Vitamin A

Although vitamin A deficiency is a much greater problem than vitamin A toxicity, particularly in developing nations, both clinical and subclinical toxicities have been associated with excessive intakes of preformed vitamin A. Many cases of hypervitaminosis A involve ingestion of a large quantity at one time by young children, who, along with the elderly, are more susceptible to toxicity.^1,2 Acute toxicity is thought to occur when, within a short period of time, adults ingest more than 100 times the recommended daily allowance (RDA) and children ingest more than 20 times the RDA. However, in addition to acute toxicity, chronic intakes of high-dose vitamin A have also been associated with harm.³ The most recognized among these is an adverse effect on bone, with observational studies suggesting an increased risk of osteoporosis and fracture. Unfortunately, assessment of vitamin A toxicity is limited by the lack of sensitive laboratory markers.

Adverse reactions to acute toxicity in children can occur with intakes as low as 1500 international units per kilogram daily (IU/kg/day),⁴ and they are usually transient. Symptoms of acute hypervitaminosis A in children given 100,000 to 300,000 IU include diarrhea, headache (possibly resulting from elevated intracranial pressure), nausea, vomiting, occasional dizziness, and fever as well as a transient bulging of the fontanelle in infants. In adults, symptoms of toxicity may also include blurred vision and lack of muscular coordination.⁵ Chronic vitamin A excesses can precipitate alopecia, arthralgias, anemia, erythema, skin peeling, thickened epithelium, and fatty liver as well as heart, kidney, and testicular defects and hypercholesterolemia. Other less commonly reported symptoms include dysphagia due to vertebral hyperostosis, and intrahepatic cholestasis.⁶ Interestingly, in a small number of case reports of dysphagia, none of the patients reported vitamin A supplementation despite high serum retinol levels, suggesting an impairment of vitamin A metabolism rather than excessive intake.⁷ Usually most of the untoward effects of excess vitamin A intake are resolved with cessation of its use.

Of all the reported adverse effects, bone abnormalities have received the most attention, with excessive intake of vitamin A suggested to have a lasting detrimental effect on bone by inducing osteoporosis.^8,9 However, the data are mixed and may be confounded by other variables, particularly vitamin D status.

One 9.5-year study involving almost 35,000 postmenopausal women with hip and other fractures found little evidence of an increased risk of fracture with higher intakes of vitamin A or retinol. There was also no evidence of a dose-response relationship in hip fracture risk with increasing amounts of vitamin A or retinol from supplements. Furthermore, the results showed no association between vitamin A ingestion from food and supplements or food only and the risk of fractures of any kind.¹⁰ Similar results were published in the American Journal of Clinical Nutrition in 2009. This was a large observational study that included over 75,000 participants from the Women’s Health Initiative. Retinol and vitamin A intake were not significantly associated with either hip or total fracture incidence among postmenopausal women. However, women in the highest quintile of retinol and vitamin A intake who also had a low intake of vitamin D did have a modest (15%-20%) increased total fracture risk.¹¹ A smaller study that enrolled Spanish postmenopausal women did find an independent risk for osteoporosis among those with the highest intake, but this risk was magnified when combined with low vitamin D levels.¹² The interaction between high vitamin A and low vitamin D levels appears biologically plausible, as vitamin A may antagonize some of vitamin D’s actions, including calcium absorption.¹³ This may be relevant not only to bone health but possibly to susceptibility to respiratory infection as well.¹⁴ A 2007 review of the bone effects of vitamin A concluded that the poor sensitivity of laboratory markers and assessment of dietary intake may contribute to the conflicting findings; it suggested that future studies incorporate superior analytic techniques, specifically stable-isotope-dilution methodology.¹⁵ Although serum retinol is often employed to screen for vitamin A toxicity, it is thought to have poor sensitivity because it is subject to homeostatic control over a wide range of intakes as well as hepatic concentrations, and thus does not necessarily represent liver stores.¹⁶ Additionally, many clinical factors interfere with its accuracy. One alternative is the measurement of fasting retinyl ester concentrations. When more than 5% to 10% of circulating vitamin A is in the form of retinyl esters, it may indicate either hepatic storage capacity or the capacity of the retinol-binding protein has been exceeded. Unfortunately, elevated retinyl esters do not necessarily indicate impaired liver function and are not sensitive to subclinical toxicity.^15,17

Although expensive and not widely available, stable isotope dilution techniques appear to correlate well with values determined by liver biopsy and may emerge as the best marker of total vitamin A stores in both deficiency and toxicity. Indeed, variations of this method may be used to determine the intake needed to maintain target body storage levels. In animal models they have shown 100% sensitivity for the diagnosis of hypervitaminosis A.^18,19 Although the deuterated retinol dilution method has been validated in both children and adults to give a quantitative estimate of internal stores, it needs further verification among diverse populations and greater accessibility.²⁰ When large doses of vitamin A are being given, careful monitoring is necessary. Rather than sudden ingestion of large doses, a gradual stepwise increase in dosage is indicated, with an evaluation of symptoms made before the dosage is increased. Usually, the first symptom of hypervitaminosis to be recognized is frontal headache. If signs or symptoms appear, supplementation should be discontinued until they disappear. Levels of liver enzymes should be determined periodically to check for hepatic damage. Typically, levels of aspartate transaminase are the first to be affected.^1,2,21,22 Patients whose liver function is compromised by viral hepatitis, protein-energy malnutrition, cirrhosis, or hemodialysis seem to be the most vulnerable to vitamin A toxicity and to require close monitoring.⁴ Vitamin A levels during pregnancy must be carefully assessed because both deficiency and excess can bring about undesirable results. Supplementation above the RDA is not warranted in pregnant or potentially pregnant women. According to one large observational study published in the New England Journal of Medicine, women consuming greater than 10,000 IU of vitamin A during pregnancy (specifically during the first 7 weeks after conception) had a 1 in 57 risk for having a child born with a birth defect.^22a

It is the opinion of this author that much if not most of the problem of vitamin A toxicity, except at very high dosages, is likely due to the high prevalence of vitamin D deficiency.

Carotenoids

Carotenoids appear to be without toxic effects at the therapeutic doses customarily used. The only effect of large dosages is an apparently benign yellowing of the skin. Although carotenoid toxicity is limited, there is concern that some individuals have difficulty converting carotenoids to vitamin A and may be more prone to vitamin A deficiency.^23–25

Several large, widely publicized therapeutic trials with synthetic beta-carotene have found that it appears to raise the risk of lung cancer in heavy smokers. It may also pose an increased risk for gastric cancer, particularly among smokers and those exposed to asbestos.²⁶ However, several factors complicate the interpretation of these results. The significance of these trials is fully discussed in Chapter 69.

Vitamin D

Significant advances have been made in understanding the role and importance of vitamin D in human health. Deficiency is now known to be widespread, with suboptimal levels much more prevalent than toxicity. The use of 25-OH vitamin D is widely accepted as a reliable biomarker, with most indicators suggesting that a level of 75 to 110 nmol/L is sufficient, although some studies indicate that even higher levels may be optimal.^27,28 The upper limits of 25-OH vitamin D are not clearly established, although levels less than 250 nmol/L are considered safe.²⁹ Therapeutic strategies should target 25-OH vitamin D levels rather than a specific supplemental dose, as the effect of supplementation on serum levels varies considerably between individuals. Doses between 2000 IU and 4000 IU will bring the majority of individuals within the range of 75 to 110 nmol/L.²⁷ Nevertheless, some will require higher dosing, and this is also a consideration for individuals with less functional vitamin D receptor polymorphisms. Doses as high as 40,000 IU per day have not been associated with toxicity.²⁸ However, very high single doses (500,000 IU in a single annual dose) have been associated with an increased risk for fracture and falls in a temporal pattern, with the highest risk in the period after administration.³⁰ Thus, lower doses given more frequently (i.e., more physiologically) are preferred. Despite ongoing controversy, vitamin D3 appears to be more potent and to produce greater storage than D2.³¹

Granulomatous diseases, such as sarcoidosis, warrant special concern because these individuals are more susceptible to hypercalcemia. Although many have low levels of 25-OH vitamin D (which appears to increase the risk for sarcoidosis), they also have elevated levels of 1,25 dihydroxyvitamin D and thus require careful management.³² Apparently there is overconversion of 25-OH vitamin D3 to 1,25(OH)2-vitamin D3 by macrophages in granulomatous disease.³³

Vitamin E

Although for many years observational studies found vitamin E supplementation to be safe, several controlled trials have been published suggesting harm with supplementation. For example, in a large meta-analysis of randomized placebo-controlled trials in which participants were given between 50 and 800 IU natural or synthetic vitamin E per day, supplementation was found to reduce the risk of ischemic stroke by 10% but to increase the risk of hemorrhagic stroke by 22%.³⁴ Similarly, a meta-analysis published in the Annals of Internal Medicine found that supplementation with more than 400 IU vitamin E increased all-cause mortality.³⁵

Although the use of synthetic versus natural vitamin E may explain some of the increase in adverse effects, the natural form of vitamin D (d-alpha tocopherol) used in many clinical trials is not without risk. For example, supplementation with natural vitamin E at 400 IU per day was associated with an increased risk of heart failure among patients with diabetes or vascular disease.³⁶

An explanation that appears more plausible is that despite the physiologic benefits of alpha-tocopherol, high-dose supplementation depletes other forms of naturally occurring vitamin E, such as beta- or gamma-tocopherol, which have greater physiologic significance. For example, in an observational study of elderly patients, higher plasma levels of beta-tocopherol were associated with a reduced risk of developing Alzheimer’s disease, whereas other forms of vitamin E were only marginally significant.³⁷ The use of both gamma- and alpha-tocopherol in patients with the metabolic syndrome was shown to be superior to either used alone. Moreover, in vitro and in vivo evidence indicates that alpha-tocopherol not only failed to demonstrate anticancer properties but also blocked the anticancer effects of gamma-tocopherol.^38,39 Gamma-tocopherol is actually more prevalent than the alpha form in the U.S. diet as well as in many plant seeds, although the vast majority of trials and available products use alpha-tocopherol.⁴⁰ Thus, it may not be “vitamin E” that has the harmful effects mentioned above but rather the isolated use of alpha-tocopherol. Additional factors are likely to have an influence as well, such as age and vitamin C intake.⁴¹ Genetics are also likely to play a role, as diabetic patients with the haptoglobin 2-2 genotype are more likely to receive benefit from supplementation.⁴² Unfortunately, most laboratory evaluations of vitamin E’s toxicity are based on alpha-tocopherol plasma or serum levels and thus may not be helpful in determining toxicity.