Throughout history, clinical observation and clinical trials have identified links between nutritional deficiency and disease. For example, scurvy was described by Hippocrates over 2000 years ago, and native cultures have known its cause and cure for centuries. The first intervention study to demonstrate the successful treatment of scurvy with citrus fruits was published in 1753 by Captain James Lind in “A Treatise of the Scurvy”. Moving forward to the 20th century, one of the resounding achievements in this field has been identification of the importance of folic acid supplements in the prevention of spina bifida, leading to an overall reduction in incidence in the Western world. The possibility that nutritional factors may play a similarly important role in the aetiology of chronic respiratory disease is therefore intriguing and has recently attracted a great deal of interest.

The aetiology of asthma remains unclear, but it is widely accepted that environmental factors play a major role and, of these, diet is a potentially important contender. Evidence for this arises from the observations that the prevalence of asthma increases as societies move from a rural subsistence towards a more Western lifestyle; this is associated, among other factors, with a change in dietary pattern including adoption of a more processed and “convenience-orientated” diet. The result of this dietary change is an overall increase in the intake of refined sugars, fats and additives, and a relative reduction in the intake of complex carbohydrates and micronutrients. This change is a relatively modern phenomenon, occurring in the UK since the end of wartime rationing and also resulting from increased industrialisation of the food supply chain.

There is now an extensive literature on the relationship between diet and respiratory disease.^1–3 In asthma, observational studies have shown encouraging evidence of a protective effect of several nutrients on disease prevalence and symptoms, including vitamin C,^4,5 vitamin E,⁶ selenium⁷ and magnesium.⁸ However, and disappointingly in view of early promise with vitamin C,⁹ these findings have not generally translated into consistently positive outcomes in intervention studies. For example, in a recent randomised placebo-controlled trial of vitamin C, magnesium or placebo in 300 patients, we found no effect of either supplement on clinical asthma control.¹⁰ Similarly, an intervention study of vitamin E in adults with asthma also showed no evidence of a benefit.¹¹ While fish oil supplements have been shown to reduce exercise-induced bronchoconstriction in asthma,¹² other studies have not shown convincing evidence of an effect on asthma symptoms or medication usage.¹³ The Heart Protection Study which randomised over 20 000 adults at high risk of heart disease to receive antioxidant vitamin supplementation or placebo looked at respiratory disease as a secondary outcome. No difference was found in spirometry or in admission to hospital for chronic obstructive pulmonary disease (COPD), asthma or any non-neoplastic respiratory disease between the two groups.¹⁴ The Cochrane collaboration reviews of a range of dietary interventions in asthma (including vitamin C supplements, sodium restriction and fish oil supplementation) all report negative or, at best, inconclusive findings.^15–17

Of the many potentially relevant nutrients identified to date, selenium is of interest because case-control studies in both New Zealand^18–20 and the UK⁷ have found a relation between low dietary selenium and an increased risk of asthma. However, the only intervention study of selenium supplementation in asthma published to date included only 24 patients with asthma and found no effect on objective markers of disease.²¹ In this issue of Thorax, Shaheen et al²² report the results of a definitive randomised controlled trial of selenium supplementation in asthma performed in 197 people living in London (see p 483). No effect of selenium supplementation was found either on quality of life or objective measures of asthma symptoms and control.

So why have observational epidemiological study findings of dietary benefit failed to translate into a positive clinical trial result? Confounding is always a potentially major problem in observational studies, and although the effects of smoking, socioeconomic status and other factors have been allowed for in many of the observational study analyses, it is always possible that these or other effects have still influenced the results. There is also the difficulty of isolating the effects of specific nutrients, given the close correlation that exists between nutrients in individual diets. For example, diets low in fruit and vegetables provide low intakes of both vitamins C and E, making it difficult to determine which, if either, single nutrient is important, and may lead to erroneous identification of nutrient effects in the observational studies. Further, the tendency for diet to track throughout life means that exposure to (or lack of) certain important nutrients identified in observational studies might also reflect effects of exposures occurring over a very long period of time, including childhood and even the prenatal period.²³ In contrast, intervention studies have mostly been performed in adults and typically last for a few weeks or, at most, months. The intervention studies may therefore be delivering too little too late in the natural history of disease to have an effect.

Another consideration is that many intervention studies have focused on single nutrient supplements, when it is plausible that a combination of factors is more likely to be effective. This argument is supported by the results of trials showing protection against ozone-induced bronchoconstriction by a combination of vitamins C and E relative to placebo in adults with asthma,²⁴ and by a combination of vitamins C, E and β-carotene in subjects without asthma.²⁵ These studies did not, however, include single nutrient supplement groups, so it is not clear whether the effect is indeed due to the combination or whether the results are specific to the ozone challenge exposure and have relatively less relevance to the clinical control of typical asthma.

Another possibility is that other as yet uninvestigated micronutrients or co-factors are more important. Taking a step further, it could be that entire foods rather than nutrients are important, the most obvious candidates being fruits, vegetables or oily fish; for example, in one case-control study the intake of apples was negatively associated with the prevalence of asthma.⁷ To date, however, there is no evidence that individual food or food group supplements are effective in improving or preventing the disease. Furthermore, it may be that dietary supplementation only works in nutritionally deplete populations and that no additional beneficial effect will occur in well-fed and consequently over-supplemented individuals. This does not account for the findings in the study by Shaheen et al,²² however, as restriction of analyses to subjects with low plasma selenium levels did not change any of the study outcomes and also did not explain the negative findings in our study of vitamin C and magnesium.¹⁰

These and other potential problems may account for the failure to find a clinically useful dietary intervention for asthma, but what are the implications for future work in this area? The optimists among us may continue to pursue the increasingly elusive missing link, exploring the roles of other possible candidate single or multi- micronutrients or even try whole food supplements. They may also hope that future improvements in understanding the scientific basis of nutrition and potential roles in airways diseases may lead to the development of new hypotheses that can be tested in interventional studies. However, the pragmatists may take the view that, since single nutrient supplements appear to be ineffective and a balanced diet which includes a range of fresh fruit and vegetables is beneficial in so many other aspects of health, the best approach, at least for the time being, is simply to recommend people with asthma to eat a healthy balanced diet.