Abstract

Endometriosis is an inflammatory disease that commonly occurs in women of reproductive age and is associated with pain and infertility. This disease can be challenging to manage given its propensity to progress and recur despite treatment. Although medical therapy is beneficial for controlling pain due to endometriosis, medical management has not proven to be effective in treating infertility resulting from endometriosis. Surgery has historically been performed to both improve pain and treat infertility in women with endometriosis. However, the optimal management of endometriosis in asymptomatic women who desire fertility is unclear. Intrauterine insemination with superovulation and IVF are other treatments that have proven to be effective in assisting women with endometriosis to conceive. As the underlying molecular mechanisms of this disease become better understood, promising new therapies for the treatment of endometriosis continue to be investigated.
Endometriosis is a benign, inflammatory disease defined by the presence of extrauterine endometrial tissue that is dependent upon estrogen [1]. It occurs frequently among women in the general population, with an estimated prevalence of 6–10% [1]. The most common symptoms of endometriosis include infertility, dyspareunia and chronic pelvic pain, and it is typically present on the pelvic peritoneum, the rectovaginal septum or the ovaries [1,2]. Although clinical findings and symptoms may suggest that a woman has endometriosis, a definitive diagnosis can only be made pathologically after surgery has been performed [3].
Various theories of pathogenesis have historically been used to explain how endometriosis develops. These include retrograde menstruation leading to implantation of refluxed endometrial tissue, the development of endometrial tissue from coelomic mesothelial cells that undergo metaplasia and lymphatic or hematogenous dissemination of endometrial cells [4]. Despite research efforts, the definite mechanisms responsible for the development of endometriosis have not been completely elucidated.
Advancements have suggested that endometriosis is not simply a local disease, but rather a chronic, multifaceted process [2]. As a result, its management can be challenging. Although traditional medical, surgical and infertility treatments continue to be employed in women with endometriosis, more recent research has focused on using therapies that target the disease at a molecular level [1,4]. This review will focus on the management of pain and infertility in women with endometriosis and will discuss the evidence behind various treatment strategies.
Pain
The most common symptom associated with endometriosis is pain. It is believed that endometriosis-related pain results from the effects of inflammatory cytokines within the peritoneal cavity, bleeding from endometriotic implants and the invasion and irritation of pelvic floor nerves [5,6]. Women with endometriosis may suffer from dyspareunia, dysmenorrhea or general cyclic or noncyclic pelvic pain. Both medical and surgical treatments are utilized to control the pain associated with endometriosis.
Medical treatment for pain
The goals with medical therapy are to minimize the production and action of estradiol, limit inflammation, inhibit the production of cyclic hormones from the ovaries and reduce menses [7]. The retrograde menstruation theory and the notion that endometriotic implants behave similarly to the normal endometrium have led to the use of medications that decrease cyclic menstruation while limiting the growth of the endometrial tissue [5]. Thus, combined oral contraceptives (COCs), progestogens, gonadotropin-releasing hormone (GnRH) agonists and danazol, a synthetic androgen, have historically been used to control symptoms. As endometriosis has become better understood at the molecular level, other medical therapies have also been studied to evaluate their effectiveness in treating endometriosis-associated pain. Ultimately, individual aspects of these medications, such as effectiveness, cost, side effects and long-term safety, influence which therapy is used.
Estrogen–progestogen contraceptives
COCs, used cyclically or continuously, are commonly prescribed as a first-line treatment for women suffering from mild endometriosis-related pain, although their mechanism of action is not completely understood. It has been proposed that by inducing a pseudopregnancy state, COCs cause decidualization and subsequent atrophy of the endometrium [8]. Other studies have suggested that alternative mechanisms may be responsible for improving the pain associated with endometriosis [9–11].
An early randomized trial by Vercellini et al. compared the effectiveness of a cyclic low-dose COC with a GnRH agonist in 57 women with moderate or severe endometriosis-related pelvic pain. At the completion of 6 months of treatment, there was a significant improvement in nonmenstrual pain in both groups of women, with no difference between the COC or GnRH agonist groups. There was also a significant decrease in deep dyspareunia in both groups, although the GnRH agonist group had more significantly improved pain scores compared with the COC group. In addition, there was a significant improvement in dysmenorrhea in the COC group. However, at the 6-month follow-up, symptoms recurred in both groups, with no significant difference between the groups [12]. In recent years, additional studies have been published evaluating the effectiveness of COCs in treating pain secondary to endometriosis. A double-blind, placebo-controlled, randomized trial of 100 women by Harada et al. compared a cyclic monophasic COC with placebo for the treatment of endometriosis-associated dysmenorrhea. The majority of women in this study only had radiologic evidence of endometriosis. After four cyclic treatments, dysmenorrhea scores were significantly decreased in both groups, but dysmenorrhea was significantly milder in the COC group compared with the placebo group [13]. More recently, in a prospective, randomized, double-blind controlled trial of 47 women, Guzick et al. compared a GnRH agonist with hormonal add-back with a continuous monophasic COC. Both treatments, administered for 48 weeks, significantly reduced pain from the baseline, but there was not a significant difference in pain improvement between the two treatment arms [14].
These studies suggest that COCs are as effective as GnRH agonists, and better than placebo, at treating endometriosis-associated pain. Their relatively low cost and long-term usability make these medications an attractive first-line therapy. However, COCs are contraindicated in certain women with coexisting medical disorders, and side effects may limit their use. To assist in determining when COCs should be avoided, the CDC adapted the WHO's Medical Eligibility Criteria for Contraceptive Use and published guidelines, which are summarized in Box 1 [101].
Progestogens
Progestogens are commonly prescribed to treat endometriosis-associated pain and are an option for women with contraindications to estrogen use. Although it has been postulated that these compounds are effective because they suppress endometrial estrogen receptors, which leads to the decidualization and subsequent atrophy of endometrial tissue [8], other studies have proposed alternate mechanisms of action that may improve endometriosis-related pain [9–11]. Suppression of matrix metalloproteinases in the endometrium may also contribute to their effectiveness [8]. Various progestogens are available for clinical use, including medroxyprogesterone acetate (MPA) and norethindrone, a 19-nortestosterone derivative [5].
The effectiveness of MPA in treating endometriosis-related pain was evaluated in a prospective, double-blind, placebo-controlled trial by Telimaa et al. A total of 59 women with mild–moderate endometriosis were randomized to continuous therapy with oral MPA, danazol or placebo for 6 months. Both MPA and danazol led to significant reductions in pelvic pain, lower back pain and pain with defecation when compared with placebo. Laparoscopy performed 6 months after treatment to evaluate peritoneal implants demonstrated partial or total resolution in 63% of the women who received MPA versus only 18% resolution in the placebo group, suggesting that MPA was superior to placebo in treating endometriosis [15]. Later, in a prospective, randomized, double-blind, placebo-controlled trial, Harrison and Barry-Kinsella also investigated the efficacy of MPA in treating endometriosis. A total of 100 women with infertility were surgically diagnosed with endometriosis and were randomized to receive either MPA or placebo for 3 months. Both groups had significant decreases in their revised American Fertility Society stages and scores at second-look laparoscopy, but there were no differences between the groups. However, there was a greater improvement in wellbeing in the women who received MPA when compared with the placebo group [16].
Contraindications to the initiation of estrogen–progestogen contraceptives.
Current breast cancer
History of DVT/PE with high risk for recurrent VTE
Acute DVT/PE
History of DVT/PE after major surgery with prolonged immobilization
Known thrombogenic mutation
Diabetic retinopathy/nephropathy/neuropathy†
Diabetes with other vascular disease or of longer than 20 years duration†
Migraine with aura at any age
History of cerebrovascular accident
Multiple risk factors for arterial cardiovascular disease†
Hypertension that is poorly controlled or with presence of vascular disease
History of, or current, ischemic heart disease
Peripartum cardiomyopathy that occurred within previous 6 months or that has led to moderately or severely impaired cardiac function
Complicated valvular heart disease
Smoking:
– ≥35 years of age
– >15 cigarettes daily
Complicated solid organ transplant
Systemic lupus erythematosus with positive or unknown antiphospholipid antibodies
Acute or flare of viral hepatitis†
Severe cirrhosis
Hepatocellular adenoma
Malignant liver tumor
Past history of breast cancer without evidence of current disease for 5 years
Less than 3 weeks postpartum and not breastfeeding
Less than 1 month postpartum and breastfeeding
History of DVT/PE with lower risk for recurrent VTE
Hyperlipidemia†
Migraine without aura in women >35 years
Adequately controlled hypertension
Peripartum cardiomyopathy more than 6 months ago
Smoking:
– >35 years of age
– <15 cigarettes daily
Inflammatory bowel disease with increased risk of VTE
Medically treated or current symptomatic gallbladder disease
Past COC-related history of cholestasis
On retroviral therapy with ritonavir-boosted protesase inhibitors
Use of certain anticonvulsants
Use of rifampicin or rifabutin therapy
History of bariatric surgery with malabsorptive procedures
† Assess according to the severity of the condition.
COC: Combined oral contraceptive; DVT: Deep vein thrombosis; PE: Pulmonary embolism; VTE: Venous thromboembolism. Data taken from [101].
Depot MPA (DMPA) has also been employed for the treatment of pain from endometriosis. Vercellini et al. randomized 80 women with moderate or severe pelvic pain to 1 year of treatment with DMPA or cyclic COCs plus low-dose danazol. After 12 months, both groups of women had significantly reduced deep dyspareunia and nonmenstrual pain scores with no significant differences noted between the groups. Dysmenorrhea was significantly reduced in both groups, but more significantly reduced in the DMPA group. Patient satisfaction was nonsignificantly higher in the DMPA group (72.5%) than in the COC plus danazol group (57.5%) [17]. In another randomized controlled trial, Schlaff et al. compared DMPA with a GnRH agonist in 274 women for 6 months. At the end of treatment, both therapies significantly reduced pain scores, and there were no significant differences between the therapies in reducing four of five pain-related symptoms. Over 60% of the women in each group had improvement in pain compared with baseline 12 months after treatment was complete, and there was no difference between groups for all five of the pain-associated symptoms [18].
The levonorgestrel-releasing intrauterine system (LNG-IUS), a derivative of 19-nortestosterone, is a newer therapy that is gaining popularity for the treatment of endometriosis-related pain. Through local release of hormone, endometrial tissue becomes atrophic and inactive [19]. In a pilot study by Vercellini et al., 40 women with moderate or severe dysmenorrhea were randomized to immediate LNG-IUS insertion or expectant management following operative laparoscopy for endometriosis. Moderate or severe dysmenorrhea recurred significantly less frequently in the women who received the LNG-IUS (10%) versus the expectantly managed women (45%). Of the women who received the LNG-IUS, 75% were satisfied or very satisfied after 1 year of treatment versus 50% of the women who did not [19]. Since then, studies have compared more traditional treatments with the LNG-IUS and have confirmed that it is a useful treatment modality [20–22]. Consequently, the LNG-IUS has become more commonly used for endometriosis-related pain, although it is not currently approved by the US FDA for this use.
Although progestogens are effective in controlling endometriosis-related pain for some women, the side effects of treatment can limit their use. Irregular bleeding, weight gain, bloating, fluid retention, breast tenderness, headaches, nausea and mood changes are among the reported side effects [23]. However, advantages, including less bone mineral density loss compared with GnRH agonists [18], improved compliance with the LNG-IUS [21] and lower cost, make progestogens an attractive option for certain women with endometriosis-associated pain.
Gonadotropin-releasing hormone agonists
GnRH agonists are similar to native GnRH, but are modified to remain bound to pituitary GnRH receptors for longer. The altered structure of these agonists prevents the stimulation from GnRH that normally occurs in a pulsatile fashion. This nonpulsatile binding results in an ultimate downregulation of the pituitary–ovarian axis. A hypoestrogenic state ensues leading to amenorrhea and endometrial atrophy [6,8]. GnRH agonists are approved by the FDA for the management of endometriosis, as they have proven to be effective in the treatment of pain due to the hypogonadotropic–hypogonadal state that is produced. GnRH agonists used in clinical practice include leuprolide, nafarelin, goserelin, buserelin and triptorelin [5]. These can be administered subcutaneously, intramuscularly or intranasally at various time intervals.
Many studies have evaluated the use of GnRH agonists for the treatment of endometriosis-associated pain. In a meta-analysis by Brown et al. that included 41 trials of 4935 women, the use of GnRH agonists for endometriosis-related pain was compared with various other strategies. One study compared GnRH agonists with no treatment and found that dysmenorrhea was significantly relieved in the treatment group (risk ratio: 3.93; 95% CI: 1.37–11.28; p = 0.01). Two additional studies compared GnRH agonists with placebo. There was a significant relief in pelvic tenderness in the GnRH agonist group (risk ratio: 4.17; 95% CI: 1.62–10.68; p = 0.003), but no significant difference in dyspareunia or defecation pressure between the two groups. There was also a significant temporary increase in symptom severity score during the stimulatory phase of therapy in the treatment group (mean difference: 2.90; 95% CI: 2.11–3.69; p < 0.001) [24].
Overall, GnRH agonists appear to improve endometriosis-related pain. However, side effects, including bone mineral depletion and vasomotor symptoms, which are related to the hypoestrogenic environment [6], limit the long-term use of these medications. The ‘estrogen threshold’ theory suggests that the amount of estrogen necessary to prevent these side effects is less than the amount needed to support endometriosis [25]. Therefore ‘add-back’ therapy using estrogen and progestogen or progestogen alone is commonly administered to help reduce these side effects and allow for longer-term use of GnRH agonists [6,8]. Hormonal add-back has proven to be effective in protecting against bone loss and suppressing vasomotor symptoms while continuing to control pain from endometriosis [26,27]. Although the usability of GnRH agonists for endometriosis is improved by add-back therapy, their universal use remains limited by their high cost and contraceptive properties.
Other treatments
Various other medical treatments have been, and continue to be, evaluated for the treatment of endometriosis-associated pain. However, due to undesirable side effects or unconfirmed effectiveness, current clinical use remains limited.
Danazol was the first drug approved by the FDA for the treatment of endometriosis. It is an isoxazol derivative of 17 α-ethinyltestosterone that creates a state of anovulation by inhibiting the LH surge [8]. Danazol also increases free testosterone and inhibits enzymes involved in steroidogenesis [8]. These effects lead to a hypoestrogenic, hyperandrogenic environment that limits growth of endometriosis [5]. A Cochrane review of five studies concluded that when compared with placebo, 6 months of danazol use resulted in significantly improved pain relief from endometriosis. However, side effects including muscle cramps, acne, edema and weight gain were more common in women using danazol [28]. These undesirable side effects result from the low estrogen, high androgen state produced by danazol and have limited its use.
Newer treatments to control pain, such as aromatase inhibitors (AIs), target specific molecular differences present in women with endometriosis. Aromatase is responsible for catalyzing the creation of estrogen in various tissues, including in endometriotic lesions where it serves to synthesize estradiol locally [2,29]. Given the essential role that estrogen plays in stimulating endometriosis, the goal of AIs is to eliminate its creation [29]. As aromatase is involved in the final step of biosynthesis of estradiol, it is a good target for inhibition [30]. The commonly used third-generation AIs are letrozole, anastrazole and examestande [30]. In premenopausal women, as AIs block extraovarian estrogen production, increases in FSH lead to the development of ovarian follicles. Therefore, additional treatments must be used in combination with AIs to downregulate the ovaries [30]. Studies that have examined the use of AIs for the treatment of endometriosis-related pain have demonstrated promising results [31–33]. Although a recent review has suggested that AIs are well tolerated overall [30] and a previous study has shown that AIs are safe for ovulation induction [34], additional trials are necessary to further confirm the efficacy and safety of these medications. In the future, these medications may have a more prominent role in the treatment of endometriosis-associated pain.
Investigational drugs that target molecular mechanisms are also being studied for the treatment of endometriosis-related pain. As the cytokine TNF-α is believed to be elevated in women with endometriosis, and likely plays a role in the increase in peritoneal macrophages that occur in this disease [8], the use of TNF-α inhibitors have been investigated. Pentoxifylline is an immunomodulator that has also been evaluated for the treatment of endometriosis-associated pain [8]. Thus far, studies that have examined the effectiveness of these drugs in treating pain from endometriosis do not provide enough evidence to support their routine use [35–37].
Surgical treatment
In certain women with endometriosis-related pain, medical treatment fails, side effects from medications become unbearable or fertility is desired. In these individuals, endometriosis is surgically managed. The goal with operative management is to restore normal anatomy, destroy visible disease and minimize recurrence [5]. Possible procedures include ablation, excision or drainage of endometriomas, lysis of adhesions, destruction of peritoneal implants through ablation and fulguration or excision [7].
Management of endometriosis
Previous trials have examined whether surgery is effective in treating women with pain from endometriosis. In a prospective, randomized, double-blind study by Sutton et al., laser ablation of minimal to moderate endometriosis was compared with diagnostic laparoscopy alone in 63 women. 6 months after surgery, there was significantly improved pain relief in women who underwent laser ablation (62.5%) when compared with those who had only diagnostic laparoscopy performed (22.6%) [38]. A follow-up study of these patients demonstrated that 90% of women whose pain was improved at 6 months had continued pain relief at 1 year [39]. After the randomization code was broken, 24 women who had originally been expectantly managed underwent laparoscopic laser ablation. Overall, a total of 56 women had laser ablation performed, and Jones et al. described longer term outcomes in 38 of these individuals. At a mean of 73 months after surgery, 73.7% of the women reported recurrence of pain, which occurred at a median time of 19.7 months. However, 55.3% of the women had satisfactory symptom relief [40]. The initial study by Sutton et al. was included in a meta-analysis of five studies that evaluated the efficacy of operative laparoscopy in the treatment of endometriosis-associated pelvic pain and found that it was beneficial when compared with diagnostic laparoscopy alone [41].
These findings suggest that operative management of endometriosis is effective in treating pain, but that recurrence is common. Although studies have confirmed that there is a role for surgery, the optimal surgical approach remains unclear [42].
Postoperative medical treatment
Some clinicians believe that postoperative medical therapy eliminates remaining disease or microscopic endometriosis that is unable to be removed surgically [7] and will treat women who suffer from more severe endometriosis-related pain with adjuvant medications. Furness et al. performed a meta-analysis to determine if postoperative medical treatment with GnRH agonists, danazol, progestogens and COCs improved endometriosis-associated pain and reduced disease recurrence. The studies suggest that postoperative medical therapy improves painful symptoms in women who are not interested in conceiving immediately after surgery, but given that medical treatments do not seem to alleviate pain after 1 year, long-term use may not be beneficial [43].
Management of endometriomas
The decision to proceed with surgical management of an endometrioma often depends on whether a woman has associated pain. If painful symptoms are present, surgery is warranted, as medical treatment is unlikely to completely resolve cysts [6]. The optimal surgical strategy for the treatment of endometriomas has been evaluated. In studies that have examined whether laparoscopic cystectomy or drainage and coagulation resulted in differences in pain relief and disease recurrence, cystectomy was the more effective surgical strategy. Pain and recurrence rates were lower after laparoscopic cystectomy when compared with other procedures [44,45]. However, the impact of cystectomy on fertility must also be considered. This subject will be discussed later in the review.
Management of deep infiltrating endometriosis
Although previous reviews have discussed the surgical management of deep infiltrating endometriosis in detail [46,47], this topic is beyond the scope of this review.
Infertility
Although a definite causal relationship has not been confirmed, endometriosis is associated with infertility. Endometriosis is found in 20–40% of infertile women, and it is suspected that multiple mechanisms contribute to decreased fertility in these women. Ovum transport may be impaired as a result of distorted pelvic anatomy. It is also theorized that the chronic inflammation that results from endometriosis may affect the receptivity of the uterus, folliculogenesis of the ovaries and proper functioning of the fallopian tubes [5]. Advanced endometriosis may impact fertility more significantly, as data suggest that an inverse relationship exists between disease severity and monthly fecundity rate [48]. Both medical therapies and surgical strategies have been employed in the treatment of endometriosis-related infertility. Other treatments, such as intrauterine insemination (IUI) and IVF, have also been used in infertile women with endometriosis.
Medical treatment
As medical therapy appears to be beneficial in controlling pain due to endometriosis, the use of medications to treat endometriosis-related infertility has also been investigated. A meta-analysis of 25 trials studied whether ovulation suppression agents improved pregnancy outcomes in women with infertility. When suppression with MPA, COCs, GnRH agonists, danazol or gestrinone was compared with placebo or no treatment, the odds ratio (OR) for pregnancy among infertile women or those trying to conceive was 1.02 (95% CI: 0.69–1.50; p = 0.22). Therefore, pregnancy outcomes did not significantly improve when women with endometriosis-related infertility were treated with ovulation suppression. In addition, as these medications inhibit pregnancy, their use actually delays a woman's ability to conceive. Thus, there is no role for medications as the sole therapy in the treatment of endometriosis-associated infertility [49].
Although other medical therapies such as AIs, progesterone antagonists, selective progesterone receptor modulators and selective estrogen receptor modulators have also been considered for the management of infertility, a recent committee opinion from the American Society for Reproductive Medicine explains that there is not enough evidence at this time to determine their efficacy [50].
Superovulation & IUI
Other strategies to treat infertility, including superovulation and IUI, have also been evaluated in women with endometriosis. In a randomized, prospective trial, Deaton et al. compared four cycles of clomiphene citrate and IUI with periovulatory intercourse in 51 women with surgically corrected endometriosis or unexplained infertility. There was a significantly higher per cycle fecundity in the women treated with clomiphene citrate plus IUI (0.095) than in women who did not receive the treatment (0.033) [51]. Tummon et al. later performed a randomized trial comparing superovulation with IUI to expectant management in 103 women with minimal or mild endometriosis. The women who underwent superovulation and IUI for four cycles had higher live-birth rates (11%) than those who were expectantly managed (2%). The OR of live birth was 5.6 (95% CI: 1.8–17.4), favoring treatment [52].
Later, a randomized controlled study of 932 infertile women, including individuals with stage I and II endometriosis, was performed to compare four cycles of intracervical insemination, IUI, superovulation with intracervical insemination and superovulation with IUI. The pregnancy rate was 33% in the superovulation/IUI group, 19% in the superovulation/intracervical insemination group, 18% in the IUI alone group and 10% in the intracervical insemination group [53]. Given these findings, superovulation with IUI can be used as one of the first-line therapies in younger infertile women with stage I/II endometriosis. In older women, superovulation or IVF may be considered [50].
IVF
Although it is not routine practice worldwide, throughout the USA, IVF is commonly performed in infertile women with endometriosis. According to data from the Society for Assisted Reproductive Technology, endometriosis was the indication for IVF in 4% of all cycles performed in 2011 [102].
The data from Society for Assisted Reproductive Technology suggest that IVF is an effective treatment for women with endometriosis-associated infertility. In women with endometriosis, 36% of retrievals resulted in live birth compared with 32% in women undergoing IVF for all indications [102]. These findings contrast results from a meta-analysis of 22 studies by Barnhart et al., which examined outcomes in women with endometriosis who underwent IVF. When women with endometriosis were compared with women with tubal factor infertility, the chance of pregnancy was significantly lower in the endometriosis group (OR: 0.56; 95% CI: 0.44–0.70). When pregnancy rates were compared between women with mild and severe endometriosis, they were significantly lower in those with severe disease (OR: 0.60; 95% CI: 0.42–0.87). Overall, the pregnancy rate in women with endometriosis who underwent IVF was 25% [54]. Despite the fact that this study demonstrated that pregnancy rates were lower in women with endometriosis than in women without endometriosis, IVF is the most successful assisted reproductive technology (ART) option that can be offered to women with this disease [54]. As it is believed that per cycle pregnancy rates in individuals with endometriosis are maximized with IVF [50], it continues to be one of the main treatment options for this group of women.
Some clinicians use medical therapy to treat endometriosis before a patient undergoes IVF. This practice has been studied to determine whether pregnancy outcomes are improved. Sallam et al. performed a meta-analysis of three randomized controlled trials of 165 women to investigate whether the use of a GnRH agonist for 3–6 months before IVF or intracytoplasmic sperm injection in women with various stages of endometriosis improved pregnancy rates. The clinical pregnancy rate per woman was significantly higher in those who received treatment when compared with those who did not (OR: 4.28; 95% CI: 2.00–9.15). The live-birth rate per woman, as reported in one study, was also significantly higher in those who received a GnRH agonist when compared with those who did not (OR: 9.19; 95% CI: 1.08–79.22) [55]. Given these results, medical pretreatment with a GnRH agonist may be warranted in women with endometriosis who will be undergoing IVF.
As IVF has developed into a more commonly performed procedure, not all women with endometriosis are surgically managed before undergoing IVF. Therefore, there may be a role for surgery in these women if they fail to become pregnant. In a retrospective case series, Littman et al. investigated whether women with endometriosis who were unsuccessful in achieving pregnancy with IVF were able to conceive after operative laparoscopy was performed. There were 29 women who opted to undergo laparoscopic treatment of endometriosis after an average of 2.2 failed IVF cycles. In this group of women, 76% conceived, with 59% of the conceptions being spontaneous. The rates of conception were highest in the women with stage I disease and lowest in stage IV disease. Pregnancy rates of these patients were significantly higher than the 37% conception rates of similarly matched women who failed to become pregnant through an average of 2.4 IVF cycles but declined surgical treatment of their endometriosis. In addition, the rates of spontaneous pregnancy were significantly higher in the surgically treated women than in the nonsurgically treated women (15%) [56]. These data suggest that surgery may increase the likelihood of conception in women with endometriosis who fail to become pregnant through IVF.
Surgical treatment
Outcomes of women who have undergone surgical management of endometriosis have been studied to determine whether fertility is improved postoperatively. The role of surgery in women with early and advanced stage disease has been evaluated. The impact of reoperation and postoperative medical therapy, as well as the benefits and risks of endometrioma removal, have also been examined.
Management of endometriosis
Two randomized controlled trials assessed whether laparoscopic surgery in infertile women with minimal-to-mild endometriosis improved pregnancy outcomes. Marcoux et al. randomized 341 women to resection or ablation of endometriosis or diagnostic laparoscopy and followed them for 36 weeks. Of the women who had their endometriosis treated surgically, significantly more carried pregnancies to at least 20 weeks (30.7%) compared with the women who only underwent diagnostic laparoscopy (17.7%) [57]. In another trial by the Gruppo Italiano, 101 women were randomized to resection or ablation of endometriosis or diagnostic laparoscopy and followed for 1 year postoperatively. There was no significant difference in pregnancy rates between the two groups, as 24% in the surgically treated group and 29% in the laparoscopy only group conceived [58]. Despite the fact that the studies reported different outcomes, a meta-analysis of these two trials determined that laparoscopic surgery was significantly beneficial in terms of live-birth rate and ongoing pregnancy rate after 20 weeks when compared with diagnostic laparoscopy in infertile women with endometriosis (OR: 1.64; 95% CI: 1.05–2.57). There was also a benefit to laparoscopic surgery for clinical pregnancy rates when compared with diagnostic laparoscopy (OR: 1.66; 95% CI: 1.09–2.51) [59].
There are no randomized controlled trials that have assessed whether operative management of advanced endometriosis in infertile women improves fecundity. However, observational data suggest that these women may benefit from surgery [50]. One study by Crosignani et al. reported pregnancy rates in 92 infertile women with severe endometriosis who actively attempted to become pregnant after undergoing surgical treatment with laparotomy or laparoscopy. After surgery, the 24-month cumulative pregnancy rates were 44.9% in the laparoscopy group and 62.7% in the laparotomy group [60].
Therefore, surgical management of endometriosis appears to benefit infertile women with earlier stage disease. Limited data suggest that there is also a role for surgery in women with more advanced disease.
Postoperative medical treatment
Postoperative medical therapy has also been evaluated to determine whether adjuvant treatment improves pregnancy rates in women with endometriosis. A meta-analysis of eight studies demonstrated that when postoperative medical treatment with GnRH agonists, MPA and danazol was compared with placebo or no treatment, there were no differences in pregnancy rates between the groups (risk ratio: 0.84; 95% CI: 0.59–1.18) [43]. Use of medical therapy also delays a woman's ability to conceive in the immediate postoperative period. Therefore, there does not appear to be a role for adjuvant medical therapy in women with endometriosis who are attempting pregnancy.
Reoperation
Recurrence of endometriosis is common, with up to a 15% rate of annual recurrence [61] and a 40–50% rate after 5 years [62]. Therefore, the fertility benefits of repeat surgery for recurrent endometriosis have been evaluated. A systematic review by Vercellini et al. reported that the pregnancy rates in women who underwent repeat surgery for recurrent endometriosis were lower than the rates in women who had primary surgery for endometriosis [63]. Therefore, the role of repetitive surgery for recurrent disease appears to be limited. In addition, evidence has suggested that IVF may be a more effective treatment strategy than a second surgery in women with recurrent endometriosis attempting to conceive [63].
Management of endometriomas
The optimal treatment strategy for women with endometriomas who desire fertility is controversial. Asymptomatic individuals may be expectantly or surgically managed, but there are concerns with both options. In addition, if surgery is performed, the preferred strategy for endometrioma removal is debatable.
The procedure resulting in the best spontaneous pregnancy outcomes was evaluated in a study by Beretta et al. In this trial, 64 women were randomized to undergo cystectomy or drainage and coagulation of endometriomas that were at least 3 cm in diameter. The cumulative pregnancy rates after 24 months were significantly higher in the cystectomy group (66.7%) than the drainage group (23.5%) [44]. Alborzi et al. also performed a prospective, randomized trial that included 62 infertile women to determine whether laparoscopic cystectomy or fenestration and coagulation of endometriomas 3 cm or greater in size resulted in different rates of spontaneous pregnancy. After 1 year of follow-up, women who underwent cystectomy had significantly higher cumulative pregnancy rates (59.4%) than the women who underwent fenestration and coagulation (23.3%) [45].
The most appropriate management for women with endometriomas who require ART is also highly debated. Possible difficulties in performing IVF in women with endometriomas include challenges with ultrasound monitoring, access issues or accidental drainage during oocyte retrieval, which may predispose patients to a pelvic infection and growth or spontaneous rupture of the endometrioma [64,65]. If endometriomas are not removed, there are also concerns about impaired fertility. An observational study by Somigliana et al. reported that women with endometriomas had a decreased ovarian response to gonadotropins [66]. However, other studies have failed to demonstrate this finding [67,68], suggesting that expectant management may not alter fertility. This was further supported by a meta-analysis by Benschop et al. that included three randomized trials that examined the effectiveness of various treatments for endometriomas before ART cycles. This study found that there was no benefit of surgical management compared with expectant management in terms of clinical pregnancy rates (aspiration vs expectant: Peto OR: 1.29; 95% CI: 0.45–3.64; cystectomy vs expectant: Peto OR: 1.15; 95% CI: 0.52–2.55) [69].
If operative management of an endometrioma is performed, the fertility risks should be considered. It has been hypothesized that surgery may be detrimental to ovarian function due to the removal of viable ovarian tissue. Subsequent inflammatory changes, as well as postoperative alterations in vascularity, may also contribute to ovarian damage [70]. Surrogate markers of ovarian function have been investigated postoperatively to evaluate the impact of surgical management. In a prospective, randomized trial, Tsolakidis et al. compared ovarian reserve damage in 20 women with endometriomas who underwent laparoscopic cystectomy or a ‘three-step procedure’. This ‘three-step procedure’ included an initial laparoscopy where the endometrioma was drained, irrigated, inspected and biopsied. Afterwards, a GnRH agonist was given for 3 months, and then a second laparoscopy was performed, where the internal wall of the cyst was vaporized with a CO2 laser. There was a significant difference in the decrease in mean anti-Müllerian hormone and the antral folic count, both markers of ovarian reserve, in the cystectomy group compared with the ‘three-step procedure’ group [71]. A systematic review by Somigliana et al. and a meta-analysis by Raffi et al. also examined the impact of surgical removal of endometriomas on ovarian reserve and demonstrated a significant decline in anti-Müllerian hormone after cystectomy [72,73]. However, when clinical outcomes in women undergoing ART were evaluated in a meta-analysis by Benschop et al., there was no difference in clinical pregnancy rate (Peto OR: 0.72; 95% CI: 0.25–2.08) or the number of mature oocytes retrieved (Peto OR: 0.60; 95% CI: −1.32–0.12) when endometriomas were managed with cystectomy or fenestration and coagulation [69].
Evidence from these studies suggests that women with endometriomas who are attempting spontaneous conception have the highest pregnancy rates when cystectomy is performed over ablation. In women undergoing ART, there is no proven benefit for endometrioma removal in terms of pregnancy outcomes. In fact, surgery may actually impair ovarian function. Therefore, the risks and benefits of endometrioma removal must be considered before an asymptomatic woman undergoes surgery.
Conclusion & future perspective
The studies reviewed here have demonstrated that there are various medical, surgical and infertility treatments available for women with endometriosis (
Summary of studies examining endometriosis-related pain.
AI: Aromatase inhibitor; COC: Combined oral contraceptive; CPP: Chronic pelvic pain; DMPA: Depot medroxyprogesterone acetate; GnRH: Gonadotropin-releasing hormone; LNG-IUS: Levonorgestrel-releasing intrauterine system; MPA: Medroxyprogesterone acetate; RCT: Randomized controlled trial.
Summary of studies examining endometriosis-related infertility.
AFC: Antral folic count; AMH: Anti-Müllerian hormone; ART: Assisted reproductive technology; CC: Clomiphene citrate; COC: Combined oral contraceptive; GnRH: Gonadotropin-releasing hormone; ICSI: Intracytoplasmic sperm injection; IUI: Intrauterine insemination; MPA: Medroxyprogesterone acetate; RCT: Randomized controlled trial.

Initial management of endometriosis.
Although conventional strategies will continue to be employed for the treatment of endometriosis, new therapies will likely be introduced in the coming years. Innovative strategies that target the disease at the molecular level continue to be developed and studied. Eventually, these newer treatments may be used adjunctively with historical therapies or even replace them. As interest in endometriosis research continues, it is also likely that the optimal surgical management of endometriomas will be further investigated. Given that endometriosis is common amongst women of reproductive age, it is imperative that quality research continues to be performed so that this disease may be better understood and optimally treated.
Executive summary
The traditional medications used to treat endometriosis-related pain, including combined oral contraceptives, progestogens, gonadotropin-releasing hormone agonists and danazol, are effective in controlling symptoms. Side-effect profiles and cost commonly dictate which therapy is used. In women with no contraindications to estrogen and progesterone therapy, combined oral contraceptives should be the first-line treatment strategy.
Women who fail oral contraceptive therapy or who have contraindications to hormonal therapy should be referred to a gynecologist for treatment.
Promising new compounds that target molecular mechanisms involved in endometriosis are actively being investigated.
Surgical management of endometriosis improves pain postoperatively, although disease commonly recurs. Postoperative medical treatment has limited use since it may improve short-term symptoms, but does not impact disease recurrence or long-term pain control.
In women with pain from endometriomas, laparoscopic cystectomy is the most successful surgical strategy in terms of pain relief and recurrence.
There is no role for traditional medical therapy in women with endometriosis-related infertility.
Women with a history of endometriosis warrant early evaluation by a reproductive specialist.
Superovulation with intrauterine insemination is an effective first-line strategy for infertile women with early-stage endometriosis.
Some evidence suggests that women with endometriosis have decreased IVF pregnancy rates compared with other infertile individuals. However, IVF is widely used in women with endometriosis and is effective in helping these individuals conceive.
Pregnancy rates are improved when women with endometriosis use gonadotropin-releasing hormone agonists before undergoing IVF.
Surgery improves fertility in women with earlier stage endometriosis. Limited data suggest that there is also a role for operative management of more advanced disease.
Postoperative medical treatment of endometriosis does not improve pregnancy rates in women attempting spontaneous conception and delays their ability to conceive.
In women with endometriomas, cystectomy results in improved spontaneous pregnancy rates when compared with cyst drainage.
The optimal treatment strategy for women with endometriomas who are undergoing IVF is unclear. Pregnancy rates are not significantly improved in women who have surgery compared with expectant management. Although data have suggested that ovarian reserves may decrease after cystectomy, other studies have shown that clinical pregnancy rates in women who have cystectomies performed do not differ from those who undergo fenestration and coagulation.
Repeat surgery provides limited benefits in infertile women with recurrent endometriosis. IVF is often a better option for women who wish to conceive.
Footnotes
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
