Multiple pregnancies display all the complications of singletons, but in highest frequencies, especially for fetal growth abnormalities. In singletons, air pollution seems to be negatively associated with fetal growth. A similar relationship in multiple pregnancies can be expected. However, multiple pregnancies are systematically excluded from the studies of the impact of air pollution on pregnancy outcomes. This study aims to analyze in a population of multiple pregnancies the relationship between fetal growth restriction (FGR), small for gestational age (SGA) and environmental chronic exposure to air pollution in moderately polluted cities.

All the multiple pregnancies of women living in the city of Besançon (121,671 inhabitants) or in the urban unit of Dijon (243,936 inhabitants) and who delivered in a university hospital between 2005 and 2009 were included. Births were classified as SGA if birth weight was<10th centile for gestational age and sex in one or more newborns of the pregnancy. FGR was defined by the association of SGA and a defect in fetal growth. FGR was retained according to the ICD10 codes in medical records. Outdoor nitrogen dioxide (NO₂) exposure was assessed using the mother's address, in front of the residential building and within 50m around the building. Several time periods were defined: each trimester, entire pregnancy and two months before delivery. All variables available in the medical records were analyzed to detect potential confounders: maternal socioeconomic characteristics, obstetrical history, complications of pregnancy and characteristics of the newborns. Multivariate logistic regression models were performed. Sensitivity analyses were conducted using different criteria to define exposure and outcome.

This study included 249 multiple pregnancies and 506 newborns. Among the 249 pregnancies, 64 presented a FGR and 94 a SGA in one or more fetus of the pregnancy. The median of NO₂ concentration within 50m around building during entire pregnancy was 23.1μg/m³ (minimum at 10.1μg/m³ and maximum at 46.7μg/m³). No association was observed between NO₂ and SGA whatever the pregnancy period considered (the odds ratio (OR) for a 10μg/m³ increase of NO₂ exposure ranged between 0.78 and 0.88). Regarding FGR, the OR associated with a 10μg/m³ increase of NO₂ exposure during the first, the second and the third trimester were respectively 1.42 (95% confidence interval (CI=0.97–2.08), 1.55 (95% CI=1.06–2.27), 1.35 (95% CI=0.92–1.98). The OR associated with a 10μg/m³ increase of NO₂ exposure during entire pregnancy and during the two last months before delivery were respectively 1.52 (95% CI=1.02–2.26) and 1.53 (95% CI=1.04–2.25). Sensitivity analyses with NO₂ concentration in front of the residential building led to similar results.

In our study, environmental exposure to NO₂ was associated with FGR in multiple pregnancies, especially during the 2nd trimester and the two last months before delivery. However, no association between NO₂ exposure and SGA was identified. These results are in line with an impact of NO₂ on fetal growth in multiple pregnancies for an exposure under the European legislation threshold.