Abstract
PURPOSE: Bone Material Strength Index (BMSi) quantifies the resistance of bone to a specified force in vivo at the mid tibia using impact microindentation (IMI). Anecdotal evidence suggests that within-participant variance in BMSi may be associated with the individual's mean BMSi. This study aimed to investigate associations between mean and variance of IMI measures in a population-based study. METHODS: Participants were men (n = 420) and women (n = 55) from the Geelong Osteoporosis Study who underwent BMSi measurement using the OsteoProbe at recent follow-up phases (men 2016-2022; women 2022-2023). Median age was 63.7 yr (IQR 53.0-71.8). BMSi standard deviation was skewed and therefore natural log transformed (referred to as ln-SD). Linear regression models were developed with ln-SD as the dependent variable and mean BMSi as the independent variable adjusting for sex, age, height and weight. RESULTS: In unadjusted models, greater BMSi was associated with lower ln-SD (β = -1.58, p = 0.042). This association was sustained after adjustment (p = 0.013), and an interaction between BMSi and age was observed (p = 0.004). In those aged 63.7 yr and over (median age), mean BMSi was inversely associated with ln-SD (β = -3.22, p = 0.002). Sex was not identified as an effect modifier. In younger participants, no BMSi*ln-SD association was observed. CONCLUSION: In older men and women, there was greater variance in low BMSi values. This suggests that standard deviation of the BMSi measure may provide additional information in the assessment of bone health and is worthy of further investigation. MINI ABSTRACT: In older men and women, greater variance is observed when BMSi values are low, reflecting potential variation in the bone surface.