Changes in ambient environmental conditions correspond to variations in in vitro fertilization outcomes during the COVID-19 pandemic.

BACKGROUND

Air pollution is associated with impaired oocyte and embryo development, but the modifying effects of large-scale societal disruptions remain unclear. The aim of this study was to examine whether in vitro fertilization (IVF)-related mature oocyte, fertilization, and pregnancy rates differ across the periods before, during, and after the coronavirus disease 2019 (COVID-19) pandemic. We further evaluated whether ambient environmental exposures modified these differences.

METHODS

We retrospectively analyzed oocyte retrieval cycles performed at a single tertiary infertility center in Seoul, South Korea (2018-2024). Complete-case analyses included 1, 871 cycles for mature oocyte rate, 1, 234 cycles for fertilization rate, and 662 cycles for clinical pregnancy. Environmental exposures (PM2.5, PM10, NO2, O3, CO, SO2, temperature, humidity) were assessed over follicular-phase, 90-day, and 365-day windows. Beta regression (mature oocyte and fertilization rates) and logistic regression (pregnancy) models were fitted with cluster-robust standard errors, adjusting for maternal age, AMH (except pregnancy), treatment indication, controlled ovarian stimulation protocol, fertilization method, and total motile sperm count.

RESULTS

The mature oocyte rate was significantly lower in the during-COVID (β = -0.32, 95% CI -0.45 to -0.19, p < 0.001) and post-COVID (β = -0.45, 95% CI -0.59 to -0.31, p < 0.001) groups than in the pre-COVID reference. The fertilization rate was higher in the post-COVID group than in the pre-COVID group in the regression model (β = 0.22, 95% CI 0.02 to 0.42, p = 0.030), whereas the during-COVID comparison was not significant (p = 0.637). Pregnancy rate did not differ significantly across the three periods (Wald χ² = 2.12, p = 0.346). Long-term (365-day) exposures exhibited significant Env × COVID interactions on the Likelihood Ratio Test for mature oocyte rate (O3: LRT p = 0.002; CO: LRT p = 0.008; NO2: LRT p = 0.009). No environmental interactions were significant for fertilization rate. For clinical pregnancy, five Env × COVID interactions reached nominal significance, including SO2 × 90-day (LRT p = 0.014), temperature × 365-day (p = 0.013), and O3 × 365-day (p = 0.020).

CONCLUSIONS

IVF outcomes varied across COVID-19 pandemic periods, and long-term ambient environmental exposures modified these effects, particularly for oocyte maturation. Recent and long-term environmental exposures should be considered when evaluating IVF success.