High Intakes of Vitamin A May Increase the Risk of Fracture
Previous studies using dietary estimates of vitamin A intake have shown that excess vitamin A consumption can be detrimental to human bone, including an increased risk of fracture. Vitamin A is stored is mobilized from the liver in the form of retinol. Retinol is realeased in target cells and converted to retinoic acid, which suppresses osteoblast activity and stimulates osteoclast formation in vitro. In a population-based, 30-year longitudinal study, Michaelsson et al use a biological marker of vitamin A, serum retinol, to assess whether increased serum retinol levels correlate with increased fracture risk.
Methods: From 1970 to 1972, 2322 men between the ages of 49 to 51 were enrolled in the study. A medical and lifestyle questionnaire and interview, serum samples and anthropometric measurements were taken at baseline. Serum retinol and beta carotene levels were measured on baseline samples only. 80% (1860) and 53% (1221) of the total cohort took part in a second and third evaluation at age 60 and 70, respectively. Dietary assessments were done in a subgroup of men (1138) as part of the third evaluation. Follow up was calculated from the date of enrollment until the date of first fracture, death, move from country, or end of study (12/31/01). Participants were categorized into one of 5 groups based on their baseline serum retinol level; <1.95 mol/liter, 1.95-2.16 mmol/liter, 2.17-2.36 mmol/liter, 2.37-2.64 mmol/liter, and >2.64 mmol/liter. Cox regression analysis was done to determine increased fracture rate ratios.
Results: The average follow up was 24 years for subjects that had fractures and 29 years for those without a fracture. During the study duration totaling 56,281 person years of observation, 266 men had one or more fractures. The overall risk of fracture increased by 26% for every increase in 1 SD in serum retinol, yielding a 1.26 multivariate rate ratio. In quintile comparisons, patients in the highest quintile for retinol had a multivariate rate ratio of 1.64 for any fracture as compared with the middle quintile. This rate ratio increased to 1.78 when the analysis was restricted to fractures at sites typical of osteoporotic fractures. When hip fractures alone were analyzed, the rate ratio for the highest quintile was 2.47 as compared to the middle quintile. Further analysis of the highest quintile revealed that patients in the 99th percentile (>3.6 mmol/liter) had a multivariate rate ratio of 7.14 (p<0.001).
111 of the 1221 men had dietary data available. Men with an estimated vitamin A intake of >1.5 mg/day had a rate ratio of 2.00 for any fracture compared with men whose dietary intake of vitamin A was <0.53 mg/day. Dietary beta carotene intake and serum beta carotene levels were not associated with the risk of fracture.
Conclusion: These data suggest that hypervitaminosis A may increase the risk of fracture in men above the age of 50.
Editorial Comments: The strengths of this study are the prospective longitudinal design and use of a serum biomarker for vitamin A instead of dietary recall alone. the weaknesses are that only one retinol measurement was made, and the prolonged lag time between this measurement and the outcome. Nonetheless, these results corroborate previous in vitro and epidemiologic studies suggesting a relationship between hypervitaminosis A and osteoporosis.
The main sources of retinoids are fish dietary products and liver, and some fortified products. The old adage, “all things in moderation”, remains good advice in the modern era.