Association between degenerative spondylolisthesis and spinous process fracture after interspinous process spacer surgery

Abstract

Spinous process fracture is a recognized complication associated with interspinous process spacer (IPS) surgery. Although occasionally identified by plain radiographs, computed tomography (CT) appears to identify a higher rate of such fractures. Although osteoporotic insufficiency fracture is considered a contraindication for IPS surgery, a formal risk factor analysis for this complication has not previously been reported. To identify risk factor(s) associated with early spinous process fracture after IPS surgery. Prospective cohort study of 39 consecutive patients with lumbar stenosis and neurogenic claudication undergoing IPS surgery at a single institution. Patients underwent preoperative dual-energy X-ray absorptiometry (DXA) scans, lumbar spine CT, and plain radiographs. Postoperatively, patients underwent repeat CT imaging within 6 months of surgery and serial radiographs at 2 weeks, 6 weeks, 3 months, 6 months, and 1 year. Preoperative CT scans were analyzed by calculating average Hounsfield units for a 1 cm(2) area of the midsagittal reconstructed image for four separate locations: midvertebral body, subcortical bone subjacent to the superior margin of the midspinous process, subcortical bone above the inferior margin of the midspinous process, and the midspinous process. Thirty-eight patients underwent IPS surgery at a total of 50 levels (38 L4-L5, 12 L3-L4; 26 one-level, 12 two-level). One patient underwent laminectomy at index surgery and was excluded from the analysis. Implants included 34 titanium X-STOP (Medtronic, Memphis, TN, USA), 8 polyaryletheretherketone X-STOP (Medtronic, Memphis, TN, USA), and 8 Aspen (Lanx, Broomfield, CO, USA) devices. Eleven spinous process fractures were identified by CT in 11 patients (22.0% of levels). No fractures were apparent on plain radiographs. The rate of spondylolisthesis observed on preoperative radiographs was 100% (11 of 11) among patients with fractures compared with 33.3% (9 of 27) of patients without fracture (p=.0001). Overall, 21 of 39 patients in this series had spondylolisthesis, and the rate of fracture in this group was 52%. Among patients without spondylolisthesis, the fracture rate was 0%. A trend was observed toward decreased DXA lumbar spine and hip T-scores among fracture patients versus nonfracture patients (0.2 ± 1.7 vs. 0.8 ± 1.7; p=.389; -1.1 ± 1.4 vs. -0.3 ± 1.4; p=.201), but these differences were not significant. Similarly, bone density based on CT measurements at four different locations revealed a trend toward decreased density among fracture patients, but these differences were not significant. Degenerative spondylolisthesis appears strongly associated with the occurrence of spinous process fracture after IPS surgery. There is a trend toward increased fracture risk in patients with decreased bone mineral density as measured by both DXA scan and CT-based volume averaging of Hounsfield units, but osteoporosis appears to be a relatively weaker risk factor. The association between spondylolisthesis and fracture observed in this study may account for the relatively poorer outcome of IPS surgery in patients with spondylolisthesis that has been reported in previous series.