Purpose: This study aims to identify risk factors for vertebral compression fracture and enhance the ability to indicate fracture risk.
Methods: A retrospective collection of clinical and imaging data was conducted for patients with vertebral compression fractures and control subjects who underwent quantitative computed tomography scans. Stepwise logistic regression analysis was employed to identify variables associated with fractures, constructing both unadjusted model and adjusted model.
Results: Compared with the non-fracture group, the fracture group showed significant differences in weight, body mass index (BMI), bone mineral density (BMD), vertebral cross-sectional area, paraspinal muscle area and right psoas major muscle fat area (all P < 0.05). Adjusted characteristics analyzed by stepwise logistic regression indicated that bone load index (BLI) (OR = 3.19, P = 0.041), paraspinal muscle fat infiltration (PMFI) (OR = 2.27, P = 0.039), and right psoas major muscle fat infiltration (RPMFI) (OR = 1.08, P = 0.005) were independent risk factors for vertebral fractures. Interaction analysis revealed a positive interaction between BLI and PMFI (OR = 1.95, P = 0.008) as well as RPMFI (OR = 1.53, P = 0.045). Compared with the unadjusted model, the diagnostic performance of the adjusted model was significantly improved (training set IDI: 19.5 %, validation set IDI: 18.4 %, P < 0.001). Correlation analysis demonstrated significant associations between BMD (r = -0.353, P = 0.002), BLI (r = 0.631, P < 0.001), PMFI (r = 0.412, P < 0.001), RPMFI (r = 0.513, P < 0.001), and the degree of vertebral compression.
Conclusion: Under conditions of bone maladaptive loading and muscle degeneration, vertebral bodies may become more susceptible to external forces, increasing the risk of vertebral compression fracture.
Keywords: Body composition; Low bone mass; Muscle degeneration; Osteoporosis; Quantitative computed tomography; Vertebral compression fracture.
Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.