R Rudäng 1, M Zoulakis 1, D Sundh 1, H Brisby 2, A Diez-Perez 3, L Johansson 1, D Mellström 1, A Darelid 1, M Lorentzon 4
Osteoporos Int. 2016 Apr;27(4):1585-1592. doi: 10.1007/s00198-015-3419-0. Epub 2015 Dec 2.
PMID: 26630975 PMCID: PMC4791463 DOI: 10.1007/s00198-015-3419-0
Reference point indentation is a novel method to assess bone material strength index (BMSi) in vivo. We found that BMSi at the mid-tibia was weakly associated with spine and hip areal bone mineral density but not with prevalent fracture in a population-based cohort of 211 older women.
Introduction: Reference point indentation is a novel method to assess BMSi in vivo. Lower BMSi has been observed in patients with prior fracture than in controls, but no association between BMSi and areal bone mineral density (aBMD) has been found. Population-based association studies and prospective studies with BMSi and fractures are lacking. We hypothesized that BMSi would be associated with prevalent fractures in older Swedish women. The aim was to investigate the associations between BMSi, aBMD, and prevalent fracture in older women.
Methods: Two hundred eleven women, mean age 78.3 ± 1.1 years, were included in this cross-sectional, population-based study. BMSi was assessed using the OsteoProbe device at the mid-tibia. Areal BMD of the hip, spine, and non-dominant radius was measured using dual-energy X-ray absorptiometry (DXA). Fracture history was retrieved using questionnaires, and vertebral fractures were identified using vertebral fracture assessment (VFA) by DXA.
Results: One hundred ninety-eight previous fractures in 109 subjects were reported. A total of 106 women had a vertebral fracture, of which 58 women had moderate or severe fractures. An inverse correlation between BMSi and weight (r = -0.14, p = 0.04) was seen, and BMSi differed according to operator (ANOVA p < 0.01). Adjusting for weight and operator in a linear regression model, we found that BMSi was positively associated with aBMD of the total hip (β = 0.14, p = 0.04), non-dominant radius (β = 0.17, p = 0.02), and lumbar spine (L1-L4) (β = 0.14, p < 0.05). Using logistic regression, we could not find any association in crude or adjusted BMSi (for age, weight, height, walking speed, calcium intake, smoking, bisphosphonate and glucocorticoid use, and operator) with prevalent fractures.
Conclusion: We conclude that BMSi is associated with aBMD but not with prevalent fracture in a population-based cohort of 211 older women.