Bone microarchitecture and estimated failure load are deteriorated whether patients with chronic kidney disease have normal bone mineral density, osteopenia or osteoporosis

Introduction: Measurement of bone mineral density (BMD) is recommended in patients with chronic kidney disease (CKD). However, most persons in the community and most patients with CKD have osteopenia, suggesting fracture risk is low. Bone loss compromises bone microarchitecture which increases fragility disproportionate to modest deficits in BMD. We therefore hypothesized that patients with CKD have reduced estimated failure load due to deterioration in microarchitecture irrespective of whether they have normal femoral neck (FN) BMD, osteopenia or osteoporosis. Methods: We measured distal tibial and distal radial microarchitecture in 128 patients with CKD and 275 age-and sex-matched controls using high resolution peripheral quantitative computed tomography, FN-BMD using bone densitometry and estimated failure load at the distal appendicular sites using finite element analysis. Results: Patients versus controls respectively had: lower tibial cortical area 219 (40.7) vs. 237 (35.3) mm(2), p = 0.002, lower cortical volumetric BMD 543 (80.7) vs. 642 (81.7) mgHA/cm(3) due to higher porosity 69.6 (6.19) vs. 61.9 (6.48)% and lower matrix mineral density 64.2 (0.62) vs. 65.1 (1.28)%, lower trabecular vBMD 92.2 (41.1) vs. 149 (43.0) mgHA/cm(3) due to fewer and spatially disrupted trabeculae, lower FN-BMD 0.78 (0.12) vs. 0.94 (0.14) g/cm(2) and reduced estimated failure load 3825 (1152) vs. 5778 (1467) N, all p < 0.001. Deterioration in microarchitecture and estimated failure load was most severe in patients and controls with osteoporosis. Patients with CKD with osteopenia and normal FN-BMD had more deteriorated tibial microarchitecture and estimated failure load than controls with BMD in the same category. In univariate analyses, microarchitecture and FN-BMD were both associated with estimated failure load. In multivariable analyses, only microarchitecture was independently associated with estimated failure load and accounted for 87% of the variance. Conclusions: Bone fragility is likely to be present in patients with CKD despite them having osteopenia or normal BMD. Measuring microarchitecture may assist in targeting therapy to those at risk of fracture.