Adult-Onset Asthma Prevention. Should We Be Investing in Antioxidant Defense?

Oxidative stress, a state of imbalance between reactive oxygen species and antioxidants, is implicated in the pathogenesis of asthma. Some experts have attributed the rising rates of asthma in developed countries to increases in oxidative stress resulting from increased exposure to air pollutants and decreased dietary intake of antioxidant-rich foods such as fruits and vegetables (1). Antioxidants protect against oxidative stress by breaking down or scavenging reactive oxygen species and fall under two main categories—enzymatic (e.g., catalase and superoxide dismutase) and nonenzymatic (e.g., glutathione and vitamins A, C, and E). Observational studies have linked deficiencies in antioxidant defense to clinical asthma (2–4). Reduced levels of vitamin C and α-tocopherol (isoform of Vitamin E) have been observed in the bronchoalveolar lavage fluid of patients with mild asthma compared with healthy control subjects (2).

As highlighted in a recent review, adult-onset asthma remains poorly characterized but disproportionately affects females, smokers, and the obese, and is less often associated with allergy and atopic disease (5). In this issue of the Journal, Larkin and colleagues (pp. 45–53) provide novel insights into the potential role that host antioxidant defense plays in the risk of adult-onset asthma (6). The authors used a nested case–control design in a population-based cohort of over 65,000 women from The Shanghai Women’s Asthma and Allergy Study (7). The cohort consisted of women between the ages of 40 and 70 years without a history of asthma followed for a median of 8 years for the development of incident asthma. Interestingly, Larkin and colleagues found that urinary levels of F2-isoprostane (a marker of oxidative stress) did not differ between matched cases and controls prior to asthma onset; however, higher systemic platelet-activating factor acetylhydrolase activity and α-tocopherol levels conferred a decreased risk of incident asthma. Based on these findings, the authors suggest that chronic inflammation may not be present prior to clinical disease presentation (given the absence of oxidative stress) and that low antioxidant levels may be a risk factor for asthma inception.

A significant strength of this study was the baseline measurement of systemic antioxidant levels and longitudinal follow-up, thus capturing incident asthma cases. Most prior studies have had important limitations (cross-sectional study design [3, 8] or antioxidant exposure based on self-reported dietary questionnaires [8, 9]). The investigators also cleverly chose a source population from Shanghai with low background rates of known risk factors for adult-onset asthma (i.e., low smoking and obesity rates) but high air pollution exposure to home in on the impact of host antioxidant defense and oxidative stress on asthma risk. As the authors have pointed out, this aspect of the study design also limits the generalizability of the findings and likely explains, at least in part, the low incidence of asthma (0.24%) observed throughout the study. Other likely reasons for the low incidence were the focus on adult-onset disease and stringent ascertainment of cases with a two-step process that involved assessment of symptoms followed by confirmatory testing with spirometry and methacholine challenge testing.

Additional strengths relate to the conduct of this nested case–control study. Nested case–control studies are ideal for evaluating rare outcomes (like incident asthma) and can provide an efficient and economical means to assess potential associations between prospectively collected predictors and an outcome. One of the recognized challenges to case–control designs relates to the selection of controls (10). Ideally, the cases are all those individuals with the outcome in a population, and controls are drawn from this same population and are at risk for the outcome (11). The current study design overcomes this major limitation by sampling controls from the same cohort. Another problematic issue found in all observational studies is confounding. Confounders are characteristics of study subjects that are associated with the exposure and the outcome and can impact the association between the predictor and the outcome. To address confounding in case–control studies, investigators can either restrict the population, match on confounders, or adjust the analysis using logistic regression or stratification with Mantel-Haenszel approaches (10, 12). Matching in case–control studies has been shown to potentially increase bias due to misclassification or overmatching (13, 14). Whenever matching is performed in case–control studies, investigators must adjust for the matching variables to ensure that no bias is inadvertently introduced (13, 15). The authors have carefully demonstrated that adjusting for matching variables did not impact their results.

A few limitations of this study should be considered prior to accepting the results at face value. The investigators examined the levels of nine antioxidants (two enzymatic and seven nonenzymatic) simultaneously without correction for multiple testing, increasing the risk for a type 1 error. Also, the authors did not adjust their analysis for local environmental exposures, which may have resulted in residual confounding. Exposure to air pollution related to living in close proximity to a major road or point source could influence both antioxidant levels and incident asthma. Lastly, a number of years could have elapsed between the timing of biospecimen collection and incident asthma, stretching possible causal inference, especially if the biomarker measurement demonstrates significant variability over time. Demonstrating that individuals with low antioxidant levels remain deficient over time could have strengthened biologic plausibility and causal inference.

With these limitations in mind, is it time to fortify our antioxidant defense system with α-tocopherol supplements or with agents that block platelet-activating factor to prevent adult-onset asthma as the authors have suggested? This study adds to a growing list of very recent longitudinal studies examining the role of vitamin E levels in the development of childhood asthma, and unfortunately, the results have been conflicting (16, 17). Prior to embarking on a large-scale interventional trial of α-tocopherol supplementation to prevent asthma—a massive endeavor—this study should be replicated in another patient population. Until then, the implications of these findings must be interpreted with caution in light of the largely disappointing results that have been obtained in randomized controlled trials conducted to date evaluating antioxidant supplements such as vitamin E in the prevention of other chronic diseases (e.g., cardiovascular disease and cancer) characterized by oxidative stress (18).