Statistics from Altmetric.com
The paper by Haby et al on the prevalence and risk factors for asthma in preschool children1 appeared in Australian newspapers under the headline “Fatty diet may double risk of childhood asthma”. But the authors have not logically established their assertion that 17% of the cases of recent asthma in Australian rural children 3–5 years old can be attributed to a high intake of polyunsaturated fats. Five reasons make this conclusion improbable.
(1) The questionnaire was sent by post and parents were not helped professionally to complete it. Only two questions were asked about diet and it seems from table 1 in the paper that there were three possible answers for bread spread and five for frying oil. Predominantly polyunsaturated fats/oils would have been in the last “other” group, presumably used by parents who weren't sure of the answer. No questions were asked about the rest of the diet, either possibly allergenic foods or about fish, the major source of omega-3 polyunsaturated fatty acids. In earlier papers some of the same authors reported an inverse relationship between total fish consumption and bronchial reactivity2 or between fresh (not canned) oily fish and asthma.3
(2) At least five randomised controlled intervention trials (some in this journal) have found no benefit in patients with asthma given high intakes of omega-3 polyunsaturated fish oils4–8 which Haby et al think of as the good oil against which omega-6 polyunsaturates such as linoleic acid could compete.
(3) Morris et al9 have given high intakes of omega-6 polyunsaturated fatty acids to patients with asthma for 8 weeks with alternating control diets containing the same amount of saturated and monounsaturated fats. There was no deterioration in symptoms, bronchodilator usage, or lung function tests.
(4) In 77 866 US nurses no relation could be found between doctor diagnosed asthma and dietary intake of fatty acids by a group of epidemiologists experienced in quantifying intake of different fatty acids.10
(5) In vitro work indicates that leukotrienes from omega-6 or omega-3 fatty acids do not have different effects on bronchial constriction,11 and those derived from eicosapentaenoic acid, an omega-3 fatty acid, do not have a dampening effect on eosinophils, the predominant effector in asthma.12
We thank Dr Truswell for his comments on our paper which investigated a wide range of risk factors for asthma in children of preschool age. We certainly agree with the caution he advises in translating findings from an observational study to clinical or public health practice. We advocate that evidence from randomised controlled trials is an essential prerequisite to such advice. However, we do not believe that the data he cites invalidate the observation we have made on the association between the use of polyunsaturated oils and spreads and the presence of asthma.
We acknowledge that our questionnaire was a crude tool for assessing dietary fatty acid intake and that the observed association with polyunsaturated oils and spreads may not be due to the balance of omega-3 and omega-6 fatty acids. Nevertheless, there was an observed association and this needs to beexplained. It is possible that the effects are attributable to differences in other related dietary constituents or to unmeasured confounding. Our view was that the relative intake of omega-3 and omega-6 fatty acids was the most plausible explanation. Our previous observation in children of primary school age that those with asthma reported a reduced intake of fish lends some support to this explanation.1
The lack of effect of short term omega-3 or omega-6 supplementation in subjects with established asthma does not preclude the potential for benefit arising from long term modification of fatty acid intake. Similarly, the lack of effect in the adults studied in the Nurses Health Study2 does not preclude the potential for adverse effects in young children.
Our findings, together with other observational data demonstrating higher consumption of polyunsaturated fats among children with atopic disease,3 should alert the scientific community to the need for further research, particularly long term randomised controlled trials, to elucidate the role of fatty acid consumption in the expression of asthma in children.