Introduction

Endometriosis is a benign gynecological pathology characterized by the existence of endometrium outside the uterine cavity and was commonly diagnosed by surgery [1]. Recently, ESHRE in 2022 stated that diagnosis of endometriosis with laparoscopy showed to be restricted to only to those with negative different imaging findings, and women with failed empirical treatment and the group stated that laparoscopy is no longer considered as the gold standard for diagnosis of endometriosis [2]. Endometriosis is an estrogen-dependent chronic inflammatory pathology that affects between 10 and 15% of women during their childbearing period. It is commonly associated with pain, infertility [3, 4], chronic stress, and anxiety [5]. Difficult conception is observed in 30 to 50% of women diagnosed with endometriosis [6], and many of them are trying to conceive through different assisted reproductive techniques [7].

Unfortunately, the pathogenesis of endometriosis associated with infertility remains unclear. Endometriosis causes infertility through various mechanisms: distortion of the normal pelvic anatomy, scarring of the fallopian tubes, inflammation of different pelvic organs with adhesion formation, alteration of immune response and hormonal environment of ova, impairment of implantation of a pregnancy, and alteration of oocytes quality [8]. Various inflammatory changes have been proposed to be the reasons behind endometriosis-associated infertility including alteration of macrophages proliferation and its phagocytic activity, increasing the numbers of malfunctioning natural killer cells and T lymphocytes with enhancement of proinflammatory and angiogenic cytokines release [9, 10]. Ovarian endometrioma decreases the volume of functioning ovarian tissue through its space-occupying action and/or the local inflammatory and immune reactions or both. This reduction in functioning ovarian tissue is aggravated by surgery. Clinical examination has low sensitivity and specificity for diagnosis of endometriosis, and laparoscopy remains the gold standard for diagnosis; however, recent studies look promising for new sonographic and magnetic resonance imaging (MRI) techniques [11].

Endometrioma could be managed by either medical or surgical modalities, and both affect the reproductive potential of the women. Women should be counselled about behavioral changes that creates optimum patients’ characteristics and healthy lifestyle before they started medical and/or surgical treatments. One of these is the maintenance of ideal body weight to decrease the endometriosis cancer risk and to improve the success of IVF [2]. The best management of endometriomas before starting IVF/ICSI trial is unknown. Medical treatment for endometrioma includes mainly hormonal therapy with progestogens, combined oral contraceptives, aromatase enzyme inhibitors, and gonadotropin-releasing hormone analogues. Their mode of action depends on their ability to decrease ovarian activity [12]. Surgical treatment includes cyst aspiration, laparoscopic ovarian cystectomy, ovarian cystectomy via laparotomy, or robotic surgery. Laparoscopic management is currently the most accepted approach being associated with lower costs and faster recovery compared to other approaches [13]. The aim of current review is to evaluate the impact of medical and surgical interventions to endometriomas prior to IVF-ET as there was no similar review to assess the effects of different treatment modalities of ovarian endometriomas on IVF outcomes.

Methodology

This study protocol was prospectively registered following the guidelines of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) with CRD42020151736 number.

Research question

  • Population: Women with unilateral or bilateral ovarian endometrioma

  • Intervention: Surgical (including cystectomy and aspiration) and medical (including GnRH, progesterone, and aromatase inhibitors) management of endometrioma before IVF

  • Comparison: Surgical treatment and medical treatment compared to no intervention

  • Outcome: Outcomes of IVF cycles

In this systematic review, the following databases: MEDLINE, Web of Science, Embase, Cochrane Library, and the clinical trial registry were searched from inception up to June 2023. The search used the keywords as follows: endometriosis; endometrioma, ovarian endometriosis, and ovarian endometrioma; ovarian endometriosis; endometriotic ovarian cyst; endometrioma/surgery; endometrioma/hormonal treatment, GnRH agonist, letrozole, aromatase inhibitors, OCPs (oral contraceptive pills)), IVF/ICSI, and in vitro fertilization; intracytoplasmic sperm injection; assisted reproductive technologies; ART; oocytes; live birth; fertilization; and live birth rate, pregnancy rate, miscarriage rate, number of oocytes, cancellation rate, randomized controlled trial(s), case-controlled studies, and cohort studies. As only few randomized controlled trials were available, the authors agreed to include other types of studies as case controlled and cohort studies whether prospective or retrospective. Review articles and case reports, editorial opinion, and commentary were excluded. We analyzed the references and citations of all available studies (both primary and secondary), narrative and systematic reviews, abstracts of gynecology, and infertility seminars. We emailed the authors directly for any missing or unclear information. If necessary, all studies written in English comparing different surgical and medical modalities against no treatment or other modalities were included. Surgical modalities include laparoscopic cystectomy, open cystectomy, or cyst aspiration. Medical modalities include progestogens, aromatase inhibitors, GnRH agonists, or oral contraceptive pills (OCPs).

Cystectomy is the stripping of the cyst wall away from the healthy ovarian tissues through either laparoscopy or laparotomy. Aspiration procedure is the transvaginal aspiration of the cyst content guided by ultrasound. Medical treatment acts through antagonizing estrogen secretion and/or action.

Two authors (A. O. and A. M.) independently assessed the titles, abstracts, and the full articles and then extracted the data of the included studies, and disagreements were discussed further with other authors. Extracted data included study type, settings, participants characteristics, details of intervention, and outcome parameters.

The risk of bias was assessed using the Cochrane Handbook of Systematic Reviews recommendations for the 3 RCTs [14]. These recommendations include random sequence generation, allocation concealment, participants and outcome assessors blinding, incomplete data of outcomes, selective reporting, and other biases.

The outcome parameters of this review included live birth, clinical pregnancy, miscarriage, cancellation rates, number of MII follicles, duration of stimulation, total dose of gonadotropin used for induction, and number of obtained embryos.

Statistical analysis

For analysis of continuous and dichotomous data, the mean difference and odd ratio with 95% confidence interval (CI) analysis were used, respectively. The random effect model was used to calculate the effect size, and I2 statistic and Cochran’s Q test were calculated to evaluate the studies heterogeneity. Significant P-value was considered when < 0.05, and significant I2 was considered when > 40%. Analyses were done using the Review Manager (RevMan) version 5.4.1 (The Nordic Cochrane Centre, Cochrane Collaboration, 2020, Copenhagen, Denmark).

Results

The PRISMA flow chart is shown in Fig. 1.

Fig. 1
figure 1

Prisma flow chart

Full size image

Study characteristics

Table 1 describes the characteristics of the included studies.

Table 1 Characteristics of the included studies
Full size table

Eighteen studies were included in this review [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]. Five were conducted in Turkey [15, 16, 18, 25, 26], 3 in Japan [19, 27, 28], and 1 in each of the following countries: Canada [31], China [21], Denmark [22], Egypt [23], France [32], Italy [30], Russia [24], Spain [17], South Korea [20], and the USA [29]. Three studies were RCTs [16, 23, 26], six were prospective cohort [21, 22, 24, 25, 30, 32], and nine were retrospective cohort studies [15, 17,18,19,20, 27,28,29, 31]. All were single center except three trials [17, 31, 32] that were conducted in two centers. Six studies compared laparoscopic surgery to no treatment [16,17,18,19, 21, 27], five compared surgery to no treatment [15, 20, 25, 26, 29], and three compared cyst aspiration to no treatment [20, 25, 27]. Medical treatment was GnRH in two studies [23, 24], OCPs in one study [15], and dienogest in one study [24], while aromatase inhibitors were studied in two trials [22, 31].

Risk of bias of included studies

Risk of bias in the three RCTs is described in Table 2. The risk of bias for non-RCTs was evaluated using the Newcastle–Ottawa scale (Table 3).

Table 2 Risk of bias
Full size table
Table 3 Risk-of-bias assessment of non-RCTs
Full size table

Synthesis of results

Eighteen studies with 3063 participants were included in our review.

Surgical intervention versus no treatment

Live birth rate was evaluated in 4 studies (565 participants). The OR effect estimate was 0.79 with [0.54, 1.18] 95% CI and 0.25 P-value (Fig. 2).

Fig. 2
figure 2

Live birth rate (surgical intervention vs. no treatment)

Full size image

Clinical pregnancy rate was evaluated in 11 studies (1447 participants). The OR effect estimate was 0.95 with [0.72, 1.26] 95% CI and 0.73 P-value (Fig. 3).

Fig. 3
figure 3

Clinical pregnancy rate (surgical intervention vs. no treatment)

Full size image

Miscarriage rate was evaluated in 6 studies (333 participants). The OR effect estimate was 0.74 with [0.33, 1.63] 95% CI and 0.45 P-value (Figure S1).

Cycle cancellation rate was evaluated in 4 studies (551 participants). The OR effect estimate was 1.62 with [0.57, 4.66] 95% CI and 0.37 P-value (Figure S2).

Number of MII oocytes was evaluated in 11 studies (1475 participants). The mean difference (MD) effect estimate was − 0.53 with [− 1.04, − 0.01] 95% CI and 0.04 P-value (Figure S3).

Number of obtained embryos was evaluated in 4 studies (569 participants). The MD effect estimate was − 0.25 with [− 0.38, − 0.11] 95% CI and < 0.001 P-value (Figure S4).

Duration of stimulation was evaluated in 6 studies (881 participants). The MD effect estimate was 0.19 with [− 0.42, 0.81] 95% CI and 0.54 P-value (Figure S5).

The total dose of gonadotropins was evaluated in 9 studies (1255 participants). The MD effect estimate was 361.14 with [168.13, 554.15] 95% CI and < 0.001 P-value (Figure S6).

Medical intervention versus no treatment

Ongoing rate was evaluated in 2 studies (270 participants). The OR effect estimate was 3.39 with [1.83, 6.26] 95% CI and < 0.001 P-value (Figure S7).

Clinical pregnancy rate was evaluated in 3 studies (330 participants). The OR effect estimate was 3.36 with [2.01, 5.63] 95% CI and < 0.001 P-value (Figure S8).

Miscarriage rate was evaluated in 2 studies (270 participants). The OR effect estimate was 1.31 with [0.46, 3.72] 95% CI and 0.61 P-value (Figure S9).

Cancellation rate was evaluated in 2 studies (270 participants). The OR effect estimate was 0.48 with [0.21, 1.10] 95% CI and 0.08 P-value (Figure S10).

Number of MII oocytes was evaluated in 3 studies (330 participants). The mean difference (MD) effect estimate was 2.04 with [0.72, 3.36] 95% CI and 0.003 P-value (Figure S11).

Subgroup analysis of all outcomes based on the type of surgical and medical intervention is shown in supplementary Table S1.

Candiani and colleagues (2020) [30] conducted a retrospective study on 142 women diagnosed with symptomatic endometrioma who were subjected to laparoscopic cyst stripping or CO2 cyst vaporization and then tried to conceive spontaneously and if failed through IVF. They reported pregnancy rates of 72.2% vs 74.3% (55.6% vs. 35.9% spontaneously;16.7% vs.38.5% through IVF) in stripping vs. laser respectively (P = 0.83), and 26.7% do not achieve pregnancy. They identified age and duration of infertility as independent indicators for pregnancy. They concluded that the pregnancy rate was not different between the two groups and suggest that CO2 laser one-step technique could be a good alternative to conventional cystectomy.

Cantor et al. in 2019 [31] conducted a retrospective study on 126 women with endometrioma and had a history of previous IVF cycle. Participants were subjected to either two doses of intramuscular 3.75-mg intramuscular depo-leuprolide with 1-month interval or daily 5-mg oral letrozole for 60 days along with the same leuprolide treatment before fresh IVF cycle. They reported a significantly higher AFC (10.3 ± 2.0 vs. 6.4 ± 2.5; P = 0.0001), lower required total doses of Gn (2079 ± 1119 versus 3716 ± 1314; P = 0.0001), higher number of mature oocytes (9.1 ± 2.4 versus 4.0 ± 1.7; P = 0.0001), clinical pregnancy rate (50% versus 22%, P = 0.003), and live birth rate (40% versus 17%, P = 0.008) in letrozole GnRH group compared to GnRH only group, respectively.

Lossl et al. in 2009 [22] conducted a prospective single arm study on 20 women candidate for IVF and diagnosed with endometrioma. They were subjected to three subcutaneous injections of 3.6 goserelin every 28 days in addition to daily oral tablet of 1-mg anastrozole for 70 days. They reported a significant decrease in endometriomal volume by 29% (P = 0.007) and serum CA125 by 61% (P = 0.001).

Discussion

In this review, women with untreated endometrioma had significantly higher numbers of MII oocytes (P = 0.04), higher number of obtained embryos (P < 0.001), and required lower doses of gonadotropins for induction (P < 0.001) compared to those who had undergone surgical management of endometrioma. However live birth (P = 0.25), clinical pregnancy (P = 0.73), miscarriage (P = 0.45), cancellation rates (P = 0.37), and the duration of stimulation (P = 0.54) did not show any significant difference between the two groups of women.

Our systematic review also confirmed that hormonal treatment of endometrioma was associated with higher ongoing pregnancy rate (P < 0.001), higher clinical pregnancy rate (P < 0.001), and higher numbers of MII oocytes (P = 0.003) when compared to women who did not receive such therapy. These effects were evident in treatment with GnRH agonists, OCPs, and dienogest, while the miscarriage (P = 0.61) and cycle cancellation rates (P = 0.08) did not show these differences.

Given the thorough search strategy and clear inclusion and exclusion criteria, this review provides a comprehensive overview of the current scientific evidence regarding surgical and medical treatment of endometriomas prior to IVF/ICSI. All 18 included studies included various comparisons, different study designs, and heterogenous reporting of data. This marked heterogeneity in study design, sample size, included population characteristics, intervention differences in type and timing, and different treatment modalities after intervention completion prevents the proper meta-analysis reporting of different outcome parameters. The review presented two different comparisons either surgical or medical treatment options.

There are contradictory observations regarding the impact of endometriosis on ovarian responsiveness to gonadotropins during controlled ovarian stimulation and IVF. Some studies reported diminished ovarian response to COH in women with unilateral endometriomas [33, 34]. On the other hand, a systematic review and meta-analysis of nine RCTs reported difference between the ovary with endometrioma compared to the contralateral nonaffected ovary regarding the numbers of retrieved oocytes, MII oocytes, and obtained embryos. All were reported to be lower from the affected ovaries. They suggested different mechanism for these findings including changes in immune markers as IL-6, VEGF, and oxidative stress markers with resultant decrease in density and diameters of primordial follicles. However, there was no differences regarding clinical pregnancy and live birth rates [35].

The impact of surgical treatment for endometrioma before IVF remains a controversy. In a retrospective trial that involved 292 women with existing endometrioma, no history of surgery who were candidate for IVF, they reported lower numbers of antral follicles and required higher doses of Gn in women with existing endometriomas compared to those with previous surgery for endometriomas and absent endometrioma at time of IVF without any significant difference in live birth rate between them [36].

In a recent meta-analysis, there was a significantly lower number of retrieved oocytes and higher rate of cycle cancellation in women with existing endometrioma during IVF cycle with similar clinical pregnancy and live birth rates when compared to women without endometriomas [37, 38]. Hamadan and colleagues suggested that the presence of ovarian endometrioma exerts a negative effect on the ovarian tissue with resultant decrease in number and quality of retrieved oocytes and increase in baseline FSH level [37].

On the other hand, a retrospective study reported a significantly higher clinical pregnancy and live birth rates trial in women who underwent laparoscopic cystectomy for endometrioma when compared to those underwent diagnostic laparoscopy without resection before IVF [39]. In a large observational study that involved 825 women diagnosed with endometriosis-related infertility, there was a significantly higher overall pregnancy rates in women who underwent endometrioma surgical resection followed by IVF compared to those who underwent surgical resection without IVF, IVF without prior resection, or no treatment [40].

Furthermore, several studies have reported the adverse effects of surgical treatment of endometrioma on ovarian reserve markers as the reduction of serum AMH levels after surgery [41]. After surgery, the level of AMH is reduced by 34% 1 week after surgery and gradually improves to reach 65% of its preoperative level 3 months after the operation. Measurement of AMH after 1 year of surgery revealed similar level to that reported after 1 month of surgery. Bilateral ovarian cystectomy is associated with more damage to ovarian reserve [42].

Regarding the medical treatment before IVF, a systematic review included 19 studies with 1709 participants compared with dydrogesterone to other hormonal therapies. It concluded that dydrogesterone caused better relieve of dysmenorrhea and improved the pregnancy rate compared to gestrinone with less side effects. They also concluded that dydrogesterone decreased the recurrence rate compared to with GnRH-a treatment. There are insufficient data to compare the efficacy of dydrogesterone, letrozole and leuprolide acetate, and the traditional Chinese medicine remains [43]. A Cochrane overview concluded that a 3-month regimen with GnRH agonist before IVF improves the pregnancy rates [44].

Strengths and limitations

This review is the first systematic review to study the clinical parameters of the effects of both surgical and medical treatment of endometrioma before IVF cycles. We included all the available studies that included all types of interventions. Adequate data extractions and proper meta-analysis when possible was done. The main limitation of this review is the low quality of evidence because of the absence of adequate numbers of RCTs and the marked heterogeneity among the included studies.

Conclusion

In conclusion, the optimal approach for treating endometrioma prior to IVF is not clear yet due to lack of well-designed randomized controlled trials.