The association between early life exposure to ambient particulate matter and long-term immune and respiratory health in children

Background: Landscape fire smoke exposure could cause adverse health outcomes, especially in children. The Hazelwood coal mine fire started on 9 February 2014 in Victoria, Australia and lasted for approximately 6 weeks. This event caused increased concentrations of fine particulate matter (PM\\(_{2.5}\\)) in the nearby area. The effect of smoke exposure during the in utero period and infancy (i.e. the first two years of life)may have implications for the development and growth of a child's immune and respiratory systems. However, there is very limited evidence regarding the associations between early life exposure to short-to-medium duration fire smoke events, which can result in severe air pollution, and the potential health outcomes in later life. Aims: This Thesis aimed to investigate possible associations between early life exposure to air pollution from the Hazelwood coal mine fire and later respiratory health and immune function. A range of complementary biomedical and epidemiological research approaches were used to address four specific research questions: 1) to evaluate current epidemiological evidence on the associations between intrauterine and infant exposure to particulate matter and subsequent development of asthma and wheezing (Chapter 2); 2) to evaluate children's lung function following infant exposure to the Hazelwood coal mine fire emissions (Chapter 3); 3) to assess the effect on health service utilisation in children after intrauterine and infant coal mine fire smoke exposure (Chapter 4), and; 4) to investigate how fire smoke-related particulate matter, and the chemical components, affect respiratory health by conducting toxicological studies in human lung cells (Chapter 5). Methods: 1) A systematic review and meta-analysis was conducted to answer Aim 1. Epidemiological data from relevant literature investigating the associations between ambient PM\\(_{2.5}\\) exposures during two time points (prenatal or the first two years of life), and wheezing or asthma throughout life was extracted from five databases. All included studies were assessed according to the Critical Appraisal Skills Programme checklists. Meta-analyses were performed if ‚Äöv¢‚Ä¢2 studies estimated the effects of continuous PM\\(_{2.5}\\). 2) To answer the second and third Aims, I collected data from the Latrobe Early Life Follow-up (ELF) Study, comprising 571 children born between 01/03/2012 and 31/12/2015 from the Latrobe Valley in Victoria, Australia. Individual exposures to 24-hour average and peak concentrations of PM\\(_{2.5}\\) during the fire were estimated using individual activity/location data, dispersion and chemical transport modelling. Lung function was measured using the forced oscillation technique (FOT), generating standardised Z scores for resistance (Rrs), reactance (Xrs) and the area under the reactance curve (AX). Data on general practitioner attendances, and dispensations of prescribed asthma inhalers, steroid skin creams and antibiotics were collected from the Australian Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme (PBS). Multiple regression analyses were used to assess the associations. 3) For Aim 4, roof space particulate matter samples from 36 different homes and their particle characteristics (i.e. size, endotoxin and chemical composition) were analysed using standardised techniques. The cytokine production of BEAS-2B cells after exposure to either media alone, 5.7 or 57 ˜í¬¿g/mL of particulate matter suspension for 4 h or 24 h, was assessed using ELISA. Principle component analysis (PCA) and linear regression analyses were employed to evaluate the associations between cytokine production and the particle composition. Results: For Aim 1 (Chapter 2), while evidence was limited and inconsistent, epidemiological literature was suggestive of an association between early life PM\\(_{2.5}\\) exposure and wheezing/asthma. Meta-analyses conducted for the associations between: (1) intrauterine exposure and asthma (n=4); (2) infant exposure and asthma (n=5); and (3) infant exposure and wheezing (n=3), found no significant associations. While meta-analysis of intrauterine exposure and wheezing (n=5) was not possible due to inconsistent exposure and outcome assessments, four studies found strong positive associations with wheeze by age 2. High heterogeneity was present among studies of intrauterine exposures and asthma, while studies of other associations showed low heterogeneity. Data in Aim 2 (Chapter 3), using FOT assessment, showed a 10 ˜í¬¿g/m\\(^3\\) increase in infant average PM\\(_{2.5}\\) exposure was significantly associated with worsening AX (˜í‚â§-coefficient, 0.26; 95%CI 0.02, 0.50), while the association between a 100 ˜í¬¿g/m\\(^3\\) increase in peak PM\\(_{2.5}\\) and AX was not significant (0.17; 95%CI -0.00, 0.33). In the analysis of MBS/PBS data (Aim 3 ‚Äö- Chapter 4), 10- and 100- ˜í¬¿g/m3 increases in average and peak PM\\(_{2.5}\\) exposure during infancy were associated with a greater incidence of antibiotics being dispensed during the year following the fire: the adjusted incidence rate ratios were 1.24 (95% CI 1.02, 1.50, p<0.05) and 1.14 (1.00, 1.31, p<0.05) respectively. No other significant associations were observed. For Aim 4 (Chapter 5), exposure to roof space particulate matter caused significant dose (IL-6, p<0.05 for all comparisons; IL-8, p<0.05 for comparisons after 24 h exposure) and time (IL-6, p<0.05 for all comparisons; IL-8, p<0.05 for all comparisons) dependent increases in cytokine production that was evident 4 and 24 h post-exposure with the exception of IL-8 production 4 h post exposure to 5.7 ˜í¬¿g/mL particulate matter which was not elevated above control levels (p>0.05). Higher concentrations of Fe, Al, Mn in particulate matter were significantly associated with increased cytokine production. Conclusions: Current evidence on the associations between early life PM\\(_{2.5}\\) exposure and adverse respiratory outcomes during childhood is limited. My analyses provided novel findings of significant associations between infant exposure to PM\\(_{2.5}\\) from coal mine fire emissions and later adverse immune and respiratory health outcomes, including worse lung reactance, and increased use of antibiotics. The underlying mechanisms might be the pro-inflammatory capacity of PM\\(_{2.5}\\) on human lung cells. Further follow-up studies are needed to confirm these findings, to investigate whether these effects persist as children develop and to further explore potential mechanisms.