Particulate matter (PM)2.5 exposure affects prenatal health and birth outcomes, including low birth weight (LBW) and preterm delivery (PTD).
To identify and explore PM2.5 exposure on adverse obstetrical effects, including preterm birth and LBW.
Four hundred and nine studies from 1982 to 2020 were identified in a search of PubMed, Embase, Scopus, Web of Science, and Science Direct. Of the 409 articles, 24 were identified as “qualitatively considered” and 7 were identified as “quantitively eligible” to be included in this meta-analysis. The pooled effect of PM2.5 exposure on LBW and PTD was calculated using a random effect model with significant heterogeneity. Seven studies were conducted in the meta-analysis, and the pooled effect of PM2.5 exposure on LBW and entire pregnancy was 1.033 (95% CI, 1.025–1.041) with significant high heterogeneity (I2 = 96.110, P = 0.000). The pooled effect of PM2.5 exposure on PTD and entire pregnancy was 1.024 (95% CI, 1.015–1.033) with significantly different low heterogeneity (I2 = 60.036, P = 0.082).
Exposure to PM2.5 during pregnancy is significantly associated with the risk of LBW, and the risk of PTD is significantly different but consistently associated with PM2.5.
Globally, PM2.5 exposure is significantly associated with serious pregnancy and birth outcomes worldwide. The emerging risks to prenatal health suggest a need for the government to influence health policies to protect maternal and pediatric health.
Introduction
Air pollutant–particulate matter (PM2.5) exposure affects prenatal health and birth outcomes. Preterm delivery (PTD) is a when a baby is born at <37 gestational weeks. Every year, 15 million deliveries occur before 37 weeks. Globally, 5% to 18% of deliveries in 184 countries are preterm. Premature births have been the leading cause of mortality among children <5 years of age, contributing to approximately 1 million deaths in 2015.1 The United Nations' target for 2030 with the Sustainable Development Goal is no more than 12 neonatal deaths per 1,000 live births and no more than 25 deaths per 1,000 live births in children under the age of 5, in all countries worldwide.2 Birth weight is the most critical indicator of infant growth during birth outcomes. One of the most significant public health problems has been low birth weight. Low birth weight (LBW) is defined as a baby born weighing less than 2,500 g (5.5 lb).3 Annually, an estimated 20 million births have been delivered as low weight births, ranging from 15% to 20% of all births worldwide. The global target in 2025 is to reduce the rates of low birth weight by 30%, leaving 14 million low-birth weight infants in all deliveries, according to the World Health Organization (WHO).3
Several studies have shown increasing associations between PM2.5 exposure and PTD with LBW. However, the results of the studies have positive and negative correlations. Particular matter (PM2.5) has been recognized as one of the risk factors with a prior history of preterm birth.4 PTD and LBW are public health issues associated with air pollutants.5,6 The risk of LBW is associated with PM2.5 in the first semester, and PM2.5 exposure is not only associated with LBW, but O3 exposure is positively associated with birth weight in the first and second trimesters.7 Both LBW and PTD are associated with PM2.5, and the most substantial effect of O3 exposure is increasing the risk of PTD and very preterm birth (VPTD) throughout pregnancy.8 Between PTD and ambient air pollution (AAP) effects, CO, O3, PM10, and PM2.5 are significantly associated with the second trimester in Wuhan, China.9 Canadian researchers found that PM2.5 exposure and LBW are associated with pregnancy in women who completed postsecondary education.10 In this study, the meta-analysis explored the association between PM2.5 exposure and birth outcomes, including PTD and LBW during different trimesters. By reducing air pollutants global health must reduce the risk of adverse obstetrical outcomes of LBW and PTD.
Patients and Methods
Our study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guideline,11 and data were collected into the endnote standard software tool from the library of Asia University. A total of 409 studies from 1982 to 2020 were identified in a search of PubMed, Embase, Scopus, Web of Science, and Science Direct using the following key words: infant, maternal, fetus, neonatal, prenatal, air pollution, PM2.5, preterm birth, and low birth weight. Only English publication articles were searched in this study. Of the 409 articles, 24 were identified and qualitatively considered, and 7 articles were quantitively eligible to be included in this meta-analysis.
This study investigated observational studies, including retrospective cohort studies, cohort studies, cross-sectional studies, and case-control studies, and only published English articles were included. Although study titles and abstracts were screened, different interventions or outcomes were initially excluded from this study. The study included the following criteria: PM2.5 exposure during the entire pregnancy, obstetric outcomes including PTD, and LBW. Findings of obstetric outcomes were performed as PTD and LBW. Preterm births1 and common births3 were defined and measured using WHO guidelines.
This study is appropriate to use for each quantitative and qualitative analysis; the quality assessment was evaluated by the New-Castle-Ottawa (NOS) (see Supplementary Materials Tables 1–3). NOS is divided into cohort, cross-sectional, and case-control studies and rated using a “star system.” Scores ranged from zero to 10, with scores indicating quality as very good, 9–10; good, 7–8; satisfactory, 5–6; and unsatisfactory, 0–4.
Each study included authors, publication year, country of analysis, study design, study period, total number of births, number of preterm births, and low birth weights with the intervention of PM2.5 exposure. Three professional experts reviewed data extraction. Under written permission, the first and second authors (O.A. and C-J.J., respectively) independently identified and extracted data from all articles. Therefore, ethics committee approval was unnecessary, and secondary data were conducted in a meta-analysis using the comprehensive meta-analysis software. The pooled effect of PM2.5 exposure and odds ratio (OR) for LBW, PTD, and 95% confidence intervals (CI) were calculated by comprehensive meta-analysis and using a random effect model with significantly high and significantly different heterogeneity, respectively.
Statistical Analysis
The pooled effect of PM2.5 exposure and LBW, PTD was calculated using a random effect model with significant heterogeneity. The heterogeneity is defined as I2, and the importance of the value of I2 follows as 0% to 40% might not be important, 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity, and 75% to 100% considerable heterogeneity. The study pooled all estimates between odd ratio (OR) as well as 95% (CI) and only one common exposure unit of effect as 2.5 in the fine particulate matter (PM) from different studies and conducted into meta-analyses and quantitively estimated the association between PTD, LBW, and PM2.5 exposure:
The meta-analysis pooled all estimations of LBW and PTD with the entire pregnancy.
In addition, the meta-analysis pooled all assessments of LBW and PTD with subgroups by first and second trimesters.
The Egger test was performed in the funnel plot symmetry.
Results
There were 409 selected studies, and 358 were screened after excluding 51 duplicated studies. Three hundred and one studies were excluded from screening because 35 were systematic reviews, meta-analyses, or case reports, 31 were animal experiments and included indoor pollutant exposure, and 178 were other birth outcomes. Fifty-seven articles were assessed for full-text articles. Thirty-three articles were further excluded: 9 did not provide specific types of pollutants, and 24 defined birth outcomes differently. After full-text articles were excluded, 24 articles were identified and qualitatively considered, and 7 articles were quantitively eligible to be included in this meta-analysis (see Figure 1).
Quantitively, 24 studies were included, of which 11 assessed PTD and LBW.8,10,12–19 Five studies assessed LBW,5,20–23 and 8 considered PTD.9,24–29 Detailed indicators are shown in Table 1.
In this study, quantitatively, 24 studies from 1996 to 2018 were included, with a large sample size of 10,502,332 births from the United States, Canada, China, India, Spain, Australia, Iran, the United Kingdom, and the Netherlands. Fifteen of 24 studies are retrospective cohort studies, 5 are cross-sectional studies, and 4 are case-control studies. Because of the findings, 11 studies evaluated both PTD and LBW, 6 considered LBW, and 7 considered PTD. Of 10,502,332 births, studies presented as few as 1285 births in India5 and as many as 2,966,705 in Canada.10 Covariates were parity, maternal age, gestational age, maternal education, maternal race/ethnicity, nationality, registered resident, marital status, employment, occupation, income, birth place, birth year, previous pregnancy, previous delivery, infant and maternal morbidity, and prenatal visits. One pollutant, PM2.5 exposure, varied in different periods of pregnancy, including trimesters.
A total of 7 studies were eligible for inclusion in the meta-analysis, with a significant association between PM2.5 exposure and birth outcomes.6,8,11,12,20,28,30 A summary of pooled effect estimates is shown in Table 2.
The association between exposure in LBW with entire pregnancy was OR = 1.033 (95% CI, 1.025–1.041) with significantly high heterogeneity (I2 = 96.110%, P = 0.000) (see Figure 2A). The significantly different association between exposure in LBW in the first trimester was OR = 1.046 (95% CI, 1.035–1.059) with significantly high heterogeneity (I2 = 0.000%, P = 0.744) (see Figure 2B). The association between exposure in LBW in the second trimester was OR = 1.119 (95% CI, 1.106–1.33) with significantly high heterogeneity (I2 = 94.147%, P = 0.000) (see Figure 2C).
The association between exposure in PTD and entire pregnancy was OR = 1.024 (95% CI, 1.015–1.033) with significantly low heterogeneity (I2 = 60.036%, P = 0.082) (see Figure 3A). Exposure of PTD during the first trimester was OR = 1.036 (95% CI, 1.010–1.062) with significantly different heterogeneity (I2 = 0.000%, P = 0.424) (see Figure 3B). Exposure of PTD during the second trimester was OR = 1.056 (95% CI, 1.030–1.080) with significantly different heterogeneity (I2 = 0.000%, P = 0.520) (see Figure 3C).
Using the Egger test, we did not reach a statistically significant publication bias; the Egger test was performed in the funnel plot symmetry (see Table 2 and Figure 4 A–E).
Discussion
A systematic review was performed to summarize the current scientific findings and quantitatively assess the association between adverse birth outcomes and maternal PM2.5 exposure. Meta-estimation was only considered a single pollutant of PM2.5, and maternal PM2.5 exposure effectively estimated the maternal exposure at the risk of LBW and PTD. Regarding the results from some studies, the high correlation of the multipollutant models makes it more complicated to interpret and disentangle the effect of each pollutant.31,32 Nevertheless, the findings are even more important in updating the prenatal guidance for adverse birth outcomes.
Maternal exposure to PM2.5 is highly related to a risk of LBW during pregnancy and the second trimester. In a New Jersey study,33 an association between maternal exposure to PM2.5 concentration and birth weight increases the risk of LBW in early and late pregnancy. This result was consistent with a chance of fetal growth restriction where PM2.5 small for gestational age estimates as first trimester 5.5% and third trimester 3.3% reduction of fetal weight.33 In addition, a previous study estimated PM2.5 concentration in the sixth gestational month was associated with a 10.3 g reduction in birth weight in those pregnancies without any complications.34 In a Massachusetts and Connecticut study, PM2.5 concentration exposure at 37 to 42 gestational weeks was associated with a –14.7 g reduction (–17.1 to –12.3) in birth weight.35 The effects of PM2.5 are not only associated with birth weight but also with head circumference as a change in head circumference –0.08 (–0.12 to –0.03) cm correlates with a change in birth weight –7 (–17 to 2) g at >37 weeks of gestation.36 During the Beijing Olympics in 2008, the greater birth weight associated with a lower level of PM2.5 concentration in the third trimester of pregnancy was a significant 7% increase in the risk of a low-birth weight baby among term birth in 2018.37 In the same study, declines in air pollutants during the 2018 Olympic games were associated with a 23-g birth weight increase in pregnancies at 8 gestational months compared with the same calendar date in 2007 or 2009.38 Maternal exposure to PM2.5 is one of the important risk factors for intrauterine inflammation (IUI), a sensitive biomarker for the developing fetus.39 IUI is a risk factor for adverse birth outcomes, including LBW and PTD, and significantly, the combined presence of maternal inflammatory response and fetal inflammatory response was more strongly associated with extreme preterm birth (<29 weeks).40
Another finding from the subgroup regarding the birth outcome concerns a small risk of PTD in our study. In many studies, maternal exposure to PM2.5 has been associated with PTD.22,41–46 However, some studies have found that the effect is positively a small, statistically nonsignificant association of risk of PTD between maternal exposure to PM2.5 during pregnancy and the third trimester.28 Similarly, however, predicted maternal exposure to PM2.5 is little evidence of association with preterm delivery risk; an increased risk of preeclampsia is associated with maternal exposure to PM2.5 in this setting.47 Preeclampsia also can lead to preterm delivery and numerous adverse outcomes, including immunologic problems and longer-term motor, cognitive, and fetal growth problems.48 However, the further finding was not associated with the risk of both LBW and PTD in the first trimester. Although prenatal exposure to PM2.5 during pregnancy is significantly associated with the risk of LBW, the risk of PTD is significantly different; however, PTD is consistently associated with PM2.5.
There were a few limitations of this study:
Some studies made several analytical decisions that caused primary cohort and case-control analyses to differ in important ways other than including the additional covariates.
One direct comparison's results significantly differed between the cohort and case-control study. Although some studies could not conduct the meta-analysis because of weak evidence of adverse effects, these studies were included in a systematic review. Therefore, the third trimester was not appropriate for meta-analysis.
Some analyses do not raise numerous questions about the comparability of the data and analysis approaches.
We did not find publication bias based on the Egger test.
Conclusion
This meta-analysis explored that PM2.5 exposure is positively associated with birth weight. Another finding from the subgroup regarding the preterm birth outcome concerns a small risk of PTD in this study. PTD is significantly different with exposure to PM2.5 and is consistently associated with PM2.5. Globally, PM2.5 exposure is significantly associated with severe consequences to prenatal care and birth outcomes in pregnancy. Health appears in emerging risks that government needs to influence health policies to pursue maternal and child health.