Occupational exposure of pregnant women to refined oil and infant wheezing: Japan environment and children's study findings.

Women currently work in similar occupations to men. However, women who are of childbearing age respond to some occupational exposures differently than men; they need suitable working environments for supporting their children's health.1-3 If pregnant women engage in occupational use of kerosene, petroleum, gasoline or benzene (refined oil; RO), they can be chronically exposed to these products.4, 5 RO may impact child development and health through exposure during pregnancy. However, few studies have examined the relationship between occupational exposure of pregnant women to RO and the development of asthma in their offspring. We investigated whether there is an association between exposure of pregnant women to RO and infant asthma within the first 12 months. Where an association was identified, we examined the crucial exposure period and a dose-dependent effect in RO for pregnant women. We used the Japan Environment and Children's Study (JECS) data set-AN-20180131, a prospective observational study. Usually in Japan, doctors, mainly home doctors, diagnose asthma according to Japanese Society of Pediatric Allergy and Clinical Immunology guidelines. Detailed information on enrolment, inclusion criteria and design of the JECS has been previously described.6 We collected prenatal and perinatal data from questionnaires distributed to pregnant women and medical record transcripts from obstetricians. JECS assessed the occupation of pregnant women at mean 14.0 weeks (SD = 5.5 weeks; early second trimester, early ST) and mean 23.3 weeks (SD = 3.8 weeks; late second trimester, late ST) by questionnaire. The occupation was classified based on the Japanese Ministry of Internal Affairs and Communication classification. The questionnaire also contained the following questions: ‘What kind of chemical substances do you handle at work?’ and ‘How often have you been exposed to chemical substances during more than half the working hours in recent months?’ Pregnant women selected one of the following responses to the question of whether they had been exposed to RO: ‘not, monthly, or weekly exposure’ at early ST and late ST. After birth, questionnaires were sent every 6 months that included information on maternal physical and psychological changes, the environmental and socio-economic status of the family, and infants' health, growth and development. We also collected information on maternal age at childbirth from the medical record and clinical history of allergy (asthma, atopic dermatitis, allergic rhinitis, allergic conjunctivitis or food allergy) and whether the infants had been diagnosed with asthma within the first 12 months using questionnaires for mothers. We classified mother and infant characteristics into three groups according to the RO exposure in early ST; additional material (https://doi.org/10.5281/zenodo.8339589). To investigate the association of RO and relevant outcomes, we used four statistical models. First, with univariate logistic analysis, we performed whether exposure to RO during pregnancy was associated with diagnosing infant asthma (Model 1). Then, we estimated whether the frequency of exposure to RO at one of the ST associated with the diagnosis of infant asthma with univariate logistic analysis (Model 2). Third, we examined both STs with RO exposure frequencies in the same logistic analysis (Model 3). Finally, in Model 4, Model 3 was adjusted with covariates (mother's age, history of allergies [asthma, atopic dermatitis, allergic rhinitis, allergic conjunctivitis and food allergy], prepregnant maternal weight, weight gain during pregnancy, mother's smoking habit, annual income, house structure, house humidity, type of heating system, mode of delivery, sex of infant and mother's smoking habit after childbirth). Results with p < .05 are considered statistically significant. Statistical analyses are performed using Stata 15.1. The JECS data set comprised information on 104,065 foetuses. However, there were several missing values and restrictions; we analysed the data of 39,736 mother–children pairs in this study. Table 1 in the additional material shows the characteristics of participants at early ST. The logistic analysis of Model 1 showed that exposure to RO during pregnancy was associated with a diagnosis of infant asthma. Results are presented in the Figure 1 and Table 2 of the additional material. Models 2 and 3 showed a dose-dependent relation between exposure to RO and the diagnosis of infant asthma. From the results of Models 3 and 4, we could observe that maternal exposure to RO, especially early ST and, to some extent, late ST, was associated with infant asthma. Lee et al.7 and Jung et al.8 reported that perinatal PM2.5 exposure was associated with developing children's asthma. They also showed a dose-dependent association between the exposure concentration. However, the crucial period of PM2.5 exposure has been broader in the study of Jung et al. than those of Lee et al. We suppose the differences between the two studied are owing to the concentration of PM2.5. The primary question is whether the mechanism of children's asthma for PM2.5 and RO is identical. Other than the DNA methylation problem,1 Hamada et al.9 examined whether prenatal exposure to residual oil fly ash could increase susceptibility to the development of asthma in offspring in early life using BALB/c mice and suggested Th2-skewed immunity might occur in offspring. We should notice the possibility that RO exposure to pregnant women causes children asthma in early life. In this study, we got every information from the questionnaire. Therefore, some participants might wrongly assess their exposure to RO and classify their occupations. The diagnosis of asthma at this age might also be inaccurate. Despite these limitations, our study provides valuable information about the relationship between occupation during pregnancy and offspring asthma development in early life. Further investigation is necessary on how exposure to RO is associated with children's asthma and what substrate in RO affects these results. The authors declare no financial relationships relevant to this article except for the Ministry of the Environment, Japan. M. K., M. S., R. S. and Y. F. provided the methodology. M. S., M. T. and Y. F. involved in statistical analysis. M. K. and M. S. prepared the original draft. M. S., T. H., K. K., K. Y. and S. O. involved in review and editing. K. K. involved in supervision. All authors have read and agreed to the published version of the manuscript. The authors declare that they have no conflict of interest except the association with the Ministry of Environment, Japan. The data that support the findings of this study are available from Japan Environment and Children's Study Program Office. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from https://www.nies.go.jp/jecs/ with the permission of Japan Environment and Children's Study Program Office.