP score 6W became greater for non-normalized patients (.55 vs 1.34, p=0.08) and 2Y (.93 vs 1.4, p=.49). PT2YNorm had lower rates of implant failure (8.9% vs 19.5%, p < .05), rod breakage (1.3% vs 13.8%, p < .05) and pseudoarthrosis (0% vs 4.6%, p < .05). Total complication rate was significantly lower for those achieving normalization (56.7% vs 66.1%, p=.02). Multivariate regressions showed 88% achieved significantly greater normalization with baseline PI-LL and PI-LL at 6W(OR .82 [.7 - .95], p < .05).
Conclusion : PT normalization occurred in almost 40% of patients by 6 weeks. Normalization is likely to occur where reconstruction addresses lumbopelvic mismatch, extends above the apex of the thoracic kyphosis, and has adequate surgical invasiveness to achieve full correction.