Objective: Sex steroids play an important role in the maintenance of bone health. Association studies on sex steroids and fractures are not consistent. Our objective was to examine whether serum oestradiol (E2) and testosterone (T) are associated with quantitative ultrasound (QUS), bone mineral density (BMD), bone turnover markers and fracture incidence. Design: The Longitudinal Ageing Study Amsterdam (LASA), an ongoing cohort study including 623 men and 634 women, aged 65-88 years. Measurements: Serum levels of E2, T, SHBG, albumin, bone turnover markers serum osteocalcin (OC) and urinary deoxypyridinoline (DPD/Cr) were measured. QUS of the heel and BMD of the hip were assessed, and a 6-year fracture follow-up was performed. Results: Men in the lowest quartile (Q1) of bioavailable E2 (bioE2) had higher levels of bone turnover and lower BMD (B = -0.09, P < 0.01) and QUS than men in the highest quartile (Q4). This also applied to Q1 of bioT. Women in Q1 of bioE2 had higher levels of bone turnover and lower BMD (B = -0.07, P < 0.01) and QUS than women in Q4. In men and women, levels of bioE2 below the median were associated with an increased risk of osteoporotic fractures after all adjustments [hazard ratio (HR) 1.53, 95% confidence interval (CI) 1.02-2.29]. In men, univariate analysis revealed that low bioT was associated with an increased fracture risk (HR 1.91, 95% CI 1.03-3.56), but after adjustment for age, this association was no longer significant. Conclusions: Low levels of bioE2 and bioT were found to be associated with high bone turnover, low QUS and BMD and high risk of osteoporotic fractures in both men and women.