Cortical bone area predicts incident fractures independently of areal bone mineral density in older men

• Published in: The journal of clinical endocrinology and metabolism Vol. 102; no. 2; pp. 516 - 524
• Main Authors: Ohlsson, Claes, Sundh, Daniel, Wallerek, Andreas, Nilsson, Martin, Karlsson, Magnus, Johansson, Helena, Mellström, Dan, Lorentzon, Mattias
• Format: Journal Article
• Language: English
• Published: Washington, DC Endocrine Society 01-02-2017; Copyright Oxford University Press; Oxford University Press
• Subjects: Absorptiometry, Photon; Aged; Aged, 80 and over; Aging - physiology; Bone density; Bone Density - physiology; Bone mass; Bone mineral density; Cancellous bone; Clinical Medicine; Computed tomography; Cortical bone; Cortical Bone - diagnostic imaging; Dual energy X-ray absorptiometry; Endocrinology and Diabetes; Endokrinologi och diabetes; Energy measurement; Femur; Fractures; Fractures, Bone - epidemiology; Health risk assessment; Humans; Klinisk medicin; Male; Medical and Health Sciences; Medicin och hälsovetenskap; Orthopedics; Ortopedi; Osteoporotic Fractures - epidemiology; Porosity; Risk assessment; Risk Assessment - methods; Risk factors; Sweden - epidemiology; Tibia; Tomography, X-Ray Computed
• ISSN: 0021-972X; 1945-7197; 1945-7197
• DOI: 10.1210/jc.2016-3177

Context:Areal bone mineral density (aBMD) measured using dual-energy X-ray absorptiometry (DXA) is used clinically to predict fracture but does not allow discrimination between the trabecular and cortical bone compartments or assessment of cortical porosity.Objective:To investigate if distinct information on cortical and trabecular bone parameters predict fracture risk independently of aBMD and clinical risk factors.Design & participants:Cortical area, cortical bone mass, trabecular bone volume fraction, and cortical porosity were measured at the tibia using high-resolution peripheral quantitative computed tomography (HR-pQCT) in 456 men (80.2±3.5 years) recruited from the general population in Gothenburg, Sweden. aBMD was measured using DXA. Incident fractures (71 men with fracture) were X-ray verified. The associations with incident fracture were evaluated using Cox proportional hazard models.Results:In models adjusted for age and BMI, cortical area (HR per SD decrease, 2.05; 95% CI, 1.58-2.65), cortical bone mass (HR per SD decrease, 2.07; 95% CI, 1.58-2.70), and trabecular BV/TV (HR, 1.62; 1.26-2.07) but not cortical porosity, were independently associated with fracture risk. These association remained essentially unaffected after adjustment for femoral neck aBMD and FRAX risk factors (area: HR 1.96 (1.44-2.66); cortical bone mass: HR 1.99 (1.45-2.74); trabecular BV/TV: HR 1.46 (1.09-1.96)). In all adjusted models, inclusion of both BV/TV and cortical area or cortical bone mass entered simultantously, only the cortical parameters remained significant predictors of fracture.Conclusion:We propose that HR-pQCT measurment of cortical area or cortical bone mass might add clinically useful information for the evaluation of fracture risk.