P–736 Elevated levels of ambient air pollutants increase the primary sex ratio in human embryos

Abstract Study question Are there any associations between ambient outdoor air pollution and the primary sex ratio (PSR)? Summary answer Short-term exposure to increased PM10, PM2.5 and NO2 levels were significantly associated with higher PSR. What is known already PSR estimates represent a backward extrapolation from data based on spontaneous or induced abortions, fetal deaths or live births and are usually male-biased. A recent study, analyzing 3- to 6-day-old embryos derived from assisted reproductive technology (ART) procedures, showed that the sex ratio at conception is unbiased (0.5). Epidemiologic studies of air pollution on secondary (birth) sex ratio showed that higher levels of particulate pollution were associated with increased rates of female birth. However, a direct association between urban levels of air pollutants and PSR has not been reported. Study design, size, duration A retrospective cohort study was carried out to assess the impact of long- or short-term exposure to six ambient outdoor air pollutants (particulate matter, PM10µm and PM2.5µm; SO2; CO; NO2; O3) on PSR (XY/XX) of couples undergoing their first IVF cycle for preimplantation genetic screening (N = 337). Data was from fixed air quality monitoring stations across the city between January 2014 and December 2018. Embryos with sex chromosome abnormalities were excluded from the analysis. Participants/materials, setting, methods Average concentrations of the pollutants for the 90 (long-term exposure) and 15 days (short-term exposure) predating oocyte retrieval represented the exposures of interest. Pollutant levels were categorized into quartiles (Q1 to Q4) and exposure risk was divided into two periods in which average concentrations and confidence intervals for the pollutants were in the upper quartile (Q4 period) or not (Q1-Q3 period). The strength association between exposure risk and PSR was performed through analysis of covariance. Main results and the role of chance The estimated means of PM10, PM2.5, SO2, NO2, O3 and CO for Q1-Q3/Q4 periods were 27.7/39.3, 16.7/23.7, 2.5/3.9, 37.0/46.4, 32.2/45.3 µg/m3 and 0.64/0.87 ppm and 26.3/43.0, 16.0/26.3, 2.4/4.2, 36.5/47.8, 31.7/50.4 µg/m3 and 0.62/0.90 ppm for long- and short-term exposures, respectively. PM10, PM2.5 and NO2 levels in the Q4 period had significantly higher PSR (138.1, 134.0 and 137.6) when compared to Q1-Q3 period (94.4, 98.1 and 96.4) for the short-term exposure (p = 0.0193; p = 0.0439; p = 0.0180, respectively). PM10, PM2.5, SO2, NO2 and CO levels in the Q4 and Q1-Q3 periods for the long-term exposure showed no significant effect on PSR. Contrastingly, O3 levels in the Q4 period had significantly lower PSR (82.6) when compared to Q1-Q3 (115.9) for the long-term exposure (p = 0.0202). A monotonic increase in PSR was observed with increased PM10 concentration in the Q4 period for the short-term exposure (F-ratio: 4.4476; p = 0.0352). Limitations, reasons for caution Some limitations of the study should be underlined, such as its retrospective nature, exposure assessment based on pollutant levels derived from a network average across city sites, and limited extrapolation of the results to the general population. Wider implications of the findings: Our data suggest that short-term exposure to environmental factors could affect the primary sex ratio in polluted seasons or cities. A monotonic effect on PSR in the case of exposure to increasing PM10 levels was identified. Trial registration number Not applicable