Particulate air pollution at the time of oocyte retrieval is independently associated with reduced odds of live birth in subsequent frozen embryo transfers.

Study question: Does exposure to particulate matter (PM) air pollution prior to oocyte retrieval or subsequent frozen embryo transfer (FET) affect the odds of live birth?

Summary answer: Live birth rates are lower when particulate matter (PM2.5 and PM10) levels are higher prior to oocyte retrieval, regardless of the conditions at the time of embryo transfer.

What is known already: Exposure to air pollution is associated with adverse reproductive outcomes, including reduced fecundity and ovarian reserve, and an increased risk of infertility and pregnancy loss. It is uncertain whether the effect on ART outcomes is due to the effects of pollution on oogenesis or on early pregnancy.

Study design, size, duration: This retrospective cohort study included 3659 FETs in 1835 patients between January 2013 and December 2021, accounting for all FETs performed at a single clinic over the study period. The primary outcome was the live birth rate per FET. Outcome data were missing for two embryo transfers which were excluded. Daily levels of PM2.5, PM10, nitric oxide, nitrogen dioxide, sulphur dioxide, ozone and carbon monoxide were collected during the study period and calculated for the day of oocyte retrieval and the day of embryo transfer, and during the preceding 2-week, 4-week, and 3-month periods.

Participants/materials, setting, methods: Clinical and embryological outcomes were analysed for their association with pollution over 24 hours, 2 weeks, 4 weeks, and 3 months, with adjustment for repeated cycles per participant, age at the time of oocyte retrieval, a quadratic age term, meteorological season, year, and co-exposure to air pollutants. Multi-pollutant models were constructed to adjust for co-exposures to other pollutants. Median concentrations in pollutant quartiles were modelled as continuous variables to test for overall linear trends; a Bonferroni correction was applied to maintain an overall alpha of 0.05 across the four exposure periods tested.

Main results and the role of chance: Increased PM2.5 exposure in the 3 months prior to oocyte retrieval was associated with decreased odds of live birth (linear trend P = 0.011); the odds of live birth when PM2.5 concentrations were in the highest quartile were reduced by 34% (OR 0.66, 95% CI 0.47-0.92) when compared to the lowest quartile. A consistent direction of effect was seen across other exposure periods prior to oocyte retrieval, with an apparent dose-dependent relationship. Increased exposure to PM10 particulate matter in the 2 weeks prior to oocyte retrieval was associated with decreased odds of live birth (linear trend P = 0.009); the odds of live birth were decreased by 38% (OR 0.62, 95% CI 0.43-0.89, P = 0.010) when PM10 concentrations were in the highest quartile compared with the lowest quartile. Consistent trends were not seen across other exposure periods. None of the gaseous pollutants had consistent effects, prior to either oocyte retrieval or embryo transfer.

Limitations, reasons for caution: This was a retrospective cohort study, however, all FETs during the study period were included and data were missing for only two FETs. The results are based on city-level pollution exposures, and we were not able to adjust for all possible factors that may affect live birth rates. Results were not stratified based on specific patient populations, and it was not possible to calculate the cumulative live birth rate per commenced cycle.

Wider implications of the findings: This is the first study to specifically analyse FETs to separate the effects of environmental exposures prior to oocyte retrieval from those around the time of embryo transfer. Our findings suggest that increased PM exposure prior to oocyte retrieval is associated with reduced live birth rate following FET, independent of the conditions at the time of embryo transfer. Importantly, the air quality during the study period was excellent, suggesting that even 'acceptable' levels of air pollution have detrimental reproductive effects during gametogenesis. At the low pollution levels in our study, exposure to gaseous pollutants did not appear to affect live birth rates. This has important implications for our understanding of the effects of pollution on reproduction, and highlights the urgent need for effective policies limiting pollution exposure to protect human health and reproduction.

Study funding/competing interest(s): No funding was provided for this study. S.J.L. is supported by the Jean Murray Jones Scholarship from the Royal Australian and New Zealand College of Obstetricians and Gynaecologists, has received educational sponsorship from Besins, Ferring, Merck, and Organon, honoraria from Hologic and Organon, consulting fees from Merck unrelated to the current study, and is a member of the Reproductive Technology Council of Western Australia. S.J.L. and R.J.H. are board members of Menopause Alliance Australia. C.S.R., M.W., and E.N. have no conflicts of interest to declare. R.J.H. is the Medical Director of Fertility Specialists of Western Australia, the National Medical Director of City Fertility Australia, and a shareholder in CHA SMG. He chairs the Western Australian Minister's Expert Panel on ART and Surrogacy. R.J.H. has made presentations for and received honoraria from Merck, Merck-Serono, Origio, Igenomix, Gideon-Richter, and Ferring, and has received support for attending meetings from Merck, Organon, and Ferring.

Trial registration number: N/A.