Is asthma prevention possible with dietary manipulation?

Most reported RCTs of primary asthma prevention use a "high-risk" selection strategy rather than a population-based strategy5 (ie, the subjects recruited have been identified as being at high risk of asthma, usually on the basis of a positive family history of asthma and/or atopy). Moreover, since the development of clinical atopy occurs in early life, studies have either recruited the subjects in utero (during the last trimester) or during the neonatal period.

A meta-analysis of maternal dietary allergen avoidance during pregnancy to prevent atopic disease in infants at high risk concluded that such avoidance is "unlikely to substantially reduce the risk of atopy in the child. Moreover, such a diet may have an adverse effect on maternal and/or fetal nutrition".6 This meta-analysis was based on three RCTs involving a total of 504 women. Although several trials suggested a lower incidence of positive allergen skin-prick testing initially, by the age of 18 months there was no benefit. Given the potential harm to the fetus (lower mean gestational weight gain), elimination diets during pregnancy are not recommended.

Another meta-analysis of three RCTs, involving a total of 209 women, assessed the efficacy of a maternal allergen-avoidance diet during lactation for preventing atopic disease in infants of women at high risk.7 Combined data from these trials suggest a strong protective effect on atopic eczema during the child's first 12–18 months of life. However, the author of the review noted substantial methodological weaknesses in all three trials.

The role of protein hydrolysed formulas on the development of atopy and asthma remains uncertain, and further studies are warranted. In a relatively small RCT (involving 177 subjects), early feeding of a protein hydrolysate (casein hydrolysate, "Pregestimil") formula to infants at risk of allergies had a long-term preventive effect on the prevalence of eczema, but not of asthma.8

The role of exclusive breastfeeding in protecting against asthma is controversial. Although most studies have shown a protective effect, several have found an increased rate of asthma in breastfed infants. A systematic review of prospective studies9 identified 12 relevant studies of adequate methodological quality. The summary odds ratio for the protective effect of at least three months' exclusive breastfeeding was 0.70 (95% CI, 0.60–0.81), indicating a 30% reduction in the development of childhood asthma. This protective effect was even greater in children with a family history of atopy (50% reduction). These results indicate a valid and important protective effect of breastfeeding on subsequent development of asthma.

Dietary supplementation with probiotics has shown encouraging results. In a double-blind, randomised, placebo-controlled trial, Lactobacillus (given prenatally to mothers and then postnatally for six months to their infants) resulted in a 50% reduction in the rate of atopic eczema at the age of two years (23% v 46%).10 As no definitive information concerning asthma can be derived at this early stage, the long-term results of this trial will be awaited with great interest. A key factor with probiotic intervention is that it does not cause any harm. Indeed, probiotics almost certainly result in a more "friendly" gut flora, which may have other beneficial effects on health apart from reducing rates of atopic disease.

Two RCTs have examined combined dietary manipulation and airborne allergen avoidance.11,12 The Isle of Wight study11 involved 120 "at risk" infants randomly allocated during pregnancy to an intervention group or a control group. The intervention group were either breastfed (with mothers excluding foods regarded as highly allergenic from their diet) or given a soy-based protein hydrolysate. In addition, exposure to house-dust mites was reduced by encasing the child's mattress and treating the home with an acaricide four times during the first nine months of life. The combined intervention resulted in a significant reduction in the rate of eczema and atopy (sensitisation to both dietary and aeroallergens) at four years of age. The rate of doctor-diagnosed asthma was also lower (24% v 35%), although this was not statistically significant. Clearly, this is an encouraging result and should be repeated in other countries, using larger numbers, and with prolonged follow-up of the subjects.

The second combined intervention RCT was conducted in Vancouver, Canada.12 This examined the effectiveness of a multifaceted intervention. A total of 545 "high-risk" infants were prenatally randomly assigned to either a multiple intervention group or a control group. The intervention involved avoidance of house-dust mite, pet allergens and environmental tobacco smoke; encouragement of breastfeeding; and supplementing the diet with a partially hydrolysed formula. The result of the intervention was a modest but statistically significant (P = 0.04) reduction in the risk of possible or probable asthma and rhinitis at the age of 12 months. The incidence of positive allergen skin-prick tests to one or more aeroallergens was similar in both groups (4.4% in the intervention group v 4.6% in the control group). Clearly, more prolonged follow-up of these infants will be essential to determine whether these differences in possible or probable asthma persist.

In the western suburbs of Sydney, a large placebo-controlled RCT (involving about 600 "high-risk" newborns recruited in pregnancy) is now in progress.13 The design is factorial, allowing two interventions to be evaluated in the one trial: a dietary supplement (omega-3 fatty acid) and comprehensive house-dust mite avoidance measures. Although this trial is only in its early stages, initial results show an impressive reduction in Der p 1 (dust mite) exposure levels in the intervention groups, plus good compliance with taking the omega-3 fatty acid supplement. When the children are old enough (ie, 5–6 years) for reliable medical outcome measures of asthma rates, this trial will supply definitive data on the effectiveness of omega-3 fatty acid supplementation alone and in combination with house-dust mite avoidance measures in the Australian environment.

In summary, RCTs have shown a consistent protective effect of a variety of dietary manipulations on subsequent asthma and/or atopy (see Box 2). Studies combining dietary manipulation with reduction in airborne allergen exposure suggest added protection. However, longer follow-up of the children in all these studies is essential to determine the long-term benefits of these preventive measures on rates of childhood asthma.