To study relationships between serum ferritin and bone metabolism in patients with hip fragility fractures. This cross-sectional study included 76 postmenopausal women with hip fracture from Feburary 2011 to June 2012. The mean age of the women was (73 ± 10) years (range, 55-93 years) and the mean duration of menstruation was (22 ± 10)years (range, 5-50 years). Serum concentrations of ferritin, transferrin, alkaline phosphatase (ALP), amino-terminal extension peptide of type I collagen (P1NP), C-terminal telopeptides of type I collagen (β-CTX)and femoral and lumbar bone mineral density by dual-energy X-ray absorptiometry were measured. Bone metabolism was compared between normal and elevated ferritin groups with t-test, Pearson linear, partial correlation and multiple regression analysis examined associations between iron- and bone-related markers. Serum ferritin concentration raised to (230 ± 146)μg/L, transferrin concentration reduced to (1.89 ± 0.33)g/L. P1NP concentration raised to (61 ± 32) ng/L when the concentration of serum ALP and β-CTX were in the normal range. T-scores for bone mineral density in the femoral neck (-2.0 ± 1.1) and lumbar (-2.1 ± 1.2) were below the normal ranges(-1.0-1.0). The subjects were divided into two groups according to serum ferritin concentration, normal group(serum ferritin concentration ≤ 150 μg/L, n = 25) and elevated group(serum ferritin concentration > 150 μg/L, n = 51). Patients of elevated group had lower bone mineral density in femoral neck and lumbar than normal group(t = 3.13,2.89, P < 0.01), and higher P1NP, β-CTX concentration (t = -2.38, -3.59, P < 0.05) . In partial correlation analysis adjusted for confounders, serum ferritin concentration was correlated negatively with bone mineral density in both femoral neck and lumbar (r = -0.335,-0.295, P < 0.05), and positively with P1NP and β-CTX (r = 0.467,0.414, P < 0.05), but not correlated with ALP (r = 0.188, P > 0.05). Transferrin concentration tended to be correlated positively with bone mineral density in both femoral neck and lumbar (r = 0.444, 0.262, P < 0.05) and negatively with ALP, P1NP and β-CTX(r = -0.326,-0.285,-0.278, P < 0.05). Iron overload has a high prevalence in postmenopausal women with fragility fracture. Increased iron stores, which might lead to bone loss and lower bone mineral density by enhancing the activity of bone turnover, could be an independent factor to take effects on bone metabolism on postmenopausal women.