Maternal pesticide exposure and risk of preterm birth: A systematic review and meta-analysis.

BACKGROUND: Maternal pesticide exposure might be associated with adverse pregnancy outcomes through triggering inflammation and oxidative stress and disrupting endocrine functions. Yet the association between prenatal pesticide exposure and risk of preterm birth remains inconclusive. OBJECTIVES: To conduct a systematic review and meta-analysis of human observational studies using the Office of Health Assessment and Translation (OHAT) framework to explore the association of per ten-fold increase of pesticide concentrations in maternal biological samples during pregnancy with risk of preterm birth and length of gestational age at birth. DATA SOURCE: Five English (PubMed, Embase, Cochrane Library, Web of Science and Scopus) and 3 Chinese databases (China national knowledge infrastructure (CNKI), Wanfang Data, and Chinese Biomedical Literature Database (CBM)) were searched till Jan 18th, 2023. STUDY ELIGIBILITY CRITERIA, PARTICIPANTS, AND INTERVENTIONS: To be included, pesticide exposure should be measured in maternal biological samples during pregnancy and in log-transformed forms. The primary outcome was preterm birth and the secondary outcome was gestational age at birth. STUDY APPRAISAL, SYNTHESIS METHODS AND CONFIDENCE ASSESSMENT: Quality of studies was evaluated using OHAT Risk of Bias Tool. Evidence was quantitatively synthesized with Correlated and Hierarchical Effects (CHE) model. The confidence rating in the body of evidence was done using OHAT. RESULTS: A total of 21 studies reported by 18 papers were included, with 7 studies for preterm birth and 19 for gestational age at birth. The meta-analysis found a ten-fold increase of pesticide concentrations was potentially associated with risk of preterm birth (pooled OR = 1.28; 95%CI: 0.93, 1.78) and shortened gestational age at birth (ß = -0.10; 95%CI: -0.21, 0.01). Sampling biospecimens in different trimesters was identified as a potential modifier in the association between pesticide exposure and length of gestational age (F = 2.77, P < 0.05). For studies that collected samples at any time during pregnancy, pesticide exposure was found to be associated with shortened length of gestational age (ß = -0.43; 95%CI: -0.81, -0.06). The confidence rating in the body of evidence was "moderate" and "very low" for preterm birth and gestational age at birth, respectively. CONCLUSION: Our result suggested moderate evidence of an association between pesticide exposure and higher risk of preterm birth. Yet more studies are still needed with larger sample size and careful considerations of confounders and accuracy of outcome measurements. Attention is also required on other pesticide compounds in addition to organochlorine and organophosphorus pesticides, and on windows of susceptibility.