Fatty acid and melatonin-enriched warming: A novel approach using vitrified oocytes and early-stage embryos for patients with poor prognosis.

The transition from slow freezing to vitrification has enabled the widespread adoption of frozen–thawed embryo transfer (FET) and oocyte cryopreservation [ 1 , 2 ]. FET improves clinical outcomes, reduces the risk of ovarian hyperstimulation syndrome, and enables the use of preimplantation genetic testing [ 3 , 4 ]. Moreover, the 2020 guideline of the European Society of Human Reproduction and Embryology (ESHRE) recognised oocyte vitrification for fertility preservation as a non-experimental procedure, underscoring its clinical significance [ 5 ]. Despite these advancements, the vitrification-warming process can compromise oocyte and embryo development by disrupting cytoskeletal integrity, altering lipid composition, and inducing oxidative stress [ 6 – 12 ]. These adverse effects are partly mediated by cryoprotectants such as dimethyl sulfoxide (DMSO). Biophysical studies have shown that DMSO interacts with lipid bilayers and alters membrane organisation, leading to increased membrane disorder and transient changes in permeability [ 13 – 14 ]. Consistent with this, cryobiological studies have associated DMSO exposure with altered membrane permeability and downstream developmental outcomes [ 15 – 17 ]. Such transient membrane alterations may provide a limited opportunity for solution-derived molecules to access intracellular compartments during the brief warming phase. Given this potential, melatonin (MEL) and fatty acids (FA) have been studied for their distinct mechanisms in protecting gametes and embryos during cryopreservation. MEL, which is present in follicular and tubal fluids, exerts antioxidant and anti-apoptotic effects during gamete maturation and early embryonic development [ 18 ]. In most studies, its use has been examined in culture media [ 19 – 22 ]; however, recent evidence suggests that supplementation during warming may reduce oxidative stress, preserve mitochondrial function, maintain oolemma permeability, and ultimately improve embryo quality and clinical outcomes [ 23 – 24 ]. Complementing the antioxidant action of MEL, FA facilitates mitochondrial β-oxidation, promoting adenosine triphosphate (ATP) production and cellular homeostasis during cryopreservation [ 25 – 27 ]. During preimplantation development, embryos undergo metabolic reprogramming toward oxidative phosphorylation, with lipid droplet remodelling and unsaturated FA biosynthesis enhancing membrane fluidity and polarity establishment [ 28 – 33 ]. Recent studies have shown that FA supplementation during warming improves blastocyst quality and clinical outcomes by restoring lipid droplet content, enhancing β-oxidation, and reducing apoptosis [ 34 – 37 ]. Although modifications to warming solutions using MEL or FA individually have been examined in human oocytes and embryos [ 23 , 24 , 34 – 37 ], their combined co-supplementation during warming has not been investigated. This combined approach may be particularly beneficial for women of advanced maternal age, who are vulnerable to cryo-induced damage from mitochondrial dysfunction and oxidative stress [ 38 , 39 ]. In addition to age-related susceptibility, thawed oocytes and early-stage embryos exhibit greater vulnerability as a result of their limited DNA repair capacity, low mitochondrial reserves, and lack of protective structures [ 40 , 41 ]. Given these biological vulnerabilities, we hypothesised that targeted modification of the warming solution could improve embryological and clinical outcomes in women of advanced maternal age with compromised developmental potential. As the combined use of FA and MEL during warming has not previously been investigated in this setting, we evaluated its impact on post-thaw survival, embryo development, and clinical pregnancy outcomes.

In the thawed embryo cohort, the control group comprised 117 patients and 338 embryos, whereas the FA + MEL group included 111 patients and 321 embryos. Both groups comprised women of advanced maternal age, with mean ages of 41.0 ± 0.32 and 41.7 ± 0.32 years in the control and FA + MEL groups, respectively. The mean anti-Müllerian hormone (AMH) level was equally low in both groups (1.6 ng/mL). All patients had undergone multiple IVF and ET cycles. No significant between-group differences were observed in baseline variables ( Table 1 ). BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; n, number; HRT, hormone replacement therapy; DOR, diminished ovarian reserve; PCOS, polycystic ovarian syndrome; IVF, in vitro fertilisation; SEM, standard error of the mean; FA, fatty acids; MEL, melatonin. Mixed factor refers to patients with ≥2 infertility factors (for example, male + uterine, or DOR + unexplained). Values are presented as the mean ± SEM. In the thawed oocyte cohort, the control group included 101 patients with 794 oocytes, and the FA + MEL group included 109 patients with 752 oocytes. The mean ages at the time of oocyte vitrification were 39.5 ± 0.42 and 38.8 ± 0.44 years in the control and FA + MEL groups, respectively. Other baseline characteristics, including BMI, AMH level, infertility duration, and history of prior IVF/ET cycles, were comparable between groups, with no significant differences ( Table 2 ). BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; n, number; DOR, diminished ovarian reserve; PCOS, polycystic ovarian syndrome; IVF, in vitro fertilisation; SEM, standard error of the mean; FA, fatty acids; MEL, melatonin. Mixed factor refers to patients with ≥2 infertility factors (for example, male + uterine, or DOR + unexplained). Values are presented as the mean ± SEM. Among these patients, ET following oocyte warming was performed in a subset. Baseline characteristics in this subgroup were generally comparable between groups, except for endometrial thickness, which was slightly lower in the FA + MEL group but remained within clinically acceptable limits. Multivariate regression analysis showed that endometrial thickness was not a significant predictor of implantation or clinical pregnancy in this subgroup ( S1 Table ). The baseline distribution of developmental stage (cleavage-stage vs. morula-stage) and morphological quality prior to warming was comparable between the control and FA + MEL groups ( P  > 0.05). Most embryos in both groups were cryopreserved at the cleavage stage. Following warming, the FA + MEL group demonstrated a significantly higher blastocyst formation rate than the control group, with blastocysts forming in 41.4% of embryos compared with 29.3% in controls ( P  = 0.024). The proportion of high-quality blastocysts was also significantly higher in the FA + MEL group than in the control group (51.1% vs. 27.3%; P  < 0.001). ICM morphological analysis revealed a significantly greater proportion of grade A embryos and fewer grade C embryos in the FA + MEL group than in the control group. TE grading showed a similar trend, with more embryos classified as grade B and fewer as grade C in the FA + MEL group. The FA + MEL group had a slightly higher proportion of grade A TE embryos; however, the difference was not statistically significant ( Table 3 ). ICM, inner cell mass; TE, trophectoderm; n, number; N, total number of embryos; FA, fatty acids; MEL, melatonin. n/N indicates the number of blastocysts or morphological grades per total embryos thawed or blastocysts formed. In total, 250 and 249 embryos were transferred in the control and FA + MEL groups, respectively, with an average of approximately 2 embryos per cycle in both groups. The distribution of embryo developmental stages at transfer was comparable between groups ( P  > 0.05), confirming that the clinical decision-making process for transfer timing was applied consistently across both cohorts. Notably, blastocyst-stage transfers were more frequent in the FA + MEL group; however, the difference was not statistically significant. The FA + MEL group showed a significantly higher implantation rate (26.1%) than the control group (14.4%) ( P  = 0.003). Similarly, the clinical pregnancy rate was significantly improved (51.4% vs. 29.1%; P  < 0.001). After adjusting for maternal age, both outcomes remained significant: the implantation rate increased by 13.2% in the FA + MEL group (95% confidence interval [CI]: 4.84–21.57; P  = 0.002), and the odds of clinical pregnancy were more than 3-fold higher (adjusted odds ratio [OR]: 3.13; 95% CI: 1.75–5.60; P  < 0.001) ( Table 4 ). CI, confidence interval; SEM, standard error of the mean; n, number; ET, embryo transfer; FA, fatty acids; MEL, melatonin. n/N values indicate ETs performed at each developmental stage per total transfer cycles; implantation events per embryos transferred; and clinical pregnancies per ET cycle. No adjustment was applied where no correlation with maternal age was found. Adjustment for maternal age was calculated using linear or logistic regression, depending on the outcome type. The post-warming oocyte survival rate was significantly higher in the FA + MEL group than in the control group (91.4% vs. 76.7%; P  < 0.001). Fertilisation and abnormal fertilisation rates did not differ significantly between groups; however, FA + MEL supplementation significantly improved embryo development following fertilisation. The proportion of embryos reaching the six-cell stage increased from 31.8% to 72.6% ( P  < 0.001), whereas the high-quality cleavage rate (24.1% vs. 51.0%; P  < 0.001) and blastocyst formation rate (10.3% vs. 24.6%; P  < 0.001) were significantly higher in the FA + MEL group than in the control group. The improvement in the high-quality blastocyst rate was more modest (25.0% vs. 33.9%; P  = 0.045); however, maternal age was identified as a confounding factor for this outcome. After adjustment for age, the FA + MEL group still demonstrated a significantly higher rate, with an estimated difference of 30.3% (95% CI: 5.42–55.16, P  = 0.018) ( Table 5 ). n, number; N, total number of fertilised oocytes; ICSI, intracytoplasmic sperm injection; CI, confidence interval; FA, fatty acids; MEL, melatonin. n/N indicates the number of embryos or blastocysts reaching a specified developmental stage per normally fertilised oocyte (2PN). Abnormal fertilisation was defined as 1PN (monopronuclear) or 3PN (tripronuclear) zygotes. No adjustment was applied where no correlation with maternal age was observed. Adjustment for maternal age was calculated using linear or logistic regression, depending on the outcome type. In a patient subset, meiotic spindle localisation was assessed using PolScope imaging prior to ICSI. Overall, 930 metaphase II oocytes were analysed: 374 and 556 from the control and FA + MEL groups, respectively. The proportion of oocytes with normal spindle localisation was significantly higher in the FA + MEL group than in the control group, accompanied by a corresponding decrease in abnormal localisation ( P  < 0.001 and P  = 0.049, respectively) ( Table 6 ). MII, metaphase II; n, number. Both groups had a mean of approximately two embryos transferred per cycle. However, the distribution of developmental stages at the time of transfer differed significantly between the groups. In the FA + MEL group, cleavage-stage transfers were significantly less frequent (48.7% vs. 83.9%, P  < 0.001), whereas morula-stage (25.6% vs. 6.5%) and blastocyst-stage (25.6% vs. 9.7%) transfers were numerically higher but did not reach statistical significance. The FA + MEL group also demonstrated significantly higher implantation (20.2% vs. 8.5%, P  = 0.016) and clinical pregnancy (38.5% vs. 16.1%, P  = 0.011) rates than the control group. Among these outcomes, only clinical pregnancy was significantly associated with maternal age. After adjustment, the FA + MEL group still exhibited a significantly higher clinical pregnancy rate, with an adjusted OR of 3.21 (95% CI: 1.17–8.82; P  = 0.024) ( Table 7 ). CI, confidence interval; SEM, standard error of the mean; n, number; ICSI, intracytoplasmic sperm injection; FA, fatty acids; MEL, melatonin; HRT, hormone replacement therapy. No adjustment was applied where no correlation with maternal age was found. Adjustment for maternal age was calculated using linear or logistic regression, depending on the outcome type.

Co-supplementation with FA and MEL during the warming of vitrified oocytes and early-stage embryos significantly improves post-thaw survival, spindle integrity, embryonic development, and clinical pregnancy outcomes in women of advanced maternal age with poor prognosis. These results suggest a synergistic effect that surpasses the benefit of either supplement alone, supporting the idea that targeted enhancement of membrane stability, redox balance, and mitochondrial metabolism can meaningfully improve developmental competence after warming. This simple, low-cost modification to existing warming protocols has substantial clinical relevance for patients with limited treatment options. Prospective studies incorporating mechanistic analyses are needed to validate these findings and further elucidate how FA and MEL support cryo-survival and early embryonic development.

This retrospective cohort study included women who underwent warming of vitrified oocytes or cleavage- and morula-stage embryos between August 2022 and December 2024 at CHA University Fertility Center, Gangnam. For research purposes, investigators accessed identifiable clinical records between 22/04/2025 and 31/05/2025; however, all data were fully de-identified immediately following the extraction process. The embryo-thawed cohort included women aged ≥35 years with recurrent poor embryo quality, defined as failure to achieve blastocyst cryopreservation in ≥2 prior in vitro fertilisation (IVF) cycles despite extended culture. The oocyte-thawed cohort comprised women aged ≥35 years at the time of oocyte vitrification. In both cohorts, women were excluded if they had endometriosis; obesity, defined as a body mass index (BMI) >25 kg/m² according to the 2022 guidelines of the Korean Society for the Study of Obesity [ 42 ]; autoimmune disease; or metabolic disorders such as hyperlipidaemia, hypertension, or diabetes mellitus. Each cohort was further classified based on whether FA and MEL (FA + MEL) supplementation was used during warming. The thawed embryo cohort comprised 228 women, with 659 thawed embryos (338 and 321 in the control and FA + MEL groups, respectively). In this cohort, vitrified embryos had originally been generated using either conventional IVF or intracytoplasmic sperm injection (ICSI), according to semen parameters. Specifically, ICSI was performed in all cases of confirmed male factor infertility, in accordance with standard clinical practice. All patients in this cohort proceeded to embryo transfer (ET), with 250 embryos transferred in the control group and 249 in the FA + MEL group. Cycles were excluded if the embryos were of poor quality at the time of vitrification, defined by the presence of >50% of blastomeres with large vacuoles (>14 μm), severe fragmentation (>35%), or multinucleation. Embryos with significantly delayed development—such as day 3 embryos with only 2–4 cells—were also excluded, as these factors could confound post-warming culture outcomes [ 43 ]. The thawed oocyte cohort included 210 women, with 794 and 752 oocytes thawed in the control and FA + MEL groups, respectively. To standardise fertilisation conditions, all surviving mature oocytes underwent fertilisation via ICSI. As a result of non-fertilisation (n = 11), poor embryo quality precluding transfer (n = 46), insufficient endometrial thickness (n = 26), or elective re-cryopreservation for embryo accumulation (n = 26), the final ET cohort comprised 62 patients in the control group and 39 in the FA + MEL group. The proportion of exclusion because of poor embryo quality was comparable between the groups (control: 23/609, 3.8% vs. FA + MEL: 23/686, 3.4%; P  = 0.79), indicating no imbalance in attrition attributable to early developmental arrest. These groups used 141 and 84 embryos, respectively, derived from the thawed oocytes ( Fig 1 ). FA, fatty acids; MEL, melatonin; FET, frozen-thawed embryo transfer; ET, embryo transfer. This study was approved by the Institutional Review Board of CHA University Fertility Center, Gangnam (IRB approval number: 2025-03-007). As this was a retrospective analysis of anonymised clinical data, the requirement for informed written or verbal consent was waived in accordance with institutional and national guidelines. Vitrification of mature oocytes and cleavage- or morula-stage embryos was performed using a two-step protocol based on ethylene glycol (EG; Sigma-Aldrich, St. Louis, MO, USA; Cat. No. 102466), DMSO (Sigma-Aldrich; Cat. No. D2650), and sucrose (Sigma-Aldrich; Cat. No. S1888), with SAGE Quinn’s Advantage Medium with HEPES (CooperSurgical, Trumbull, CT, USA) as the base medium. To ensure consistent quality, all vitrification solutions were freshly prepared under standardised laboratory conditions. All procedures were conducted at 25 °C for oocytes and at 37 °C for embryos. Equilibration was performed under strictly timed conditions in an equilibrium solution consisting of 7.5% EG and 7.5% DMSO in HEPES medium supplemented with 20% human serum albumin (HSA). Equilibration was carried out for 7.5 min for oocytes at room temperature and for 2.5 min for cleavage- and morula-stage embryos at 37 °C. Subsequently, the samples were transferred to a vitrification solution containing 15% EG, 15% DMSO, and 0.5 M sucrose for 1 min (oocytes) and 20 s (embryos). After equilibration, samples were loaded onto gold electron microscopy grids (Gilder, Westchester, PA, USA) and rapidly plunged into liquid nitrogen for vitrification. Vitrified oocytes and cleavage- or morula-stage embryos were warmed using a stepwise sucrose dilution protocol in HEPES-buffered medium supplemented with 20% HSA (CooperSurgical, Trumbull, CT, USA). For cleavage- and morula-stage embryos, a four-step dilution protocol was used (0.5 M, 0.25 M, 0.125 M, and 0.0 M sucrose), with 2.5 min of exposure at each step. For oocytes, a five-step protocol was applied (1.0 M, 0.5 M, 0.25 M, 0.125 M, and 0.0 M sucrose), also with 2.5 min of exposure at each step. In the FA+ MEL group, all warming solutions were supplemented with 1% FA (Gibco, Thermo Fisher Scientific, Waltham, MA, USA; Cat. No. 11905-031) and 10 μM MEL (Sigma-Aldrich, St. Louis, MO, USA; Cat. No. M5250). These supplemented warming solutions were freshly prepared under standardised laboratory conditions. Prior to clinical use, each batch underwent quality control, including verification of osmolarity and pH within predefined acceptable ranges. All solutions were prepared using sterile techniques, passed through 0.22 μm filters, and used within a specified time frame according to laboratory standard operating procedures. Following warming, mature oocytes underwent ICSI under standard conditions. Cleavage-and morula-stage embryos were cultured in SAGE Quinn’s Advantage Blastocyst Medium under reduced oxygen (5% O 2 , 6% CO 2 , 37 °C) using the oil-drop culture method in Oosafe four-well dishes. Cleavage-stage embryos were cultured for 2 additional days before transfer, whereas morula-stage embryos were cultured overnight. Post-warming survival was defined as the presence of 100% viable blastomeres ( Fig 2 ). (a) Mature oocytes were warmed in control medium or medium supplemented with 1% fatty acids and 10 μM MEL, followed by ICSI. A subset was assessed for meiotic spindle angle using PolScope before ICSI. Resulting embryos were cultured for 120–168 h to evaluate developmental progression. (b) Cleavage- and morula-stage embryos were warmed in control or FA+ MEL-supplemented medium and cultured for 24–48 h to assess further development. FA, fatty acids; MEL, melatonin; ICSI, intracytoplasmic sperm injection. All patients underwent controlled ovarian stimulation using either gonadotropin-releasing hormone (GnRH) agonist or antagonist protocols for pituitary suppression. Stimulation regimens were individualised based on patient age, ovarian reserve markers, and previous responses to gonadotropins. Ovulation was triggered with 250–500 μg recombinant human chorionic gonadotropin (Ovidrel®, Serono, Modugno, Italy) combined with 0.1–0.2 mg GnRH agonist (Decapeptyl®, Ferring, Sweden) when at least three follicles ≥17 mm or two follicles ≥18 mm in diameter were observed. Oocyte retrieval was performed 34–36 h after trigger administration. Endometrial preparation for FET, including ET using previously cryopreserved oocytes, was performed using one of the following protocols: (i) hormone replacement therapy (HRT) with oral oestradiol valerate 2 mg thrice daily (Progynova®, Merck, Poland), initiated on day 2 or 3 of the menstrual cycle; (ii) a modified natural cycle using letrozole 5 mg daily (Femara®, Novartis, Switzerland), also starting on day 2 or 3; or (iii) a true natural cycle. In HRT cycles, the timing of ET was determined by endometrial thickness and the initiation of luteal phase support. In modified and natural cycles, transfer timing was based on the estimated day of ovulation. The specific transfer day (Day 3, 4, or 5) was individualised through multidisciplinary consultation between clinicians and embryologists, integrating prior IVF cycle outcomes with embryological development. Given the cohort’s history of recurrent developmental arrest, a ‘rescue transfer’ strategy was adopted: embryos showing slow progression were transferred at the cleavage or morula stage, whereas those with robust potential were cultured to the blastocyst stage. Personalised luteal phase support was provided in all thawed ET cycles. Given the patients’ history of poor clinical outcomes, a maximum of two blastocysts or three cleavage- or morula-stage embryos were transferred following warming. Normal fertilisation was confirmed by the presence of two pronuclei and a second polar body 16–17 h after insemination, in accordance with the updated 2025 ESHRE/Alpha Scientists in Reproductive Medicine (ALPHA) Istanbul Consensus [ 44 ]. In oocyte-thawing cycles, the fertilisation rate was calculated as the proportion of surviving oocytes that underwent normal fertilisation. Meiotic spindle angle assessment was performed in a subgroup of the thawed oocyte cohort before ICSI using the PolScope spindle view system (Oosight TM META Imaging System; CRI, Woburn, MA, USA) in conjunction with glass-bottom culture dishes (MatTek, Ashland, MA, USA). The orientation of the meiotic spindle relative to the first polar body has been used as a non-invasive marker of oocyte integrity and developmental competence [ 45 ]. In this study, oocytes with meiotic spindle angles 30° or not visible were classified as abnormal. Embryo quality was assessed using established morphological criteria. To minimise subjective bias, two senior embryologists (each with >15 years of experience) performed all evaluations following standardised laboratory protocols based on the Istanbul Consensus. Regular inter-observer calibration was conducted to ensure scoring consistency. For cleavage-stage embryos, assessment included cell number, degree of fragmentation, and blastomere symmetry; embryos with ≥6 cells and <20% fragmentation were classified as good quality [ 46 , 47 ]. Morula-stage embryos were evaluated based on the degree and completeness of compaction, developmental timing, and absence of abnormalities. Good-quality morula-stage embryos were defined as those exhibiting full or near-complete compaction involving most of the embryo volume, with minimal fragmentation and appropriate developmental timing [ 44 ]. Blastocyst-stage embryos were assessed using the Gardner and Schoolcraft grading system [ 48 , 49 ], which evaluates the degree of blastocoel expansion, inner cell mass (ICM), and trophectoderm (TE) morphology. Blastocysts graded AA, AB, BA, or BB were classified as morphologically good quality [ 47 , 50 ]. Clinical outcomes included implantation and clinical pregnancy rates. Implantation was confirmed by the presence of a gestational sac on transvaginal ultrasonography at 5–6 weeks of gestation. The implantation rate was calculated as the number of gestational sacs observed divided by the number of embryos transferred. Clinical pregnancy was defined as the presence of foetal cardiac activity confirmed by transvaginal ultrasonography at 7–8 weeks of gestation, and the clinical pregnancy rate was calculated per ET cycle. Data distribution normality was assessed using the Shapiro–Wilk test, which revealed non-normal distributions; therefore, non-parametric tests were applied. Continuous variables are expressed as the mean ± standard error of the mean and were compared using the Mann–Whitney U test. Categorical variables are presented as percentages and compared using the Chi-square test. Univariate and multivariate regression analyses identified age as the only significant covariate. Accordingly, clinical outcome comparisons between groups were adjusted for age, using linear and multivariate logistic regression for continuous and categorical variables, respectively. Because embryo stage at transfer in the thawed-oocyte cohort represents post-warming developmental progression, it was not included as a baseline covariate in the primary models to avoid over-adjustment of a variable lying on the causal pathway. All statistical analyses were conducted using IBM SPSS Statistics version 29.0 (IBM Corp., Armonk, NY, USA). A P -value < 0.05 was considered statistically significant.

BMI, body mass index; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; n, number; DOR, diminished ovarian reserve; PCOS, polycystic ovarian syndrome; SEM, standard error of the mean; FA, fatty acids; MEL, melatonin. Mixed factor refers to patients with ≥2 infertility factors (for example, male + uterine, or DOR + unexplained). Values are presented as the mean ± SEM. (DOCX) ICM, inner cell mass; TE, trophectoderm; n, number; N, total number of blastocysts assessed; FA, fatty acids; MEL, melatonin. n/N indicates the number of blastocysts assigned to each morphological grade per total number of blastocysts evaluated. (DOCX)