Subsequent Vertebral Fractures Post Cement Augmentation of the Thoracolumbar Spine: Does it Correlate With Level-specific Bone Mineral Density Scores?
Related Experiment Video
Updated: Apr 6, 2026

Minimally Invasive Treatment for Thoracolumbar Burst Fracture Using Sagittal Alignment Screws and A Trauma Reduction Device
Published on: November 8, 2024
Hwee Weng Dennis Hey, Dennis Hey Hwee Weng1, Jun Hao Tan
1*Department of Orthopaedic Surgery, National University Health System, Singapore, Singapore†Saw Swee Hock School of Public Health, National University of Singapore, Singapore, Singapore‡Pinnacle Spine & Scoliosis Centre, Mount Elizabeth Medical Centre, Singapore, Singapore.
Study Design:
A case-control study.
Objective:
In this study, we investigated the correlation between level-specific preoperative bone mineral density and subsequent vertebral fractures. We also identified factors associated with subsequent vertebral fractures.
Summary Of Background Data:
Complications of cement augmentation of the spine include subsequent vertebral fractures, leading to unnecessary morbidity and more treatment. Ability to predict at-risk vertebra will help guide management.
Methods:
We studied all patients with osteoporotic compression fractures who underwent cement augmentation in a single institution from November 2001 to December 2010 by a single surgeon. Association between level-specific bone mineral density T-scores and subsequent fractures was assessed. Multivariable analysis was performed to identify significant factors associated with subsequent vertebral fractures.
Results:
93 patients followed up for a mean duration of 25.1 months (12-96) had a mean age of 76.8 years (47-99). Vertebroplasty was performed in 58 patients (62.4%) on 68 levels and kyphoplasty in 35 patients (37.6%) on 44 levels. Refracture was seen in 16 patients (17.2%). The time to subsequent fracture post cement augmentation was 20.5 months (2-90). For refracture cases, 43.8% (7/16) fractured in the adjacent vertebrae. Subsequently fractured vertebra had a mean T-score of -2.860 (95% confidence interval -3.268 to -2.452) and nonfractured vertebra had a mean T-score of -2.180 (95% confidence interval -2.373 to -1.986). A T-score of -2.2 or lower is predictive of refracture at that vertebra (P = 0.047). Odds ratio increases with decreasing T-scores from -2.2 or lower to -2.6 or lower. A T-score of -2.6 or lower gives no additional predictive advantage. After multivariable analysis, age (P = 0.049) and loss of preoperative anterior vertebral height (P = 0.017) are associated with refracture.
Conclusion:
Level-specific T-scores are predictive of subsequent fractures and the odds ratio increases with lower T-scores from -2.2 or less to -2.6 or less. They have a low positive predictive value, but a high negative predictive value for subsequent fractures. Other significant associations with subsequent refractures include age and anterior vertebral height.
Level Of Evidence:
4.
