Recurrent implantation failure (RIF) is defined as failure to achieve pregnancy after transfer of good quality embryos in women below 40 years following several in vitro fertilization (IVF) treatment cycles [1]. RIF may be due to declined endometrial receptivity, defective embryonic development, acquired thrombophilia (i.e. antiphospholipid syndrome) and other factors such as endometriosis, hydrosalpinx and myomas distorting the uterine cavity [2]. Factors which might cause defective endometrial receptivity include uterine polyps, septa, leiomyomata, adhesions and thin endometrium [3-7].

Hysteroscopy is the gold standard technique for diagnosis of intrauterine macroscopic lesions. When ultrasound, saline infusion sonography, and hysteroscopy for endometrial pathology were compared, the relative sensitivities and specificities were 89% and 56%, 91.8% and 60%, and 97.3% and 92%, respectively [5, 6]. Ultrasound is more effective in evaluating myometrial and extramural uterine disease such as type III-VII fibroids and ovarian diseases, but is less sensitive with cornual disease, sessile polyps, intrauterine adhesions, and chronic endometritis. Three-dimensional ultrasound is more sensitive and specific than transvaginal and saline infusion sonography, identifying 100% of submucosal myomas and müllerian anomalies. It is still of lower sensitivity and specificity for polyps (61.1% and 91.5%, respectively) relative to hysteroscopy [7].

Several approaches were tried to improve outcome of IVF in RIF patients such as low molecular weight heparin, immunotherapy with intravenous immunoglobulin (IVIG), assisted hatching, endometrial stimulation and office hysteroscopy [3, 8]. Transvaginal ultrasound guided embryo transfer may significantly increase the percentage of pregnancies per transfer [9].

Office hysteroscopy is a well-tolerated outpatient procedure that can diagnose uterine cavity and endometrial pathology and is considered to be the gold standard for evaluating the endometrial cavity [10]. It is used to correct abnormalities in the uterine cavity in order to improve the pregnancy rates. Intrauterine pathology affects reproductive outcomes, and advances in technology allow many opportunities for better diagnosis and management through office hysteroscopy [11]. Office hysteroscopy may be a method of endometrial scratch. Endometrial scratch may be beneficial in RIF patients to increase the odds of implantation, clinical pregnancy and live birth rate. Endometrial scratch is a type of endometrial injury during hysteroscopy using instruments such as grasper forceps. There is no specific time for this procedure. Most of the publicatins were in the preceeding luteal phase. Till now there is no solid conclusion on its efficacy [10, 12]. The current study aimed to assess the role of office hysteroscopy in the diagnosis and management of the cause of RIF and its impact on pregnancy rate.

Data were fed to the computer and analyzed using IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp) Qualitative data were described using numbers and percents. The Kolmogorov-Smirnov test was used to verify the normality of distribution Quantitative data were described using range (minimum and maximum), mean, standard deviation and median. The Chi-square test was used to compare categorical variables. Significance of the obtained results was judged at the 5% level.

The study consisted of 175 subjects with RIF. Eligibility and follow up of the cases is illustrated in the flowchart (Figure 1). The frequency of ICSI failures was 2 in 98 patients, 3 in 36 patients, 4 in 24 patients and more than 4 in 17 patients. The patient’s characteristics including mean age, levels of estradiol (E2) on basal day 3, anti-mullerian hormone (AMH), follicle stimulating hormone (FSH), leutinizing hormone (LH), prolactin and thyroid stimulating hormone (TSH) are presented in Table 1.

Figure 1.

— Flow chart.

The frequency of overall pathological findings was 78.3%. Out of 175, 40 (22.8%) patients (This adds up to 101.1% so needs clarification) did not show any abnormal findings on hysteroscopy. Of these 8(20%) women with correct endometrial pathology became pregnant after hysteroscopy. Sixty-six (37.7%) patients were noted to have endometritis, septum, polyp and adhesions while 69 (39.4%) had combined abnormalities. Endometritis 43 patients (65%) and partial septate uterus (U2a) 10 patients (15%) were the most common hysteroscopic abnormalities observed with 13 (19.6%) women becoming pregnant after office hysteroscopy (Table 2).

Table 3 describes the incidence of combined pathologies in 69 patients out of 175. Thirty-four (49%) were noted to have partial septate uterus with endometritis. Seven (10.4%) were reported with three pathologies (septum + polyp + endometritis). Among 69 women with combined pathologies, only 13 (19.7%) had clinical pregnancies after office hysteroscopy. Of 34 (49%) patients with septum and endometritis, 10 (29.4%) became pregnant. None of the patients with three pathologies obtained pregnancy. Among 175 patients with different numbers of ICSI failures 33 women became pregnant following office hysteroscopy as given in Table 4.

Recurrent implantation failure (RIF) is a complex problem that may be affected by embryo quality, endometrial receptivity, endometrial cavity problems and systemic disorders like thrombophilia [2, 14]. By adopting the definition of RIF, we excluded the embryo quality as a factor. Thrombophilia can be easily diagnosed and after thrombophylaxis, pregnancy rates improved. Patients who received low molecular weight heparin for thromboprophylaxis had a significant increase in the implantation and pregnancy rates that are comparable with the findings of Clark et al. [15].

After exclusion of thrombophilia, uterine cavity and endometrial receptivity will be the expected factors affecting RIF. Office hysteroscopy is a safe and easy tool for diagnosing cavity problems and endometrial pathologies [16]. Uterine cavity abnormalities may be present in up to 50% of women with recurrent implantation failure. Findings include polyps, endometritis, adhesions and congenital uterine anomalies. Of these, chronic endometritis plays a pivotal role in etiology of RIF and is in accordance with the conclusions of MacAnanny et al. [17]. In our study, hysteroscopic abnormal findings were detected in 78.3% of cases out of 175 which was higher than earlier studies by Fatma. (Reference) Aletebi reported abnormalities in 38% out of 132 and Cenksoy et al. found hysteroscopic abnormalities in 44.9% patients out of 156 [18, 19].

Several reports confirmed that hysteroscopic corrections of uterine abnormalities improved pregnancy rates. In logistic regression analysis, doing hysteroscopy before the first IVF cycle without correcting anatomic anomalies was not an independent predictor of live birth rate (OR: 0.72, 95% CI: 0.310-1.68, p = 0.45) [20]. In this study, out of 175 cases, 40 women were without endometrial abnormalities, 135 were reported with different hysteroscopic findings on office hysteroscopy and 33 patients achieved pregnancy. This result was comparable with the earlier reports of Elmorsy et al. [21]. In addition, Aletebi et al. in his findings reported that 132 patients with repeated RIF, 50 patients (38%) had uterine abnormalities on hysteroscopy and 55% of patients with abnormal uterine pathologies got pregnant in subsequent IVF cycles after hysteroscopy [18]. Office microhysteroscopy is well tolerated, safe and simultaneous operative correction make it an ideal and routine procedure in order to diagnose and manage missed intrauterine pathologies, especially in women with infertility resulting in additional improvement of the pregnancy rate [22].

Endometritis was the most common noted pathology either single (65%) or combined (83.7%). This finding necessitates the value of office hysteroscopy in RIF if 3D ultrasound is normal. Endometritis showed a non-satisfactory pregnancy rate after complete cure (16.3%). This pregnancy rate is lower in a non-significant manner than cases with normal finding (21.1%) and adhesions (20%). This may denote the neutralized effect of antibiotic therapy in endometritis, but the overall low pregnancy rate may reflect micro-environmental affection by endometritis or underlying cell mediated immunity abnormalities. This theory may affect the endometrial receptivity and endometrial response to hormonal changes. This finding was magnified in cases with uterine polyps which showed the worst pregnancy rate and decreased the same in cases of mullerian anomalies if combined with polyps. This may reflect endometrial hyperplasia or asynchronous endometrial growth with altered levels of inflammatory mediators making the endometrium non receptive. In addition to specifically altering receptivity, uterine abnormalities may induce or be associated with local endometrial changes. Significant decrease in endometrial leukaemia inhibitory factor mRNA expression in abnormal endometrial canal during the midsecretory phase, indicates that RIF in women with endometrial canal defects could have a molecular basis [23]. A defect of both vascular endothelial growth factor transmembranous receptors (KDR and Flt-1) has also been described in endometrium covering uterine septum and thought to be responsible for uterine abnormalities [24]. It was also established that mast cells and other humoral factors are dysregulated in endometrium near uterine polyps, which adds an immunologic component to this pathological entity [25].

The occurrence of mullerian duct anomalies also differs significantly, with findings ranging from 0.16-10% [26, 27]. They were either congenital mullerian anomalies (U1, U2, U3) or acquired lesions like endometritis, polyps, adhesions or combined lesions. In the present study, 15 cases were reported with varied classes of uterine anomalies (U1, U2a, U3a), with 10 having a class U2a or partial septate uterus (15%). After office hysteroscopy 4 women became pregnancy and the improved percentage was similar to other studies [28]. Alonso Pacheco conducted a study on a total of 63 women with uterine anomalies with 60 trying to conceive after metroplasty. Clinical pregnancy rate after metroplasty was 83.3% (n = 50/60) (p < 0.001), and the LBR was 63.3% (n = 38/60) (p < 0.001). Cesarean section rate was 26.3%. No pregnancy complications potentially related to metroplasty were reported. The spontaneous abortion rate was 12% (n = 6/50) (p < 0.001) [28].

In the current study, 4 (6%) of the patients were diagnosed with polyps, 14 (20%) with endometrial polyps and 11 (15%) with partial septate polyps as uterine abnormalities. None of the women with single pathology, and two women with combined pathology of polyps became pregnant after hysteroscopy. This low pregnancy rate might be due to increased levels of glycodelin in the ovulatory phase in women that inhibits sperm-oocyte binding thus preventing the process of implantation and angiogenesis [29]. It is also postulated that endometrial polyps decrease mRNA levels of HOXA10 and HOXA11, which are well-known molecular markers of endometrial receptivity [30]. In contrast to our results, previous studies of Aletebi et al. observed a significant improvement in pregnancy rates after hysteroscopic polypectomy [18].

In the present study of 175 women, 98 patients with two failed ICSI trails and 77 patients with three and more than that failed ICSI trails were reported. Pregnancy was observed in 33 women (19.8%). In the above cases after doing office hysteroscopy before the next ICSI attempt. Pregnancy rate did not show any improvement in cases with previous three failed ICSI denoting hidden etiology rather than the gross picture. Repeated ICSI and repeated uterine instrumentation may be a cause of endometritis and its harmful effects on implantation. This suggest the need for contemporary investigation of molecular indicators of endometrial receptivity [31]. In the study of Achache et al. he compared patients with repeated IVF failure with fertile controls and revealed that reduced prostaglandin synthesis in the human endometrium may also lead to poor endometrial receptivity in patients with RIF [32]. The results of the present study reinforce the view that office hysteroscopy is beneficial to women with RIF prior to further infertility treatment.

In conclusion, office hysteroscopy is an essential procedure for infertility investigation before ICSI. It demonstrates and corrects abnormalities in uterine cavities as it is cost effective and the gold standard technique for assessment of the endometrial cavity. It was considered as a positive prognostic factor in improving the clinical pregnancy rate in recurrent miscarriage patients with ICSI. Further studies need to be carried out to confirm the conclusions of this study. The limitations of this study were that it was not a randomized control trial and the environment is different between the Egypt and Libya (place of the study). The strength of the study it was conducted on a cohort of cases with high order of RIF which carry the worst prognosis in IVF and need a tool for both diagnosis and management.

Alexandria university, faculty of medicine 0304338. Registered 17 June 2019 - Retrospectively registered, http://iclas.org/members

ME analyzed and interpreted the patient data regarding the findings in office hysteroscopy. HM, IF and ME performed the office hysteroscopy and the ICSI cycles, and were contributors in writing the manuscript. All authors read and approved the final manuscript.

This study was approved from the Ethics committee, faculty of medicine, Alexandria university. The number is: 0304338.

Written informed consent for publication of their clinical details and/or clinical images was obtained from the patient. A copy of the consent form is available for review by the Editor of this journal.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

The Misurata national IVF center team for assistance in the work.

The authors declare no conflict of interest.