No increase in risk of hip fracture at high serum retinol concentrations in community-dwelling older Norwegians: the Norwegian Epidemiologic Osteoporosis Studies

卷:102

Holvik, Kristin ; Ahmed, Luai A. ; Forsmo, Siri ; Gjesdal, Clara G. ; Grimnes, Guri ; Samuelsen, Sven Ove ; Schei, Berit ; Blomhoff, Rune ; Tell, Grethe S. ; Meyer, Haakon E.

关键词: hip fracture;    vitamin A;    retinol;    elderly;    Norway;    case-cohort;   

DOI  :  10.3945/ajcn.115.110528

学科分类：食品科学和技术

【 摘 要 】

Background: Norway has the highest hip fracture rates worldwide and a relatively high vitamin A intake. Increased fracture risk at high intakes and serum concentrations of retinol (s-retinol) have been observed in epidemiologic studies. Objective: We aimed to study the association between s-retinol and hip fracture and whether high s-retinol may counteract a preventive effect of vitamin D. Design: We conducted the largest prospective analysis of serum retinol and hip fracture to date in 21,774 men and women aged 65-79 y (mean age: 72 y) who attended 4 community-based health studies during 1994-2001. Incident hip fractures occurring up to 10.7 y after baseline were retrieved from electronic hospital discharge registers. Retinol determined by high-pressure liquid chromatography with ultraviolet detection in stored serum was available in 1154 incident hip fracture cases with valid body mass index (BMI) data and in a subcohort defined as a sex-stratified random sample (n = 1418). Cox proportional hazards regression weighted according to the stratified case-cohort design was performed. Results: There was a modest increased risk of hip fracture in the lowest compared with the middle quintile of s-retinol (HR: 1.41; 95% Cl: 1.09, 1.82) adjusted for sex and study center. The association was attenuated after adjustment for BMI and serum concentrations of a-tocopherol (HR: 1.16; 95% CI: 0.88, 1.51). We found no increased risk in the upper compared with the middle quintile. No significant interaction between serum concentrations of 25-hydroxyvitamin D and s-retinol on hip fracture was observed (P = 0.68). Conclusions: We found no evidence of an adverse effect of high serum retinol on hip fracture or any interaction between retinol and 25-hydroxyvitamin D. If anything, there tended to be an increased risk at low retinol concentrations, which was attenuated after control for confounders. We propose that cod liver oil, a commonly used food supplement in Norway, should not be discouraged as a natural source of vitamin D for fracture prevention.