A hyaluronan-enriched embryo transfer medium widely marketed as an in vitro fertilization add-on showed no advantage over a standard transfer medium in a large real-world clinical cohort, according to a retrospective study published in Archives of Gynecology and Obstetrics. The medium, sold under the brand name EmbryoGlue, is enriched with 0.5 milligrams per milliliter of hyaluronan and reduced recombinant human albumin, and is designed to support embryo implantation by promoting adhesion between the embryo and the endometrial lining. In the new study, clinical pregnancy rates were identical in cycles using the enriched medium and cycles using a conventional medium, and the enriched medium was not an independent predictor of pregnancy in a multivariate regression model.
The research team, led by Pelin Coştur Filiz and Serdar Filiz at a university-based assisted reproduction unit, analyzed 335 embryo transfer cycles performed between June 2025 and April 2026. The cohort was deliberately weighted toward frozen embryo transfer, which accounted for 87.5 percent of all cycles, reflecting the growing dominance of frozen transfer in modern assisted reproduction practice. Cycles were divided into a hyaluronan-enriched transfer medium group of 171 cycles and a control group of 164 cycles in which Irvine NX medium was used throughout. Allocation to transfer medium was not randomized; it was determined through shared clinical decision-making between the treating clinician and the patient during routine pre-transfer consultation, based on clinical history, prior transfer outcomes, patient preference, and clinician judgment. Neither patients nor clinicians were blinded to the allocation.
The primary outcome was the clinical pregnancy rate per transfer cycle, defined as the presence of at least one intrauterine gestational sac confirmed by transvaginal ultrasound at six to seven weeks of gestation. Serum beta-hCG was measured ten days after transfer in day-5 cycles and twelve days after transfer in day-3 cycles, with a value of at least 25 mIU/mL considered positive. The overall clinical pregnancy rate across the cohort was 56.7 percent, or 190 of 335 cycles. When the two groups were compared directly, the rates were exactly matched: 56.7 percent, or 97 of 171 cycles, in the hyaluronan-enriched group and 56.7 percent, or 93 of 164 cycles, in the control group. The corresponding odds ratio was 1.00 with a 95 percent confidence interval of 0.65 to 1.54 and a p value greater than 0.999.
The investigators also carried out exploratory subgroup analyses stratified by transfer type, embryo developmental stage, patient age, and the number of previous transfer attempts. None of these comparisons reached statistical significance. In the frozen embryo transfer subgroup of 293 cycles, clinical pregnancy rates were 60.0 percent in the enriched-medium group and 61.5 percent in the control group, with an odds ratio of 0.94. Among blastocyst transfers, rates were 67.7 percent versus 72.4 percent, yielding an odds ratio of 0.80. In women aged 35 years or older, rates were 39.0 percent versus 38.5 percent, and among cycles with two or more previous transfer attempts, rates were 54.9 percent versus 55.6 percent. The authors emphasized that these subgroup analyses were exploratory, were not adjusted for multiple comparisons, and were limited by small sample sizes, particularly for fresh embryo transfer.
Multivariate logistic regression, performed on 333 cycles with complete covariate data, identified three factors independently associated with clinical pregnancy. Frozen-thawed transfer was associated with markedly higher odds than fresh transfer, with an odds ratio of 4.32 and a 95 percent confidence interval of 2.00 to 9.31. Blastocyst-stage transfer, compared with transfer at other stages, carried an odds ratio of 2.96 with a confidence interval of 1.82 to 4.82. Younger patient age was also independently associated with pregnancy, with an odds ratio of 0.92 per year of age. Hyaluronan-enriched medium use, by contrast, was not an independent predictor, with an odds ratio of 0.89, a confidence interval of 0.55 to 1.46, and a p value of 0.652. The number of embryos transferred and a history of previous transfer attempts were likewise not independently associated with the outcome. The model showed good discrimination, with an area under the curve of 0.734, and borderline acceptable calibration on the Hosmer-Lemeshow test.
A sensitivity analysis restricted to the first transfer cycle contributed by each woman, covering 313 cycles, produced consistent results. Clinical pregnancy rates were 54.3 percent in the enriched-medium group and 56.3 percent in the control group, and the medium was again not independently associated with pregnancy. This analysis addressed a methodological concern: although only 21 women, or 6.7 percent of the 313 women in the cohort, contributed more than one cycle, repeat transfers from the same woman were not modeled as clustered observations in the primary cycle-level analysis.
One baseline imbalance between the groups deserves particular attention. The hyaluronan-enriched group contained a significantly higher proportion of cycles with two or more previous transfer attempts, 41.5 percent versus 27.4 percent, a difference the authors attributed to the intuitive clinical practice of escalating add-on interventions in patients who have not responded to standard treatment. This pattern of confounding by indication would be expected to bias the observed association of the enriched medium downward, meaning a true benefit could be underestimated. The authors cautioned that the null result should therefore not be read as evidence that the medium is harmful. At the same time, the enriched-medium group was enriched for precisely the population in which a benefit has most often been claimed, and no advantage appeared there, although only 116 cycles involved two or more previous attempts and the study was not powered to detect effects within subgroups.
The mechanistic rationale behind hyaluronan-enriched media rests on solid biological ground. Hyaluronan, or hyaluronic acid, is a glycosaminoglycan naturally present at high concentrations in the endometrium, follicular fluid, and oviductal secretions of the female reproductive tract. Its principal receptor, CD44, is expressed on the surface of human embryos from the oocyte stage through blastocyst development, and experimental work has shown that hyaluronan supplementation promotes embryo attachment to endometrial epithelial cells in vitro and enhances fetal development in animal models. Meta-analyses of randomized trials suggest a modest overall benefit on clinical pregnancy and live birth rates, but individual studies remain heterogeneous. Notably, the only double-blind randomized controlled trial conducted exclusively in frozen-thawed embryo transfer cycles found no improvement in live birth or clinical pregnancy rates with the enriched medium.
The new findings align with several recent negative studies, including a prospective randomized trial of 581 cycles that found no beneficial effect of hyaluronan enrichment on clinical pregnancy, implantation, or delivery rates, and retrospective analyses reporting no significant benefit. They contrast, however, with a systematic review and meta-analysis that reported increased clinical pregnancy and live birth rates with hyaluronan-enriched medium in autologous oocyte cycles, with estimated live birth rates of 32 percent in control groups and 39 percent in enriched groups, and with a retrospective historical-cohort study of 726 frozen transfer cycles in which clinical pregnancy rates were higher with the enriched medium. The authors noted that the high baseline clinical pregnancy rate of 61.5 percent in their control frozen transfer cycles, achieved with standardized hormone replacement protocols, may make any incremental benefit of an adjunct statistically harder to demonstrate.
The study carries important limitations. Its retrospective, nonrandomized, unblinded design introduces the risk of selection bias, and residual confounding cannot be excluded despite adjustment for key covariates. Infertility diagnosis was not systematically recorded and could not be included in the regression model, and stage-specific embryo grading was not harmonized or entered into the model, leaving possible confounding by embryo quality. The single-center setting limits generalizability, and clinical pregnancy was the only endpoint analyzed, because miscarriage, ongoing pregnancy, and live birth were not recorded in a structured, retrievable form at this regional referral center. The study was powered to detect an absolute difference of 15 percentage points, so smaller but potentially clinically meaningful differences cannot be excluded, and the absence of a statistically significant difference should not be interpreted as proof that no clinically relevant effect exists. The authors concluded that their findings do not provide evidence supporting the routine use of EmbryoGlue in unselected frozen embryo transfer populations, and that prospective randomized trials with live birth as the primary endpoint remain warranted to determine whether specific patient subgroups might benefit.
Subject of Research: Clinical pregnancy outcomes of hyaluronan-enriched embryo transfer medium in predominantly frozen IVF cycles
Article Title: Hyaluronan-enriched transfer medium in routine clinical practice: a retrospective cohort study showing no improvement in pregnancy outcomes among predominantly frozen embryo transfer cycles
Article References: Coştur Filiz, P., & Filiz, S. (2026). Hyaluronan-enriched transfer medium in routine clinical practice: a retrospective cohort study showing no improvement in pregnancy outcomes among predominantly frozen embryo transfer cycles. Archives of Gynecology and Obstetrics, 313(1), Article 300. https://doi.org/10.1007/s00404-026-08592-z
Image Credits: AI Generated
DOI: 10.1007/s00404-026-08592-z
Keywords: EmbryoGlue, hyaluronan, embryo transfer medium, frozen embryo transfer, IVF, clinical pregnancy rate, implantation, retrospective cohort study, ART add-ons, multivariate logistic regression, infertility, assisted reproduction
News Source: Harold Sullivan. (October 29, 2026). EmbryoGlue shows no pregnancy benefit in real-world IVF cohort. Scienmag.
