L26/O-020 Adenomyosis and pregnancy outcomes in women undergoing assisted reproductive technology (ART): a prospective multicentre cohort study (AdAPT-ART)

Abstract Study question What is the impact of adenomyosis and its phenotypes on pregnancy outcomes in women undergoing ART? Summary answer Adenomyosis is associated with poorer pregnancy outcomes, from embryo implantation to term live birth, across autologous and donor oocyte cycles and euploid embryo transfers. What is known already Adenomyosis has been associated with adverse pregnancy outcomes, although evidence remains conflicting, particularly for donor-oocyte and euploid embryo transfers. It is unclear whether these associations reflect an intrinsic uterine pathology or are confounded by gamete- or embryo-related factors. Most studies assess adenomyosis as a binary diagnosis, with limited evaluation of phenotype-specific effects. Stratification by key ART-related modifiers, such as oocyte source, embryo ploidy status, and phenotype defined by cavity distortion, myometrial involvement, and severity score, remains inadequate. Study design, size, duration This prospective multicentre cohort study was conducted over 20 months (June 2022–February 2024) across eight assisted conception clinics. In total, 16,042 women were screened for adenomyosis using two- and three-dimensional transvaginal ultrasound (2D–3D TVS). Of these, 813 eligible women were included: 363 with adenomyosis (452 initiated cycles, 559 transfers) and 450 without adenomyosis (504 initiated cycles, 549 transfers). Participants/materials, setting, methods Participants aged 18 to 45 years undergoing ART were screened. Those with co-existing fibroids, uterine malformations, untreated hydrosalpinx, or prior myomectomy or adenomyomectomy were excluded. Adenomyosis was diagnosed by the presence of at least one Morphological Uterus Sonographic Assessment (MUSA) feature on pre-treatment ultrasound. Modified Poisson regression with multivariable analyses was used to estimate effect sizes per participant, initiated cycle and transfer. Stratified and subgroup analyses were performed by phenotype and ART-related modifiers. Main results and the role of chance Transfers in women with adenomyosis were associated with lower adjusted relative risk (aRR) of pregnancy (0.70, 95% confidence interval [CI] 0.62 – 0.79), clinical pregnancy (0.65, 95% CI 0.57 – 0.74), live birth ( 0.36, 95% CI 0.30 – 0.44) and higher miscarriage (4.73, 95% CI 3.26 – 6.86) and pregnancy loss (3.58, 95% CI 2.67 – 4.81) compared to controls. This association remained consistent when data were analysed per participant and per initiated cycle. The live birth rate remained lower in donor oocyte transfers (aRR: 0.42, 95% CI 0.26 – 0.68) and in euploid embryo transfers (aRR: 0.38, 95% CI 0.24 – 0.61). The miscarriage risk remained approximately 5 times higher across donor oocyte (aRR: 4.68, 95% CI 1.37 – 16.03) and euploid embryo transfers (aRR: 4.69, 95% CI 1.39 – 15.81). The association between adenomyosis and adverse pregnancy outcomes was consistent across fresh vs. frozen transfers and downregulated protocols (all interaction p-values > 0.05). Phenotypes most strongly associated with adverse outcomes included junctional zone adenomyosis, irregular or interrupted junctional zone, moderate–severe cavity distortion on 3D ultrasound, >50% myometrial involvement, and moderate–severe disease by proposed classification. Limitations, reasons for caution Due to high rates of unsuccessful transfers and pregnancy loss in the adenomyosis cohort, the final sample of 540 clinical pregnancies was lower than the pre-specified target of 828. Deep infiltrating endometriosis was not systematically excluded on all pre-treatment scans, potentially resulting in residual confounding. Wider implications of the findings Adenomyosis is associated with poorer pregnancy outcomes across ART cycle types and analytical denominators. Beyond routine screening, phenotype-based risk stratification, including site, type, severity, cavity distortion, and myometrial involvement, is essential for individualised counselling and for informing the development of targeted, personalised therapeutic interventions. Trial registration number No