The Efficacy of Medical Treatment on Endometrioma: A Systematic Review and Meta-Analysis
Abstract
Background: Ovarian endometriomas have been shown to have a negative effect on fertility. There is a dilemma on the timing for surgery either before or after the fertility treatment. By delaying surgical treatment for infertile patients with endometriomas, medical treatment has become an important choice for these patients. The objectives of this review were to compare the efficacy of medical treatment on endometrioma size and endometriosis-associated pelvic pain (EAPP).
Methods: We performed a systematic review and meta-analysis examining women who have endometrioma and underwent medical therapy for endometrioma size reduction. The primary outcome measure was endometrioma size reduction. Secondary outcome measures EAPP.
Results: We included 14 studies for the meta-analysis. We performed a systematic review and meta-analysis examining women with endometrioma who underwent medical therapy for endometrioma size reduction. The primary outcome measure was endometrioma size reduction. Secondary outcome measure is EAPP. The majority of the studies were non-randomized controlled trials (RCTs; 9/14), and five were RCTs. Women who received medical treatment have a significant endometrioma size reduction in diameter compared to women not receiving any medical treatment or placebo (MD −9.66mm; 95% CI [−13.85, −5.46], three studies, 467 women, I2=96%) and a reduction in visual analog scale (VAS) for EAPP (MD −2.64; 95% CI [−3.31, −1.97], two studies, 338 women, I2=0%). Women who received dienogest (DNG) treatment have a significant endometrioma size reduction in diameter (MD −4.61mm; 95% CI [−9.08, −0.15], three studies, 220 women, I2=96%). Compared to women receiving other medical treatments, women receiving DNG treatment had more VAS reduction of EAPP (MD −0.46; 95% CI [−0.62, −0.31], four studies, 451 women, I2=85%).
Conclusions: The use of medical therapy is associated with endometrioma size reduction and a reduction in endometriosis-related pain when compared with no medical treatment given. With the availability of various medical options, surgery can thus be avoided to minimize the risk of damage to the ovarian reserves.
INTRODUCTION
Endometrioma is one of the primary types of endometriosis that is present in 17%–44% of women with endometriosis (Redwine, 1999; Vercellini et al., 2003). Studies have shown that ovarian endometriomas can have a negative effect on fertility by interrupting the rate of spontaneous ovulation, reducing the amount of follicular number and activity in the adjacent ovarian tissue (Benaglia et al., 2009) or by structural tissue alterations in the ovarian cortex leading to reduced ovarian reserve (Lee et al., 2020). Meta-analyses from our group showed that when endometrioma coexists during IVF/ICSI, it reduces the chance of pregnancy (Hamdan et al., 2015). This is partly caused by surgical treatment that has a deleterious effect on ovarian reserves as proven by a reduction in the mean AMH level in a study by Hirokawa et al. (2011) and a meta-analysis by Raffi et al. (2012).
Over time, there has been a significant change in how ovarian endometriomas are managed, moving from primarily surgical to medical treatment. The key challenge lies in deciding when to opt for complete excision surgery for ovarian endometriomas, typically after fertility treatment concludes. With the delay in surgical treatment, medical treatment has become an important modality in managing patients with endometriosis and endometriomas.
Medical treatment is generally effective in relieving endometriosis-associated pelvic pain (EAPP). Various types—combined oral contraceptive pills (COCP), progestins, danazol, aromatase inhibitors, and GnRH agonists—have been used for pain control, with varying side effects (Agarwal and Foster, 2015; Davis et al., 2007; Prentice et al., 2000a,b; Selak et al., 2007). Several studies (Agarwal and Foster, 2015; Harada et al., 2009; Taniguchi et al., 2015, 2017; Tsujioka et al., 2009) showed a reduction in the size of endometriomas after ovarian suppression with GnRH-agonists, danazol, aromatase inhibitors, COCP, or progestins. The decrease in the size of endometrioma varied from 11% to 50% in diameter or 35%–75% in volume. Recent studies exploring the use of dienogest (DNG) (Angioni et al., 2020; Del Forno et al., 2019; Harada et al., 2017; Shume et al., 2021; Takenaka et al., 2015) compared with other medical treatments also noted a reduction in the size of endometriomas by 5%–40% in diameter and up to 75% in volume reduction.
The availability of different types of medical treatment and the effects on endometrioma size reduction require further clarification. Hence, we performed a systematic review and meta-analysis with the following objectives: (i) to compare the effect of medical treatment on endometrioma size and (ii) to compare the effect of medical treatment on EAPP. The primary outcome was endometrioma size reduction either in diameter size or volume of endometrioma; the secondary outcome was EAPP reduction based on the visual analog scale (VAS).
METHODS
Criteria for considering studies for this review
Type of studies and participants
Published cohort or case-control studies (retrospective or prospective) and randomized controlled trials (RCTs) were eligible for inclusion. The included studies had: (i) women diagnosed with endometrioma, (ii) have one or more endometrioma, (iii) unilateral or bilateral, (iv) newly diagnosed endometrioma or recurrent endometrioma, and (v) underwent medical therapy for endometrioma size reduction. The endometriomas were diagnosed by imaging modalities or through surgery. Studies were excluded if: (i) the participants had received any known non-surgical treatment (medical management or alternative treatment) before medical therapy or (ii) no control groups. We considered appropriate control groups to be: (i) women with endometrioma who received no treatment/placebo versus any medical treatment or (ii) women with endometrioma who received either one of the two medical therapies.
Types of interventions and outcome measures
Trials comparing active intervention versus no treatment/placebo or comparison between two different medical therapies will be eligible for inclusion. The primary outcome measure was endometrioma size reduction either in diameter size or volume of endometrioma. Secondary outcome measures EAPP reduction based on the VAS in the scale of 10.
Search methods for the identification of studies
A systematic search was performed on all published studies from January 1990 to December 2023 on medical treatment of endometrioma on and the size reduction outcome, without language restriction. The following electronic databases, trial registers, and websites were searched: MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, CINAHL, and PubMed. A search strategy was carried out based on the following keywords and/or medical subject heading (MeSH) terminology: endometriosis, endometrioma, and medical treatment. Reference lists of all primary and review articles were handsearched, and experts in the field were contacted to obtain additional articles not captured in the electronic searches. Relevant journals and conference abstracts that were not covered in the databases were also handsearched. The study was registered with PROSPERO (registration ID CRD42024497876).
Data collection and analysis
Selection of studies
After an initial screen of titles and abstracts retrieved by the search (conducted by Y.C.L.), the full texts of all potentially eligible studies were retrieved. Two review authors (Y.C.L. and S.H.) independently examine these full-text articles for compliance with the inclusion criteria and select studies eligible for inclusion in the review. Study investigators were contacted if clarification was needed for study eligibility. Any disagreements were resolved by a third author (M.H.). The selection process is documented with a “PRISMA” flowchart (Fig. 1).

Fig. 1. PRISMA flow diagram.
Data extraction and management
Two review authors (Y.C.L. and S.H.) independently extracted data from eligible studies using a data extraction form designed and pilot-tested by the authors on two different occasions. Any disagreements were resolved by discussion with a third author (M.H.). Data extracted included study characteristics and their outcome data. Both reviewers counterchecked these extracted data repeatedly. Where studies had multiple publications, the latest and main trial report was used as the reference and additional details were scanned from the secondary or earlier papers. Study investigators were contacted for further data and/or results, as required. All the available data were extracted into RevMan5 for further analysis.
Assessment of risk of bias in included studies
Two review authors (Y.C.L. and S.H.) independently assessed RCTs in the included studies for risk of bias using the Cochrane risk of bias assessment tool (www.cochrane-handbook.org) to assess: allocation (random sequence generation and allocation concealment); blinding of participants and personnel, blinding of outcome assessors; incomplete outcome data; selective reporting; and other bias. Any disagreements were resolved by discussion with a third author (M.H.). All judgments were described fully and conclusions were presented in the risk of bias table.
Quality assessment of the data
Two review authors (Y.C.L. and S.H.) assessed the methodological quality of the studies and extracted relevant data such as diagnosis of endometrioma, medical treatment, selection of controls, and definition of primary and secondary outcomes. Where available, we extracted statistical data from the original papers or calculated missing parameters by using the data provided. The quality of individual studies was assessed in accordance with the Newcastle–Ottawa Scale. By using the Newcastle–Ottawa Scale (Wells, 2010), non-randomized studies were rated according to eight items categorized in three domains: study group selection, comparability of the groups, and ascertainment of outcome (maximum scores of 4, 2, and 3, respectively). Scores were represented with stars for each quality item to provide a visual assessment. Studies were awarded up to nine stars if they fulfilled all the quality items.
Comparative analysis
Where available, for all outcomes, we compared medical treatment versus no treatment/placebo and different types of medical treatment.
Assessment of heterogeneity
We considered whether the clinical and methodological characteristics of the included studies were sufficiently similar for meta-analysis to provide a clinically meaningful summary. Statistical heterogeneity was assessed by the measure of the I2. Scores below 50% were considered to represent low or moderate heterogeneity, whereas I2 equal to or greater than 50% was taken to indicate substantial heterogeneity. The random effects model was chosen a priori to pool the results from individual studies given the increased clinical heterogeneity of the population assessed and the duration of treatment.
Data synthesis
Measures of treatment effects
For continuous data (e.g., endometrioma size reduction, VAS score), if all studies report exactly the same outcomes, mean differences (MDs) among treatment groups were calculated. And 95% confidence intervals for all outcomes were presented.
RESULTS
Results of the search
The search retrieved 3,346 articles; however, 3,263 articles were excluded because it was clear from the title that they did not fulfill the selection criteria. Another 46 articles were excluded after 83 articles were screened through the abstract. Thirty-seven studies were potentially eligible and were retrieved in full text. We further excluded 23 studies that had no available control groups for comparison. A final number of 10 studies were included in the meta-analysis as they had similar groups of treatments for comparison.
Description of the studies and participants
The majority of the included studies (Table 1) were non-RCT (Angioni et al., 2020; Del Forno et al., 2019; Ferrero et al., 2014; Piacenti et al., 2021; Porpora et al., 2013; Shume et al., 2021; Takenaka et al., 2015; Tsujioka et al., 2009; Xholli et al., 2020; n=9) and the remaining studies (n=5) were RCTs (Hamid et al., 2014; Harada et al., 2009, 2017, 2022; Takeuchi et al., 2000).
![]() |
Twelve (n=12) studies compared two different medications used. Seven (n=7) studies compared DNG and various medical treatments used: (i) Levonorgestrel+ethinylestradiol (LNG/EE; n=1), (ii) DNG+EE (n=2), (iii) norethindrone acetate (NETA; n=1), (iv) leuprorelin (n=1), (v) intranasal buserelin (n=1), and (vi) cabergoline (n=1). Another five (n=5) studies compared (i) relugolix and leuprorelin (n=1), (ii) NETA and NETA+letrozole (n=1), (iii) cabergoline and decapeptyl (n=1), (iv) Danazol and GnRH analog, and (v) buserelin and leuprorelin (n=1). Three (n=3) studies compared women receiving medical treatment versus no treatment or placebo. These medical treatments were; (i) hormonal therapy containing DNG (or DNG/EE; n=1), (ii) EE+drospirenone (n=1), and (iii) N-acetylcysteine (n=1).
Outcomes of study
Among all the included studies, reported outcomes on endometrioma size reduction were as follows: seven studies reported reduction in diameter, four studies reported reduction in volume, and three studies reported reduction in both diameter and volume. Outcome on VAS on endometriosis-associated chronic pelvic pain was reported in 10 studies. Outcome on number of endometriomas, Anti-Mullerian hormone level, antral follicle count, revised-ASRM score, number of capsules containing follicles, mean number of follicles attached to a cyst, estradiol level, CA 125 level, and bone mineral density were reported in individual studies, however, these outcomes were unable to be compared.
Twenty-three studies were excluded from the review for the following reasons: all the studies reported no comparison in interest (see Appendix).
Quality assessment of studies
Systematic risk assessment of methodological bias of five included RCTs revealed that one study has a high risk of bias in the randomization process (Takeuchi et al., 2000). In bias due to deviations from intended intervention, two studies (Hamid et al., 2014; Takeuchi et al., 2000) revealed to have a high risk of bias. There were some concerns in the bias due to missing outcome data in one study (Takeuchi et al., 2000). In summary, overall judgment for one study (Hamid et al., 2014) showed some concerns, one study (Takeuchi et al., 2000) showed high risk of bias, and the remaining three studies (Harada et al., 2009, 2017, 2022) were low risk (Fig. 2a). By using the Newcastle–Ottawa Scale (Wells et al., 2010), majority of the non-randomized studies were awarded with eight stars, whereas one study was awarded with seven stars (Table 2).

Fig. 2. Risk of bias using the Cochrane risk assessment tool for RCT: (A) traffic light plot and (B) summary plot.
![]() |
Impact of medical treatment on endometrioma size reduction
i. Medical treatment versus no active intervention or placebo
Women who received medical treatment have a significant endometrioma size reduction in diameter compared to women not receiving any medical treatment or receiving placebo (MD −9.66mm; 95% CI [−13.85, −5.46], three studies, 467 women, I2=96%). There was no significant reduction in endometrioma volume in women receiving medical treatment compared to no treatment/placebo (MD −17.01mL; 95%CI [−40.17, 6.15], two studies, 208 women, I2=98%; Fig. 3.1).

Fig. 3.1. Forest plot of comparison: medical treatment versus no treatment/placebo, outcome: endometrioma size reduction.
ii. Dienogest versus different types of medical treatment
Women who received DNG treatment have a significant endometrioma size reduction in diameter (MD −4.61mm; 95% CI [−9.08, −0.15], three studies, 220 women, I2=96%). There was no difference in endometrioma size in volume (MD −3.14mL; 95%CI [−15.00, 8.72], two studies, 151 women, I2=96%) when comparing DNG treatment with other medical treatments (Fig. 3.2).

Fig. 3.2. Forest plot of comparison: DNG versus other treatments, outcome: endometrioma size reduction.
Impact of medical treatment on EAPP
i. Medical treatment versus no active intervention or placebo
Women who received medical treatment had less pelvic pain compared to those without intervention or placebo (MD −2.64; 95% CI [−3.31, −1.97], two studies, 338 women, I2=0%; Fig. 4.1).

Fig. 4.1. Forest plot of comparison: medical treatment versus no treatment, outcome: VAS on endometriosis-associated pelvic pain.
ii. Dienogest versus different types of medical treatment
Compared to women receiving other medical treatments, women receiving DNG treatment had more VAS reduction of EAPP (MD −0.46; 95% CI [−0.62, −0.31], four studies, 451 women, I2=85%; Fig. 4.2).

Fig. 4.2. Forest plot of comparison: DNG versus other treatment, outcome: VAS on endometriosis-associated pelvic pain.
DISCUSSION
This meta-analysis demonstrated that there is a role in medical management to reduce endometrioma size compared to no active intervention. When comparing different types of medical treatments, there is a role of DNG in reducing the diameter of endometrioma size. Undoubtedly, the pain is much improved with medical treatment compared to those without treatment, with DNG treatment leading to more VAS reduction in EAPP.
Medical treatment remains a treatment option in managing endometriosis and endometrioma, especially for women who are keen to avoid surgery and still have further fertility wishes. These treatments consist of COCP, progestins, aromatase inhibitors, GnRH-analogs, cabergoline, danazol, and DNG. The availability of various treatment options allows clinicians to have more choices and at the same time, having dilemma from choosing the most suitable and effective treatment.
The use of medical treatment for EAPP has been well-established for many years. Various meta-analyses on different medical options proved the effectiveness of medical treatment in relieving the VAS score of dysmenorrhea, dyspareunia, and chronic pelvic pain associated with endometriosis. Samy et al. (2021) in their meta-analysis on medical treatment options for endometriosis-related pain summarized that DNG, combined hormonal contraceptives, and elagolix were the highest-ranked interventions for reducing the severity of pelvic pain at 3 months; while at 6 months, GnRH-analogs, levonorgestrel-releasing intrauterine system (LNG-IUS) and DNG were linked to more reduction in pelvic pain when compared with placebo. However, the authors did not compare the effects between two different groups of medical therapies to determine which treatment was the most superior. Another similar meta-analysis by Lin et al. (2021) looking into efficacy and safety of DNG in the treatment of endometriosis reported that DNG was more effective than placebo in alleviating endometriosis-related pain and is superior to GnRH-analogs in relieving pain with lesser effects in loss of bone mineral density, lower incidences of headaches, and hot flushes; however, they did not compare with other common medical therapies such as combined hormonal contraceptives or other progestins, which are also effective in relieving pain and with lesser side effects. Medical therapy is also equally effective when compared with surgical treatment as reported by Chaichian et al. (2017) in their systematic review and meta-analysis comparing the efficacy of surgery and medical therapy for pain management in endometriosis.
In this review, we found that women who were on medical treatment were associated with a slight endometrioma size reduction when compared to no medical treatment or placebo. On the other hand, when comparing DNG with other medical treatments, DNG in not superior to other medical treatments in reducing the size of endometrioma. Some articles only published the percentage of size reduction or the calculated differences and did not provide the measurements. Attempts were made to contact the authors via email to retrieve raw data, however, none responded/replied to the emails.
With regards to endometriosis-related pelvic pain, we found that medical treatment leads to a reduction in VAS score on endometriosis-related pain when compared to no treatment or placebo. However, the Forrest plot revealed that treatment favors other medical treatments when compared with DNG. This conclusion differs from the literature search that we found as the studies we used to compare were less when compared with other meta-analyses. In this study, the VAS score results were reported in two different scales (10mm vs. 100mm). The 100mm scale was divided by 10 to allow comparisons to be made among studies. The EAPP was also reported differently. Some studies reported as general EAPP while other studies reported in detail, that is, dysmenorrhea, dyspareunia, and chronic pelvic pain. We decided to choose chronic pelvic pain to compare the data among studies with EAPP.
The difficulty found in this meta-analysis is the heterogeneity in the data and the treatment used. Further studies comparing similar groups should be encouraged in the future to allow better comparisons to be made to produce more quality evidence. The emergence of newer agents in the management of endometriosis, which includes oral GnRH antagonists (elagolix and relugolix) can also be included in future studies.
CONCLUSIONS
The use of medical therapy is associated with endometrioma size reduction and a reduction in endometriosis-related pain when compared with no medical treatment given. With the availability of various medical options, surgery can thus be avoided to minimize the risk of damage to the ovarian reserves.
AUTHORS’ ROLES
M.H. conceived and designed the study, and drafted and revised the manuscript. Y.C.L. and S.H. developed the search strategy for the identification of articles, identified the articles, and acquired and analyzed the data. Y.C.L. drafted the manuscript. M.H. and N.R. revised the manuscript. All authors approved the final version of the manuscript.
CONFLICT OF INTEREST
None declared.
FUNDING
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
DATA AVAILABILITY
Data can be shared according to data protection legislation upon reasonable request to Y.C.L.
ORCID
Y. C. Lim https://orcid.org/0000-0001-6436-1858
S. Hassan https://orcid.org/0009-0002-8605-081X
N. Razali https://orcid.org/0000-0003-4667-9314
M. Hamdan https://orcid.org/0000-0002-1006-2614
APPENDIX
Characteristics of all the excluded studies in the systematic review. |
![]() |