Distribution of fat-soluble compounds, including dietary carotenoids, tocopherols and retinol, within the ovarian follicular environment is incompletely characterized in relation to in vitro fertilization (IVF) outcomes. Prior studies reported inconsistent associations between these micronutrients and reproductive endpoints. The present monocentric non-interventional study aimed to describe concentrations of fat-soluble micronutrients in plasma and follicular fluid (FF) of women undergoing IVF, and to analyze adjusted associations with ovarian function, embryo development and pregnancy outcomes.
The study enrolled 82 women undergoing their first IVF cycle in 2021–2022 at a single center in France. At oocyte puncture both plasma and FF samples were collected. Lifestyle data covering the preceding three months were recorded. The analysis was observational and non-interventional; the trial is registered under NCT04866329 and the study received ethical approval from the Comité de Protection des Personnes of Île‑de‑France.
Eleven fat-soluble compounds were quantified in both plasma and FF: two tocopherols (including γ-tocopherol and α-tocopherol), three xanthophylls (including β-cryptoxanthin), five carotenes (including α-carotene, lycopene, phytoene, phytofluene) and retinol. All compounds were detected in plasma and FF except phytoene, which was undetectable in FF. Overall concentrations were lower in FF than in plasma.
Plasma and FF α-tocopherol concentrations were both positively associated with plasma estradiol measured before oocyte puncture (both p < 0.01). Among xanthophylls, only plasma β-cryptoxanthin showed a positive association with plasma progesterone concentrations (p = 0.02). In contrast, FF α-carotene and lycopene were inversely associated with plasma progesterone (p = 0.01 and p = 0.02, respectively). These findings indicate micronutrient associations with circulating steroid hormones in the peri‑puncture period.
Different micronutrients demonstrated opposing associations with oocyte maturity. Plasma phytofluene and plasma phytoene were positively associated with mature oocyte rate (p = 0.03 and p = 0.01, respectively). Conversely, FF retinol was negatively associated with mature oocyte rate (p = 0.03). This pattern suggests potential compound‑specific effects on oocyte maturation, with some carotenoid species linked to higher maturity and FF retinol linked to lower maturity.
Several fat‑soluble compounds were associated with subsequent embryo outcomes. Carotenes, tocopherols and retinol showed inverse associations with later IVF indicators: fertilization rate was inversely associated with plasma γ-tocopherol (p < 0.001) and with FF retinol (p = 0.02). Plasma phytofluene was inversely associated with the proportion of top-quality embryos (p = 0.02). Additionally, both plasma and FF γ-tocopherol were positively associated with embryo fragmentation (both p < 0.05), indicating a relation between γ-tocopherol levels and fragmentation severity.
Micronutrient concentrations in plasma were associated with measures of implantation and pregnancy. Plasma α-tocopherol and α-carotene were inversely associated with biochemical pregnancy at day 7 after embryo transfer (p = 0.05 and p = 0.02, respectively). Clinical pregnancy was inversely associated with plasma α-tocopherol (p = 0.03) and plasma phytoene (p = 0.01). Live birth was inversely associated with plasma α-tocopherol (p = 0.04) and plasma phytoene (p = 0.02). These inverse associations across later outcomes contrast with some positive associations observed earlier in the IVF sequence.
Across measured stages of IVF, associations between tocopherols, carotenes, retinol and outcomes were heterogeneous. Some carotenoids (phytofluene, phytoene) showed positive associations with oocyte maturity, whereas tocopherols and retinol were linked inversely to fertilization, embryo quality and pregnancy endpoints. The authors interpret these patterns as consistent with a potentially beneficial effect limited to early outcomes and a complex, context-dependent role of these fat-soluble micronutrients in female reproduction. No causal inferences can be drawn from this observational study.
The study is monocentric and observational; results may be influenced by unmeasured confounding and cannot establish causality. The dataset analyzed is embargoed until main publications by the research team; however, data are available from the corresponding author upon reasonable request. Ethical approval was provided by the Comité de Protection des Personnes of Île‑de‑France and patient consent procedures were followed as reported.
Funding sources declared include Région Centre‑Val de Loire, APRIA 2019‑00134937, PERFIDE Centre INRAE Val de Loire and CHRU of Tours. The authors declared no competing interests. The manuscript was submitted to PLOS ONE on August 19, 2026.