Among women with poor ovarian response and repeated implantation failure, endometriosis was not linked to statistically significant differences in embryo development or the early clinical outcomes measured after assisted reproductive treatment. The result came from a small retrospective cohort, and the authors stressed that it should not be read as proof that the disease has no effect.
The study looked at women aged 35 to 44 who met the Bologna criteria for poor ovarian response, had at least three previous failed embryo-transfer cycles and underwent fresh embryo transfer using SV-ICSI at a single fertility center between January 2023 and December 2024.
What the researchers compared
The final analysis included 144 women. Thirty-six had laparoscopically confirmed endometriosis, while 108 age-matched women formed the comparison group. The control participants had no previous endometriosis diagnosis or ultrasound evidence suggesting the disease, and were randomly selected from the eligible non-endometriosis pool.
Researchers examined embryo-related and clinical ART outcomes. They compared continuous measures with unpaired t-tests and categorical measures with chi-square or Fisher exact tests. A multivariable logistic regression model, which accounts for several factors at once, adjusted for age, body mass index, AMH level, the number of oocytes retrieved and the number of embryos transferred.
The measured outcomes were similar
The embryo figures were close. Fertilization was 76.2% in the endometriosis group and 78.9% in the control group. Cleavage, the early division of an embryo, was 96.3% and 96.6%, respectively. Good-quality cleavage embryo rates were 55% and 55.6%, while blastocyst formation rates were 23.9% and 21.7%. None of these differences was statistically significant.
The reported clinical outcomes showed the same pattern. Implantation rates were 13% and 11.7% in the endometriosis and control groups. Clinical pregnancy rates were 19.4% and 18.5%, while clinical abortion rates were 28.6% and 40%. None of the group differences was statistically significant.
After adjustment for the measured factors, endometriosis was not independently associated with clinical pregnancy. The estimated odds ratio was 0.795, with a 95% confidence interval from 0.284 to 2.224 and a P-value of 0.662. The wide interval underscored the uncertainty around the estimate.
Why the result remains uncertain
Because the study was retrospective, it can show an association in these records but cannot establish cause and effect. The authors also said the small endometriosis group and limited number of clinical pregnancy events may have reduced statistical power. They cautioned that the adjusted estimates had limited stability and precision.
Live birth was not prospectively assessed. Instead, implantation and clinical pregnancy served as surrogate reproductive outcomes, so the study leaves live-birth outcomes untested.
Beyond the sample size, the evidence is narrow. The study was conducted at one fertility center and included only fresh embryo-transfer cycles in the specified eligibility group. The authors did not regard the finding as definitive.
The question is still open
The authors described the findings as preliminary and hypothesis generating, rather than definitive evidence that endometriosis has no effect on ART outcomes. They said larger prospective investigations incorporating live-birth outcomes are needed.
The research received no external funding, and the authors declared no conflict of interest.
Paper data and sources
Original title: Influence of endometriosis on the outcomes of assisted reproductive technology in poor ovarian responders with previous implantation failures.
Authors: Soojin Oh, Seula Lee, Sungwook Chun et al.
Journal/Repository: Obstetrics & gynecology science
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.5468/ogs.26164
Original paper · Full text