点击显示 收起
Linus Pauling Institute
Oregon State University
Corvallis, OR 97331-6512
E-mail: maret.traber{at}oregonstate.edu
Dear Sir:
My recent editorial in the Journal (1) emphasized the difficulty in setting the daily -tocopherol requirement and was not intended as advocacy for high-dose vitamin E supplementation. Evidence-based medicine is not based only on randomized clinical trials (RCTs) and meta-analyses but also takes into account all relevant evidence. The scientific evidence that vitamin E is essential for human health is overwhelming. -Tocopherol insufficiency results in a sensory neuropathy, which has been documented in patients with ataxia and vitamin E deficiency (AVED). Symptoms are secondary to a genetic defect in the hepatic -tocopherol transfer protein (-TTP) (2). The plasma -tocopherol concentrations of persons with AVED are one-tenth of normal, and their nerves become -tocopherol–depleted before symptom onset (3); -tocopherol supplements reverse or halt symptom progression (2). Thus, the nervous system is vulnerable to inadequate -tocopherol status.
Hemilä and Miller refer to peroxide-dependent erythrocyte hemolysis as "a surrogate endpoint that has not been validated against any clinically relevant outcome." However, more than 30 y ago, this test was used clinically to show that children with cystic fibrosis were vitamin E deficient (4). These children absorbed vitamin E poorly and thus had low plasma -tocopherol concentrations, anemia, and increased erythrocyte turnover—symptoms that were reversed by -tocopherol supplements (4). The Food and Nutrition Board (FNB) used peroxide-dependent erythrocyte hemolysis data to set the current recommended dietary allowance (RDA)—15 mg -tocopherol (5)—which is lower than the "current US recommendation for vitamin E" cited by Hemilä and Miller. The US RDA uses the daily value (DV), which is defined from the 1968 FNB recommendation (also based on erythrocyte hemolysis) of 30 IU (30 mg dl--tocopheryl acetate); %DV is used on food labels.
Hemilä and Miller stated, "Most blood concentrations of micronutrients, including antioxidants, are collinear." This statement is incorrect with respect to vitamin E. An appreciation of the complex pharmacokinetics of -tocopherol is essential to understanding its disposition and human vitamin E status. High plasma -tocopherol concentrations may reflect high -tocopherol intakes. However, hyperlipidemia also elevates plasma -tocopherol, because -tocopherol concentrations are collinear with circulating lipids. In normolipidemic subjects, low plasma -tocopherol concentrations reflect inadequate vitamin E intakes. When inadequate amounts of -tocopherol are consumed, plasma concentrations are maintained by -TTP, whereas peripheral tissue -tocopherol depletion occurs (3). To assess vitamin E status, one should measure plasma -tocopherol and lipid concentrations and, ideally, tissue -tocopherol concentrations.
-Tocopherol is not found in most high-antioxidant foods, such as fruit and vegetables. Low-fat diets decrease -tocopherol intakes because the fat-soluble vitamin is largely present in high-fat foods. Therefore, substantial changes in the kinds of foods Americans eat are needed for them to obtain 15 mg -tocopherol/d from dietary sources, such as seeds, nuts, spinach, and safflower oil.
What is the downside to consuming a less-than-optimal -tocopherol intake? It is difficult to determine, because it takes decades for symptoms of suboptimal vitamin E status to become readily apparent. It took 40 y for symptoms to be detectable in a patient with chronic fat malabsorption and -tocopherol deficiency (7). Such a delay in the first appearance of symptoms shows the fallacy of concluding, after an observation of only a relatively short time (eg, 5 y), that there is no harm to inadequate vitamin E intakes.
The Alpha-Tocopherol Beta-Carotene Cancer Prevention (ATBC) Study provides an interesting contrast. Analysis of baseline serum -tocopherol concentrations in 29 000 men, nearly one-half of whom are now dead, showed a significant correlation between high serum -tocopherol status and lower chronic disease mortality (8), which suggested that long-term dietary habits that provide higher -tocopherol intakes are beneficial. In contrast, supplementation for only 5–8 y with 50 mg all-rac--tocopheryl acetate (22 mg 2-R--tocopherol or 1.5 times the RDA) showed no such relation (8). Given that clinical symptoms take decades to appear in humans with various chronic diseases, the effects of correcting suboptimal vitamin E intakes cannot be assessed by using RCTs that last only years, rather than decades. Therefore, the suggestion by Hemilä and Miller to carry out RCTs seems impractical, if not unethical, given the potential for inadequate -tocopherol intakes in the "placebo" group to deplete tissue, especially nervous system tissue, of -tocopherol.
Hemilä and Miller contend that high-dose -tocopherol is dangerous, but they specify no mechanism for any adverse effect. Miller et al (9), in a meta-analysis analyzing the relation between dose and mortality, found a benefit of 4% when vitamin E supplements were provided in the range of dietary requirements. This outcome contradicts their widely publicized claim of vitamin E supplement harm, a claim that was criticized in many letters to the editor in the journal that published the report of Miller et al (see the July 2005 issue of Annals of Internal Medicine). A systematic review sponsored by the National Institutes of Health concluded that the evidence was insufficient to prove the "presence or absence of benefits" for vitamin E supplements (usually 400 IU) for the prevention of cancer or chronic disease (10). The Cache County Study found that vitamin E supplements had "no effect" on mortality, but their conclusion was based on a combination of outcomes "[in which] increased mortality was observed in subjects with severe cardiovascular disease and a possible protective effect in those without" (11). This latter finding is of interest because the Women's Health Study, a primary prevention trial with vitamin E supplements (600 IU every other day for 10 y) in 40 000 healthy women, concluded that vitamin E had no effect on the occurrence of heart disease or cancer (12). However, in subgroup analysis, vitamin E supplements decreased cardiac mortality by 49% in women >65 y old—ie, those who are at greater risk of heart disease than are younger women (12). Taken together, these studies suggest that, in healthy persons, a generous -tocopherol intake for a prolonged period is beneficial, not harmful. Thus, intakes in the range of the RDA—15 mg -tocopherol/d—obtained from a healthy diet, from a multivitamin, or as an -tocopherol supplement, appear to me to be a prudent public health recommendation.
ACKNOWLEDGMENTS
The author had no personal or financial conflict of interest.
REFERENCES