Effect Modification of Serum Omega-3 Fatty Acids on the Associations between Urinary Phthalate Biomarkers Mixture and Pregnancy Outcomes among Women Seeking Fertility Care

Abstract

Background: Phthalate exposures are ubiquitous and have been associated with pregnancy complications. Interaction between serum long-chain n-3 polyunsaturated fatty acids (n3PUFA) and phthalate biomarkers is biologically plausible because both can bind to human peroxisome proliferator-activated receptors (PPARs), which are involved in placenta development. However, evidence of this interaction in humans is lacking.

Objective: This study evaluated whether serum n3PUFA modifies the associations of biomarkers of phthalate exposure on pregnancy outcomes.

Methods: Among 351 women undergoing in vitro fertilization in the Environment and Reproductive Health study (2004-2017), we evaluated the effect modification of eicosapentaenoic acid (EPA) and serum docosahexaenoic acid (DHA) on the association of pregnancy outcomes with the mixture of urinary concentrations of phthalate biomarkers by quantile g-computation. All models were adjusted for age, body mass index, prior smoking, infertility diagnosis, treatment year, and urinary specific gravity.

Results: Concentrations of the phthalate biomarkers mixture were associated with higher adjusted probabilities of pregnancy loss and lower estimated probabilities of live birth among women with serum $\text{EPA}+\text{DHA}$ in the lowest tertile ($<2.66\%$ of total fatty acids), but not among women with middle-to-high serum $\text{EPA}+\text{DHA}$ ($p\text{interactions}=0.06$ and 0.15, respectively). Among women in the lowest tertile of serum $\text{EPA}+\text{DHA}$, the adjusted probability [95% confidence interval (CI)] of pregnancy loss for women in the lowest and highest quartile of phthalates mixtures was 5% (95% CI: 2%, 16%) and 44% (95% CI: 23%, 85%), respectively ($p\text{trend}=0.01$). The corresponding estimates were 14% (95% CI: 5%, 41%) and 11% (95% CI: 3%, 42%) among women with serum $\text{EPA}+\text{DHA}$ in the highest tertile ($\ge 3.78\%$ of total fatty acids) ($p\text{trend}=0.81$). Similar trends were observed for live birth but not for implantation and clinical pregnancy.

Conclusions: This study suggests adverse effects of phthalate exposure on pregnancy loss and live birth may be attenuated by intakes of n3PUFA. These results, if replicated, could inform clinical practice reducing the burden of infertility by phthalate exposure among the general population and improving pregnancy outcomes among subfertile couples. https://doi.org/10.1289/EHP15942.

Figure 1.

Estimated probability of pregnancy loss (A) and live birth (B) across quartiles of phthalate biomarkers mixture stratified by tertiles of serum $\text{EPA}+\text{DHA}$ fatty acids concentrations among 351 women from the Environment and Reproductive Health (EARTH) Study (2004–2017). Data are presented as probability (95% CI) of outcome, corresponding to Table S5. Quantile g-computation models adjusted for age, BMI, ever smoked, primary infertility diagnosis, year of treatment, and urinary specific gravity. Wald test was used to test for interaction and test for trend was based on quartiles of biomarker mixture. After stratification by serum $\text{EPA}+\text{DHA}$ (in percentage of total fatty acids), tertile 1 (1.05%–2.65%), tertile 2 (2.66%–3.77%), and tertile 3 (3.78%–35.24%) consisted of 116, 116, and 119 women, respectively. Note: BMI, body mass index; CI, confidence interval; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; $\text{EPA}+\text{DHA}$, the sum of EPA and DHA concentrations.