357 - Maternal exposure to environmental greenness, air pollution and temperature and neonatal outcomes of preterms

Publication Number: 357.5702

Prakesh S. Shah, Mount Sinai Hospital, Toronto, ON, Canada; Alice Aveline, Imperial college, London, England, United Kingdom; Nicole Bando, Unity Health Toronto, Toronto, ON, Canada; Mohammad Noaeen, Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada; Jie Yang, Fudan Children's Hospital, Shanghai, Shanghai, China (People's Republic); Thuy Mai Luu, Centre Hospitalier Universitaire Sainte-Justine, Montreal, PQ, Canada; Marc Beltempo, McGill University Faculty of Medicine and Health Sciences, Monreal, PQ, Canada; Abhay Lodha, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada; Jeffrey Brook, University of Toronto, Toronto, ON, Canada

Background: Outcomes of preterm neonates of < 29 weeks’ gestation in relation to maternal exposure to greenness, pollutants and ambient temperature are unknown.

Objective: To examine the associations between maternal exposure to greenness, air pollution and ambient temperature, during pregnancy and survival without major morbidity (SWMM) in neonates < 29 weeks’ gestation in Canada.

Design/Methods: In this retrospective cohort study, we studied infants born between 2010-2020 using data from the Canadian Neonatal Network linked to environmental data from the Canadian Urban Environmental Health Research Consortium (CANUE). Data on the level of greenness (during the growing season and annual average), air pollutants (ozone, nitrogen dioxide, smoke, PM2.5, sulfur dioxide) and ambient temperature at the mother’s residential postcode, were extracted from the CANUE database. Data covered the nine months prior to each infant’s birth (encompassing duration of pregnancy and between 7-17 weeks of pre-pregnancy period depending upon gestation at birth). Data were converted into an index of levels of greenness, two air pollution indices (one for ozone levels; and one for levels of fine particulate matter [PM2.5], NO2 and smoke) and an index of temperature using principal components analysis. Association with SWMM (Grade 3/4 IVH or PVL, NEC stage 2/3, nosocomial infection, stage 3 or higher ROP, or BPD) were assessed using logistic regression analyses.

Results: Of the 14,748 infants included 32% SWMM. BPD was the most common morbidity (Table 1). Infants from mothers exposed to low levels of greenness had lower odds of SWMM compared to infants from mothers exposed to moderate levels of greenness (adjusted OR 0.85, 95% CI 0.76-0.94). Low levels of ozone pollution were associated with higher odds of SWMM compared to either moderate or high levels (moderate vs low tertile OR 1.12, 95% CI 1.03-1.22; high vs low tertile OR 1.20, 95% CI 1.05-1.37). Maternal exposure to lower temperatures had lower odds of SWMM compared to moderate temperatures (moderate vs low tertile OR 0.85, 95% CI 0.76-0.96). Table 2 reports the odds of SWMM in infants exposed to singular or combinations of the environmental variables. Areas of Canada were identified with none, single, two and all three exposures (Figure 1).

Conclusion(s): Maternal exposure to low levels of greenness, high ozone and low temperature during pregnancy are associated with lower odds of SWMM in preterm neonates of < 29 weeks’ gestation.

Table 1: Key characteristics of the cohort (n = 14,748)
1IQR = Interquartile range
2Quintiles of dependency (1 = least dependent, 5 = most dependent)
3Quintiles of ethnic concentration (1 = least ethnically concentrated, 5 = most ethnically concentrated)
4Quintiles of instability (1 = least unstable, 5 = most unstable)
5Quintiles of deprivation (1 = least deprived, 5 = most deprived)
6BPD = Bronchopulmonary dysplasia, NEC = Necrotizing enterocolitis, LOS = Late onset sepsis, IVH = Intraventricular hemorrhage or periventricular leukomalacia, ROP = Retinopathy of prematurity

Table 2: Association between exposure to combinations of environmental stressors and odds of survival without major morbidity
1 Infants were classified into exposure groups as a function of whether they were in the lowest tertile of the greenness index, highest tertile for ozone related air pollution and/or lowest tertile for ambient temperature.
2 Odds are calculated using logistic regression, adjusted for gestational age at birth, birthweight, birth epoch, maternal diabetes, maternal antenatal drug use, maternal antenatal hypertension, infant small for gestational age, sex, SNAP-II and neighborhood risk indices in tertiles.

Figure 1: Geographic distribution of environmental risk factors (high levels of ozone related air pollution, low temperatures and low levels of greenness) in Canada