Metabolic factors and hip fracture risk in a large Austrian cohort study

To explore the association of incident hip fractures with metabolic syndrome (MetS) and its single components, we designed a prospective cohort study of hip fracture incidence among 117,053 participants of a populationbased health surveillance program in Vorarlberg, the westernmost Austrian province. Incident hip fractures were recorded between 5 and 10 years after inclusion at baseline from 2003 to 2009. Applying Cox proportional hazard models for each MetS component and for a composite z-score for MetS, hazards for fracture were estimated in quintiles, as continuous z-score variables, and as pathological cut off values. Mean age was 50.1 +/- 15.6 years at baseline, 5-10 years after which 947 incident hip fractures occurred. An association of a higher composite MetS score with decreased hip fracture risk was observed in women (HR 0.80, 95%-CI 0.88-0.96, p < 0.01) which disappeared upon adjustment for BMI. BMI was inversely associated with hip fracture risk in women and men (HR for the highest compared with the lowest quintile: 0.83 (95%-CI: 0.63-1.10, ptrend < 0.05) and 0.55 (95%-CI: 0.38-0.79, ptrend < 0.001), respectively). Only in women, hip fracture risk was reduced at high cholesterol levels (HR for the highest relative to the lowest quintile: 0.64, 95%-CI: 0.48-0.84, ptrend < 0.05) and in hypercholesterolemic patients (HR 0.82, 95%-CI: 0.67-0.99, p < 0.05), but elevated in hyperglycemic patients (HR 1.33, 95%-CI: 1.05-1.70, p < 0.05). Hypertriglyceridemia was associated with increased male hip fracture risk (HR 1.33, 95%-CI: 1.03-1.72, p < 0.05). The inverse association between the MetS and hip fracture risk is mainly driven by one single component, namely BMI.