Safety

Vitamin A Toxicity: Retinol, Beta-Carotene, and the Asymmetric Risks

May 12, 2026 · 4 min read ·

Vitamin A is one of the few vitamins where genuine, dose-dependent supplement toxicity is well documented — and one of the few where the two main supplement forms (retinol and beta-carotene) have different safety profiles. The story is complicated by the fact that supplementation has been tested for cancer prevention in large randomised trials, and the results were not what investigators expected.

Retinol — the preformed vitamin A

Retinol and retinyl esters are the active form, found in liver, dairy fat, and fish liver oils. The Tolerable Upper Intake Level for adults is 3,000 µg/day (10,000 IU) [1]. Chronic intakes above the UL produce hypervitaminosis A — headache, dry skin, hair loss, hyperlipidaemia, raised intracranial pressure, hepatotoxicity, and at extremes, liver fibrosis. Acute toxicity at very high single doses can produce vomiting, vertigo, and desquamation.

Teratogenicity

The most clinically important toxicity is teratogenic. Retinol intakes above 10,000 IU/day in early pregnancy are associated with increased risk of cranial-neural-crest defects in the fetus. Multiple prospective studies have established this signal at supplementation levels routinely consumed by women not specifically planning pregnancy [2]. The mechanism — retinoid signalling in neural-crest patterning — is the same pathway exploited by isotretinoin, which has an absolute pregnancy contraindication.

Beta-carotene — the provitamin form

Beta-carotene from plant foods is converted to retinol with a regulated efficiency that drops as body stores rise; this is why food sources of carotenoids essentially cannot cause vitamin A toxicity. High-dose beta-carotene supplements bypass this regulation but do not produce hypervitaminosis A. They do, however, have their own safety signal.

The ATBC and CARET trials — lung cancer in smokers

Two large trials of beta-carotene supplementation tested whether the antioxidant could prevent lung cancer in high-risk populations. The Alpha-Tocopherol, Beta-Carotene (ATBC) trial randomised 29,133 Finnish male smokers to beta-carotene 20 mg, vitamin E, both, or placebo for 5–8 years. Lung cancer incidence was 18% higher in the beta-carotene group than placebo [3]. The Beta-Carotene and Retinol Efficacy Trial (CARET) randomised 18,314 smokers and asbestos workers to beta-carotene 30 mg + retinol vs placebo, and was stopped early when interim analysis showed a 28% increase in lung cancer in the active arm [4].

What this means in 2026

High-dose beta-carotene supplementation (≥20 mg/day) in current or former smokers is associated with increased lung cancer risk and is contraindicated. The risk does not extend to dietary beta-carotene from food sources. The mechanism is uncertain — possibly pro-oxidant effects of high beta-carotene concentrations in a chronically oxidatively stressed pulmonary milieu — but the trial signal is robust [5].

Vitamin A in eye disease — the AREDS distinction

The original Age-Related Eye Disease Study (AREDS) included beta-carotene in its formula. The AREDS2 trial replaced beta-carotene with lutein/zeaxanthin precisely because of the smoker lung cancer signal. Modern AMD supplements should not contain beta-carotene; check labels [6].

Practical takeaways

For non-pregnant adults without smoking history, a daily multivitamin containing 1,500–3,000 µg retinol or 5–10 mg mixed carotenoids is safe. Pregnant or potentially pregnant women should keep total retinol intake below 10,000 IU/day and avoid liver and high-dose retinol products. Current and former smokers should avoid beta-carotene supplements above food-level intakes. Patients with chronic liver disease are at higher risk of retinol-induced hepatotoxicity. Vitamin A as a cancer or general-prevention supplement is not supported by current evidence and is potentially harmful in specific groups.

Carotenodermia — the harmless beta-carotene side effect

A useful clinical distinction: very high intake of carotenoid-containing foods or supplements produces a yellow-orange tinge to the palms, soles, and sometimes the face, called carotenodermia. It is reversible, harmless, and entirely different from jaundice (which spares the sclerae and lacks the orange hue concentrated in fatty tissues). Carotenodermia indicates high carotenoid status but not vitamin A toxicity, because the body downregulates beta-carotene → retinol conversion at high stores. The take-home: orange palms from supplement use is not a toxicity signal; it is a saturation signal, and the supplement can be continued or reduced without medical concern. The cancer risk in smokers is a separate issue from the carotenodermia phenotype.

Sources

  1. NIH Office of Dietary Supplements. "Vitamin A and Carotenoids — Health Professional Fact Sheet." Updated 2022.
  2. Rothman KJ, Moore LL, Singer MR, et al. "Teratogenicity of high vitamin A intake." N Engl J Med, 1995;333(21):1369-1373. PMID: 7477116. DOI: 10.1056/NEJM199511233332101.
  3. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. "The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers." N Engl J Med, 1994;330(15):1029-1035. PMID: 8127329. DOI: 10.1056/NEJM199404143301501.
  4. Omenn GS, Goodman GE, Thornquist MD, et al. "Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease." N Engl J Med, 1996;334(18):1150-1155. PMID: 8602180. DOI: 10.1056/NEJM199605023341802.
  5. Bjelakovic G, Nikolova D, Gluud LL, Simonetti RG, Gluud C. "Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases." Cochrane Database Syst Rev, 2012;(3):CD007176. PMID: 22419320. DOI: 10.1002/14651858.CD007176.pub2.
  6. Chew EY, Clemons TE, Sangiovanni JP, et al. "Lutein + zeaxanthin and omega-3 fatty acids for age-related macular degeneration: the Age-Related Eye Disease Study 2 (AREDS2) randomized clinical trial." JAMA, 2013;309(19):2005-2015. PMID: 23644932. DOI: 10.1001/jama.2013.4997.