Antioxidant supplements – can they do athletes more harm than good?

Oxygen is amazing stuff. Thanks to its special chemical reactivity, it provides us with the energy required to sustain life, including the ability to power movements and muscular contraction. This explains why oxygen – and the ability to absorb, transport and use it – is so important to endurance athletes, who need lots of the stuff to sustain maximum power and work outputs.

However, the oxygen molecule is a double-edged sword. For this same chemical reactivity can also wreak cellular havoc by means of the transient, highly reactive and potentially extremely destructive molecular species called free radicals, which are produced unavoidably as a consequence of harnessing the chemical energy of oxygen within the body.

Without getting into the chemical fine detail, a free radical is simply a molecule that contains an unpaired electron. Why is that important? Well, the laws of physics dictate that electrons are only really ‘happy’ and stable when paired up with a partner, which explains why stable, or nonreactive, chemical molecules nearly always have chemical bonds containing a pair of shared electrons.

An atom or molecule containing a single, or unpaired, electron is distinctly ‘unhappy’; it has a lot of energy, is very unstable and is highly reactive, eager to snatch an electron from somewhere else in order to form a stable electron pair. This is what free radicals are: molecules or molecule fragments containing unpaired electrons, desperate to snatch electrons from other chemical bonds in order to form a stable electron pair.

Chaos of free radical chain reactions

But by doing this, and stealing a single electron from a molecule already containing an electron pair, a second radical is created, which can itself go on to snatch an electron from elsewhere.

When free radicals are generated in the body, a chain reaction is set up in which thousands of molecules are robbed of an electron and then obliged to pinch one from somewhere else! Remember what it was like at school when the teacher gave out 29 textbooks to 30 classmates? If you were the unlucky one without, you nicked a book from someone else; when he found out, he nicked one from someone else, and so on. Although the class was only ever one book short, the result was often a chain reaction of thefts, causing utter chaos in the classroom!

Free radical chain reactions are very fast. One free radical can easily produce a chain of a hundred billion reactions in the time it takes to blink. Each individual free radical in that chain has only a very fleeting existence, perhaps lasting for just one hundred millionth of second before snatching back an electron from another chemical bond. For this reason, you could never go and collect a bottle of free radicals. But the important thing about them is the trail of damage they leave behind in the cell. If electrons are being ripped out of chemical bonds that hold together structures like cell walls or DNA, irreparable damage to the cell and/or its genetic material may be the end result. And this damage is now thought to be one of the root causes of degenerative diseases, inflammation and the ageing process in general.

The good news, though, is that human body comes equipped with a number of systems capable of deactivating the free radicals produced as a result of aerobic metabolism, and dissipating their energy harmlessly. Collectively known as the antioxidant defence system, these systems use both antioxidant enzymes (large protein molecules manufactured in the body) and antioxidant nutrients (consumed in the diet) to mop up unwanted free radical activity, ‘soak up’ the energy of these unpaired electrons and break the chain of free radical reactions, thereby minimising damage to the body.

In recent years, there has been much speculation that athletes, who not only consume more oxygen than others to fuel their training but also frequently train at or near their maximum oxygen uptakes, might be at increased risk of free radical-induced damage, or ‘oxidative stress’. Athletes don’t just process a larger volume of oxygen than their sedentary counterparts – they also process it at a higher rate; during training, the rate of oxygen processing by the mitochondria (the energy-producing furnaces in the cells) can rise by a factor of 20, placing exceptionally high demands on antioxidant defence systems. The fact that free radical generation does increase during vigorous exercise is no longer in doubt(1-5). However, considerable confusion remains about the implications of this increased free radical generation. There are three key questions:

  1. Does this increased oxidative stress actually lead to significant biological damage within the cells of athletes?
  2. Can the body of an athlete adapt to this increased oxidative stress by manufacturing higher levels of the antioxidant enzymes?
  3. Can an athlete’s antioxidant defences be fortified by ingesting increased dietary amounts of the antioxidant nutrients, including beta carotene, vitamin C, vitamin E and the mineral selenium?

Free radical activity during exercise

The answer to the first question is not yet clear. Two powerful techniques, known as ‘electron spin resonance’ and ‘paramagnetic resonance spectrometry’ now enable scientists to directly measure the concentration of free radicals during exercise and can be used to detect the ‘superoxide’ radical, one of the most reactive and damaging radical species. However, most of these studies have been carried out in animals rather than humans; moreover, it is not possible to prove conclusively that the increased production of superoxide radicals automatically leads to free radical damage.

An alternative approach is to look for signs of free radical damage, rather for the presence of the free radicals themselves. One of the commonest current methods is to measure how much lipid peroxidation has occurred. When oxygen-free radicals attack the lipid membranes around cells, molecules called peroxides are formed. These peroxides are not produced in other metabolic pathways, so an increase in peroxide concentration is a sure sign that more oxidative stress has occurred. Other techniques look for signs or fragments of oxygen radicaldamaged DNA, such as 8-hydroxyguanine.

However, it is important to realise that in humans these tests are subject to error. Many of these oxidative stress markers are very fragile and readily degrade before analysis, while other substances can interfere with the testing reagents, producing false positive readings. Relying on a single marker to measure oxidative stress in humans is, therefore, fraught with difficulties and probably explains some of the conflicting results that have emerged from clinical trials.

Conflicting results on oxidative stress

For example, increases in blood levels of a molecule called malondialdehyde (MDA), which is formed in the body when lipids are damaged by oxygen radicals, have been found after:

  • an 80k race(6);
  • a 30-minute treadmill test at 60% and 90% of maximal oxygen uptake(7);
  • downhill running(8);
  • incremental cycling tests to exhaustion in sedentary and moderately trained men(9,10).

By contrast, no increases in MDA were found after:

  • a half-marathon(11);
  • 60 minutes of bench-stepping exercise(12);
  • maximal cycle ergometry exercise(13);
  • incremental cycle ergometry exercise in elite athletes(14).

The implication of these conflicting results is that tests for oxidative stress and damage in humans need be interpreted with caution, especially when a single marker is used.

The human body can adapt to many environmental and metabolic stressors, so can it adapt to oxidative stress? On balance, the evidence suggests that it can. A number of studies have compared the antioxidant defence systems of athletes before and after a period of increased training intensity or duration and have found that both increased volume and intensity of training stimulate the production of antioxidant enzymes in the body, including glutathione peroxidase and superoxide dismutase(15,16,17). Moreover, some studies have also shown that this increase in antioxidant enzymes can reduce the levels of oxidative stress markers in the blood after training, so apparently offering protection against oxidative damage(18).

However, these results still need to be interpreted with caution because many of the studies have used different markers of antioxidant status and different training levels of subjects. More importantly, it is highly debatable whether the increased production of antioxidant enzymes observed is sufficient to combat the increased oxidative stress of heavy training loads, which has led to suggestions that athletes should take further steps to boost their defences by supplementing their diet with antioxidant nutrients.

This is where the story begins to get really tangled. Some studies have demonstrated that certain antioxidant nutrients can reduce apparent oxidative stress when supplemented at higher levels than would ordinarily be found in the diet. For example, a selenium-supplemented group of healthy males produced significantly higher levels of glutathione peroxidase (one of the body’s main antioxidant enzymes) in response to a mixture of treadmill running and cycling at different intensities (65-100% VO2max) than a control group(19).

 Similar beneficial effects have been observed for vitamin E. In a long-term study on endurance racing cyclists, a group supplemented with vitamin E at 10 times the normal rate showed a smaller increase in blood MDA after strenuous exercise than a placebo group(20). The supplemented cyclists also had lower levels of blood creatine kinase (a protein normally found in muscle, which can leak into the blood after membrane damage), suggesting a protective effect of vitamin E on muscle damage induced by oxidative stress.

Benefits of vitamins C and E

Other studies have also indicated that vitamin E supplementation may help reduce oxidative damage during exercise. When cyclists were supplemented with vitamin E at 40 times the RDA, the amount of pentane they breathed out from their lungs (pentane is a gas produced by lipid peroxidation) dropped significantly(21). There is also some evidence, albeit rather less convincing, that vitamin C offers antioxidant protection, particularly when given in combination with vitamin E. For example, 400IU of vitamin E and 200mg of vitamin C taken for four weeks before a marathon run resulted in reduced levels of blood MDA immediately after the event and persisted for 24 hours(22).

However, other well-conducted studies have cast doubt on the efficacy of antioxidant nutrient supplementation. Athletes ingesting either 2,000mg per day of vitamin C or a carbohydrate placebo were asked to run 27k, after which their blood levels of dienes (a marker of lipid peroxidation) was measured. No differences were observed between the groups(23). Another study comparing athletes supplemented with a combination of antioxidant nutrients (294mg vitamin E, 1,000mg vitamin C and 60mg of coenzyme Q10) and placebo before a 31k run found that the blood antioxidant potential (a measure of total antioxidant activity) was raised substantially in the supplemented group; however, there was no corresponding reduction in the amount of LDL diene conjugation (a measure of oxidative stress inflicted on molecules called low-density lipoproteins, which circulate in the bloodstream)(24).

A recent fascinating American study examined the effects of supplemental vitamin C (500mg per day) and vitamin E (400IU per day) for two months on oxidative damage to DNA by measuring the levels of a marker substance called 8-hydroxy-2’-deoxyguanosine (8-OHdG) excreted in the urine(25). They also collected detailed dietary information from each of the 184 subjects in the study. The researchers found that, by comparison with placebo, neither vitamin reduced the level of excreted 8-OHdG, suggesting no effect on oxidative damage to DNA.

Intriguingly, however, the researchers found that higher intakes of fruit and vegetables did reduce the amount of excreted 8-OHdG. They also found that the greater the level of exercise, the lower the level of damaged DNA marker, supporting the hypothesis that the body can upregulate its antioxidant defence systems in response to increased oxidative stress.

Although the increased intake of fruit and vegetables correlated with an increase in dietary vitamin C intake (fruit and vegetables being particularly rich in this vitamin), the researchers did not believe that these higher vitamin levels were responsible for the reduction in DNA damage (otherwise this same reduction should have been seen in the supplemented group, which it wasn’t). Rather they concluded that there there must be other biologically active substances in fruit and vegetables responsible for this protective effect (something we’ll return to later).

Staying on the safe side

Given current uncertainties about the effectiveness of antioxidant nutrient supplementation, wouldn’t it be wisest for athletes to take a supplement containing a mixture of the antioxidant nutrients ‘just to be on the safe side’? Perhaps not, because a new study suggests that, far from being synergistic, some antioxidant nutrients may actually work against each other(26)! Seven trained male cyclists were treated with four different supplementation regimes, as follows:

  • placebo;
  • 1,000mg of vitamin C per day;
  • 400IU of vitamin E per day;
  • 1,000mg of vitamin C plus 400IU of vitamin E.

After completing a steady-state ride and performance ride on the ergometer, blood samples were drawn and analysed for MDA (a lipid peroxidation marker). As expected, there were no differences in terms of performance benefits between the different supplementation regimes. In line with other studies, it was also found that the combination of vitamins C and E reduced blood levels of MDA. However, the researchers were surprised to discover that vitamin E supplementation alone reduced preexercise blood MDA levels far more than the combined supplement – by around 40% – and also substantially reduced post-exercise MDA levels!

More worrying, though, was the finding that, by comparison with placebo, vitamin C supplementation alone actually elevated MDA levels; in other words, it acted as a pro-oxidant rather than an antioxidant. The researchers concluded that, while 400IU daily of vitamin E did offer protection, 1,000mg daily of vitamin C appeared to promote cellular damage. This is certainly a plausible theory because, taken in excess, vitamin C is known to exhibit a phenomenon known as ‘Fenton chemistry’, where it acts as a catalyst to stimulate the production of the highly damaging hydroxyl radical from minerals (such as iron) and naturally occurring substances (such as hydrogen peroxide) in the body.

Oxidative stress may be essential

Although appropriate levels of antioxidant supplementation may offer some long-term protection to athletes, and although there is some limited evidence that vitamin C may help reduce post-exercise muscle damage, there is no real evidence to date that antioxidant nutrients can boost short-term performance in athletes. Indeed, some scientists have even proposed that excessive antioxidant supplementation may be counterproductive because oxidative stress and some degree of free radical damage may actually be an essential part of the adaptation process within muscles.

Additionally, recent animal studies lend support to the notion that ‘more isn’t always better’. In one of these, greyhounds were treated with three different supplementation regimes, as follows(27):

  • placebo;
  • 1,000mg vitamin C daily with food;
  • 1,000mg administered orally one hour before racing on race days and with food on non-race days.

The results demonstrated that, regardless of when the vitamin C was administered, supplemented dogs ran 0.2 seconds slower over 500m than their non-supplemented counterparts – a small but statistically significant difference. These results appear to support those from an earlier study, which showed that, while a modest daily dose of 100IU of vitamin E didn’t affect running performance, a higher dose of 1,000IU caused greyhounds to run more slowly(28).

Other recent studies seem to indicate that high doses of antioxidant nutrients may actually harm performance. For example, rats fed high doses of vitamin E were not able to produce as much muscle force as their unsupplemented counterparts during low frequency stimulation(29); and in a human study, vitamin C and N-acetyl cysteine (another antioxidant) given during the acute phase inflammatory response to an eccentric arm injury increased the amount of oxidatively damaged lipids, resulting in transiently increased tissue damage(30).

The best advice for athletes

Faced with this bewildering array of information, what’s the best advice for athletes seeking maximum performance today and optimal protection for tomorrow? First, the evidence is that on balance, while not improving short-term performance, modest doses of antioxidant nutrients do appear to offer some protection. However, more is not necessarily better and higher doses may actually increase oxidative damage and could even lead to reduced performances.

Secondly, because antioxidant nutrients work together synergistically, both with each other and with the antioxidant enzymes of the body, any supplementation should be in the form of a complex (for example containing beta-carotene, vitamin C, vitamin E and selenium) rather than single nutrients. Although it is difficult to make hard and fast recommendations, the evidence suggests that total daily vitamin C intake should not exceed 500mg per day, with 300-400mg per day the upper supplementation limit for most people.

Although there is less evidence for detrimental effects of high vitamin E supplementation, many studies suggesting a protective effect have used around 400IU per day, and it seems prudent not to exceed this figure. The UK Foods Standards Agency suggests a safe upper limit of 350mcg per day for selenium supplementation, but in the absence of a proven deficiency most studies have shown little or no benefit to exceeding 200mcgs per day. The safe upper limit for beta-carotene is set at 7mg per day.

Finally – and perhaps most importantly of all – don’t forget about fruit and vegetables. In recent years, there has been an explosion of research into naturally-occurring substances in plants (often responsible for giving the plant its characteristic colours and flavours) called phytochemicals. Many of these compounds display remarkable antioxidant capacities, sometimes tens or even hundreds of times greater than the antioxidant nutrients. Example include the carotenoid family found in red and green fruits and vegetables, the flavenoid family found in citrus fruits, the tocotrienol family found in nuts, seeds and wheatgerm, and a number of sulphur-containing compounds, such as sulphorane, found in broccoli, and allicin found in garlic.

As a rule of thumb, the more colourful the fruit or vegetable, the higher its phytochemical content will be. It was almost certainly the higher phytochemical intake of those fruit and vegetable lovers in the study on DNA damage(25) that afforded them the real protection So if you’re serious about obtaining maximum protection, make sure you’re getting at least the recommended levels of those fruit and veg portions a day – if not more!


  1. Med Sci Sports Exerc 1993;25:218–24
  2. J Sports Sci 1997;15:353–63
  3. Ann N Y Acad Sci 1998;854:102–7
  4. Eur J Appl Physiol 1998;77:498–502
  5. Proc Nutr Soc 1998;57:9–13
  6. Eur J Appl Physiol 1988;57:60–3
  7. J Appl Physiol 1993;74:965–9
  8. Muscle Nerve 1989;12:332–6
  9. Eur J Appl Physiol 1987;56:313–6
  10. Int J Biochem 1989;21:835–38
  11. Arch Biochem Biophys 1990;82:78–83
  12. Free Radic Res Commun 1993;19:191–202
  13. J Appl Physiol 1994;76:2570–7
  14. Med Sci Sports Exerc 1984;16:275–7
  15. Eur J Appl Physiol 1988;57:173–6
  16. Clin Sci 1991;80:611–8
  17. Int J Sports Med 1984;5:11–4
  18. Jpn J Phys Fitness Sports Med 1996; 45:63–70
  19. Biol Trace Element Res 1995;47:279–85
  20. Int J Sport Nutr 1994;4:253–64
  21. J Appl Physiol 1978;45:927–32
  22. Acta Physiol Scand 1994;151:149–58
  23. J Sports Med Phys Fitness 1999, 38(4): 281-5
  24. Am J Clin Nutr 1997, 65(4): 1052-6
  25. Cancer Epidemiol Biomarkers Prev 2000, 9(7): 647-52
  26. J Strength Cond Res 2003, 17(4): 792-800
  27. J. Nutr. 2002, 132:1616S-1621S
  28. FASEB J, 2001 15:A990
  29. J. Appl. Physiol. 2001, 90:1424- 1430
  30. Free Radic. Biol. Med. 2001, 31:745- 753
Share this

Follow us