Diagnosis of vertebral fractures using a low-dose biplanar imaging system.

Abstract

Vertebral fractures (VFs) are independent risk factors for new fractures. However, spine radiographs cannot be used as a screening method. EOS® has a good diagnostic value for the diagnosis of VF with a better legibility of upper thoracic spine and a higher concordance between readers compared to vertebral fracture assessment (VFA). Vertebral fractures (VFs) are risk factors for new fractures. However, spine radiographs cannot be used as a screening method for both cost and radiation concerns. EOS® X-ray imaging system which allows the acquisition of biplane images in an upright weight-bearing position with low radiation dose was used. The objective of this study was to compare EOS® to VFA for the diagnosis of VF. We conducted a cross-sectional study in subjects aged above 50 years with indication for spine imaging. EOS® and VFA of the spine were performed the same day. Sensitivity (Se), specificity (Sp), negative predictive value (NPV), and the interobserver precision of EOS® were compared to VFA for the diagnosis of VF. Two hundred patients (mean age 66.2 years) were included. At the vertebral level, 2.4 and 3.6 % of vertebrae were not legible using EOS® and VFA, respectively (p = 0.0007). The legibility of spine was significantly affected by scoliosis (odds ratio (OR) = 2.8, p < 0.0001, for EOS®, and OR = 1.8, p = 0.0041, for VFA). Sixty-six patients (33.0 %) and 69 (34.5 %) had at least one VF using VFA and EOS®, respectively. At patient level, Se, Sp, and NPV for the diagnosis of VF of EOS® were 79.7, 91.6, and 99 %, respectively. Concordance between both observers was very good for EOS® (kappa-score = 0.89), higher than for VFA (κ = 0.67). This study shows that EOS® has a good diagnostic value for the diagnosis of VF with a better legibility of upper thoracic spine and a higher concordance between readers compared to VFA.