Vitamin A and Bone Loss: When Retinol Intake Raises Fracture Risk
Vitamin A is famous as the cause of acute hypervitaminosis in Arctic explorers who ate polar bear liver. The more relevant chronic toxicity for modern supplement users is subtler: a body of observational evidence and mechanistic data linking sustained high preformed retinol intake — from supplements, cod liver oil, and liver-rich diets — to increased risk of hip fracture and osteoporosis in older adults. Beta-carotene does not carry this risk because conversion to retinol is regulated.
The Swedish observational signal
Melhus and colleagues 1998 reported that women in the highest quintile of retinol intake (more than ~1,500 mcg/day) had a doubled risk of hip fracture and significantly lower bone mineral density than women in the lowest quintile [1]. A follow-up Swedish cohort study by Michaëlsson and colleagues showed a graded increase in fracture risk above a serum retinol of approximately 75 mcg/dL [2]. The U.S. Iowa Women's Health Study and the Nurses' Health Study reported similar associations between preformed vitamin A intake and hip fracture.
The mechanistic basis
Retinoic acid, the active metabolite of vitamin A, accelerates osteoclast differentiation and stimulates bone resorption while inhibiting osteoblast activity. Animal studies in rats fed high vitamin A doses reproducibly show reduced cortical bone area and increased fracture susceptibility. Vitamin A and vitamin D act in partially opposing fashion on osteoblastic gene transcription, which may explain why the bone effects are most pronounced when vitamin D status is marginal [3]. The animal data are clearer than the human observational data because the doses studied are larger and the populations more uniform.
Confounding and uncertainty
Not all studies show the association — particularly when retinol intake in the population is moderate. The U.S. NHANES did not consistently replicate the Swedish signal, possibly because U.S. supplement-driven intake distributions are different. Reverse causation is also possible: people with low bone mass may be more likely to take supplements containing vitamin A. The 2003 IOM (now NASEM) review concluded that the human evidence was suggestive but not conclusive, and the U.S. tolerable upper intake level (UL) of 3,000 mcg/day for preformed retinol equivalents in adults rests partly on the bone-toxicity evidence [4].
Where the risk actually lives
Most adults eating a mixed diet do not approach problematic intakes. The risk concentrates in: regular cod liver oil users (one tablespoon contains roughly 4,500 mcg retinol depending on brand), high-potency multivitamins listing 5,000-10,000 IU preformed vitamin A daily, organ meat consumers eating beef liver several times a week, and people taking retinol-containing prenatal vitamins beyond pregnancy without adjusting the formulation. Beta-carotene supplements do not carry this risk because the conversion to retinol is downregulated when stores are sufficient [5].
How to keep vitamin A in the safe zone
For most adults, dietary vitamin A from mixed sources is adequate and unproblematic. People wanting a long-term insurance multivitamin should choose products that provide most or all of their vitamin A as beta-carotene rather than as preformed retinol, or that keep retinol at or below 50 percent of the RDA (so the supplement plus diet rarely exceeds the UL). Cod liver oil users should check the retinol content and not combine it with a separate multivitamin containing vitamin A. Pregnant women should not take retinol above ~3,000 mcg/day due to teratogenicity, which is a separate (and much more acute) concern than chronic bone effects [6].
The bigger picture
Vitamin A excess is a chronic-toxicity story without acute warning signs at the doses that matter. The fracture signal is one of the better examples in nutritional epidemiology where a fat-soluble micronutrient at supraphysiologic doses produces opposite effects from its low-dose physiological role. Calcium and vitamin D get most of the bone-health discussion in supplement aisles; vitamin A's role belongs in that conversation too — on the side of restraint.
Sources
- Melhus H, Michaëlsson K, Kindmark A, et al. "Excessive dietary intake of vitamin A is associated with reduced bone mineral density and increased risk for hip fracture." Ann Intern Med, 1998;129(10):770-778. PMID: 9841582. DOI: 10.7326/0003-4819-129-10-199811150-00003.
- Michaëlsson K, Lithell H, Vessby B, Melhus H. "Serum retinol levels and the risk of fracture." N Engl J Med, 2003;348(4):287-294. PMID: 12540641. DOI: 10.1056/NEJMoa021171.
- Conaway HH, Henning P, Lerner UH. "Vitamin a metabolism, action, and role in skeletal homeostasis." Endocr Rev, 2013;34(6):766-797. PMID: 23720297. DOI: 10.1210/er.2012-1071.
- National Institutes of Health Office of Dietary Supplements. "Vitamin A and Carotenoids: Fact Sheet for Health Professionals." Updated 2023. https://ods.od.nih.gov/factsheets/VitaminA-HealthProfessional/
- Feskanich D, Singh V, Willett WC, Colditz GA. "Vitamin A intake and hip fractures among postmenopausal women." JAMA, 2002;287(1):47-54. PMID: 11754708. DOI: 10.1001/jama.287.1.47.
- Rothman KJ, Moore LL, Singer MR, Nguyen US, Mannino S, Milunsky A. "Teratogenicity of high vitamin A intake." N Engl J Med, 1995;333(21):1369-1373. PMID: 7477116. DOI: 10.1056/NEJM199511233332101.