Hysteroscopic Endometrial Fundal Incision versus Hysteroscopy Only in Oocyte Recipients: A Randomized Controlled Trial Assessing The Reproductive Outcomes.

Successful implantation is a multistep process which still remains challenging and not totally understood ( 1 , 2 ). The two main determinants of the implantation are high embryo quality and endometrial receptivity leading both in a great coordination ( 3 ). Initially, fertility researchers enormously managed to ameliorate the number and quality of the embryos, investigating different stimulation protocols, immunological approaches, laboratory culture conditions and preimplantation genetic screening ( 4 ). Then, endometrial receptivity led to the treatment of intrauterine endometrial pathology, different hormonal treatment mainly in cases with thin endometrium and the introduction of a mechanical endometrial injury known as endometrial scratching (ES) ( 5 , 6 ). Although, many fertility experts adopted these techniques, still only 30-35% of all the embryotransfers (ET) lead to successful embryo implantation ( 7 ). Endometrial receptivity is estimated to be responsible for 2/3 of the implantation failures ( 8 ). This limiting factor in embryo implantation may be overwhelmed by different approaches including diverse ES techniques ( 9 - 11 ). The pipelle catheter or canula, Novak curette and hysteroscopy with or without induced endometrial injury are some approaches for this minimally invasive procedure, which are considered to improve implantation by inflammatory reaction ( 12 - 15 ). Natural killers of the uterine cells, receptivity genes, leukemia inhibitory factor and other unknown inflammatory molecules are hypothesized to be increasingly synthesized by this so called “scratching” triggering ( 16 ). According to published data, 83% of fertility specialists in Australia, the UK and New Zealand provide or advise ES, which is cost-effective for the patients ( 17 ). A randomized controlled trial assessed ES with pipelle in 1,364 women undergoing an IVF cycle and found no improvement in live birth rates, thus, giving more ground to the devotees of the hysteroscopic approach for this intervention ( 18 ). Notably, the accurate and standardized approach of ES under real-time visualization of the endometrial cavity during hysteroscopy provides safety to the operator in contrast with the pipelle scratching, which is a blind procedure. The study's objective was to assess the impact of the endometrial fundal incision (EFI) during hysteroscopy, a unique ES approach developed by our team, on reproductive outcomes in oocyte recipients.

During the study period, 206 oocyte recipients were assessed for eligibility and 73 were excluded from the randomization process as 41 were diagnosed with intrauterine pathology during hysteroscopy, 6 had a history of a previous intrauterine intervention, 14 refused to participate and 12 deferred ET. Therefore, a total of 133 oocyte recipients were randomized. The EFI group initially consisted of 66 women, of whom 4 did not complete the ET cycle as 3 had a thin endometrium and 1 had high levels of progesterone. Accordingly, the control group originally consisted of 67 patients, of whom 5 were not included in the analysis as 3 did not manage to synchronize during their ET cycle and 2 were lost in follow-up. Figure 1 shows the course of the 34-month trial. Flowchart of the study. ET; Embryo transfer and EFI; Endometrial fundal incision. No significant differences in the age, duration of infertility, smoking habit, previous live birth, and duration of HRT were identified between the two groups. In terms of stimulation time, estrogen administration method and dosage, or endometrial thickness, there were also no differences between the two groups. Prior to ET, hormone blood levels (estradiol and progesterone) were likewise comparable ( Table 1 ). Additionally, none of the participants received second hysteroscopy before the first ET. Regarding reproductive outcomes, the pregnancy test was higher (79%, n=49/62) in the EFI than in the non-hysteroscopy group (59.7%, n=37/62, P=0.019). Moreover, the live birth rate was higher, but not statistically significant in the EFI (58.1%, n=36/62), compared to 51.6% (n=32/62) in the office hysteroscopy only group (P=0.470, Table 2 ). Demographics of the recruited patients Data are presented as mean ± SD or n (%). EFI; Endometrial fundal incision, BMI; Body mass index, and HRT; Hormone replacement treatment. Reproductive outcomes in the study population EFI; Εndometrial fundal incision, β-hCG; Beta-Human chorionic gonadotropin, and *; Statistical significant.

Our results demonstrated that EFI during hysteroscopy, one to three months before ET, in oocyte recipients without intrauterine pathology, improved pregnancy rates, but without affecting the live birth rates. These findings could affect current ART practices; EFI could be considered as an extra tool for IVF specialists to improve pregnancy rates in oocyte recipients with a normal intrauterine cavity. Even though the gold standard for early detection and if necessary, treatment of intrauterine pathology is hysteroscopy, there is still disagreement on the role of routine hysteroscopy in the treatment of infertile women without any identified or definite intrauterine pathology ( 23 ). Hysteroscopy has two main drawbacks: first, it is an invasive operation and second, there is still disagreement over the true impact of the detected intracavitary pathology on fertility ( 24 , 25 ). However, it was reported that up to 26% of the patients with normal hysterosalpingography may have aberrant hysteroscopic findings ( 26 ). Moreover, despite having greater potential than transvaginal scans and saline infusion sonography, three-dimensional ultrasound still falls short of hysteroscopy in terms of accuracy (sensitivity: 68.2% and specificity: 91.5%) ( 27 ). Consequently, whether or not to have hysteroscopy before the first IVF cycle remains debatable, particularly in Greece, where assisted reproductive procedures are mainly offered in private IVF clinics. Additionally, the importance of hysteroscopic evaluation before ET has been previously highlighted in oocyte recipients ( 28 ), as the identified rate of intrauterine abnormalities was much greater (21.7%) than the estimated one (10-15%) ( 29 ). As already mentioned, a receptive endometrium, an euploid embryo and adequate communication among the semiallotypic embryo and the female's endocrinological and immunological systems should all take place for implantation to be successful ( 1 ). This finding gave rise to the idea that endometrial damage could enhance implantation and occasionally other researchers investigated the subject in human samples, obtaining encouraging results ( 12 , 30 , 31 ). Interestingly, the results were conflicting and made ES a topic of debate and one of the most contentious and interesting add-on procedures in the field of reproduction ( 32 ). Several mechanisms by which injury may increase pregnancy rates have been proposed; i. Increasing the endometrial inflammatory response, ii. Increasing the synthesis of growth factors that promote injury reconstruction, iii. Encouraging decidualization, and iv. Increasing the expression of specific endometrial genes ( 33 ). Of note, the increase of leukemia inhibitory factor after endometrial injury during hysteroscopy may contribute to all of these fertility promoting effects ( 16 ). Pipelle ES has been previously proposed as a simple, cost-effective and minimally invasive technique to increase endometrial response during IVF cycles ( 18 ). As per our proposed technique, unlike the pipelle, in which the blinded catheter scrapes the posterior or anterior uterine wall but never the fundus, the injury is directed to the uterine fundus and the operator can even adjust the degree of the induced injury. Furthermore, in cases with arcuate uterus (type U2a), scratching is also therapeutic since it restores this congenital variation of the uterine fundus, which was previously thought to be physiological with no impact on implantation. Other methods for ES have been previously described; Jayakrishnan et al. ( 34 ) specifically used the hysteroscope for the treatment of any identified uterine disease, apart from local injury. Additionally, curettage of the fundus and the posterior wall following hysteroscopy was carried out in another study ( 15 ), while monopolar needle forceps were used by Seval et al. ( 13 ) to inflict injury. First-time ET attempts in patients were the subject of a meta-analysis by Vitagliano et al. ( 35 ), who found no association between ES and pregnancy rates of initial ART cycles. This is in contrast to our findings; we found increased pregnancy rates following EFI, which could be attributed to an increase in endometrial receptivity caused by mechanically inducing endometrial damage and associated regeneration of new tissue. Furthermore, our findings could be explained by the fact that, in order to limit the systemic bias, we limited the study population to recipients without repeated implantation failure or intrauterine pathology. According to a recent meta-analysis, ES had a positive impact on CPR (OR: 1.34; 95% CI: 1.14- 1.58), live birth rate (OR: 1.30, 95% CI: 1.06-1.60), while also increased the multiple pregnancy rate (OR: 1.35, 95% CI: 1.07-1.71) ( 36 ). Regarding the timing of intervention, a trial conducted by Karimzade et al. showed significant decrease in pregnancy rates when ES was performed in the day of oocyte retrieval during ART ( 37 ). It was also found that performing ES during the follicular phase of the current cycle increases the miscarriage rate (25 vs. 8%, P=0.032) ( 38 ). Our results on the timing of EFI are also encouraging; performing the EFI during the follicular phase 1-3 months before the ET increased the pregnancy rates in our sample. According to findings from previous studies, where the effect of ES during hysteroscopy was investigated, there was no difference in miscarriage rate ( 13 , 15 ). Moreover, a recently published meta-analysis found no statistically significant difference on miscarriage rate following ES (RR: 0.80, 95% CI: 0.52-1.22) ( 6 ). The same findings were also identified following ES with pipelle in donor egg recipients ( 19 , 22 ). Interestingly, in our study, the miscarriage rate was higher in the EFI group, but without statistical significance, which could be attributed to the enormous induction of endometrial decidualization caused by our technique. To our knowledge, this study managed to examine the effects of whether to undergo or not a unique hysteroscopy-induced ES the so called “EFI”, on oocyte donation cycles. The primary strength of this study is the consistency of the endometrial preparation process and embryo quality because the embryos transferred are from young donors with good reproductive backgrounds, assuring a low bias in pregnancy outcomes between groups. Second, because of the strict exclusion criteria, every patient included in the investigation had a normal intrauterine cavity. In addition, the entire data set came from a single center and all hysteroscopies were conducted by the same reproductive medicine consultants to reduce interobserver disparities. With regard to the limitations, the sample size was relatively small to reach definitive conclusions on the impact of EFI on reproductive outcomes. The primary criticism of our proposed technique is that hysteroscopy and EFI should be used under general anesthesia compared to other pipelle scratching techniques, which are of low cost and usually performed in an outpatient basis.

Performing EFI during hysteroscopy one to three months prior to ET seems beneficial for implantation in oocyte recipients where endometrial pathology was not identified, according to the findings of this study. The idea that site-specific mechanical damage during hysteroscopy could enhance reproductive results was set to evaluation by our study. In order to clarify the mechanism of action and further evaluate the impact of the technique, more longitudinal well-designed studies utilizing EFI in chosen groups are encouraged.

Informed consent was acquired from each patient. Τhis trial was registered on 2 October 2020 in ClinicalTrials. gov with an ID: NCT04580056 . The study protocol was also approved by the Institutional Review Board of Assisting Nature IVF Unit (registration number: 0210201405). No incentives were provided to the patients and all the hysteroscopic procedures were offered without extra cost. Different treatment protocols and populations of hysteroscopic ES studies may be associated with high risk of bias in the interpretation of their results; thus, ovum donation cycles should be investigated. These cycles should use donated blastocysts from young, viable oocytes and because of this, the receivers’ chances of becoming pregnant are unaffected by their age or the low-quality oocytes ( 19 ). Furthermore, heterogenicity among recipients would be reduced, as all of them will have a substituted cycle with the same protocol to prepare the endometrium for embryo implantation. Therefore, performing EFI during the cycle preceding ET would be an effective procedure by itself. This was a randomized controlled trial carried out at the Third Department of Obstetrics and Gynecology, School of Medicine, Faculty of Health Sciences, Aristotle University of Thessaloniki, Greece in cooperation with “Assisting Nature Centre of Reproduction and Genetics”, which is a private IVF Unit located in Thessaloniki, Greece. Patient recruitment was conducted from October 2020 to June 2023. All hysteroscopies were offered during the follicular phase of the 1-3 preceding menstrual cycles, before initiating the endometrial preparation. All patients undergoing hysteroscopy prior to oocyte donor IVF treatments were eligible for the trial. Eligibility criteria for oocyte recipients were: i. Age between 25 and 50 years, ii. Only frozen blastocysts could be offered, iii. Absence of serious endometrial pathology, i.e. submucosal fibromas classified as FIGO 0-2 ( 20 ) or polyps diagnosed during 2D and 3D ultrasonography, hysterosalpingography or office hysteroscopy, iv. Endometrial thickness >7 mm and blood progesterone levels <1.5 pg/ ml the day before progesterone supplementation during hormone replacement therapy (HRT) preparation and v. EFI was performed with the use of endoscopic scissor only, without the use of electrocautery method. Exclusion criteria were: i. Women who had an office hysteroscopy within six months before receiving oocyte treatment, ii. Severe male factor with a sperm count <15×106 /ml, total motility <40%, or normal forms <4% according to the World Health Organization criteria ( 21 ), iii. Those who had undergone uterine surgery in the past, iv. Free fluid in the endometrial cavity during HRT preparation, v. Unilateral or bilateral hydrosalpinx, vi. Severe adenomyosis, vii. Müllerian malformations, and viii. Patients with a body mass index (BMI) >35 kg/m 2 due to associated decreased implantation rates. The clinical trial comprised of two distinct treatment arms (1:1), namely the intervention EFI group that underwent EFI during hysteroscopy and the control group that underwent a conventional treatment protocol (office hysteroscopy only) without any form of endometrial manipulation ET. Using a computer-generated code with the use of random permuted blocks of randomly varying size generated by a web-based application (random.org), we constructed the randomization chart (IP: 2a02:1388:409 6:c4ee:8196:3154:592e:e2a4 and timestamp: 2020-10-17 10:04:54 UTC). After assigning patients to one of two treatment groups (group A for the intervention and group B for the control; the assignment was done before the initiation of each protocol), the patients were allocated to the associated protocol. Patients in both groups received a video recording of their hysteroscopy, so blinding was not possible. The same hormone endometrial preparation protocol was used for all frozen ET; beginning on day 2 of the cycle, if an ultrasound revealed quiet ovaries and basal hormone levels (estradiol-E2 80 pg/ml and progesterone 1.5 ng/ml), the woman could receive HRT. Before progesterone supplementation, estrogen supplementation was given for a minimum of 10 days and a maximum of 20 days in the form of 17-b estradiol. In particular, we started on day 2 with 2 mg (1×1) per day until day 5, then switched to 4 mg (1×2) till day 6, 6 mg (1×3) for the following 3 days until day 8 and finally 8 mg (2×2) until the pregnancy test. Between days 10 and 11, we measured the blood levels of progesterone, luteinizing hormone (LH) and estradiol (E2), as well as the endometrial thickness, using ultrasonography. The therapy was continued for 3 more days if the endometrial thickness was less than 7 mm. The ET was arranged six days after the ideal endometrial thickness (>7 mm) was reached. At that point, daily progesterone was offered. All the blastocysts transferred were day 5 ones. Nine days after embryo transfer or 14 days after starting progesterone supplementation, the levels of beta-human chorionic gonadotropin (β-hCG) were measured. One to three months before starting a new HRT cycle, at the early follicular phase, all recipients received standard assessment. In order to improve cavity visualization, women who were scheduled for hysteroscopy started taking the contraceptive pill on day 3. Between days 6 and 13 of the menstrual cycle, a hysteroscopy was carried out using vaginoscopic approach. Sedation was achieved with standard analgesics. 0.9% normal saline was utilized with a rigid hysteroscope (Stortz Bettochi® 4.8 mm hysteroscope; continuous flow; 30° forward oblique view). Systematic inspection was done after the uterine cavity had been sufficiently distended. All hysteroscopies were carried out by N.P., R.N. or E.P., three reproductive medicine consultants. Using a 2 mm Wolf® endoscopic scissor, EFI was carried out according to our randomization list. Through one fallopian ostium to the next, the EFI was conducted in a single straight line at the uterine fundus. As for the depth of the incision, it was made until the first vessels could be seen through the connective tissue. The primary outcomes were clinical pregnancy rates (CPR) and live birth rates, defined as the birth of a live fetus after 24 weeks of gestation, which survived for at least one month. With regards to secondary outcomes, pregnancy rates, defined as the proportion of women who had a quantitative serum β-hCG test above 10mIU/ml nine days following the transfer of the blastocyst, early and late miscarriages and biochemical pregnancy were considered. Fetal heart activity, starting at around 6-7 weeks, was used to diagnose clinical CPR. Moreover, all miscarriages before the 5th week were defined as biochemical pregnancies. Pregnancy loss prior to 12 weeks of gestation (including biochemical pregnancies) was defined as early miscarriage, while pregnancy loss between 12 and 24 weeks was defined as late miscarriage. According to earlier research on ES during hysteroscopy ( 13 , 15 ) and on ES in oocyte recipients ( 19 , 22 ), implementing ES increased CPR by 10-30%. We expected that a 25% variation in CPR would be clinically significant because the typical CPR in egg donor IVF cycles at our facility is 55-60%. To detect a mean difference of 25% after hysteroscopic EFI with a two-sided significance level of 5% and power of 80% with equal allocation to two arms would require 62 oocyte acceptors in each arm of the trial. We considered a 5% dropout probability since egg donor IVF cycles have a low cancellation rate, primarily because of insufficient endometrial preparation or a lack of healthy embryos to transfer. Only the initial ET was taken into consideration and each patient was only randomly assigned once. The continuous variables’ values were presented as means (SD) and absolute frequencies (%), where appropriate. The independent samples t test was used to compare the between-group differences. Statistical analysis of the classified data was performed using Fisher’s exact and Pearson’s Chi-square tests. The cutoff point for statistical significance was 0.05. IBM SPSS v25.0 (IBM, USA) was utilized for all data analyses, per protocol.