Abstract
BACKGROUND: Vertebral size and especially cross-sectional area (CSA) are independently associated with vertebral fracture risk. Previous studies have suggested that physical activity and especially high-impact exercise may affect vertebral strength. We aimed to investigate the association between high-impact exercise at 31 and 46 years of age and vertebral dimensions in midlife. METHODS: We used a subsample of 1023 individuals from the Northern Finland Birth Cohort 1966 study with records of self-reported sports participation from 31 and 46 years and MRI-derived data on vertebral dimensions from 46 years. Based on the sports participation data, we constructed three impact categories (high, mixed, low) that represented longitudinal high-impact exercise activity in adulthood. We used linear regression and generalized estimating equation (GEE) models to analyse the association between high-impact exercise and vertebral CSA, with adjustments for vertebral height and body mass index. RESULTS: Participation in high-impact sports was associated with large vertebral CSA among women but not men. The women in the 'mixed' group had 36.8 (95% confidence interval 11.2-62.5) mm(2) larger CSA and the women in the 'high' group 43.2 (15.2-71.1) mm(2) larger CSA than the 'low' group. CONCLUSIONS: We suggest that participation (≥ 1/week) in one or more high-impact sports in adulthood is associated with larger vertebral size, and thus increased vertebral strength, among middle-aged women.