Ambient temperature reduces oocyte yield in assisted reproductive technology via a partially sex hormone-mediated...

Are higher ambient temperatures before oocyte retrieval associated with adverse oocyte-related outcomes in women undergoing their first IVF or ICSI cycle?

Higher pre-retrieval temperatures are associated with reduced oocyte yield, an effect partially mediated through LH and partially counteracted by compensatory increases in FSH and estradiol (E2).

Evidence on the impact of ambient temperature on IVF/ICSI outcomes is emerging but limited. A recent study reported weak negative associations between higher ambient temperatures during folliculogenesis and oocyte yield, though the effect was modest and findings across populations remain inconsistent. However, these findings are further limited by small sample sizes, specific clinical subpopulations or geographic regions, which prevents the establishment of a generalizable dose-response relationship and leaves the underlying biological mechanisms largely unexplored.

A retrospective cohort study of 58 468 women undergoing their first IVF/ICSI cycle with oocyte retrieval between 1 January 2018 and 31 May 2024.

Women undergoing their first IVF/ICSI cycle were included. The primary outcomes were total oocytes retrieved, mature oocytes retrieved, and 2PN (two pronuclei) fertilized oocytes. Secondary outcomes included rates of oocyte maturation, normal fertilization, and blastocyst formation. Basal sex hormone levels were measured during the early follicular phase (cycle days 2-4). Environmental exposure data, including both ambient temperature and apparent temperature (integrating air temperature and humidity), were assessed over biologically relevant exposure windows. Generalized linear mixed-effects models were used to estimate associations between temperature exposures and oocyte-related outcomes. Restricted cubic spline models assessed nonlinearity. A counterfactual causal mediation analysis examined the roles of basal sex hormones. Stratified analyses by age, BMI, antral follicle count, anti-Müllerian hormone, season, COVID-19 period, and the distance to ART center, along with sensitivity analyses, were performed to evaluate robustness.

Each 1°C increase in apparent temperature was associated with significant decreases in total oocytes retrieved (-0.20%; 95% CI: -0.28%, -0.12%), mature oocytes retrieved (-0.17%; 95% CI: -0.26%, -0.09%), and 2PN fertilized oocytes (-0.19%; 95% CI: -0.28%, -0.09%); consistent associations were observed for ambient temperature (-0.25%, -0.22%, and -0.22% per 1°C, respectively). For rate-based outcomes, no significant associations were observed for the normal fertilization rate or blastocyst formation rate; the mature oocyte rate showed small positive associations in longer-term exposure windows (3-week and 90-day). Mediation analyses indicated that LH partially mediated these associations, accounting for up to 5.23% of the total effect, while both FSH and E2 exhibited suppression effects.

The observational design cannot definitively establish causality. Unmeasured confounding (e.g. individual indoor climate control, occupational heat exposure) and residential address-based exposure assessment are potential sources of bias. Restricting the primary analysis to completed cycles may introduce selection bias, although sensitivity analyses incorporating canceled cycles yielded consistent estimates.

Higher ambient temperatures represent a modifiable environmental risk factor for oocyte yield in ART treatment, with basal sex hormone pathways, including LH as a partial mediator and compensatory shifts in FSH and E2, providing a plausible biological basis for this effect. In the context of rising global temperatures, these findings suggest that ambient thermal exposure during the peri-folliculogenesis period warrants clinical attention, and that monitoring basal sex hormone levels may help identify patients at risk of a suboptimal ovarian response in high-temperature settings.

This study was supported in part by the Technology Innovation and Research & Development Project of the Chengdu Science and Technology Bureau (No. 2024-YF05-02159-SN), the Open Fund of Chongqing Maternal and Child Disease Control and Public Health Research Center (No. CQFYSJ01001), the Program of Inheritance and Innovation of Traditional Chinese Medicine in Chongqing (Chongqing Traditional Chinese Medicine [2022] No. 33), and Chongqing Yuzhong District Natural Science Foundation Project (No. 20240118).

All authors declare that they have no conflicts of interest related to this study.

N/A.