Risk Factors for Refracture following Primary Osteoporotic Vertebral Compression Fractures.

ABSTRACT

BACKGROUND: In the aging population, osteoporosis and related complications have become a global public health problem. Osteoporotic vertebral compression fractures are among the most common type of osteoporotic fractures and patients are at risk of secondary vertebral compression fracture. OBJECTIVES: To identify risk factors for secondary vertebral compression fracture following primary osteoporotic vertebral compression fractures. STUDY DESIGN: Retrospective study. SETTING: Department of Orthopedic, an affiliated hospital of a medical university. METHODS: This retrospective cohort study evaluated the risk factors for secondary vertebral compression fracture in 317 consecutive patients with systematic osteoporotic vertebral compression fractures who received percutaneous vertebroplasty and kyphoplasty or conservative treatment. Patients were divided into secondary vertebral compression fracture (n = 43) and non- secondary vertebral compression fracture (n = 274) groups. We retrospectively analyzed clinical characteristics and radiographic parameters, including gender, age, body mass index, number of primary fractures, primary treatment (percutaneous vertebroplasty and kyphoplasty or conservative treatment), nonspinal fracture history before primary fracture, primary fracture at the thoracolumbar junction, steroid use, bisphosphonate therapy, and Hounsfield units value of L1. RESULTS: Comparison between the groups showed significant differences in age (P = 0.001), nonspinal fracture history (P < 0.001), and Hounsfield units value of L1 (P < 0.001). The receiver operating characteristic curves demonstrated that the optimal thresholds for age and Hounsfield units value of L1 were 75 (sensitivity 55.8%; specificity 67.5%) and 50 (sensitivity 88.3%; specificity 67.4%), respectively. In multivariate logistic regression analysis, nonspinal fracture history (OR = 6.639, 95% CI = 1.809 - 24.371, P = 0.004) and Hounsfield units value of L1 < 50 (OR = 15.260, 95% CI = 6.957 - 33.473, P < 0.001) were independent risk factors for secondary vertebral compression fracture. LIMITATIONS: The main limitation is the retrospective nature of this study. CONCLUSION: Patients with low Hounsfield units value of L1 or non-spinal fracture history are an important population to target for secondary fracture prevention.