Use of proton density fat fraction MRI to predict the radiographic progression of osteoporotic vertebral compression fracture

Objective This study evaluated the diagnostic performance of the proton density fat fraction (PDFF) in predicting the progression of osteoporotic vertebral compression fractures (OVCFs). Methods The cohort in this retrospective study consisted of 48 patients with OVCFs who underwent spine MRI that included PDFF between December 2016 and June 2018. The patients were divided into two groups (with versus without OVCF progression, based on the radiographic results obtained at the 6-month follow-up examination). Two musculoskeletal radiologists independently calculated the PDFF of the fracture and the PDFF ratio (fracture PDFF/normal vertebrae PDFF) using regions of interest. The mean values of these parameters were compared between the two groups, and the receiver operating characteristic curves were analysed. Results The mean age was significantly higher in the group with OVCF progression (71.6 ± 8.4 years) than in the group without (64.8 ± 10.5 years) ( p = 0.018). According to reader 1, the PDFF ratio was significantly lower in the group with OVCF progression versus that without OVCF progression (0.38 ± 0.13 vs 0.51 ± 0.20; p = 0.009), whereas the difference in the PDFF itself was not statistically significant. The PDFF ratio [area under the curve (AUC) = 0.723; 95% confidence interval (CI), 0.575–0.842] had a larger AUC than did the PDFF (AUC = 0.667; 95% CI, 0.516–0.796). The optimal cut-off value of the PDFF ratio for predicting OVCF progression was 0.42; this threshold corresponded to sensitivity, specificity, and accuracy values of 84.0%, 60.9%, and 72.9%, respectively. Conclusion The age and PDFF ratio can be used to predict OVCF progression. Key Points • Chemical shift-encoded magnetic resonance imaging provides quantitative parameters for predicting OVCF progression. • The PDFF ratio is significantly lower in patients with OVCF progression. • The PDFF ratio is superior to the PDFF for predicting OVCF progression.