Discriminating value of HR-pQCT for fractures in women with similar FRAX scores: A substudy of the FRISBEE cohort
S I Cappelle,M Moreau,R Karmali,L Iconaru,F Baleanu,V Kinnard,M Paesmans,S Rozenberg,M Rubinstein,M Surquin,P-H Blard,R Chapurlat,J J Body,P Bergmann
DOI: https://doi.org/10.1016/j.bone.2020.115613
IF: 4.626
Bone
Abstract:Areal bone mineral density (aBMD) has a low sensitivity to identify women at high fracture risk. The FRAX algorithm, by combining several clinical risk factors, might improve fracture prediction compared to aBMD alone. Several micro-architectural and biomechanical parameters which can be measured by high-resolution peripheral quantitative computed tomography (HR-pQCT) are associated with fracture risk. HR-pQCT in combination or not with finite element analysis (FEA) may be used to improve bone strength prediction. Our aim was to assess whether HR-pQCT measurements (densities, cortical and trabecular microarchitecture, biomechanical proprieties assessed by FEA) had an added value in predicting fractures in a subgroup of women belonging to the Belgian FRISBEE cohort. One hundred nineteen women who sustained a fracture (aged 60 to 85 years) during the initial follow-up of our cohort had a radius and tibia examination by HR-pQCT and were compared with controls matched for their FRAX score at baseline. We found that low distal radius total (OR = 1.41 [1.07-1.86] per SD, p < 0.05) and trabecular densities (OR = 1.45 [1.10-1.90], p < 0.01), trabecular number (OR = 1.32 [1.01-1.72], p < 0.05), intra individual distribution of separation (OR = 0.73 [0.54-0.99], p < 0.05) as several FEA parameters were significantly associated with fractures. At the distal tibia, impaired cortical density (OR = 1.32 [1.03-1.70] per SD, p < 0.05) and thickness (OR = 1.29 [1.01-1.63], p < 0.05) and apparent modulus (OR = 1.30 [1.01-1.66], p < 0.05) were significantly correlated with fractures. A low ultra distal radial aBMD (UDR) measured at the time of HR-pQCT was significantly associated with fractures (OR = 1.67 [1.22-2.28], p < 0.01). Women from both groups were followed further after the realization of the HR-pQCT and 46 new fractures were registered. In this second part of the study, low UDR aBMD (OR = 1.66 [1.18-2.35], p < 0.01), total (OR = 1.48 [1.08-2.03], p < 0.05), cortical (OR = 1.40 [1.04-1.87], p < 0.05) and trabecular (OR = 1.37 [1.01-1.85], p < 0.05) densities or apparent modulus (OR = 1.49 [1.07-2.05], p < 0.05) at the radius were associated with a significant increase of fracture risk. At the tibia, only the cortical density was significantly associated with the fracture risk (OR = 1.34 [1.02-2.76], p < 0.05). These results confirm the interest of HR-pQCT measurements for the evaluation of fracture risk, also in women matched for their baseline FRAX score. They also highlight that UDR aBMD contains pertinent information.