These authors contributed equally to this study and share first authorship. This article is based on the doctoral dissertation of the first author (Soo Yeon Yang) from the Department of Medical Device Management and Research, Samsung Advanced Institute for Health Sciences & Technology (SAIHST), Sungkyunkwan University.
Hysteroscopy can be used both to diagnose and to treat intrauterine pathologies. It is well known that hysteroscopy helps to improve reproductive outcomes by treating intrauterine pathologies. However, it is uncertain whether hysteroscopy is helpful in the absence of intrauterine pathologies. This study aimed to confirm whether hysteroscopy improves the reproductive outcomes of infertile women without intrauterine pathologies.
We conducted a systematic review of 11 studies retrieved from Ovid-MEDLINE, Ovid-Embase, and the Cochrane Library. Two independent investigators extracted the data and used risk-of-bias tools (RoB 2.0 and ROBINS-I) to assess their quality.
Diagnostic hysteroscopy prior to in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) was associated with a higher clinical pregnancy rate (CPR) and live birth rate (LBR) than non-hysteroscopy in patients with recurrent implantation failure (RIF) (odds ratio, 1.79 and 1.46; 95% confidence interval, 1.40–2.30 and 1.08–1.97 for CPR and LBR, respectively) while hysteroscopy prior to first IVF was ineffective. The overall meta-analysis of LBR showed statistically significant findings for RIF, but a subgroup analysis showed effects only in prospective cohorts (odds ratio, 1.40 and 1.47; 95% confidence interval, 0.62–3.16 and 1.04–2.07 for randomized controlled trials and prospective cohorts, respectively). Therefore, the LBR should be interpreted carefully and further research is needed.
Although further research is warranted, hysteroscopy may be considered as a diagnostic and treatment option for infertile women who have experienced RIF regardless of intrauterine pathologies. This finding enables nurses to educate and support infertile women with RIF prior to IVF/ICSI.
It is well known that hysteroscopy helps to improve reproductive outcomes by facilitating the treatment of intrauterine pathologies. However, it is uncertain whether hysteroscopy is helpful even in the absence of intrauterine pathologies.
Our study showed that diagnostic hysteroscopy alone prior to in vitro fertilization, compared with non-hysteroscopy, may improve reproductive outcomes even in the absence of intrauterine pathologies in women who have experienced recurrent implantation failure (RIF).
Recognizing that hysteroscopy may be considered as a diagnostic and treatment option for infertile women who have experienced RIF regardless of intrauterine pathologies, may be helpful in the education and advocacy of infertile women with RIF. Endometrial biopsy during hysteroscopy can also be considered.
Infertility is defined as the failure to establish a clinical pregnancy after 12 months of regular, unprotected sexual intercourse, due to an impairment of an individual’s capacity to reproduce either alone or with his or her partner [
Assisted reproductive technology (ART) has been developed and distributed worldwide to help infertile couples, but despite its high cost, its success rate remains low [
There are various reasons for implantation failure, including embryo quality and endometrial receptivity, but in many cases, the cause is unknown [
Hysteroscopy is the gold-standard test for assessing intrauterine conditions [
Treating intrauterine pathologies through hysteroscopy has been found to lead to improvements in reproductive outcomes, since intrauterine lesions can negatively affect the implantation rate [
This systematic review was performed to reflect the latest results on whether diagnostic hysteroscopy prior to IVF improves the reproductive outcomes, including the LBR, of infertile women without intrauterine pathologies compared to infertile women who do not undergo hysteroscopy.
Ethics statement: This study is a literature review of previously published studies and was therefore exempt from institutional review board approval.
On January 28, 2020, a search was conducted for relevant articles regarding hysteroscopy in infertile women in the following databases: Ovid-MEDLINE, Ovid-Embase, and the Cochrane Library (the Cochrane review and trials database).
Combinations of the following Medical Subject Heading keywords were used for the searches: “hysteroscopy,” “minihysteroscopy,” “infertility,” “subfertility,” “intrauterine insemination,” “assisted conception,” “ICSI,” “fertilization in vitro or IVF,” “embryo transfer (ET),” “conception,” “miscarriage or abortion,” and “IVF-ET.”
Two reviewers (SYY and SHL) independently screened the titles and abstracts of the studies extracted from the databases. The full text was subsequently reviewed to identify potential relevant articles. Studies were selected regardless of whether they reported experiences of recurrent implantation failure (RIF), and we included both RCTs and NRSs. Studies that reported the following were included: (a) infertile women who were scheduled to use ART (IVF/ICSI) for infertility treatment; (b) hysteroscopy in infertile women; and (c) the CPR or LBR in infertile women without intrauterine pathologies who underwent hysteroscopy. Additionally, only papers published within the last 20 years were included. The following types of studies were excluded: (a) animal studies; (b) articles not in English; and (c) conference posters, study protocols, review articles, cost-effectiveness analysis studies, and abstracts.
We defined the outcomes of interest before the systematic review. The primary outcome measures were the CPR and LBR, and the secondary outcome measures were the implantation and abortion rates, as well as adverse events related to hysteroscopy.
In cases of disagreement between the reviewers, discussions were held to resolve the issue. The principle was set that in cases where a consensus was not reached between the two reviewers, the third reviewer would intervene; however, all conflicts were resolved without the intervention of a third reviewer.
Two reviewers (SYY and SHL) independently conducted quality assessments using the Cochrane’s risk of bias tool, ver. 2 (RoB 2.0; August 22, 2019 version) for RCTs [
The RoB 2.0 tool includes five domains; bias arising from the randomization process, bias due to deviations from the intended intervention, bias due to missing outcome data, bias due to outcome measurement, and bias due to the selection of the reported results. Each criterion for the RoB 2.0 tool was evaluated as either “low risk,” “high risk,” or “some concerns.” The ROBINS-I tool includes seven domains; bias due to confounding, bias due to the selection of the participants, bias in the classification of the interventions, bias due to deviations from the intended interventions, bias due to missing data, bias in measurement of outcomes, and bias in selection of the reported result. Each item was graded as “low risk,” “moderate risk,” “serious risk,” “critical risk,” or “no information.” Disagreements regarding the quality assessments between the reviewers were resolved through discussion.
Two reviewers (SYY and SHL) independently extracted data from the studies selected according to the selection criteria. Disagreements between the reviewers were resolved through discussions. The following data were extracted for each of the 11 selected studies: author; year of publication; title; country in which the study was conducted; study design, and group; number and ages of the patients; experiences of RIF; previous investigations (diagnostic tests performed before participation in the study such as transvaginal ultrasonography [TVS] or hysterosalpingography [HSG]); descriptions of the participants (inclusion and exclusion criteria, type of infertility); details of the intervention (hysteroscopy or no hysteroscopy); whether endometrial stimulation was performed; the method used to attempt pregnancy; the author’s conclusion; the main outcome measures; intergroup differences; and adverse events of hysteroscopy.
The authors of the selected studies were contacted to provide missing or unclear information on the trial methods or data. We used the meta-analyses of observational studies in epidemiology reporting guidelines [
The pooled odds ratio (OR) was extracted for categorical data. Meta-analysis was undertaken where there were two or more studies. From each study, binary data were extracted in 2×2 tables and the results were pooled and expressed as ORs with 95% confidence intervals (CIs) using a random-effects model, as appropriate [
The process of study selection is summarized in
A total of 2,048 studies were initially identified. After excluding duplicates, 1,705 studies remained. A total of 120 studies were selected upon initial screening. After the full-text review, 111 studies were excluded and nine studies were included, with two studies additionally included based on a hand search (March 10, 2020). Ultimately, a total of 11 studies were included [
Six RCTs [
Of the 11 studies included in our systematic review, two (18.2%) performed endometrial stimulation during hysteroscopy [
In the hysteroscopy intervention group, ART (IVF/ICSI) was performed after hysteroscopy in the initial proliferative phase. In the non-hysteroscopy group, the attempt to use ART was made immediately in 10 studies, with the exception of one study [
Regarding embryo transplantation, fresh embryos were transplanted in nine studies [
A 2.9- to 5.5-mm-diameter hysteroscope was used in the intervention group. Four and three studies (36.4% and 27.3%, respectively) used a 4-mm and 5-mm-diameter hysteroscope, respectively [
Upon quality assessment, three of the six RCTs [
Of the five NRSs, four [
Diagnostic hysteroscopy vs. non-hysteroscopy was analyzed by subgroup according to IVF attempts.
The same seven studies (four RCTs [
The overall meta-analysis of the seven studies showed that the RIF group [
Eight of the 11 studies (three RCTs [
The overall meta-analysis of the eight studies showed that the RIF group [
Diagnostic hysteroscopy vs. non-hysteroscopy was analyzed by subgroup according to whether endometrial stimulation was performed during hysteroscopy.
Seven of the 11 studies (four RCTs [
The results of the seven studies showed significant differences in the CPR regardless of whether endometrial stimulation was performed in the diagnostic hysteroscopy group without intrauterine pathologies before IVF/ICSI when compared with the non-hysteroscopy group (OR, 1.67, 95% CI, 1.42-1.97; I2=0%,
As reported above, the same eight studies (three RCTs [
The results of the eight studies showed significant differences in the LBR regardless of endometrial stimulation in the hysteroscopy group without intrauterine pathologies before IVF/ICSI when compared with the non-hysteroscopy group, but the degree of significance was not as high as it was for the CPR (OR, 1.34; 95% CI, 1.09–1.64; I2=38%,
The implantation rate was reported for the hysteroscopy groups, but no study separately reported the implantation rate of infertile patients without intrauterine pathologies (diagnostic hysteroscopy), so this parameter was excluded from the analysis.
Three of the 11 studies (2 RCTs, 1 NRSs [
A subgroup analysis was performed with RCTs and NRSs, as high heterogeneity was found (
The results of the three studies did not show a significant difference in the miscarriage rate in the diagnostic hysteroscopy group without intrauterine pathologies compared with the non-hysteroscopy group (OR, 1.22; 95% CI, 0.57–2.58; I2=60%,
Seven studies (63.6%) did not mention any adverse events relating to hysteroscopy [
This study is the first systematic review and meta-analysis to compare the reproductive outcomes of infertile patients without intrauterine pathologies who underwent hysteroscopy (diagnostic hysteroscopy) and groups of infertile patients who did not undergo hysteroscopy (non-hysteroscopy) since the systematic review conducted by El-Toukhy et al. [
This study showed that performing diagnostic hysteroscopy prior to IVF/ICSI may improve the CPR and LBR even in patients without intrauterine pathologies, as opposed to not performing hysteroscopy, especially in patients with RIF; however, hysteroscopy prior to the first IVF attempt was found to be ineffective. A subgroup analysis was conducted according to whether endometrial stimulation was performed during hysteroscopy to determine whether endometrial biopsy affects reproductive outcomes when diagnostic hysteroscopy is performed in infertile women without intrauterine pathologies. Regardless of endometrial stimulation, the hysteroscopy group showed greater improvement in the CPR and LBR than the non-hysteroscopy group.
Regarding the number of IVF attempts, our study showed that the CPR after diagnostic hysteroscopy was effective in patients who had experienced RIF without intrauterine pathologies (in comparison to no hysteroscopy), but not in infertile women without intrauterine pathologies attempting IVF for the first time (OR, 1.79 and 1.51; 95% CI, 1.40–2.30 and 0.97–2.36 for RIF and first attempts, respectively). The CPR was assessed in seven studies with 3,152 participants. Our findings are supported by recent systematic reviews by Cao et al. [
This study also showed that performing diagnostic hysteroscopy prior to IVF/ICSI for women with RIF may improve the LBR even in the absence of intrauterine pathologies compared with the non-hysteroscopy group, whereas hysteroscopy prior to the first IVF attempt was found to be ineffective (OR, 1.46 and 1.16; 95% CI, 1.08–1.97 and 0.86–1.56 for RIF and first attempts, respectively). However, the subgroup analysis showed effectiveness only in prospective cohorts (OR, 1.40 and 1.47; 95% CI, 0.86–1.56 and 1.04–2.07 for RCTs and prospective cohorts, respectively). The LBR was also assessed in eight studies with 4,372 participants. Regarding the effects on the LBR in women with RIF, the results of previous systematic reviews are discordant. Cao et al. [
Di Spiezio Sardo et al. [
Studies classified as having some concerns in RoB 2.0 [
With regard to endometrial stimulation during hysteroscopy, this study showed improvements in the CPR and LBR regardless of endometrial stimulation (OR, 1.67 and 1.34, 95% CI, 1.42-1.97 and 1.09–1.64 for CPR and LBR, respectively). This result is consistent with the systematic review of Kamath et al. [
A subgroup analysis was performed according to whether endometrial stimulation during hysteroscopy. The degree of improvement in IVF outcomes observed after endometrial stimulation during hysteroscopy seemed to be higher than that after no endometrial stimulation during hysteroscopy (OR, 1.96 and 1.59; 95% CI, 1.36–2.83 and 1.32–1.92 for the CPRs after endometrial stimulation and no endometrial stimulation, respectively; OR, 2.15 and 1.23; 95% CI, 1.35–3.44 and 1.04–1.45 for the LBRs after endometrial stimulation and no endometrial stimulation, respectively). The CPR was assessed in seven studies with 3,152 participants and the LBR was also assessed in eight studies with 4,372 participants, but only two RCTs investigated endometrial stimulation during hysteroscopy [
Our study showed that diagnostic hysteroscopy alone prior to IVF may improve reproductive outcomes even in the absence of intrauterine pathologies, compared with patients who did not undergo hysteroscopy. In addition to the hypothesis of cytokine and growth factor release due to the injury induced by hysteroscopy, three hypotheses have been proposed to explain the improvement of reproductive outcomes resulting from diagnostic hysteroscopy even if an intrauterine pathology is not corrected. First, the saline used during hysteroscopy mechanically removes the harmful anti-adhesive glycoprotein molecules involved in endometrial receptivity from the endometrial surface (cyclooxygenase-2, mucin-I, and integrin αVβ3) [
Despite our findings, this study has several limitations. First, some studies did not separately investigate infertile women with intrauterine pathologies after hysteroscopy regarding the CPR, LBR, implantation, and miscarriage rates separately; therefore, not all of the data were limited to infertile women without intrauterine pathologies who underwent hysteroscopy before ART compared with the non-hysteroscopy group. We tried to contact authors to obtain this information, but no response was received. Nonetheless, this study is meaningful as it is the first systematic review to quantify the effect of hysteroscopy on both the CPR and LBR in infertile women without intrauterine pathologies. Second, heterogeneity was shown when pooling results for the LBR from the eight included studies (
In conclusion, this systematic review and meta-analysis showed that performing diagnostic hysteroscopy prior to IVF/ICSI may improve the CPR and LBR as opposed to not performing hysteroscopy, even in the absence of intrauterine pathologies, especially in patients with RIF; however, hysteroscopy prior to the first IVF attempt was found to be ineffective. In addition, stimulation of the endometrium during hysteroscopy may improve reproductive outcomes. However, large-scale randomized studies are needed to provide stronger evidence in the future. Although further research is needed, hysteroscopy may be considered as a diagnostic and treatment option for infertile women who have experienced RIF regardless of the presence of intrauterine pathologies, and endometrial biopsy could be considered when performing hysteroscopy. Hysteroscopy has few adverse events, as confirmed in this systematic review, but infertile women may feel fear and anxiety before hysteroscopy and might doubt whether hysteroscopy can improve reproductive outcomes. If infertile women who have experienced RIF are scheduled for hysteroscopy before IVF/ICSI, nurses can not only provide emotional support by telling patients that adverse effects of hysteroscopy are rare, inform them that hysteroscopy may have a beneficial effect on reproductive outcomes even if there is no intrauterine pathology to be treated, may also alleviate their fears.
Further details on supplementary materials are presented online (available at
Characteristics of the included studies (N=11)
Risk of bias. (A) Randomized controlled studies; summary. (B) Randomized controlled studies; by study. (C) Non-randomized studies; summary. (D) Non-randomized studies; by study.
Conceptualization, Validation: all authors; Methodology, Software, Visualization: Yang SY; Formal analysis: Yang S, Lee S; Supervision: Lee S, Chon S; Writing–original draft, Yang S, Lee S; Writing–review & editing: all authors.
The authors declared no conflict of interest.
None.
Please contact the corresponding author for data availability.
None.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.
CPR: Clinical pregnancy rate; ICSI: intracytoplasmic sperm injection; IVF: in vitro fertilization; LBR: live birth rate; NRS: non-randomized study; RCT: randomized controlled trial.
Diagnostic hysteroscopy vs. non-hysteroscopy according to the number of in vitro fertilization attempts. (A) Clinical pregnancy rate. (B) Live birth rate.
df: Degree of freedom; M-H: Mantel-Haenszel; RCT: randomized controlled trial; RIF: recurrent implantation failure.
Diagnostic hysteroscopy vs. non-hysteroscopy in patients who underwent endometrial stimulation during hysteroscopy. (A) Clinical pregnancy rate. (B) Live birth rate.
df: Degree of freedom; M-H: Mantel-Haenszel; RCT: randomized controlled trial.
Meta-analysis of the miscarriage rate.
df: Degree of freedom; M-H: Mantel-Haenszel; RCT: randomized controlled trial.
The characteristics and effectiveness of the reviewed interventions (N=11)
First author (year) | Intervention |
Comparator | Endometrial irritation (I only) | Method of pregnancy attempt (both I and C) | Embryo / day of ET | Authors’ conclusion | Main outcome measures | Intergroup differences | Adverse events of hysteroscopy | |
---|---|---|---|---|---|---|---|---|---|---|
Timing | Hysteroscope | |||||||||
Tanacan (2019) [ |
In the early to midfollicular phase of the menstrual cycle (1–3 months before the start of IVF) | Not specified | Without diagnostic hysteroscopy prior to the first IVF cycle | No scratching | IVF | Fresh embryo / day 3 or day 5 | OH before the first IVF treatment cycle did not improve fertility outcomes in patients without previously detected pathology of the uterine cavity. | (1) Implantation rate | (1) .840 | Not specified |
Distension medium: not specified | Routine usage of hysteroscopy should not be offered to patients in their first IVF cycles. | (2) CPR | (2) .541 | |||||||
(3) LBR | (3) .420 | |||||||||
Alleyassin (2017) [ |
Between the 18th and 22nd day of their menstrual cycles (mid-luteal phase) before ICSI cycles | 4-mm diameter diagnostic sheath, continuous flow, rigid, 30° view (Karl Storz, Tuttlingen, Germany) | Did not undergo OH before ICSI cycles | No scratching | ICSI | Fresh embryo / day 3 | Routine OH before ICSI cycles provided direct evaluation of uterine cavity. | (1) CPR | (1) .004 | Not specified |
Distension medium: Saline | CPR improved after correction of endometrial cavity abnormalities. | (2) Miscarriage rate | (2) NS | |||||||
El-Toukhy (2016) [ |
Before controlled ovarian stimulation for IVF | 2.9-mm diameter, rigid 30° view, with an atraumatic tip (TROPHY scope; Karl Storz) | Immediate controlled ovarian stimulation for IVF/ICSI | No scratching | IVF (with or without ICSI) | Fresh embryo / | Routine OH did not improve IVF outcomes in women with RIF who had a normal uterine ultrasound scan. | (1) Pregnancy rate | (1) .86 | No hysteroscopy-related adverse events |
Within 14 days of menstruation | Distension medium: saline | When it is considered top quality (day 2 or days 3–4 or days 5–6) | (2) CPR | (2) .65 | ||||||
(3) LBR (after 1 cycle of IVF) | (3) .96 | |||||||||
Smit (2016) [ |
In the early-mid follicular phase of a menstrual cycle (days 3–12) | 5-mm outer diameter continuous flow hysteroscope with a 5-Fr working channel and a 30° direction of view | Immediate start of IVF | No scratching | IVF | Fresh embryo / not specified | Routine OH before the first IVF or ICSI treatment cycle did not improve fertility prospects in infertile women with a normal TVS of the uterine cavity who had not had a previous hysteroscopy. | (1) Implantation rate | (1) .23 | One (<1%) woman: endometritis after hysteroscopy |
1–3 months before the start of IVF treatment | Distension medium: saline | (2) CPR | (2) .71 | |||||||
(3) OPR | (3) .69 | |||||||||
(4) LBR | (4) .75 | |||||||||
Pabuçcu (2016) [ |
In early follicular phase (1–6 months before the beginning of a new cycle) | 4-mm outer diameter, rigid, continuous flow; 30° forward and oblique view | Immediately started a new ART cycle | No scratching | IVF/ICSI | Fresh embryo / day 3 or day 5 | Unrecognized intrauterine pathologies can be easily detected and concurrently treated during the OH procedure with high success rates. | (1) Implantation rate | (1) .38 | Not specified |
Distension medium: saline | The overall beneficial impact in terms of reproductive outcomes seems to depend on the extent of the pathology. | (2) Chemical pregnancy rate | (2) .08 | |||||||
(3) LBR | (3) .06 | |||||||||
(4) Miscarriage rate | (4) .26 | |||||||||
Hosseini (2014) [ |
In the menstrual cycle just before ovarian stimulation or endometrial preparation | 4-mm rigid, continuous flow, 30° forward, and oblique view | Hysteroscopy was not performed | No scratching | ART IVF/ET | Fresh or frozen embryo / day 3 | OH before fresh cycles and frozen-thawed cycles in women experiencing RIF with apparently normal uterine cavity significantly increased the pregnancy rates. | (1) Chemical pregnancy rate | (1) <.001 | Not specified |
Distension medium: saline | (2) CPR | (2) .001 | ||||||||
(3) Delivery rate | (3) .026 | |||||||||
Kilic (2013) [ |
Assessed prior to IVF | 4-mm (Karl Storz) | Underwent IVF without OH evaluation | No scratching | IVF | Not specified | OH before IVF can detect and treat intrauterine pathologies, with positives effect on pregnancy outcomes. | (1) LBR | (1) <.05 | Not specified |
Follicular phase (days 5–7 of menstrual cycle) | Distension medium: saline | |||||||||
Shawki (2012) [ |
The early postmenstrual period before controlled ovarian stimulation for ICSI | 3.5 mm with a 0° grade (Versascope; Gynecare, Ethicon, Sommerville, NJ, USA) | Immediate controlled ovarian stimulation for ICSI | Endometrial biopsy | ICSI | Fresh embryo / not specified | Improvement in implantation and CPR were observed after OH prior to ICSI. | (1) CPR | (1) <.05 | Not specified |
Optic Illumination (250-W Xenon light source) | Routine OH should be an essential step of the infertility workup before ART even in patients with normal HSG and/or TVS. | (2) Implantation rate | (2) <.05 | |||||||
Distension medium: saline | ||||||||||
Makrakis (2009) [ |
Less than 12 months before the first IVF attempts | 2.9-mm, 30° angle, external sheath of 5.5-mm diameter providing inflow and outflow (Karl Storz) | Matched control (no hysteroscopy before IVF cycles) | No scratching | IVF | Fresh or frozen embryo / day 3–5 | Hysteroscopy could be seen as a positive prognostic factor for achieving a subsequent IVF pregnancy in women with a history of two consecutive implantation failures. | (1) CPR | (1) .04 | Not specified |
Shortly after cessation of menses | Distension medium: saline | (2) OPR | (2) .06 | |||||||
Rama Raju (2006) [ |
The early proliferative phase before controlled ovarian stimulation for IVF treatment | 5-mm diameter, 1.9-mm miniature, 30° view, 3 mm Bettocchi continuous flow sheath with an incorporated 5-Fr working channel (Karl Storz) | Immediate controlled ovarian stimulation for IVF treatment | Endometrial biopsy | IVF | Fresh embryo / day 3 | Patients with recurrent IVF-ET failures after normal HSG should also be reevaluated using hysteroscopy prior to commencing IVF-ET cycles in order to enhance the CPR. | (1) CPR | (1) <.05 | No further complications |
Distension medium: glycine | (2) Miscarriage rate | (2) NS | ||||||||
(3) LBR | (3) <.05 | |||||||||
Demirol and Gurgan (2004) [ |
The early proliferative phase before controlled ovarian stimulation for IVF treatment | 5-mm continuous flow, lens diameter 2.9-mm, 30° view, 5-mm diameter sheath, Bettocchi, size 5 (Karl Storz) | Immediate controlled ovarian stimulation for IVF treatment | No scratching | IVF | Fresh embryo / day 3 | Patients with normal HSG but recurrent IVF-ET failure should be evaluated prior to commencing IVF-ET cycles to improve the clinical PR. | (1) Number of clinical pregnancies | (1) <.05 | Mild pain resembling menstrual cramps |
(2–6 months after the last failed IVF cycles) | Distension medium: saline | (2) Number of first trimester abortions | (2) NS |
ART: Artificial reproductive technology; C: control; CPR: clinical pregnancy rate; ET: embryo transfer; Fr: French guage; HSG: hysterosalpingography; I: intervention; ICSI: intracytoplasmic sperm injection; IVF: in vitro fertilization; LBR: live birth rate; NS: not significant; OH: office hysteroscopy; OPR: ongoing pregnancy rate; PR: pregnancy rate; RIF: recurrent implantation failure; TVS: transvaginal sonography.