An Approach to Improve Endometrial Receptivity: Is It Beneficial to Flush The Uterine Cavity with Follicular Fluid and Granulosa Cells? A Phase III Randomised Clinical Trial.

One of the significant obstacles to infertility treatment in assisted reproduction technology (ART), is the reduction of the embryo implantation rate ( 1 ). A receptive endometrium is one arm of embryo implantation. The prevalence of abnormal endometrial receptivity is not indicated directly in the literature; however, it is considered to contribute to two-thirds of cases afflicted with implantation failure ( 2 ). Endometrial receptivity is a complicated process that can be influenced by a variety of conditions. Polyps, adenomyosis, fibroids, and chronic endometritis are frequent gynaecologic diseases that might affect endometrial receptivity as well as the effectiveness of ART ( 3 ). Various treatments have been proposed to promote endometrial receptivity in patients with these pathological conditions ( 4 ). Endometriosis patients, for instance, may benefit from laparoscopic ablation or resection of endometrial implants ( 5 ) and those who suffer from leiomyoma and hydrosalpinges may benefit from myomectomy and salpingectomy, respectively ( 6 ). However, the rate of implantation in ovulation stimulation cycles remains low even for women who do not have these disorders. Approaches in the different studies that have been used to boost the rate of implantation in these patients include endometrial scratching, endometrial receptivity array (ERA) to pinpoint the precise time of endometrial receptivity, platelet-enriched plasma therapy as a potential management for thin endometria, and endometrial flushing with follicular fluid (FF) ( 1 , 7 , 8 ). The application of a low-complication strategy to enhance endometrial thickness and receptivity reduces the need to postpone embryo transfer cycles, improves fertility, and reduces the burden of treatment for patients ( 9 ). During the physiological processes of ovulation, FF enters the fallopian tubes together with the oocytegranulosa cells ( 10 ). The abundance of growth factors and cytokines in the FF of mature oocytes affects embryo implantation in a paracrine and autocrine manner. FF can alter the composition of fallopian tube secretions, alter the frequency of ciliary beats, and promote the in vitro growth of endometrial cells. Therefore, fallopian and uterine tube tissues are affected locally by a variety of substances present in the FF ( 11 , 12 ). Blood plasma crosses the blood-follicular barrier and the dynamic secretion of granulosa and theca cells creates FF, which serves as a crucial milieu for oocyte development ( 13 ). An important factor in determining oocyte quality, its capacity for fertilisation, and the potential for embryonic development are biochemical characteristics of the FF ( 14 ). FF surrounds the granulosa-oocyte complex and mediates the communication between cells within the follicle. On the other hand, the FF of mature oocytes contains growth factors and cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor (TGF), epidermal growth factor, and insulin-like growth factor (IGF), which are crucial to natural fertilisation success and may have paracrine or autocrine effects on embryo implantation ( 15 - 17 ). TGF regulates apoptosis and improves implantation, VEGF induces angiogenesis, and IGF promotes endometrial development ( 18 ). FF, on the other hand, has an immunosuppressive effect by producing interleukin-1 (IL-1) and IL-2 inhibitors and decreasing IL-2 receptor α (CD25) expression, which facilitates embryo implantation in the uterine cavity ( 19 ). The rationale behind research of flushing the uterine with FF accompanied by granulosa cells is based on the properties of these components and their potential impact on fertility ( 20 , 21 ). FF is discharged into the fallopian tube during the physiological events of ovulation in the menstrual cycle and enters the uterine cavity ( 22 ). Granulosa cells that surround the oocytes regulate follicular development, maturation, and atresia. These cells can be isolated from the FF of women undergoing intracytoplasmic sperm injection (ICSI) treatment. In vitro, GCs can produce progesterone and are viable for several days ( 23 ). By combining these components in a uterine flush, researchers hope to create a more favourable environment for implantation and pregnancy. However, more research is needed to fully understand the potential benefits and mechanisms of this approach. Therefore, this study aimed to evaluate the rate of implantation and clinical pregnancy following intrauterine flushing with FF plus granulosa cells in the ICSI cycle.

We accessed the medical records of 160 patients for potential participation in this study. From these, 20 patients did not meet the inclusion criteria: six females declined to participate, eight couples had severe sperm parameters at the time of OPU, and six women did not provide suitable, blood-free FF. Therefore, 140 women were randomly allocated to the control and intervention groups (n=70 per group). In both groups, after oocyte puncture, three women were excluded due to the risk of OHSS; therefore, we assessed 67 subjects from each group. In addition, four patients in the intervention group and two patients in the control group were not assessed for clinical outcomes because their embryos failed to develop and were not appropriate for transfer ( Fig .1 ). CONSORT diagram demonstrates study enrolment and participants. The Embryologist decided not to transfer the grade D embryos that were produced for these individuals. The quality of the oocytes or sperm cells is one of the potential explanations for the lack of embryo development during IVF. Table 1 lists the demographic and clinical characteristics of the two groups. There were no statistically significant differences between the intervention and control groups in terms of age, body mass index (BMI), duration of infertility, hormonal profile, number of IVF/ICS, and number of transferred embryos. The thickness of the endometrium is the most common marker for endometrial receptivity ( 2 ). Our assessment showed no significant difference in endometrial thickness according to vaginal sonography measurement on the day of the hCG injection between the control (10.16 ± 1.72 mm) and intervention (10.14 ± 1.53 mm) groups (P=0.937). Table 1 shows no significant difference in the clinical outcome of ART (implantation and clinical pregnancy per embryo transfer) between the two groups. However, the number of two pronuclei (2PN) was significantly higher in the study group compared to the control group (P=0.037). There was no statistically significant difference between the two groups in terms of pregnancyrelated complications which included ectopic pregnancy, blighted ovum, and abortion. Demographic and clinical information of the study patients Values are reported as mean ± standard deviation (SD) or numbers (percentage). P<0.05 indicates statistical significance. a ; Mann-Whitney U, b ; Student’s t test, c ; X 2 (Chi-square), d ; Fisher’s exact test, ET; Embryo transfer, AMH; Anti-Müllerian hormone, FSH; Follicle stimulating hormone, LH; Luteinizing hormone, hCG; Human chorionic gonadotropin, TE; Transferred embryo, BMI; Body mass index, MII; Metaphase II oocyte, MI; Metaphase I oocyte, GV; Germinal vesicle, PN; Pronuclei, No ET; Embryo transfer was not performed, and EP; Ectopic pregnancy. As shown in Table 2, univariate logistic regression analysis showed that only oestradiol levels had an association with clinical pregnancy. Univariate logistic regression analysis assessment of clinical pregnancy predictors *; P<0.05 is considered significant. BMI; Body mass index, AMH; Anti-Müllerian hormone, FSH; Follicle stimulating hormone, LH; Luteinizing hormone, hCG; Human chorionic gonadotropin, MII; Metaphase II oocyte, MI; Metaphase I oocyte, GV; Germinal vesicle, PN; Pronuclei, and ET; Embryo transfer.

Based on the current study, uterine cavity flushing with FF from mature follicles following oocyte retrieval had no effect, either positive or negative, on clinical pregnancy or implantation rates in women with moderate male factor infertility. The effects of FF on endometrial receptivity are multifaceted. While some studies have suggested that FF components can positively affect endometrial receptivity by modulating gene expression, cell proliferation, and differentiation ( 16 ), others have implicated FF in the development of endometriosis and ovarian cancer. There is some evidence that human FF can alter endometrial receptivity. In this context, Hashish et al. ( 20 ) published a randomised controlled trial (RCT) on 100 sub-fertile women who underwent ICSI. They demonstrated that injection of two millilitres of one mature oocyte FF into the uteri of sub-fertile women three days before ET (n=50) resulted in improved clinical pregnancy and implantation rates. However, this increase was not significantly different between the control (without uterine flushing) and intervention groups (50 women per group). They hypothesised that the absence of substantial changes between the two groups was due to the inadequate FF extracted from a mature oocyte and suggested that FF derived from several mature oocytes should be injected. However, in our study, the FF from 2-3 dominant follicles were pooled to eliminate the abovementioned limitation. Hamdi et al. ( 21 ) published an RCT, but they flushed the endometrium of 55 sub-fertile women who had been matched for age and other demographic characteristics using FF from 2 to 3 follicles as opposed to 55 sub-fertile women who were not flushed. They reported no significant differences between the intervention and control groups in terms of clinical or chemical pregnancy rates or implantation rates. The remarkable point of the them study was that the participants encountered a variety of aetiological causes of infertility, including fallopian tube obstruction, ovulation dysfunction, moderate endometriosis, low ovarian reserve, and moderate male dysfunction. In our study, we excluded female pathological conditions from the current investigation because some of the altered FF compositions (e.g., endometriosis patients) may not be safe and exert a negative impact on the uterine milieu ( 25 ). The compositions of FF that are based on biological functions include hormones, growth factors, cytokines, and immunological substances. There is a wide range of hormones in FF during different stages of folliculogenesis. These include gonadotropins (FSH, LH, hCG), prolactin, oestrogen, progesterone, androgens, corticoids, and growth hormones ( 26 , 27 ). In addition, human FF contains multiple different classes of growth factors. Activin and inhibin levels in FF have been proven to accurately predict the quality of oocytes and embryos on days two or three as well as the pregnancy rate ( 28 ). Inhibin and activin improve endometrial receptivity and embryo implantation ( 29 ). Other growth factors include EGF, TGF-α, leptin, AREG, BDNF, LIF, BMP-15, IGF-I, IGF-II, VEGF, and FGF ( 15 , 28 , 30 ). Proinflammatory cytokines are produced in the ovaries and FF before and during ovulation. For example, IL-1 β is synthesized locally from luteal granulosa cells. Fertilisation rates are associated with high cytokine levels. Other IL found in FF, including IL-2, IL-6, IL-10, and IL-12, have also been demonstrated to boost the embryo's implantation potential ( 31 , 32 ). FF is composed of T lymphocytes, B lymphocytes, dendritic cells, natural killer cells, and macrophages ( 33 , 34 ). The endometrium requires all of the mentioned factors to become receptive. Development of a receptive endometrium is accomplished by successive exposure to oestrogen, progesterone, immune cells, and expressions of VEGF ( 35 , 36 ). On the other hand, since FF contains a high concentration of potent mitogen factors for cell proliferation ( 37 ), several studies have focused on investigating its ability to up-regulate endometrial receptivity genes. The effect of FF on the expressions of key genes involved in endometrial receptivity with proven involvement in the process of implantation, including LIF, ITGAV, HOXA10-A11, and ITGB3 in endometrial stromal cells, was studied in vitro. Endometrial stromal cells exposed to FF showed increased expression of all of the examined genes. On the other hand, incubation with 20% FF for 72 hours showed no dose- or time-dependent cytotoxic effects on stromal cells of the endometrium ( 38 ). In contrast, FF is a carcinogenic factor ( 39 ). Bahar-Shany et al. ( 39 )showed that FF exposure increased the expression of genes involved in inflammation and induced doublestranded DNA breakage. They considered ovulation, the key non-genetic risk factor for ovarian cancer. In addition, significant cell proliferation induced by FF in vitro indicated a favourable environment for the proliferation of endometrial cells. FF obtained from endometriosis patients has been shown to induce a higher endometrial cell proliferation than those derived from women without this disorder ( 40 ). Molecular and in vitro approaches can be used to investigate the effects of FF on endometrial tissue and fallopian tubes. It is important to determine the concentration that has the most impact on implantation. Alternatively, the immunological components present in FF and how they affect implantation should be studied independently. More considerations should be taken into account for using FF to determine its cost-benefit in clinical practice. We attempted to consider all potential confounding factors in this analysis; however, some limitations may have affected the results. In the physiological state of fertilisation, FF enters the uterus through the fallopian tube, which can prepare the uterus for fertilisation by secreting a series of factors ( 12 ). It was not possible to simulate physiological conditions and inject fluid into the fallopian tube in this study. In addition, there is no goldstandard diagnostic procedure for determining endometrial receptivity; hence, we used both implantation and clinical pregnancy as substitute outcomes to validate the presence of a receptive endometrium. However, failed clinical pregnancy may also be due to poor embryo quality or other factors, and not related to abnormal endometrial receptivity. In this case, multiple confounding variables should be taken into consideration. Finally, it was not practical to blind the physicians throughout the injection process.

FF flushing into the endometrial cavity had no effect, either positively or negatively, on clinical pregnancy or implantation rates; however, it is important to emphasize that FF should be utilized with extreme caution.

This triple-blind randomised clinical trial, blinded to participants, researchers, and analysts ( NCT04077970 ) was retrospectively registered in September 2019. This study investigated the implantation and clinical pregnancy rates after intrauterine flushing of infertile patients with FF and granulosa cells compared to a control group. The study was approved by the Ethics Committee of Royan Institute (IR.ACECR.ROYAN. REC.1396.135). The procedure and rationale of interventions had been explained before the ovum was picked up and informed consent had been received from the selected women. A total of 140 infertile women who referred to the Infertility Clinic at Royan Institute (Tehran, Iran) from 2018 to 2021 were randomly allocated to either a case group that received 2 ml of clear FF from two dominant follicles, which consisted of two cumulus-oocyte complexes (COC) or a control group that received uterine cavity catheterisation. The cause of infertility in the couples was moderate male factor. We used the balanced block randomisation method to assign patients to the intervention or control group. An epidemiologist designed the block randomisation strategy using STATA software (version 13, StataCorp LLC, College Station, TX, USA), taking into account a total of four blocks. Prior to randomisation, the treatment allocation was concealed in sealed, numbered envelopes. Before the patient entered the operating room, a nurse opened the envelope and classified the patients into one of the two groups. Inclusion criteria consisted of: age 20-38 years; normal ovarian reserve based on antral follicle count (AFC) and anti-Müllerian hormone (AMH) levels; normal hormone profiles that included follicle-stimulating hormone (FSH) and luteinizing hormone (LH); regular menstrual cycles; and presence of at least two oocytes (or COC) in the aspirated dominant FF. Patients with the following criteria were excluded from the study: history of submucosal leiomyoma; intramural and subserosal leiomyomas larger than five cm; myoma with a compression effect or submucosa myometrium; endometriosis; endometrioma; hydrosalpinx; tuberculosis; tubal factor infertility; cervicitis; ovarian hyperstimulation syndrome (OHSS); low or poor response; untreated thyroid disorder; diabetic and hepatitis diseases; and severe male factor infertility (azoospermia). Controlled ovarian hyperstimulation was performed using a standard regimen comprised of a gonadotropin hormone-releasing hormone (GnRH) agonist or antagonist. In the agonist-long protocol, oral contraceptive pills were administered during the menstrual phase of the patient’s previous menstrual cycle, then pituitary down-regulation was processed during the luteal period with buserelin injections for 12- 21 days. Desensitisation was subsequently confirmed by the absence of dominant follicles as visualised by sonography, endometrial thickness ≤6 mm, and plasma estradiol (E2) levels of ≤50 pg/mL. After confirmation, ovarian stimulation was initiated on the second or third day of the menstrual cycle by administration of follitropin alfa, as recombinant FSH (Cinnal-F, CinnaGen, Tehran, Iran). The dose of Cinnal-F was determined according to the patient's age, AFC, and AMH level (150-225 IU). When at least two follicles reached a diameter of 18-20 mm or more, human chorionic gonadotropin (hCG, Ovitrelle®,10 000 IU) was administered. After 34-36 hours, FF puncture was performed via transvaginal ultrasound-guided needle aspiration. Endometrial thickness and pattern (triple line or striated) were checked during the last ultrasonography on the day of the hCG injection. In the GnRH antagonist protocol, the Cinnal-F (150- 225 IU) dose was determined according to the patient’s age, AFC, and AMH levels, and administered on days 2-3 of the patient’s menstrual period. The size and number of developing follicles were checked by transvaginal ultrasound. When the follicles reached a diameter of 13 mm or more, ovarian stimulation was accompanied by daily administration of a GnRH antagonist, cetrorelix (Cetrotide, Asta Medica, Netherlands; 0.25 mg/dL, s.c.) on the sixth day after stimulation until at least two follicles reach a diameter of 18 mm or more. Afterwards, hCG (Ovitrelle®, 10 000 IU) was administered, and follicle fluid puncture and aspiration were performed 34- 36 hours later with a needle under transvaginal ultrasound guidance ( 24 ). On the day of ovum pick-up, the FF and oocyte with surrounding granulosa cells that were greater than 20 mm in diameter were harvested by ultrasound-guided transvaginal follicular aspiration from the intervention group patients 34-36 hours following the hCG injection. Avoiding contamination during in vitro fertilization (IVF) treatment in the embryology laboratory is crucial for the success of the procedure and for patient safety. Sterilised equipment was used during the procedure and all samples (FF and culture media) were carefully handled by trained personnel to minimise the risk of cross-contamination. FF from the first 2 to 3 dominant follicles that contained at least two COCs were obtained. We chose the FF that contained COCs with a score of 1 (expanded cumulus cells and radiating corona cells). Next, we visually inspected the FF in order to identify the presence of any contamination by blood because it is a relatively reliable and fast method for the identification of blood-contaminated fluid. Clear, uncontaminated FF were selected for the next procedure. A total of two millilitres of pooled FF per patient was loaded into an intrauterine catheter (Labotect GmbH, Germany) and endometrial flushing was performed with the intent to minimise all risks of potential infection following this procedure and prevent inconvenience for the study participants. The participant had the capacity to make the decision. The physicians or midwives disclosed information on the procedure of treatment, including the expected risks or benefits. The participants comprehend the relevant information, and voluntarily granted consent, without coercion or duress. To limit biased interpretation of treatment, the study was designed as triple-blind trial (blinded participants/blinded researcher/ blinded analyst). However, the physician was not blinded to the patient's assignment in order to perform the endometrial flushing procedures. The control group included 70 randomly selected women who only underwent uterine cavity catheterisation. Embryo transfer was carried out 2-3 days later in both groups. The day 3 embryos were graded as follows: grade A embryos had at least eight blastomeres of equal size and no or<10% cytoplasmic fragmentation; grade B had seven to eight blastomeres and 11-25% cytoplasmic fragmentation; grade C had two to six blastomeres and 26-35% cytoplasmic fragmentation, and grade D had four to eight blastomeres with >35% fragmentation. Grades C and D were considered to be low-quality embryos, whereas grades A and B were considered highquality. Patients with symptoms of ovarian hyperstimulation were followed and the embryo transfer was cancelled. Subsequently, their embryos were vitrified in an equilibration solution for seven minutes and vitrification medium for a maximum of one minute. The basic solution consisted of human tubal fluid (HTF) supplemented with human serum albumin (10 µg/ml). We placed two or three of the embryos in cryotops for storage in liquid nitrogen. Patients returned to the clinic 24 to 48 hours after oocyte pick-up (OPU) for physical examination, including vital signs, in addition to assessment of signs and symptoms of allergic reactions or fever. Luteal phase support was initiated the day after OPU. Patients received 400 mg of vaginal progesterone suppositories (Cyclogest®, Actoverco, Iran) twice daily for 16 days, which was continued for up to ten weeks in case of pregnancy. Chemical pregnancy was confirmed by a positive serum β-hCG test measured by a standard kit (Elecsys Reagent Kit, Roche Cobas) where a cutoff of more than 50 mIU/mL indicated a positive result. Transvaginal ultrasonography further confirmed the pregnancy. Implantation rate was measured as the gestational sac number observed by transvaginal ultrasound divided by the number of transferred embryos on days 35-42 post OPU. A standard first-trimester vaginal ultrasound examination was performed to evaluate the presence, size, location, and number of gestational sacs, as well as the presence of a yolk sac and embryo within the gestational sacs. The clinical pregnancy rate was defined as the observation of a gestational sac with cardiac activity/foetal heartbeat on ultrasound examination at two to three weeks after a positive serum β-hCG (four to six weeks after embryo transfer) over the number of embryo transfer cycles. Participants were followed for up to 12 weeks of gestation as routine follow-up for pregnant patients. In addition, vaginal ultrasonography was performed to confirm the presence of an intrauterine pregnancy, diagnose multiple gestations, and rule out ectopic pregnancy. Once the serum level of β-hCG reached 1000 mIU/ml, a gestational sac could be visualised within the uterus by transvaginal sonography. If the location of the gestational sac was not seen after serial β-hCG assessment, the pregnancy was deemed ectopic. Ectopic pregnancy is defined as an empty uterus, an extrauterine gestational sac containing a yolk sac with or without an embryo. The rate of ectopic pregnancy was determined by dividing the number of pregnancies outside the uterine cavity by the total number of pregnancies. Multiple pregnancy rate was obtained by dividing the number of pregnancies that contained more than one embryo by the total number of pregnancies. A blighted ovum or anembryonic pregnancy occurs when the embryo fails to grow or lacks a heartbeat at an early stage in a gestational sac as monitored by transvaginal sonography. Statistical analysis was carried out using the Statistical Package for the Social Sciences, version 20 (SPSS, Inc., Chicago, IL, USA). The Kolmogorov‐Smirnov test was used to test for normal data distribution, and the appropriate parametric or non-parametric statistical tests were performed. The independent t test and Mann-Whitney U test were used to compare differences between both groups for normal and non-normal continuous variables, respectively. The chi-square and Fisher’s tests were used to compare the categorical variables between groups. Descriptive data are presented as mean standard deviation (SD) and qualitative data are presented as frequency (percentage) where indicated. Univariate logistic regressions were performed to determine independent relationships between patient characteristics and clinical pregnancy. Statistical significance was set at P<0.05.