Overview of bone microstructure, and treatment of bone fragility in chronic kidney disease
Bone is type 1 collagen impregnated with crystals of calcium hydroxyapatite.3 The collagen confers flexibility enabling energy absorption by deformation. The mineral confers rigidity for loading. If under‐mineralized bone (osteomalacia) deforms excessively, it may fail. If homogeneously and fully mineralized bone becomes brittle, it may fail. The non‐collagenous helical proteins (e.g. osteopontin and osteocalcin) participate in collagen–mineral interaction providing ‘hidden length’ by unfolding, which minimizes stress on hydroxyapatite crystals during loading.4 Accumulation of advanced glycation end products (e.g. pentosidine) during ageing, renal disease and antiresorptive therapy may compromise bone's ‘toughness’ – its ability to absorb energy – predisposing to fractures.5
Negative remodelling balance and increased remodelling rate (exacerbated by secondary hyperparathyroidism) compromise microstructure.3 Unbalanced remodelling upon the intracortical surface of Haversian canals enlarge the canals causing cortical porosity, the source of 70% of all appendicular bone loss (because 80% of the skeleton is cortical bone).6
High porosity is a consistent finding in chronic kidney disease, a consistent observation in animal models of renal disease, and is the likely explanation of the high predictive value of cortical volumetric bone mineral density (BMD) for fracture (this is not found for trabecular bone or BMD).7 Stiffness is a seventh power function of porosity so a small increase in porosity with modest bone loss disproportionately reduces stiffness.8 Porosity is likely to be an important predictor of fracture and a target for therapy. Unbalanced trabecular remodelling erodes trabeculae yet high trabecular density is reported in some animal models of chronic kidney disease. The reasons for this are obscure. One possibility is measurement errors in segmenting cortical from trabecular bone using thresholding. High intracortical remodelling produces intracortical porosity and cortical fragments which look like trabeculae. The imaging algorithm may incorrectly apportion the cortical fragments to what seems to be an enlarged medullary canal leading to an over estimate of ‘trabecular’ density and an underestimate of cortical porosity.