Background/objectives: High vitamin A intake may be associated with a decreased bone mineral density (BMD) and increased risk of fractures. Our objectives were to study whether dietary intake of vitamin A (total, retinol or beta-carotene) is associated with BMD and fracture risk and if associations are modified by body mass index (BMI) and vitamin D.
Subjects/methods: Participants were aged 55 years and older (n=5288) from the Rotterdam Study, a population-based prospective cohort. Baseline vitamin A and D intake was measured by a food frequency questionnaire. BMD was measured by dual-energy X-ray absorptiometry at four visits between baseline (1989-1993) and 2004. Serum vitamin D was assessed in a subgroup (n=3161). Fracture incidence data were derived from medical records with a mean follow-up time of 13.9 years.
Results: Median intake of vitamin A ranged from 684 retinol equivalents (REs)/day (quintile 1) to 2000 REs/day (quintile 5). After adjustment for confounders related to lifestyle and socioeconomic status, BMD was significantly higher in subjects in the highest quintile of total vitamin A (mean difference in BMD (95% confidence interval (CI))=11.53 (0.37-22.7) mg/cm(2)) and retinol intake (mean difference in BMD (95% CI)=12.57 (1.10-24.05) mg/cm(2)) than in the middle quintile. Additional adjustment for BMI diluted these associations. Fracture risk was reduced in these subjects. Significant interaction was present between intake of retinol and overweight (BMI >25 kg/m(2)) in relation to fractures (P for interaction =0.05), but not BMD. Stratified analysis showed that these favourable associations with fracture risk were only present in overweight subjects (BMI >25 kg/m(2)). No effect modification by vitamin D intake or serum levels was observed.
Conclusions: Our results suggest a plausible favourable relation between high vitamin A intake from the diet and fracture risk in overweight subjects, whereas the association between vitamin A and BMD is mainly explained by BMI.