Upper instrumented vertebra–femoral angle and correlation with proximal junctional kyphosis in adult spinal deformity

Introduction Although matching lumbar lordosis (LL) with pelvic incidence (PI) is an important surgical goal for adult spinal deformity (ASD), there is concern that overcorrection may lead to proximal junctional kyphosis (PJK). We introduce the upper instrumented vertebra–femoral angle (UIVFA) as a measure of appropriate postoperative position in the setting of lower thoracic to pelvis surgical correction for patients with sagittal imbalance. We hypothesize that a more posterior UIV position in relation to the center of the femoral head is associated with an increased risk of PJK given compensatory hyperkyphosis above the UIV. Methods In this retrospective cohort study, adult patients undergoing lower thoracic (T9–T12) to pelvis correction of ASD with a minimum of 2-year follow-up were included. UIVFA was measured as the angle subtended by a line from the UIV centroid to the femoral head center to the vertical axis. Patients who developed PJK and those who did not were compared with preoperative and postoperative UIVFA as well as change between postoperative and preoperative UIVFA (deltaUIVFA). Results Of 119 patients included with an average 3.6-year follow-up, 51 (42.9%) had PJK and 24 (20.2%) had PJF. Patients with PJK had significantly higher postoperative UIVFA (12.6 ± 4.8° vs. 9.4 ± 6.6°, p = 0.04), deltaUIVFA (6.1 ± 7.6° vs. 2.1 ± 5.6°, p < 0.01), postoperative pelvic tilt (27.3 ± 9.2 vs. 23.3 ± 11, p = 0.04), postoperative lumbar lordosis (47.7 ± 13.9° vs. 42.4 ± 13.1, p = 0.04) and postoperative thoracic kyphosis (44.9 ± 13.2 vs. 31.6 ± 18.8) than patients without PJK. With multivariate logistic regression, postoperative UIVFA and deltaUIVFA were found to be independent risk factors for PJK (p < 0.05). DeltaUIVFA was found to be an independent risk factor for PJF (p < 0.05). A receiver operating characteristic (ROC) curve for UIVFA as a predictor for PJK was established with an area under the curve of 0.67 (95% CI 0.59–0.76). Per the Youden index, the optimal UIVFA cut-off value is 11.5 degrees. Conclusion The more posterior the UIV is from the femoral head center after lower thoracic to pelvis surgical correction for ASD, the more patients are at risk for PJK. The greater the magnitude of posterior translation of the UIV from the femoral head center from preop to postop, the greater the likelihood for PJF.

The impact of preoperative supine radiographs on surgical strategy in adult spinal deformity

OBJECTIVE Preoperative planning for adult spinal deformity (ASD) surgery is essential to prepare the surgical team and consistently obtain postoperative alignment goals. Positional imaging may allow the surgeon to evaluate spinal flexibility and anticipate the need for more invasive techniques. The purpose of this study was to determine whether spine flexibility, defined by the change in alignment between supine and standing imaging, is associated with the need for an osteotomy in ASD surgery. METHODS A single-center, dual-surgeon retrospective analysis was performed of adult patients with ASD who underwent correction of a thoracolumbar deformity between 2014 and 2018 (pelvis to upper instrumented vertebra between L1 and T9). Patients were stratified into osteotomy (Ost) and no-osteotomy (NOst) cohorts according to whether an osteotomy was performed (Schwab grade 2 or higher). Demographic, surgical, and radiographic parameters were compared. The sagittal correction from intraoperative prone positioning alone (sagittal flexibility percentage [Sflex%]) was assessed by comparing the change in lumbar lordosis (LL) between preoperative supine to standing radiographs and preoperative to postoperative alignment. RESULTS Demographics and preoperative and postoperative sagittal alignment were similar between the Ost (n = 60, 65.9%) and NOst (n = 31, 34.1%) cohorts (p > 0.05). Of all Ost patients, 71.7% had a grade 2 osteotomy (mean 3 per patient), 21.7% had a grade 3 osteotomy, and 12.5% underwent both grade 3 and grade 2 osteotomies. Postoperatively, the NOst and Ost cohorts had similar pelvic incidence minus lumbar lordosis (PI-LL) mismatch (mean PI-LL 5.2° vs 1.2°; p = 0.205). Correction obtained through positioning (Sflex%) was significantly lower for in the osteotomy cohort (38.0% vs 76.3%, p = 0.004). A threshold of Sflex% < 70% predicted the need for osteotomy at a sensitivity of 78%, specificity of 56%, and positive predictive value of 77%. CONCLUSIONS The flexibility of the spine is quantitatively related to the use of an osteotomy. Prospective studies are needed to determine thresholds that may be used to standardize surgical decision-making in ASD surgery.