Abstract Study question Do exogenous gonadocorticoids protect against early pregnancy loss (EPL) following frozen embryo transfer (FET)? Summary answer Exogenous gonadocorticoids conferred a 59.5% increased risk of EPL (p<.05) without any advantage to biochemical pregnancy rates. This corresponded to a significantly thinner pre-FET endometrium. What is known already During a medicated FET cycle, exogenous progestogens and/ or oestrogens are administered with the intention of inducing endometrial maturation prior to implantation. However, the concomitant administration of both these gonadocorticoids is known to produce secretory changes alongside endometrial proliferation. It is unclear whether this non-physiological morphology is truly favourable for implantation, with some authors reporting similar EPL rates between medicated and “natural” FET cohorts while others have found exogenous gonadocorticoids to reduce the risk of EPL compared to alternative protocols. Study design, size, duration Retrospective, single-centre cohort study assessing the rates of biochemical pregnancy and EPL following 819 FETs. All FETs were conducted between January 2019 and November 2022 within a single British centre. Participants were followed up at 6-8 weeks post-FET to assess for biochemical pregnancy and EPL. The former was defined as a positive urinary human chorionic gonadotropin; the latter was defined as the absence of pregnancy on transvaginal ultrasound despite previous confirmation of biochemical pregnancy. Participants/materials, setting, methods Participants comprised 586 women undergoing medicated FET cycles and 233 women undergoing “natural” cycles. Electronic healthcare records (EHRs) were accessed to record pregnancy outcomes, hormonal therapies and endometrial thickness at 48hrs prior to FET. Binomial variables were compared using Pearson’s chi-squared test; continuous variables were analysed via T-tests and logistic regressions. Cut-offs of 8mm, 9mm and 10mm were applied to endometrial thickness in sub-group analyses. Statistical test were conducted using STATA v18.0 (StataCorp LLC, Texas). Main results and the role of chance Women undergoing medicated FET cycles achieved 16.9% higher rates of biochemical pregnancy compared to women undergoing “natural” cycles (57.7% versus 49.4%, p<.05). However, they also suffered a 59.5% higher rate of EPL (28.8% versus 18.0%, p<.05), which corresponded to significantly thinner endometrium at pre-FET ultrasound in comparison to women on “natural” cycles (9.13mm versus 9.48mm, p<.05). Logistic regressions revealed that, within the medical FET cohort, each 1mm increase in thickness conferred a 13.8% increased risk of EPL (95%CI: 3.1%-25.5% p<.01) but no advantage in biochemical pregnancy rates. “Natural” cycles demonstrated no such association. Endometrial thickness was then dichotomised using three thresholds. Women undergoing medicated FET cycles whose endometrium exceeded 9mm or 10mm had 62.1% and 71.4% greater risk of EPL, respectively, compared to those below either threshold. Biochemical pregnancy rates did not differ. Women undergoing “natural” FET cycles did not exhibit any significant difference using either threshold, suggesting that the disparate EPL rate in medicated cycles may result from non-physiological endometrial stimulation by exogenous gonadocorticoids. The 8mm cut-off was insignificant in both cohorts. Although these results cannot confirm a causal relationship between exogenous gonadocorticoids, endometrial thickness and EPL, they are sufficiently striking to warrant further investigation into their pathophysiological basis. Limitations, reasons for caution EHRs failed to report endometrial thickness for 125 participants (15.3%), limiting statistical power in logistic regressions. Our results cannot be confidently applied to other centres until further studies are conducted. Additionally, our design did not allow for the disparity in EPL rates to be attributed to a particular pathophysiological mechanism. Wider implications of the findings This data generates further uncertainty within the literature with regards to the relative risk of EPL in medicated versus “natural” FET cycles. The empirical evidence to support one protocol over the other is insufficient at present, and thus clinical recommendations should be based on the individual preferences of each patient. Trial registration number not applicable