EUVIMEDEUVIMED entdecken
Kohortenstudie

L26/P-183 Do female infertility factors compromise oocyte competence and AI-based quality assessment? Results from a large retrospective cohort study

A. Iaconelli, M. Del Collado, D. Braga, A. Setti, D. Hernadez, J. Fjeldstad, P. Mojiri, L. Vanzella, E. Borges

Human Reproduction · 01.07.2026 · 41 (Supplement_1)

DOI: 10.1093/humrep/deag083.519 ↗

Kurz gefasst

Kohortenstudie in der Zeitschrift Human Reproduction, online erschienen am 08.07.2026. Verfasst von A. Iaconelli und 8 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Study question Do female infertility factors associated with reduced oocyte quality influence oocyte competence and the predictive value of an AI-based oocyte score for embryo developmental outcomes? Summary answer Female infertility factors other than advanced maternal age did not meaningfully impair oocyte competence. The AI-based oocyte quality scores remained predictive across female clinical conditions. What is known already Advanced maternal age (AMA) is the female infertility factor most consistently associated with diminished oocyte quality and adverse reproductive outcomes due to increasing chromosomal impairment with age. Evidence regarding the impact of polycystic ovary syndrome (PCOS), endometriosis, and body mass index (BMI) on intrinsic oocyte quality or competence remains heterogeneous and debated within the scientific community. AI-based tools have emerged to estimate oocyte and embryo quality, but their robustness across different female infertility conditions is not well established. It remains unclear whether AI-derived oocyte quality metrics maintain their predictive relevance in the presence of various female infertility factors. Study design, size, duration Retrospective cohort study performed between 2020 and 2025 at a university-affiliated IVF center. Analysis included: (i) comparison between a control group (1,017 cycles; 5,658 oocytes) and women with PCOS (102 cycles; 1,238 oocytes), endometriosis (265 cycles; 1,842 oocytes), overweight (BMI≥25–<30; 416 cycles; 2,790 oocytes), obesity (BMI≥30; 136 cycles; 926 oocytes), or AMA (≥38 years; 589 cycles; 2,910 oocytes); and (ii) evaluation of predictive robustness of an AI-derived oocyte quality score. Participants/materials, setting, methods Laboratory outcomes AI-derived oocyte and embryo quality scores, fertilization, blastocyst and top-quality blastocyst rates and embryo morphokinetics. Clinical outcomes were implantation, clinical pregnancy, and miscarriage. In the comparative analysis, embryo-level outcomes were analyzed using generalized estimating equations and clinical outcomes using generalized linear models. Mixed-effects logistic regression models assessed MS associations with fertilization and blastocyst formation and tested effect modification by female factors. Linear mixed-effects models examined associations between oocyte and embryo scores. Main results and the role of chance In the comparative analysis, female infertility factors other than AMA were not associated with clinically meaningful impairment of oocyte or embryo competence. AMA was associated with lower oocyte (6.14±3.10 vs 5.65±2.88; p = 0.005) and embryo (5.26±2.51 vs 4.96±2.38; p = 0.008) quality scores, without significant differences in morphokinetic parameters. Blastocyst formation (54.0% vs 46.4%; p = 0.012) and top-quality blastocyst rates (48.2% vs 41.0%; p < 0.001) also differed between groups. Implantation, clinical pregnancy, and miscarriage rates were similar across all comparisons. Oocyte score consistently predicted fertilization (OR = 2.13 per SD increase; p < 0.001), blastocyst formation (OR = 1.85 per SD increase; p < 0.001), and top-quality blastocysts (OR = 1.30 per SD increase; p < 0.001) across maternal age, BMI, and PCOS or endometriosis diagnoses, with no significant effect modification and largely parallel predicted probability curves. Maternal age and, to a lesser extent, BMI showed expected baseline effects on specific outcomes, without attenuating its predictive value. The oocyte score was positively associated with embryo score (β = 0.38 per SD; p < 0.001) across all female factor strata. This association was preserved across the maternal age and BMI spectrum and showed modest attenuation in PCOS, where it remained statistically significant albeit weaker (β = 0.19; p = 0.046). Limitations, reasons for caution The retrospective, single-center design limits causal inference. Although statistical models were extensively adjusted, residual confounding and unmeasured factors cannot be completely excluded. Wider implications of the findings Despite the biological impact of AMA, female infertility factors commonly associated with reduced oocyte quality do not meaningfully compromise oocyte competence. Importantly, the AI-driven oocyte scores remained predictive across female infertility conditions, supporting its use for individualized oocyte-level assessment and relevance for treatment planning and decision making in IVF. Trial registration number No

Zitieren

APA

Iaconelli, A., Collado, M. D., Braga, D., Setti, A., Hernadez, D., Fjeldstad, J., Mojiri, P., Vanzella, L., & Borges, E. (2026). L26/P-183 Do female infertility factors compromise oocyte competence and AI-based quality assessment? Results from a large retrospective cohort study. Human Reproduction, 41(Supplement_1). https://doi.org/10.1093/humrep/deag083.519

Vancouver

Iaconelli A, Collado MD, Braga D, Setti A, Hernadez D, Fjeldstad J, et al. L26/P-183 Do female infertility factors compromise oocyte competence and AI-based quality assessment? Results from a large retrospective cohort study. Human Reproduction. 2026;41(Supplement_1). doi:10.1093/humrep/deag083.519

BibTeX
@article{Iaconelli2026,
  author = {Iaconelli, A. and Collado, M. Del and Braga, D. and Setti, A. and Hernadez, D. and Fjeldstad, J. and Mojiri, P. and Vanzella, L. and Borges, E.},
  title = {L26/P-183 Do female infertility factors compromise oocyte competence and AI-based quality assessment? Results from a large retrospective cohort study},
  journal = {Human Reproduction},
  year = {2026},
  volume = {41},
  number = {Supplement_1},
  doi = {10.1093/humrep/deag083.519}
}
Kostenlos starten →