This Cyberounds^® will discuss unlabeled use of low-dose estrogen, norethindrone and aromatase inhibitors.

Epidemiology

Endometriosis remains one of the most complex pathologies faced by OBGYNs and primary care physicians. The overall prevalence of endometriosis in reproductive age women has been estimated to be between 3-10%, but can vary depending on the population studied.(1)(2)(3)(4) Endometriosis affects 12-32% of reproductive age women with pelvic pain and 9-50% of women who are infertile.(5)(6)(7)(8)(9) It is a benign, yet often chronic, disease defined by the presence of endometrial glands and stroma that have implanted outside the uterine cavity. Endometriosis is more prevalent among first-degree relatives of affected women.(10)(11)(12)(13) While several candidate genes and polymorphisms have been associated with elevated endometriosis risk, clearly defined genetic underpinnings of the disease are lacking.

Etiology

Many alternate theories exist regarding the pathogenesis of endometriosis including coelomic metaplasia, vascular/lymphatic dissemination and direct transplantation. These theories are predominantly based on clinical observation. The classic theory supporting the pathogenesis of endometriosis has been one of “retrograde flow,” whereby endometrium flows into the peritoneal cavity through the fallopian tubes and implants. This has been supported by several observations: (1) surgically injected endometrial cells induce endometriosis in non-human primates; (2) endometriosis typically develops in the dependent portions of the pelvis; and (3) endometrial cells can grow in cell culture and attach to mesothelium.(14)

Yet, given that endometriosis is only manifested in a small fraction of the 75-90% of women in which retrograde menstruation occurs,(15)(16)(17) researchers continue to seek other disease initiation models. This quest has brought more attention to the “altered immunity” theory, best described by the notion that women with endometriosis have an impaired immune system, which cannot adequately scavenge and manage the ectopic endometrial glands that make their way into the peritoneal cavity.

A large retrospective survey of members of the Endometriosis Association has suggested a link between endometriosis and syndromes associated with autoimmunity like Sjogren’s, rheumatoid arthritis, lupus and multiple sclerosis because of the higher prevalence of these syndromes in people with endometriosis as compared to controls.(18) This study, however, has been criticized because of inherent biases in the study design and furthermore another large study did not confirm the connection.(19) Current evidence cannot reliably support a higher prevalence of autoimmune diseases in those patients with endometriosis.

On the other hand, various immunologic abnormalities have been described in patients with endometriosis. Differences in certain cytokine and growth factor expression profiles in both serum and peritoneal fluid have generated efforts toward the development of a non-invasive, diagnostic test for endometriosis. Despite these initiatives, serum analyte testing remains challenging. Ultimately, it remains unclear whether altered immunity predisposes one to endometriosis or is a consequence of the disease itself.

Despite all of the above advances, endometriosis remains a poorly understood condition and, in all likelihood, various aspects of genetics, anatomy, altered immunity, and the environment have a role in the development of the disease. Patients with endometriosis typically desire two things: pain relief and pregnancy, but often, the treatments are in conflict with each other. The approach to managing pain and/or infertility in patients with endometriosis is very complex. Because endometriosis varies in both its severity and patient response to treatment, it is easy for patients and physicians to get frustrated. Therefore, it is paramount for physicians to keep in mind the patient’s primary goals – both short- and long-term – when managing the disease.

With a backdrop of the current state of endometriosis, we look to highlight, through the use of clinical vignettes, common decisions OBGYNs have to make when managing patients with known or suspected endometriosis. We outline five crossroads in the evaluation and management strategy of one patient with suspected endometriosis. We draw from our experience and the current literature to provide the physician and his/her team with the tools to adequately counsel and treat patients with endometriosis who ultimately desire pregnancy. We will limit the scope of our discussion to reproductive age women who ultimately desire future fertility.

Clinical Case A 23-year-old nulliparous woman presents with worsening cyclic dysmenorrhea and deep dyspareunia. She has minimal relief with over-the-counter ibuprofen. She is also looking for reliable, effective contraception.

The Workup
Although seeing lesions at the time of laparoscopy remains the “gold standard” for diagnosing endometriosis, the history and physical exam can increase the sensitivity of a clinical diagnosis. Historical elements that suggest endometriosis include onset of pelvic pain after menarche, urinary or bowel dysfunction or discomfort, pain with intercourse and response to any NSAIDs or hormone treatment in the interim.

Patients with endometriosis may have induration or nodularity present in the posterior cul-de-sac or the uterosacral ligaments on bimanual exam and may have palpable adnexal masses. The exam is often uncomfortable, but findings are predominantly non-specific and many patients with endometriosis have an unremarkable exam. A transvaginal ultrasound is recommended in all cases of pelvic pain, particularly in the case of adnexal fullness or suspected endometriomas, which are a sign of advanced stage endometriosis.

Making the Diagnosis

Medically
While not definitive, a clinical response to medical management with a trial of NSAIDs combined with oral contraceptive pills (OCPs) has been used as a non-invasive surrogate. A three-month course of a gonadotropin-releasing hormone (GnRH) agonist has also been used empirically as a diagnostic tool, but other abdominopelvic conditions (i.e., Irritable bowel syndrome) have been found to respond to GnRH agonists and therefore one cannot reliably conclude that a clinical response is diagnostic of endometriosis.(20)

Surgically
Traditionally, laparoscopy with pathologic evaluation of lesions is referred to as the diagnostic gold standard, though histologic evaluation of biopsies confirms the diagnosis 50-65% of the time. Surgical evaluation should not be performed within three months of hormonal treatment. Classic endometriotic lesions have a black blue “powder burn” appearance; however, the majority of lesions are atypical: white and opaque, flame-like and vesicular.(21)(22)(23)(24)(25) Endometriomas increase disease severity and are typically identified on ultrasound by their “ground glass” homogeneity.

Many different classification systems exist for qualifying the severity of endometriosis. The American Society for Reproductive Medicine (ASRM) is the most widely accepted but is limited by studies that suggest poor correlation with lesion site and pain(26)(27)(28) as well as pregnancy outcomes.(29) To date, predicting pain symptoms based on lesion site, depth of invasion, number, appearance and pain location remains nebulous.(30) Recently, the Endometriosis Fertility Index (EFI) was developed to predict cumulative pregnancy rate and remains the only endometriosis classification system to predict a clinical outcome.(31)(32)

Initial Treatment Options
Oral contraceptives or NSAIDS were taken first in over 80% of patients with known or suspected endometriosis.(33)

Attacking inflammation
Endometriosis is an inflammatory process driven by (1) various cytokines and growth factors (IL1, TNF-alpha, PGE₂, RANTES, MCP, VEGF) mediated by macrophages; (2) the direct effect of bleeding in the peritoneum; and (3) the irritation of nerves that infiltrate the lesions. Although there are many prospective trials studying NSAIDs in patients with primary dysmenorrhea, there are few randomized controlled trials of NSAIDs versus placebo in the treatment of endometriosis. Most of the utility of NSAIDs comes from their low cost and observational studies of patients with minimal pain that get relief with use. Although the research literature for primary dysmenorrhea suggests that the fenemate class of NSAIDs (mefenamic acid) may be better than the phenylproprionate class (ibuprofen, naproxen), whether or not this observation holds true in patients with endometriosis is unknown. Fenemates have the benefit of blocking prostaglandin action as well as synthesis. Though often due to familiarity, availability and sometimes cost, phenylproprionates are prescribed first.

Recommendation
Typically, pain associated with endometriosis is linked to onset of menses, however this varies widely and some patients depending on the severity of their disease will have non-cyclical symptoms. In those in whom a temporal relationship is noted, we recommend starting with ibuprofen 600-800 mg TID on the day before onset of symptoms, through 2-3 days post cessation of menses (adjusting based on symptom onset) for 2-3 cycles.

A trial of mefenamic acid, 500 mg, with onset of symptoms and then 250 mg q6 hours for 3 days is also reasonable, especially if there has been no previous response to ibuprofen or naproxen in the past.

In addition to these first line pharmacologic recommendations it may be useful to give nonpharmacologic advice. The most common things recommended for these women are heat wraps and exercise/increasing physical activity especially in pain-free times to improve any musculoskeletal components to pelvic pain.

Attacking the Hormone-mediated Effect of Endometriosis (First Line)
Estrogen comes from three places in women with endometriosis: (1) ovary, (2) peripheral conversion of androgens and (3) locally produced by the lesion which houses steroidogenic enzymes including aromatase. Decreasing the estrogen environment can be achieved by different delivery routes. We recommend combined oral contraceptive (COC) pills as initial therapy for women who are not seeking pregnancy, citing a 75-90% response rate(34)(35)(36) No one formulation is superior to the other, although a recent study suggests that pill users may have more satisfaction then patch users.(37) Additionally, there is no evidence favoring cyclic or continuous delivery of pills. We typically start with cyclic administration of a monophasic preparation but have a low threshold for recommending longer periods of continuous administration if breakthrough bleeding is not an issue for the patient. Given their effectiveness, some have proposed the use of COCs as a primary prevention option for patients at risk for endometriosis (i.e., family history); however a recent meta-analysis did not overwhelmingly support this practice.(38)

Progestins (pill, injectable, implantable formulations) can be used to treat endometriosis because they inhibit the growth of the endometrium and induce decidualization. Higher doses will induce an anovulatory, amenorrheic state, which may result in pain relief. The levonorgestrel intrauterine device may have a role in patients with severe recto-vaginal endometriosis; however, experience remains limited to recommend this as a first line treatment.

To Scope or NOT To Scope (Part 1)

The big question is whether or not this patient should have a diagnostic laparoscopy. In our patient case, in the absence of any findings on physical exam or ultrasound and no previous hormonal therapy, a course of 2-3 months of OCPs would be beneficial. If pain remains, a diagnostic procedure might be warranted. If the patient had ultrasound findings suspicious of an endometrioma, surgical resection of the endometrioma would be recommended in certain situations (discussed further below).

Clinical Case Our patient tries the OCPs but does not get relief and eventually seeks a second opinion where she undergoes laparoscopy and has endometriotic lesions ablated. She states that she was told she has severe endometriosis. She continues the OCPs but her pain is worse and she returns to you one year later saying that her pain is back and she would like to avoid surgery. She is currently not trying to conceive nor does she anticipate doing so during the next two to three years.

Operative Laparoscopy and Post Operative Treatment

Surgical treatment varies depending on the severity of the disease. In those patients with minimal disease, surgery is probably equal to medical treatment in terms of pain relief; however, recurrence after excision or ablation can occur in 10-20% of women within the first year.(39)(40) Restoring normal anatomy is the principle tenet when dealing with severe endometriosis-related pain, and often requires the consult of a gynecologic surgeon experienced in endometriosis surgery. If endometriotic lesions are visible during laparoscopy, excision or ablation can be performed and no evidence favors one method over the other. We typically use unipolar ablation; however, when adjacent to vital structures or a diagnosis is needed, excision is preferred.

Postoperative suppressive treatment is usually reserved for those patients with moderate or severe endometriosis-associated pain, though these are often individualized treatment plans based on a patient’s medical issues and pregnancy desires. The goal of adjuvant hormone therapy is to increase the recurrence-free interval. Studies have shown greater pain free intervals when surgery is followed by treatment with GnRH agonists, danazol, OCPs, and progestins, while others have not. In patients for whom pain relief is the only concern, as in our patient above, postoperative suppression with OCPs is reasonable. One group reported a five-fold reduction in time to recurrence with the use of a GnRH agonist plus an aromatase inhibitor compared to GnRH alone.(41)(42)(43)(44)

Attacking the Hormone-Mediated Effect of Endometriosis (Second Line)

Many believe that a first line treatment has “failed” if no relief occurs after six months. In this patient’s case she has had a surgical intervention and, despite postsurgical therapy with OCPs, she falls into the 10-20% of women who have a probable disease recurrence. At this point the patient is a candidate for more aggressive hormone therapy; however, the provider must always consider the patient’s plans for pregnancy. In this case, the patient is not interested in conception.

Danazol was the first FDA-approved drug for the treatment of endometriosis in the United States. It works by creating a high androgen, low estrogen state, which can be effective at keeping the disease under control. Yet, the highly androgenic environment results in weight gain, oily skin, decreased breast size, hirsutism and other related symptoms in up to 80% of women, although only 10% are sufficient enough to cause discontinuation of the medicine.(45) This side effect profile paved the way for GnRH agonists to replace danazol as a second line treatment.

Our patient does not foresee pregnancy in the next couple of years and is interested in maximizing medical therapy in order to minimize pain and avoid more aggressive surgery. Treatment with leuprolide 3.75 mg monthly plus 5 mg of norethindrone acetate daily is a proven regimen for patients with chronic pelvic pain that is refractory to OCPs.(46)(47) This regimen, commonly referred to as “GnRH plus add back” is approved for 12 months without evidence of a decrease in bone mineral density, but small cohort studies suggest that therapy can be continued for as many as 10 years.(48)(49) Although this combination suppresses ovulation, it is not an approved form of contraception and patients should use a back up contraceptive method. Finally, women on this therapy should always be counseled about adequate calcium, vitamin D, nutrition and exercise recommendations to protect bone health long term.

As mentioned before, patients may not tolerate the side effect profile of high-dose progestins, in which case a low-dose estrogen (0.625 CEE) can be added daily to the 5 mg of norethindrone. This regimen has not been approved by the FDA but it is equally effective and supported by clinical trials. It should be reserved for those who cannot tolerate the GnRH plus progestin-only add-back option. It is rare that these patients will experience vaginal bleeding but they must be counseled that it may occur. Bone mineral density surveillance is recommended, though the exact interval between testing periods in these young women remains somewhat unclear. An annual fasting lipid profile is recommended for those on norethindrone due to its effect on cholesterol levels.

From multiple case reports and ongoing experience, the use of aromatase inhibitors has become much more popular as an off-label treatment for endometriosis.(50)(51)(52)(53)(54) Their effectiveness may be best applied in the postsurgical setting. The use of aromatase inhibitors must be combined with another drug (i.e., oral contraceptives, GnRH agonist or norethindrone acetate) in premenopausal women to offset the development of multiple follicular cysts. In one study, which combined letrozole treatment with norethindrone, the combination therapy was more effective than norethindrone alone in those with severe endometriosis. This study was hampered by the observation that user satisfaction was not improved and the letrozole group had more side effects.(55) Similar bone loss worries exist with aromatase inhibitors as listed above and consultation with an OBGYN familiar with the drug should be obtained prior to recommendation.

Clinical Case Our patient received one year of GnRH plus add-back with norethindrone acetate only and obtained moderate relief of pain. She recently married and is hoping to conceive. She returns for a preconception visit as a 25-year-old G0.

Endometriosis and Infertility

Until now in our presentation, we have focused on pelvic pain related to endometriosis and its management. Endometriosis is strongly associated with infertility with up to 40% of infertile women having the disease.(14) Observations among infertile patients undergoing infertility treatments have further strengthened the belief that endometriosis contributes to infertility, and patients should be counseled about this when endometriosis is suspected or confirmed. Evidence is lacking, however, that “suppressing” endometriosis with medical management will improve future fertility. Patients with concurrent endometriosis and infertility might find medical management a detriment, as it suppresses ovulation, and they sacrifice time, which is their enemy when trying to conceive, particularly in older women whose fecundability declines with age.

At this point, the patient should stop her medications and begin timed intercourse with appropriate preconceptual counseling. If she is not pregnant within one year (though some would consider a work-up at 6 months given her known endometriosis risk factor for infertility), she deserves a workup with hysterosalpingogram and semen analysis, assuming she is ovulating monthly. Laparoscopic involvement may be deferred until the tubal evaluation and semen analysis are performed.

To Scope or NOT To Scope (Part 2)

Questions about the need for surgical intervention in a patient with known endometriosis who is attempting pregnancy often arise. In this case, our patient has never attempted pregnancy and one could argue that she deserves an expectant management approach, particularly given her young age. If the situation were different and she had already attempted pregnancy for at least 6-12 months or was older, more aggressive management might be appropriate.

A widely cited randomized controlled trial (RCT) suggests that patients with unexplained infertility and minimal to mild endometriosis are twice as likely to conceive over a three-year period after laparoscopic excision/ablation than those whose lesions are left alone at the time of laparoscopy.(56) When the results of this study were combined with a smaller study(57) that showed no difference in outcome between treatment and expectant management, there was no difference in pregnancy rates overall. Using a “number-needed-to-treat” (NNT) analysis (i.e., the number of patients needed to undergo laparoscopy in order to achieve one successful pregnancy), Marcoux et al. calculated the number to be 9 – nine patients with unexplained infertility need to undergo a laparoscopy to achieve one live birth. When the data from the Marcoux study was combined with that of a smaller study, the NNT was 12.(56)(57)(58)

Remember, our patient had moderate endometriosis as reported in her operative note. No prospective studies have examined pregnancy rates in those with moderate or severe endometriosis undergoing surgery versus no treatment. Case series suggest a 30-50% cumulative pregnancy rate 1-3 years after surgery in women with moderate to severe endometriosis.(59)(60)(61)

Some contend that the number needed to treat is probably greater than 12 since the diagnosis of endometriosis is not known prior to the intervention.(62) Therefore, the decision to undergo surgery should be weighted greater by the amount of pain that the patient has. If the patient remains asymptomatic and has not conceived after one year, laparoscopy may be offered, but we generally proceed with empiric therapy consisting of clomiphene citrate and intrauterine insemination. In the case of the patient with a longer history of infertility and severe disease, we would consider proceeding directly to in vitro fertilization (IVF).

Medical and Surgical Interventions Before IVF

A widely cited meta-analysis showed that patients with endometriosis had less success achieving pregnancy with IVF than those with tubal factor infertility (OR 0.57; 95% CI 0.44, 0.97), particularly if the patients had severe disease.(63) Furthermore, ovarian response and fertilization rates were less optimal, which implies a biological effect on the gamete may also be contributing to the association between endometriosis and infertility. A follow-up review using the CDC’s registry of IVF clinic outcome data suggested that there was no difference, however.(14)(64)(65) As mentioned above, medical suppression for endometriosis should not be recommended as a means for improving pregnancy rates in patients with minimal or mild disease. Although suppression may increase the symptom-free interval after surgery for patients with endometriosis-related pain, there are no data to support this practice as a strategy to improve fertility in these patients.(62)

Clinical Case Our patient on her own, for two years, attempted unsuccessfully to become pregnant. Her infertility workup was unremarkable and she began clomiphene citrate/intrauterine insemination (CC/IUI). During a cycle, she noted increasing abdominal pain that was limiting her ability to sleep at night. An ultrasound revealed a 4.5 cm x 4.5 cm right ovarian complex cyst suggestive of an endometrioma. She has completed two cycles of CC/IUI but has not yet become pregnant.

The Endometrioma Question
This vignette highlights one of the biggest controversies in the management of patients with concurrent endometriosis and infertility. While endometrioma cyst excision has been shown to improve pregnancy rates,(66)(67) the surgery remains controversial because resection of the endometrioma may compromise the ovary(68)(69)(70) and perhaps subsequent ovarian oocyte pool and response to gonadotropin stimulation(69)(71)(72) if this patient were to proceed with IVF. In patients with an asymptomatic endometrioma, one could argue that the only reason to pursue surgery would be to confirm the diagnosis. Experts representing the European Society of Human Reproduction and embryology (ESHRE) recommend resection of an endometrioma prior to IVF if it is greater than 4 cm, citing a 2% chance of malignancy and an improvement in ovarian access.(62)(73)The presence of an endometrioma has not been demonstrated to adversely affect IVF outcome. Although aspiration during oocyte retrieval can be performed, it may be associated with a risk, albeit low, of ovarian abscess (0-1.7%), and recurrence rates are high without resection of the cyst wall.(74)

Finally, the degree of pain drives the intervention. Because the pain is disrupting her daily life, our “patient” may benefit from laparoscopic resection of the endometrioma. If her pain was not bothersome and she was preparing for IVF, we would likely proceed with IVF. Ultimately, the physician needs to help the patient understand that without surgery the endometrioma will not regress and an extensive risk/benefit discussion should be had prior to any intervention.

Summary

Whereas surgical management was formerly the mainstay for patients with pain, current options for medical management are more numerous and better tolerated. Birth control pills and NSAIDs provide frontline treatment, while the use of GnRH agonists with add-back therapy are second line treatments best prescribed by gynecologists familiar with these interventions.

When endometriosis is suspected in a reproductive age woman, it is critical for the physician to identify the patient’s reproductive goals and incorporate them into the management plan. In patients with minimal pain and unexplained infertility, there may be a benefit to performing laparoscopy to excise/ablate endometriotic lesions. How to manage endometriosis in patients undergoing IVF remains controversial, though, particularly as it relates to severe disease manifested by endometriomas. These cases are best managed individually in consult with a reproductive endocrinologist.

Despite its poorly understood pathogenesis, new paradigms, like that of altered immunity, may help us better understand the management and treatment of this disease. Current research in both genetic and immunologic etiologies will hopefully advance our understanding of the underlying pathogenic mechanisms, and open new possibilities for improved diagnosis and treatment. The management of pain and infertility associated with endometriosis has drastically evolved over the last 20 years with new treatments and we expect the next 20 years to provide better insight into non-invasive diagnosis and even better targeted therapies.