Fragility Score: a REMS-based indicator for the prediction of incident fragility fractures at 5 years
Paola Pisani, Francesco Conversano, Maurizio Muratore, Giovanni Adami, Maria Luisa Brandi, Carla Caffarelli, Ernesto Casciaro, Marco Di Paola, Roberto Franchini, Davide Gatti, Stefano Gonnelli, Giuseppe Guglielmi, Fiorella Anna Lombardi, Alessandra Natale, Valentina Testini, Sergio Casciaro
Abstract
BACKGROUND: Accurate estimation of the imminent fragility fracture risk currently represents a challenging task. The novel Fragility Score (FS) parameter, obtained during a Radiofrequency Echographic Multi Spectrometry (REMS) scan of lumbar or femoral regions, has been developed for the non-ionizing estimation of skeletal fragility. AIMS: The aim of this study was to assess the performance of FS in the early identification of patients at risk for incident fragility fractures with respect to bone mineral density (BMD) measurements. METHODS: Data from 1989 Caucasians of both genders were analysed and the incidence of fractures was assessed during a follow-up period up to 5 years. The diagnostic performance of FS to discriminate between patients with and without incident fragility fracture in comparison to that of the BMD T-scores measured by both Dual X-ray Absorptiometry (DXA) and REMS was assessed through ROC analysis. RESULTS: Concerning the prediction of generic osteoporotic fractures, FS provided AUC = 0.811 for women and AUC = 0.780 for men, which resulted in AUC = 0.715 and AUC = 0.758, respectively, when adjusted for age and body mass index (BMI). For the prediction of hip fractures, the corresponding values were AUC = 0.780 for women and AUC = 0.809 for men, which became AUC = 0.735 and AUC = 0.758, respectively, after age- and BMI-adjustment. Overall, FS showed the highest prediction ability for any considered fracture type in both genders, resulting always being significantly higher than either T-scores, whose AUC values were in the range 0.472-0.709. CONCLUSION: FS displayed a superior performance in fracture prediction, representing a valuable diagnostic tool to accurately detect a short-term fracture risk.