Granulocyte-macrophage colony-stimulating factor-containing medium for blastocyst recovery culture: study protocol for a randomized controlled trial (GSET-study).

Study question Does the post-thaw use of a GM-CSF-containing medium improve blastocyst transfer outcomes in all blastocysts in a frozen-thawed embryo transfer cycle? Summary answer The use of a GM-CSF-containing medium at post-thawing especially improves the live birth (LB) rate of morphologically poor blastocysts in a frozen-thawed embryo transfer cycle. What is known already GM-CSF, a cytokine secreted by the epithelial cells of the female reproductive tract, plays an important role in embryonic development, implantation, and subsequent development in humans and animals. In humans, GM-CSF increases the blastocyst developmental rate and decreases the chances of miscarriage. Previously, we reported that the use of a GM-CSF-containing medium for blastocyst recovery culture after thawing improves the clinical pregnancy (CP) rate in a frozen-thawed blastocyst transfer cycle (ESHRE, 2019). However, it is unclear whether GM-CSF improves embryo transfer outcomes in all blastocysts. In addition, it is necessary to accumulate information regarding its effects on neonatal outcomes. Study design, size, duration We performed a retrospective observational study to compare two groups: a GM-CSF group (GM-CSF-containing medium; SAGE-1step GM-CSF, Cooper Surgical) and a control group (GM-CSF-free medium; ONE STEP Medium, NAKA Medical). We analyzed 566 blastocyst transfer cycles in patients aged 30–39 years who underwent frozen-thawed single embryo transfer at Takahashi Women’s Clinic (Japan) from February 2018 to February 2019. Chromosomal analysis was not performed. Participants/materials, setting, methods We used a control medium for blastocyst culture and a Cryotop safety kit for blastocyst vitrification. After thawing, we cultured blastocysts in a GM-CSF-containing medium or control medium for 3–5 h until transfer. Embryo transfer outcomes were compared. We performed the multivariate logistic regression analysis(MVRA) to adjust confounding bias. A subgroup analysis was also performed of morphological grade according to Gardner’s criteria (excellent: ≥AA, good: blastocysts containing B, poor: blastocysts containing C). Main results and the role of chance There were no difference in patient background between the two groups. The CP and LB rates in the GM-CSF group and control group were 54.3% vs. 42.6% and 42.9% vs. 31.1%. The MVRA adjusted by confounding factors(patient age, BMI, basal AMH, blastocyst grade, day of vitrification, number of previous failed ETs, and assisted hatching) demonstrated that CP (p = 0.0193; adjusted odds ratio [aOR], 1.55) and LB rate (p = 0.0080; aOR, 1.67) were significantly higher in GM-CSF group than that of control group. Moreover, the CP and LB rates of the GM-CSF group and control group were: excellent-blastocysts at 62.0% vs. 58.8% (p = 0.5955; OR, 1.14), 52.7% vs. 45.6% (p = 0.2466, aOR:1.33), good-blastocysts 52.1% vs. 37.6% (p = 0.0561; OR, 1.80), 38.0% vs. 26.6% (p = 0.1072; OR, 1.69), and poor-blastocysts 38.9% vs. 17.9% (p = 0.0115; OR, 2.92), 25.9% vs. 9.0% (p = 0.0164; OR, 3.56). A GM-CSF-containing medium significantly improved the CP and LB rates of poor-grade blastocysts. There were no significant differences between the GM-CSF group and control group in the male ratio (52.7％ vs. 51.0%, p = 0.8057), pregnancy duration (38.8±1.4 weeks vs. 38.5±1.8 weeks, p = 0.2558), cesarean section rate (38.2% vs. 40.8%, p = 0.6979), birth weight (3133±466g vs. 3037±437g, p = 0.1281), and congenital anomaly rate (0.91% vs. 2.04％, p = 0.6026). Limitations, reasons for caution This was a single-center, retrospective study. Chromosomal abnormalities in embryos were not considered; however, the LB rate among babies was analyzed. The basic chemical composition of the culture medium (salt concentration, glucose concentration, etc.) used in the control group was different from that of the GM-CSF-containing medium. Wider implications of the findings We found that the use of a GM-CSF-containing medium improved the clinical pregnancy and live birth rates of poor-grade blastocysts without affecting the babies. This may be an effective therapeutic strategy for some patients as it may allow for the effective use of poor-grade euploid blastocysts. Trial registration number not applicable

Day 2 embryo transfer (ET) and day 3 ET afford similar reproductive outcomes in the poor responder

Study question Does endometrial stimulation with culture medium containing granulocyte macrophage colony-stimulating factor (GM-CSF) promote implantation potential of vitrified-warmed blastocysts? Summary answer Endometrial stimulation with culture medium containing GM-CSF prior to transfer of vitrified-warmed blastocysts may promote embryo implantation and improve pregnancy continuation. What is known already Human embryonic development and implantation in vivo is precisely regulated by various cytokines, including growth factors. GM-CSF is a cytokine that plays an important role in reproductive function. We reported the possibility that the use of culture medium containing GM-CSF in cleavage stage vitrified-warmed embryo transfers (ETs) promotes embryo implantation (ESHRE2018). However, it has been reported that endometrial stimulation with culture supernatant up to blastocyst stage (SEET method) improves embryo implantation, but the effect of endometrial stimulation with culture medium containing GM-CSF (G-SEET method) on embryo implantation is not clear. Study design, size, duration This study was conducted at a single in vitro fertilization (IVF) center from June 2020 to February 2022. The G-SEET procedure was performed at each patient’s request. All study participants provided informed consent, and the study design was approved by the ethics committee of IVF Nagata Clinic, Fukuoka, Japan. Participants/materials, setting, methods We examined 702 cycles of vitrified-warmed blastocyst transfer. Warmed blastocysts were incubated in recovery culture for several hours and transferred on day 5 (control group). For the G-SEET method, only culture medium containing GM-CSF was injected into the uterus two days before ET, which was performed as in the control group (G-SEET group). The human chorionic gonadotropin (hCG) positive, clinical pregnancy, miscarriage, and ongoing pregnancy rates were compared between the two groups. Main results and the role of chance Of the 702 cycles of single vitrified-warmed blastocyst transfers, the control and G-SEET groups had 459 and 243 cycles, respectively. The G-SEET group had a higher patient age (37.5±4.2 vs. 36.2±3.8, p < 0.01), a thinner endometrial thickness (10.3±1.9 vs. 10.9±2.0, p < 0.01), a lower percentage of good blastocyst transfers (65.4% vs. 79.5%, p < 0.01), and a higher cumulative number of implantation failures (3.6±2.5 vs. 1.9±1.9, p < 0.01) than the control group. The G-SEET method was performed at the patient’s request, so the number of cycles and patient background varied. Multivariate analysis was performed using female age, male age, number of miscarriages, endometrial thickness, presence of good blastocysts, and cumulative number of implantation failures as confounding factors. Compared with the control group, the G-SEET group exhibited the following: hCG positivity rate [odds ratio (OR): 1.14, 95% confidence interval (CI): 0.79–1.64, p = 0.481], clinical pregnancy rate (OR: 1.81, 95% CI: 1.21–2.71, p = 0.003), miscarriage rate (OR: 0.735, 95% CI: 0.30–1.76, p = 0.491), and ongoing pregnancy rate (OR: 1.79, 95% CI: 1.20–2.69, p = 0.004). Clinical pregnancy and ongoing pregnancy rates were significantly higher in the G-SEET group compared with the control group. Limitations, reasons for caution The study was limited by the study size and lack of data about live birth rates after ET. In addition, intrauterine infusion of GM-CSF-free culture medium should be used as a control, but this was difficult in the private clinic setting. Wider implications of the findings This study suggests that endometrial stimulation with culture media containing GM-CSF prior to transfer of vitrified-warmed blastocysts promotes embryo implantation and enhances the ongoing pregnancy. The G-SEET method may be clinically useful as a new ET method. Trial registration number not applicable

To determine whether granulocyte-macrophage colony-stimulating factor (GM-CSF)-containing medium could improve embryo-transfer outcomes in frozen-thawed blastocyst transfer. Patients who underwent frozen-thawed blastocyst transfer (430 women, aged 30-39years, 566 cycles) were analyzed. Frozen-thawed blastocysts were cultured in GM-CSF-containing medium or control medium for 3-5h, followed by transfer to the uterus. The embryo-transfer outcomes in the two groups were measured and compared, and a propensity score matching (1:1) method was used to balance the differences in baseline characteristics. We analyzed 213 matched samples. In patients who underwent frozen-thawed blastocyst transfer with GM-CSF, the percentage of human chorionic gonadotropin-positive cases, biochemical pregnancies, clinical pregnancies, ongoing pregnancies, and live birth rates was 60.6%, 7.98%, 52.6%, 42.9%, and 40.9%, respectively, as compared with 45.1%, 3.29%, 41.8%, 31.1%, and 30.5%, respectively, for the control groups. The rates of human chorionic gonadotropin positivity (odds ratio [OR]: 1.87, 95% confidence interval: [CI]: 1.27-2.75), biochemical pregnancy (2.55, 1.04-6.29), clinical pregnancy (1.54, 1.05-2.27), ongoing pregnancy (1.64, 1.13-2.41), and live birth (1.67, 1.14-2.45) were significantly higher in the GM-CSF group than the control group. The incidence of pregnancy loss (22.3% vs. 27.0%) did not significantly differ between the groups. The use of a GM-CSF-containing medium for blastocyst-recovery culture improved the live birth rate as a result of increased implantation rate in the frozen-thawed blastocyst-transfer cycle. The use of GM-CSF-containing medium following blastocyst thawing could be an effective choice for improving the blastocyst-transfer outcomes.

Study question Does endometrial thickness (EMT) predict live birth (LB) after fresh and frozen-thawed embryo transfer (ET) and is there a lower EMT cut-off for ET? Summary answer Once intracavitary pathology and inadvertent progesterone exposure is excluded, EMT is not predictive for LB. EMT is not linearly associated with probability of LB. What is known already EMT is commonly used as a marker of endometrial receptivity and in turn, assisted reproductive technology treatment success. ET is often cancelled or postponed if EMT is below an arbitrary cut-off. However, the available evidence on the relationship between EMT and LB rates is conflicting and too dubious to hold such strong stance. An overwhelming majority of the studies on the subject are retrospective, they use different arbitrary cut off values ranging between 6 to 9 mm with heterogeneous stimulation and transfer protocols. Study design, size, duration Records of all women who underwent fresh or frozen-thawed ET in Koc University Hospital Assisted Reproduction Unit between October 2016 - August 2019 were retrospectively screened. All women who underwent fresh or frozen-thawed blastocyst transfer during the study period were included. Every woman contributed to the study with only one transfer cycle for each category, i.e., fresh ET and frozen-thawed ET. Participants/materials, setting, methods After ruling out endometrial pathology, EMT was measured on the day of ovulation trigger for fresh ET cycles, and on the day of progesterone commencement for frozen-thawed ET. ET was carried out, regardless of EMT, if there was no suspicion of inadvertent progesterone exposure, i.e., due to follicular phase progesterone elevation in fresh or premature ovulation in frozen ET cycles. Main results and the role of chance 560 ET cycles, 273 fresh and 287 frozen-thawed, were analyzed. EMT varied from 4mm to 18mm. EMT were similar between women who achieved a LB and who did not after fresh ET [10.5 (9.2 – 12.2) mm and 9 (8 – 11) mm, respectively, p = 0.11]. Ovarian stimulation characteristics and proportion of women who received a single embryo were similar (69% vs 68.3%, respectively, p = 0.91). Women who achieved a LB was significantly younger than those who did not [35 (32–38) and 37 (33–41), respectively, p < 0.01]. Women who had a LB and who did not after frozen-thawed ET had similar EMT of 8.4 (7.4 – 9.7) mm and 9 (8 – 10) mm, respectively (p = 0.38). Women who achieved a LB were significantly younger than those who did not [32 (29–35) vs 34 (30–38) years, p = 0.04]. The proportion of women who received a single ET was similar between women who achieved a LB and who did not after a FET [86/95 (90.5%) vs 181/192 (94.3%), respectively, p = 0.26]. Area under curve values of EMT for predicting LB in fresh, frozen-thawed and all ET were 0.56, 0.47 and 0.52, respectively. EMT and LB rate were not linearly correlated in fresh or frozen-thawed ET cycles. Limitations, reasons for caution Although our study is retrospective, no women was denied ET due to EMT in our center. Only patients undergoing ET were included in the analysis, which may introduce bias due to the selection of couples who were competent enough to produce at least one blastocyst fit for transfer. Wider implications of the findings: Since women with thin endometrium had reasonable chance for LB even in the absence of a cut-off for EMT in this unique dataset, delaying or denying ET for any given EMT value alone does not seem justified. Further studies in which ET is carried out regardless of EMT are needed. Trial registration number Not applicable

P-208: Results comparison between frozen-thawed blastocyst transfer using ultra-rapid vitrification method with a cryoloop and fresh blastocyst transfer

Study question Are cycle outcomes different in frozen embryo transfer in obese and overweight women compared to normal weight cases? Summary answer As BMI increases, although implantation rates are similar, miscarriage rates increase and live birth rates decrease, especially in cases whose BMI value is above 30. What is known already Obesity has adverse effects on the reproductive system. Obesity causes ovulatory dysfunction and menstrual cycle disorders, thus reducing fertility. As BMI increases, the implantation, pregnancy and ongoing pregnancy rates decrease and the rate of clinical losses increases. In donor oocyte cycles, the BMI of the recipient has a statistically significant detrimental effect on obstetrics outcomes such as pregnancy rate and live birth rate. However, very few studies evaluate the impact of BMI on frozen-thawed single blastocyst transfer. Study design, size, duration This retrospective study was conducted at Istanbul Memorial Hospital, ART and Reproductive Center between 2011 and 2020. A total of 5642 frozen-thawed single blastocyst transfer cycles were examined. Patients were grouped according to the World Health Organisation BMI classification system: Group I (BMI 25–29.9) (n = 1663); Group II (BMI 30–34.9) (n = 598); Group III (BMI 35–39.9) (n = 150); Group IV (BMI>40) (n = 30); Control Group (BMI 18.5–24.9) (n = 3201). Participants/materials, setting, methods: Patients between 17–42 years old were included. Preimplantation genetic diagnosis (PGD) was performed for patients >37 years old. Exclusion criteria were: repeated pregnancy losses, Mullerian abnormalities, intrauterine adhesions, endometrial thickness <7mm during frozen embryo transfer (FET) cycle. For endometrial preparation, modified natural cycle or artificial cycle were used. Single top/good quality blastocysts were transferred. Main results and the role of chance A total of 5642 FET cycles were analyzed. There was no significant difference in patient characteristics in terms of mean age, endometrial thickness on embryo transfer day and Anti Mullerian Hormone levels between the groups. Cycle outcomes were analysed according to the BMI groups. Mean age of groups were 32.1(17–42), 32(18–42), 32.6(20–42), 32.8(23–42) in groups I to IV respectively and 32(18–42) in the control group. Pregnancy rates were 70.9%(n = 1180) 70.7%(n = 423), 76%(n = 114) and 54.8%(n = 17) in groups I to IV respectively and 72.4% in the control group(n = 2321) (p > 0,05). Clinical pregnancy losses rates were 18.9%(n = 196), 23.9%(n = 86), 23.1%(n = 22) and 23%(n = 3) in groups I to IV respectively and 15.1% in the control group(n = 316) (p < 0,05). The live birth rates were 49.9%(n = 830) 45.1%(n = 270), 47.3%(n = 71) and 33.3%(n = 10) in groups I to IV, respectively and 54.9% (n = 1759) in the control group (p < 0,05). There was no statistically significant difference in implantation rates between the groups but clinical pregnancy losses rates were higher in obese patients (groups II-III-IV) whereas live birth rates were lower compared to group I (overweight) and the control group. Limitations, reasons for caution The limitation of the study is its retrospective nature. Wider implications of the findings: Our study shows that, there is a significantly higher risk of negative cycle outcomes in obese patients. Pre-treatment counselling is therefore needed to increase patient awareness of the risks and to provide advice on weight loss. Trial registration number Not applicable

Study question To determine the appropriate timing for starting low-dose aspirin (LDA) among women undergoing frozen-thawed embryo transfer (ET). Summary answer Starting LDA prior to embryo implantation may decrease clinical pregnancy rates after frozen-thawed blastocyst transfer. What is known already Anticoagulant therapy including LDA is frequently performed during ET in IVF treatment. Previous meta-analysis including 17 studies, 16 of which evaluated pregnancy outcomes in fresh ET cycles, demonstrated that LDA slightly increased clinical pregnancy rate but the clinical effects of LDA on embryo implantation still remains controversial in frozen-thawed ET. Furthermore, discussions regarding the appropriate timing of aspirin initiation among women undergoing frozen-thawed ET have still been ongoing. Study design, size, duration A cross-sectional study was conducted on 885 infertile women who underwent thrombophilia screening between 2020 and 2023. We recruited first frozen-thawed blastocyst transfer cycles in 553 consecutive women aged <40 years. Participants/materials, setting, methods LDA was started on the day of ET from 2020 to 2021 in 79 women (day 0 group) and at 5 days after ET from 2021 to 2023 in 215 women (day 5 group). We also recruited 259 women who underwent first frozen-thawed blastocyst transfer without LDA treatment from 2020 to 2023 (control). We compared pregnancy outcomes after frozen-thawed ET between the three groups. Main results and the role of chance Clinical pregnancy and livebirth rates after frozen-thawed ET in the day 0 group were significantly lower than those in the other two groups (clinical pregnancy rates: 57.5%, 40.5%, and 61.4%, p = 0.005 and livebirth rates: 48.6%, 34.2%, and 54.0%, p = 0.01 in the control, day 0, and day 5 groups, respectively). Multivariable logistic regression analysis showed that livebirth rate in the day 0 group was significantly lower than those in the other groups (odds ratio [OR]: 0.54, 95% confidential interval [CI]: 0.31 − 0.95); however, no significant difference in livebirth rates was found between the day 5 and control groups (OR: 1.13, 95%CI: 0.70 − 1.80). Limitations, reasons for caution We did not compare pregnancy outcomes between women with thrombophilia who did and did not receive LDA. We changed the timing of LDA initiation due to low pregnancy rates in the day 0 group; therefore, the number of patients in the day 0 group was insufficient owing to study feasibility. Wider implications of the findings Although LDA may suppress an overactive proinflammatory response after ovarian stimulation in fresh ET cycles, our findings showed that aspirin did not improve pregnancy outcomes in frozen-thawed ET cycles with physiological hormone condition. As such, unnecessary LDA use should be avoided during IVF treatment. Trial registration number No

Twelve of 15 studies contributed data that could be included in meta-analyses. The quality of the evidence ranged from moderate to very low. Only three of the 15 studies reported data for live birth, although the data for ongoing pregnancy and clinical pregnancy are consistent with the live birth data, suggesting no difference between Day three and Day two embryo transfer for these outcomes. There was no evidence of a difference identified between Day three and Day two embryo transfer for multiple pregnancy, miscarriage or ectopic pregnancy per woman randomised. No data were reported for complication rate, fetal abnormality or woman's evaluation of the procedure. The current evidence has not identified any evidence of differences in pregnancy outcomes between Day two and Day three embryo transfers. Any further studies comparing these timings of embryo transfer are unlikely to alter the findings and we suggest that this review no longer be updated.

Does female obesity affect live birth rate after frozen-thawed blastocyst transfer? Live birth rate was not statistically different between obese and normal weight patients after frozen-thawed blastocyst transfer (FBT). Obesity is a major health problem across the world, especially in women of reproductive age. It impacts both spontaneous fertility and clinical outcomes after assisted reproductive technology. However, the respective impact of female obesity on oocyte quality and endometrial receptivity remains unclear. While several studies showed that live birth rate was decreased in obese women after fresh embryo transfer in IVF cycle, only two studies have evaluated the effects of female body mass index (BMI) on pregnancy outcomes after frozen-thawed blastocyst transfer (FBT), reporting conflicting data. This retrospective case control study was conducted in all consecutive frozen-thawed autologous blastocyst transfer (FBT) cycles conducted between 2012 and 2017 in a single university-based centre. A total of 1415 FBT cycles performed in normal weight women (BMI = 18.5-24.9kg/m2) and 252 FBT cycles performed in obese women (BMI ≥ 30kg/m2) were included in the analysis. Endometrial preparation was standard and based on hormonal replacement therapy. One or two blastocysts were transferred according to couple's history and embryo quality. Female and male age, smoking status, basal AMH level and type of infertility were comparable in obese and normal weight groups. Concerning FBT cycles, the duration of hormonal treatment, the stage and number of embryos (84% single blastocyst transfer and 16% double blastocysts transfer) used for transfer were comparable between both groups. Mean endometrium thickness was significantly higher in obese than in normal weight group (8.7 ± 1.8 vs 8.1 ± 1.6mm, P< 0.0001). Concerning FBT cycle outcomes, implantation rate, clinical pregnancy rate and live birth rate were comparable in obese and in normal weight groups. Odds ratio (OR) demonstrated no association between live birth rate after FBT and female BMI (OR = 0.92, CI 0.61-1.38, P= 0.68). Anthropometric parameters such as hip to waist ratio were not used. Polycystic ovarian syndrome status was not included in the analysis. Our study showed that live birth rate after frozen-thawed blastocyst transfer was not statistically different in obese and in normal-weight women. Although this needs confirmation, this suggests that the impairment of uterine receptivity observed in obese women after fresh embryo transfer might be associated with ovarian stimulation and its hormonal perturbations rather than with oocyte/embryo quality. No external funding was received. There are no competing interests. N/A.

ObjectiveTo study the earliest timepoint at which serum human chorionic gonadotropin (hCG) can predict live birth after blastocyst frozen embryo transfer, and to determine the cutoff values for hCG levels on days 7–11 after frozen-thawed embryo transfer with blastocysts.DesignSingle-center retrospective cohort study.SubjectsWe investigated 18,491 cycles of frozen-thawed blastocyst transfers at our hospital between January 2016 and December 2022.ExposureNot applicable.Main Outcome MeasuresThe hCG cutoff values for live birth on days 7–11 after frozen-thawed blastocyst transfer.ResultsReceiver operating characteristic analysis between serum hCG levels and live birth after frozen-thawed blastocyst transfer showed area under the curve values of 0.962, 0.958, 0.954, 0.957, and 0.954, for days 7, 8, 9, 10, and 11, respectively. Serum hCG cutoff values from the 7th to 11th days after frozen-thawed blastocyst embryo transfer were 31.05 mIU/mL, 46.6 mIU/mL, 72.05 mIU/mL, 95.4 mIU/mL, and 159.9 mIU/mL, respectively.ConclusionWe determined the hCG cutoff values for live births after frozen-thawed blastocyst transfer. We suggest that it is possible to predict pregnancy outcomes as early as the 7th or 8th day after frozen-thawed blastocyst transfer.

To study whether serum estradiol (E2) levels prior to progesterone administration in the artificial endometrial preparation (AEP) of frozen-thawed blastocyst transfer affect the live birth rate. Retrospective cohort study. Tertiary-care academic medical centre. A total of 3857 frozen-thawed blastocyst transfer cycles were divided into three groups: <200 pg/ml (n = 1676); 200-399 pg/ml (n = 1296); and ≥400 pg/ml (n = 885), based on the 25th (182.3 pg/ml) and 75th percentile (390.2 pg/ml) of serum E2 level prior to progesterone administration. Univariable and multivariable logistic regression analysis was performed. The primary outcome of the study was the live birth rate and the secondary outcomes included clinical pregnancy rate, pregnancy loss rate, neonatal birthweight, Z-score, and small for gestational age (SGA). Compared with the reference group, accounting for major covariates, the live birth rate significantly decreased in the '≥400 pg/ml' group (adjusted OR 0.71, 95% CI 0.59-0.85). Compared with the reference group, there was an association between the E2 level in the '≥400 pg/ml' group and a decrease in the clinical pregnancy rate (adjusted OR 0.74, 95% CI 0.61-0.89). Compared with the reference group, the pregnancy loss rate significantly increased in the '≥400 pg/ml' group (adjusted OR 1.45, 95% CI 1.08-1.93). The E2 levels did not affect neonatal birthweight, Z-score, and SGA among singletons. High serum E2 levels prior to progesterone administration in AEP are associated with a decreased live birth rate after frozen-thawed blastocyst transfer. High serum E2 levels prior to progesterone administration in artificial FET are associated with a decreased live birth rate after frozen-thawed blastocyst transfer.

GM-CSF (granulocyte macrophage colony-stimulating factor) is a growth factor that is used to supplement culture media in an effort to improve clinical outcomes for those undergoing assisted reproduction. It is worth noting that the use of GM-CSF-supplemented culture media often adds a further cost to the price of an in vitro fertilisation (IVF) cycle. The purpose of this review was to assess the available evidence from randomised controlled trials (RCTs) on the effectiveness and safety of GM-CSF-supplemented culture media. To assess the effectiveness and safety of GM-CSF-supplemented human embryo culture media versus culture media not supplemented with GM-CSF, in women or couples undergoing assisted reproduction. We used standard methodology recommended by Cochrane. We searched the Cochrane Gynaecology and Fertility Group Trials Register, CENTRAL, MEDLINE, Embase, CINAHL, LILACS, DARE, OpenGrey, PubMed, Google Scholar, and two trials registers on 15 October 2019, checked references of relevant papers and communicated with experts in the field. We included RCTs comparing GM-CSF (including G-CSF (granulocyte colony-stimulating factor))-supplemented embryo culture media versus any other non-GM-CSF-supplemented embryo culture media (control) in women undergoing assisted reproduction. We used standard methodological procedures recommended by Cochrane. The primary review outcomes were live birth and miscarriage rate. The secondary outcomes were clinical pregnancy, multiple gestation, preterm birth, birth defects, aneuploidy, and stillbirth rates. We assessed the quality of the evidence using GRADE methodology. We undertook one comparison, GM-CSF-supplemented culture media versus culture media not supplemented with GM-CSF, for those undergoing assisted reproduction. We included five studies, the data for three of which (1532 participants) were meta-analysed. We are uncertain whether GM-CSF-supplemented culture media makes any difference to the live-birth rate when compared to using conventional culture media not supplemented with GM-CSF (odds ratio (OR) 1.19, 95% confidence interval (CI) 0.93 to 1.52, 2 RCTs, N = 1432, I2 = 69%, low-quality evidence). The evidence suggests that if the rate of live birth associated with conventional culture media not supplemented with GM-CSF was 22%, the rate with the use of GM-CSF-supplemented culture media would be between 21% and 30%. We are uncertain whether GM-CSF-supplemented culture media makes any difference to the miscarriage rate when compared to using conventional culture media not supplemented with GM-CSF (OR 0.75, 95% CI 0.41 to 1.36, 2 RCTs, N = 1432, I2 = 0%, low-quality evidence). This evidence suggests that if the miscarriage rate associated with conventional culture media not supplemented with GM-CSF was 4%, the rate with the use of GM-CSF-supplemented culture media would be between 2% and 5%. Furthermore, we are uncertain whether GM-CSF-supplemented culture media makes any difference to the following outcomes: clinical pregnancy (OR 1.16, 95% CI 0.93 to 1.45, 3 RCTs, N = 1532 women, I2 = 67%, low-quality evidence); multiple gestation (OR 1.24, 95% CI 0.73 to 2.10, 2 RCTs, N = 1432, I2 = 35%, very low-quality evidence); preterm birth (OR 1.20, 95% CI 0.70 to 2.04, 2 RCTs, N = 1432, I2 = 76%, very low-quality evidence); birth defects (OR 1.33, 95% CI 0.59 to 3.01, I2 = 0%, 2 RCTs, N = 1432, low-quality evidence); and aneuploidy (OR 0.34, 95% CI 0.03 to 3.26, I2 = 0%, 2 RCTs, N = 1432, low-quality evidence). We were unable to undertake analysis of stillbirth, as there were no events in either arm of the two studies that assessed this outcome. Due to the very low to low quality of the evidence, we cannot be certain whether GM-CSF is any more or less effective than culture media not supplemented with GM-CSF for clinical outcomes that reflect effectiveness and safety. It is important that independent information on the available evidence is made accessible to those considering using GM-CSF-supplemented culture media. The claims from marketing information that GM-CSF has a positive effect on pregnancy rates are not supported by the available evidence presented here; further well-designed, properly powered RCTs are needed to lend certainty to the evidence.

How does the duration of estrogen (E2) treatment prior to frozen-blastocyst transfers affect the live birth rate (LBR)? Prolonged E2 exposure as part of artificial endometrial preparation (AEP) significantly decreases the LBR after autologous frozen-thawed blastocyst transfer. One effective method for endometrial preparation prior to frozen embryo transfer is AEP, a sequential regimen with E2 and progesterone, which aims to mimic the endocrine exposure of the endometrium in a normal cycle. Nevertheless, the optimal duration of E2 administration prior to transfer remains unknown. An observational cohort study was conducted in a tertiary care university hospital between 01/07/2012 and 31/12/2015. The main inclusion criteria was having a single frozen-thawed blastocyst transfer with an AEP using exogenous E2. A total of 1377 frozen-thawed blastocyst transfers were assigned to four groups according to the duration of the E2 administration prior to the embryo transfers. These comprised a '≤21 days' group (n = 330), a '22-28 days' group (n = 665), a '29-35 days' group (n = 289) and a '36-48 days' group (n = 93). The '≤21 days' group' was taken as the reference group. The main measured outcome was the LBR following frozen-thawed blastocyst transfers. Statistical analysis was conducted using univariate and multivariate logistic regression models. LBR significantly decreased when the E2 exposure prior to the frozen-thawed blastocyst transfer exceeded 28 days: OR = 0.66; 95% CI [0.46-0.95]; P = 0.026 and OR = 0.49 [0.27-0.89]; P = 0.018, respectively, for the '29 to 35 days' group and for the '36 to 48 days' group compared to the reference group. Early pregnancy loss rates significantly increased when the E2 exposure lasted more than 35 days prior to the frozen-thawed blastocyst transfer (OR = 2.37 [1.12-5.05]; P = 0.025 vs. the reference group). After multivariate logistic regression, E2 exposure lasting more than 28 days prior to the frozen-thawed blastocyst transfer was associated with a decrease in the LBR, for the '29-35 days' group (OR = 0.65; [0.45-0.95]; P = 0.044) as for the '36-48 days' group (OR = 0.49; [0.26-0.92]; P = 0.035), vs. the reference group. One limitation is linked to the observational design of this study. In order to give patients the best chance to obtain a live birth after frozen-thawed blastocyst transfer, the length of E2 exposure prior to the frozen-blastocyst transfer should not exceed 28 days. This study provides new insight in regard to endometrial preparation using AEP prior to frozen-blastocyst transfer. No funding and no competing interest.

Noninvasive preimplantation genetic testing for aneuploidy in spent culture medium as a substitute for trophectoderm biopsy