Prophylactic vertebral cement augmentation at the uppermost instrumented vertebra and rostral adjacent vertebra for the prevention of proximal junctional kyphosis and failure following long-segment fusion for adult spinal deformity

Abstract

Background ContextProximal junctional kyphosis (PJK) and proximal junctional failure (PJF) are common problems after long-segment (>5 levels) thoracolumbar instrumented fusions in the treatment of adult spinal deformity (ASD). No specific surgical strategy has definitively been shown to lower the risk of PJK as the result of a multifactorial etiology. PurposeThe study aimed to assess the incidence of PJK and PJF in patients treated with prophylactic polymethylmethacrylate (PMMA) cement augmentation at the uppermost instrumented vertebrae (UIV) and rostral adjacent vertebrae (UIV+1). Study Design/SettingThis is a retrospective cohort-matched surgical case series at an academic institutional setting. Patient SampleEighty-five adult patients over a 16-year enrollment period were identified with long-segment (>5 levels) posterior thoracolumbar instrumented fusions for ASD. Outcome MeasuresPrimary outcomes measures were PJK magnitude and PJF formation. Secondary outcomes measures were spinopelvic parameters, as well as global and regional sagittal alignment. MethodsThe impact of adjunctive PMMA use in long-segment (≥5 levels) fusion for ASD was assessed in adult patients aged 18 and older. Patients were included with at least one of the following: lumbar scoliosis >20°, pelvic tilt >25°, sagittal vertical axis >5 cm, central sacral vertical line >2 cm, and thoracic kyphosis >60°. The frequency of PJF and the magnitude of PJK were measured radiographically preoperatively, postoperatively, and at maximum follow-up in controls (Group A) and PMMA at the UIV and UIV+1 (Group B). ResultsEighty-five patients (64±11.1 years) with ASD were identified: 47 control patients (58±10.6) and 38 patients (71±6.8) treated with PMMA at the UIV and UIV+1. The mean follow-up was 27.9 and 24.2 months in Groups A and B, respectively (p=.10). Preoperative radiographic parameters were not significantly different, except the pelvic tilt which was greater in Group A (26.6° vs. 31.4°, p=.03). Postoperatively, the lumbopelvic mismatch was greater in Group B (14.6° vs. 7.9°, p=.037), whereas the magnitude of PJK was greater in controls (9.36° vs. 5.65°, p=.023). The incidence of PJK was 36% (n=17) and 23.7% (n=9) in Groups A and B, respectively (p=.020). The odds ratio of PJK with vertebroplasty was 0.548 (95% confidence interval=0.211 to 1.424). Proximal junctional kyphosis was observed in 6 (12.8%) controls only (p=.031). The UIV+1 angle, a measure of PJK, was significantly greater in controls (10.0° vs. 6.8°, p=.02). No difference in blood loss was observed. No complications were attributed to PMMA use. ConclusionsThe use of prophylactic vertebral cement augmentation at the UIV and rostral adjacent vertebral segment at the time of deformity correction appears to be preventative in the development of proximal junctional kyphosis and failure.