Abstract Study question How anxiety and stress questionnaires punctuations and endometrial steroid biosynthesis metabolites levels are associated with embryo implantation and pregnancy outcomes in IVF patients? Summary answer Those patients with high cortisol and low estradiol endometrial levels, were stressed according to STAI State punctuations and showed lower rates of implantation and pregnancy. What is known already Although IVF treatments are known to be a source of stress, its relevance in reproductive success remains controversial, mainly due to the lack of accurate biomarkers. Steroid hormones, such as progesterone and estrogens, are essential for reproductive physiology, and directly related to common stress biomarkers such as cortisol. Nowadays, clinical stress evaluation is exclusively performed via highly subjective psychological questionnaires, not being standardized in the clinical setting. In this study, we measured endometrial levels of main steroid hormones -including cortisol-, stress psychological questionnaires, and reproductive outcomes for determining the relationship among potential stress biomarkers and their influence on fertility. Study design, size, duration In this prospective cohort study, a total of 55 IVF patients were included (<45 years old, no uterine or systemic pathologies and good quality embryos). All endometrial biopsies were collected in mid-secretory phase for metabolites measurement between 2019 and 2021. Also, these patients were psychologically evaluated by stress related questionnaires the same day of biopsy collection or days later in the same cycle. Reproductive outcomes were compared between groups established according to stress biomarkers thresholds. Participants/materials, setting, methods Concentration of eleven selected steroid metabolites was measured by ultra-performance liquid chromatography-tandem mass spectrometry. A psychological questionnaire was also assessed to measure patients’ stress and anxiety (STAI). According to metabolites’ concentration distribution, Barnard’s test was applied to compare the proportion of patients with successful and unsuccessful treatments at different thresholds. A Wilcoxon test was performed for mean metabolites’ levels comparison between stressed and unstressed patients. A p-value of 0.05 was stablished as significant. Main results and the role of chance All patients with high endometrial cortisol levels (>12.5 ng/g; n = 3) failed to implant (100% vs 45.8% non-implanted in > 12.5 ng/g and <12.5 ng/g patients respectively, p-value=0.047), while the totality of patients with high endometrial estradiol levels (>0.9 ng/g; n = 5) achieved embryo implantation and on-going pregnancy (100% vs 50% implanted in > 0.9 ng/g and <0.9 ng/g patients respectively, p-value=0.038). Those patients with high cortisol levels, all not pregnant, had also low estradiol levels and high stress state punctuation (STAI State punctuations ≥ 60). Conversely, patients with high estradiol levels, all pregnant, had also low cortisol levels and were not stressed (STAI State punctuations < 60). Stratification of patients into stressed (STAI State>60) and unstressed (STAI State<60) independently of their implantation results, revealed that 27% of stressed women presented high cortisol levels, while 100% of unstressed patients had low cortisol levels. In the same way, 42% of unstressed patients had high estradiol levels, while 100% of stressed patients presented low estradiol values. Hence, mean values differences of cortisol (10.54 vs 3.05 ng/g) and estradiol (0.55 vs 4.10 ng/g) between stressed and unstressed patients were not significant (p-value=0.08862 and p-value=0.2327, respectively). No relevant associations were observed with stress as a trait. Limitations, reasons for caution Due to the heterogeneous and non-normal metabolites’ concentration distribution among patients, this study would need a largest sample size for having a representative population in the different subgroups. Although we have used molecular and psychological validated stress biomarkers, a causal relationship between stress and reproductive outcomes should be further explored. Wider implications of the findings This study highlighted cortisol and estradiol as potential endometrial biomarkers that better reflect pregnancy prognosis. Further investigation is needed to find if other psychological questionnaires molecularly recognize stressed patients as a noninvasive diagnosis associated to reproductive outcomes. These patients would benefit from psychological counseling to improve their reproductive success. Trial registration number Not applicable