BMC Musculoskeletal Disorders (Jan 2022)
Vacuum phenomenon as a predictor of kyphosis after implant removal following posterior pedicle screw fixation without fusion for thoracolumbar burst fracture: a single-center retrospective study
Abstract
Abstract Background Posterior pedicle screw fixation without fusion has been commonly applied for thoracolumbar burst fracture. Implant removal is performed secondarily after bone union. However, the occurrence of secondary kyphosis has recently attracted attention. Secondary kyphosis results in poor clinical outcomes. The purpose of this was to determine predictors of kyphosis after implant removal following posterior pedicle screw fixation without fusion for thoracolumbar burst fracture. Methods This retrospective study reviewed 59 consecutive patients with thoracolumbar burst fracture who underwent implant removal following posterior pedicle screw fixation without fusion. Inclusion criteria were non-osteoporotic fracture and T11-L3 burst fracture. Old age, sex, initial severe wedge deformity, initial severe kyphosis, and vacuum phenomenon were examined as factors potentially associated with final kyphotic deformity (defined as kyphotic angle greater than 25°) or loss of correction. Logistic regression analysis was performed using propensity score matching. Results Among the 31 female and 28 male patients (mean age 38 years), final kyphotic deformity was found in 17 cases (29%). Multivariate analysis showed a significant association with the vacuum phenomenon. Loss of correction was found in 35 cases (59%) and showed a significant association with the vacuum phenomenon. There were no significant associations with other factors. Conclusions The findings of this study suggest that the vacuum phenomenon before implant removal may be a predictor of secondary kyphosis of greater than 25° after implant removal following posterior pedicle screw fixation without fusion for thoracolumbar burst fracture, but that old age, sex, initial severe kyphosis, and initial severe wedge deformity may not be predictors.
Keywords