• Asthma

Respiratory disease taken as a whole is a major global burden contributing significantly to morbidity and mortality at all stages of life. According to WHO estimates, asthma is the most prevalent respiratory condition with over 300 million people affected globally and an additional 100 million diagnoses forecast by 2025. The second most prevalent condition is COPD, with an estimated 210 million suffering globally. Pneumonia and acute lower respiratory illnesses are the leading causes of death in children under 5 years of age, responsible for an estimated 1.4 million deaths annually. It is currently understood that low lung function is a major risk factor linking all these entities, increasing susceptibility to these global health problems.

Lung function is known to track throughout childhood,1 ,2 so to a very real extent the lung function one is born with is a major determinant of future respiratory health.3–5 However, lung function may grow better6 or less well7–9 than anticipated depending on environmental exposures, especially those occurring in early life. Airway branching pattern develops during the first 16–18 weeks of fetal development, with alveolar development beginning later and continuing in early postnatal life. Thus, the respiratory system is vulnerable to adverse environmental exposures in both prenatal and postnatal life. The most compelling data for prenatal exposures influencing lung function at birth10 and beyond11 are for fetal exposure to tobacco smoking, especially from the mother smoking during pregnancy. However, other prenatal exposures to environmental toxicants are thought to limit respiratory development in utero, as exemplified by maternal exposures to household chemicals during pregnancy.12 There are two potential pathways by which prenatal exposures might result in low lung function at birth; by direct effects on lung development or via effects on somatic growth of the fetus. Birth weight has a small but significant direct effect on adult lung function, independent of factors known to influence adult lung function, including parental and personal heights and tobacco smoking.13 However, to date, the direct effects of prenatal exposures on lung function are presumptive as no study has been conducted using a technique that directly measures airway calibre or airway mechanics at or soon after birth.

Morales et al14 report on the effects of intrauterine and early postnatal exposure to ambient air pollution on lung function in preschool-aged children (mean age 4.5 years). They have used sophisticated modelling to estimate individual exposures to benzene and nitrogen dioxide to indicate exposure to traffic-related pollution and found significant reductions in FEV₁ with both pollutants. A dose–response relationship was demonstrated with the risk of low lung function, defined as FEV₁ <80% predicted, increasing with increasing pollutant exposures. In addition, Morales et al14 were able to demonstrate that the effects of exposure were most pronounced during the second trimester. These data are credible and consistent with previous reports of maternal exposures to ambient air pollution decreasing fetal somatic growth15 and infant lung function measured during tidal breathing.16 Hansen et al,15 using less sophisticated modelling of pollution exposure than used by Morales et al,14 also showed that effects of exposures varied during gestation; however, in their study the impacts were more variable. There are also numerous reports of the adverse effects of ambient air pollution, either from traffic-related or industrial sources, on lung function in older children.9 ,17 ,18 What the study from Morales et al14 adds to the literature is the increased precision of the exposure estimates and the age of the children studied. They provide convincing evidence that prenatal exposures to traffic-related pollution have long-term effects on lung function in otherwise healthy children.

Despite the value of this study, it does have limitations. The model validation used passive samplers distributed over the study area, rather than in the individual children's homes and the exposure estimates were averaged over various periods (annual, pregnancy and each trimester) to represent mean levels for each period. There are no data on which to judge whether such mean levels better represent the potential for ‘damage’ than peak levels, the number of occasions above a threshold value or some type of integrated exposure, such as area under an exposure–time curve. Clearly personal monitoring in a study of the scope (both in time frame and in numbers of subjects studies) undertaken by Morales et al14 would not be possible and they have probably done as much as they could. Another limitation is the use of spirometry in preschool-aged children. This test is not easy to perform in young children and indeed, as shown in their Table 2, Morales et al14 were only able to obtain reproducible spirometry in fewer than 50% of the children tested. Another issue with spirometry in this age group is the physiological interpretation of FEV₁. The utility and validity of standard spirometry performed during maximal forced expiration are that expiratory flow limitation is induced so that the flow obtained is independent of effort and, as such, reflects the mechanical properties of the airways and lungs. However, young children are unlikely to be able to maintain flow limitation to very low lung volumes; this is likely to be one of the reasons for the normal fall in FEV₁/FVC ratio reported through childhood.19 The FEV₁/FVC ratio reported by Morales et al14 of 93%–94% is normal for this age range,20 but very different from the 80% that is taken to be the lower limit of normal in older children or adults. Young children's lungs are essentially empty within 1 s and thus FEV₁ and the FEV₁/FVC ratio need to be interpreted differently in young children. Morales et al14 may have been better to have measured lung function with the forced oscillation technique, which has both a higher success rate in young children and measures the mechanical properties of airways and lungs with a similar physiological interpretation in all ages, with the exception of infants measured via a face mask where the impedance of the nasal pathways is also included.

Despite the study limitations, Morales et al14 should be lauded for producing convincing data in a large number of children highlighting the long-term consequences of exposure to traffic-related pollution and the lifelong increased risk of respiratory disease. This exposure is potentially avoidable and can be influenced by regulation and legislation. Policy makers need to heed data such as those presented by Morales et al14 as limiting exposure to traffic-related pollution during fetal development and early postnatal life is one way that the burden of respiratory disease can be decreased.