O18. DXA Measures of Bone Mass in Mother, Father and Offspring: Results from the Southampton Women’s Survey

Abstract

Background: Although there is evidence that measures of bone size, mineralization and density may be partly inherited, there are scant data available from which to elucidate independent influences of mother and father. In this study we aimed to document such independent relationships between indices of bone mass in childhood and corresponding measures in the mother and father, using the Southampton Women's Survey (SWS). Method(s): The SWS is a prospective, population-based cohort of 12,583 initially non-pregnant women aged 20-34 years, from which 3156 subsequent singleton pregnancies were followed. In a subset of participants, DXA assessment (Hologic Discovery) of bone mass was obtained in the child (at 8-9 years) and in both parents. Measurements included bone area (BA), BMC, areal BMD (aBMD) and size-corrected BMC (BMC adjusted for BA, height and weight; scBMC) at whole body minus head (WB), lumbar spine and total hip sites. Correlation and linear regression methods were used to assess independent relationships between bone indices of the child and of each parent. In each case, as the predictor and outcome variables had the same units, the regression coefficients were dimensionless and thus effect sizes could be compared across associations. Result(s): Complete data were available for 126 mother-father-offspring trios. There were no significant correlations between maternal and corresponding paternal DXA measures (all P>0.05). However, strong positive associations were observed between parental WB, total hip and lumbar spine measures (BA, BMC, BMD) and the corresponding indices in the offspring (maternal: beta=0.17-0.39; all P<0.001; paternal: beta=0.09-0.21; all P<0.01). Associations for scBMC at each site were somewhat weaker, but remained statistically significant (all P<0.01). In multivariate modelling, independent relationships were observed between either parent and the offspring (mother-child beta=0.10-0.33, P<0.05; father-child beta=0.09-0.18, P<0.05). Notably larger effect sizes were observed for maternal than paternal-offspring relationships for WB BMD (beta=0.21, P<0.01 vs beta=0.12, P<0.05 respectively), lumbar spine bone area (beta=0.30, P<0.001 vs beta=0.15, P<0.05 respectively), and total hip bone area (beta=0.33, P<0.001 vs beta=0.11, P<0.05 respectively). Conclusion(s): We observed independent associations between indices of bone mass in the offspring, and corresponding measures in the mother and father, with evidence of a greater maternal than paternal effect size for whole body BMD and bone area at the lumbar spine and total hip sites. Although direct genetic inheritance offers one mechanistic explanation, the low proportion of variance in bone indices explained by known genetic polymorphisms, increasing understanding of epigenetic mechanisms, and disparity between maternal and paternal associations suggest that such relationships could be in part underpinned by gene-environment interactions in early life.