Dr. Robert L. Barbieri
Kate Macy Ladd Professor of Obstetrics, Gynecology and Reproductive Biology, Harvard Medical School; Chief of Obstetrics and Gynecology, Brigham and Women's Hospital; Chairman of Obstetrics, Gynecology and Reproductive Biology, Harvard Medical School


Endometriosis is the presence of tissue that resembles normal endometrium at site outside of the uterus. The anatomical areas most commonly affected by endometriosis are the ovaries, the pelvic peritoneum, the uterosacral ligaments, the fallopian tubes, the appendix and the bowel serosa. Endometriomas, or "chocolate cysts" are cysts of endometriosis within the ovary. The "gold standard" for diagnosing endometriosis is laparoscopy, with visual recognition of endometriosis lesions. The severity of endometriosis is defined by the American Society for Reproductive Medicine using a surgical staging system based on the size and location of endometriosis implants and the severity of pelvic scarring. The stages are: Stage I-minimal, Stage II-mild, Stage III-moderate and Stage IV-severe.

The prevalence and incidence of endometriosis have not been well characterized. The prevalence of endometriosis is probably in the range of 5% of women of reproductive age (13 to 50 years of age). The peak incidence is approximately 25 to 30 years of age. Hormonal (estrogen), mechanical (retrograde menstruation) and immunological factors play important roles in the etiology of endometriosis. Endometriosis lesions are absolutely dependent on estrogen for survival and growth. In the absence of estrogen, endometriosis lesions regress. Interventions which reduce estrogen production, such as surgical removal of the ovaries or hormone treatment with a GnRH analogue, are associated with improvement in pelvic pain in most women with endometriosis. Retrograde menstruation of viable endometrial cells from the uterus into the pelvic cavity is believed to be an important "mechanical factor" that is an initiating event in the development of endometriosis. Women with severe stenosis of the cervical os and functioning endometrium menstruate in a retrograde manner back through the fallopian tubes into the pelvic cavity. These women develop endometriosis in almost 100% of cases. In normal women immunological mechanisms probably clear the small amount of endometrium, which reaches the pelvic cavity from retrograde menstruation. In women with endometriosis there is indirect evidence that abnormalities in the immune system reduce the normal clearance of endometrium that enters the pelvic cavity via retrograde menstruation. A family history of endometriosis is associated with a 2-fold increased risk, suggesting a genetic predisposition.

Most women with endometriosis present for medical care with a "chief complaint" of pelvic pain, infertility and/or an adnexal mass (ovarian endometrioma).

Endometriomas - Endometriosis Cysts of the Ovary:

Endometriomas are benign endometriosis cysts of the ovary. Endometriomas are monoclonal tumors. The monoclonal nature of endometriomas suggests that they arise from a somatic mutation in a precursor cell. No medical therapy is available to treat endometriomas. Endometriomas typically present as adnexal masses palpable on physical exam or detected on pelvic ultrasound. In gynecology, a general rule is that persistent, complex adnexal masses that are larger than 3 cm (as determined by pelvic ultrasound) should be surgically resected. The basis of this general rule is that surgical removal of all large, persistent, complex, ovarian cysts will increase the detection and effective treatment of ovarian cancer at an early stage. Ovarian cancer, a leading cause of death, can only be successfully treated when it is detected at an early stage. Surgical resection of an endometrioma will treat the tumor and provide definitive pathological diagnosis (proving that it is not an ovarian cancer). After surgery, about 10% of women develop a second endometrioma. Following surgical treatment of an endometrioma, long-term treatment with an estrogen-progestin oral contraceptive reduces the risk of recurrence of endometriosis lesions in the ovary. In women who have not completed their family, prevention of new endometriosis lesions in the ovary protects future fertility by reducing the risk of the need for additional ovarian surgery.


Infertility is defined as the inability of a couple to achieve a pregnancy after trying for 12 months. Infertility treatment is complex, and in most situations, couples with infertility should be referred to infertility specialists for evaluation and treatment. There is evidence that immediate referral of an infertile couple to an infertility specialist results in less resource consumption than a pattern of referral from a primary care physician to a general gynecologist and then finally to an infertility specialist. A standard approach to the initial diagnosis of infertility is to perform a semen analysis, to document ovulation (serum progesterone, basal body temperature chart), and to demonstrate patency of the fallopian tubes (hysterosalpingogram). If the initial work-up of the infertile couple demonstrates normal ovulatory function, normal semen analysis and patent fallopian tubes, there is a 40% chance that the female partner has endometriosis. The sequential treatment of these women often includes one surgical procedure to determine if endometriosis is present. If the laparoscopy demonstrates the presence of endometriosis, then an attempt to resect all the lesions and adhesions can be performed at the same surgery. If surgery is not followed by pregnancy within the next 12 months, then empirical induction of multi-follicular ovulation with clomiphene or gonadotropins in combination with intrauterine insemination is commonly recommended. If this approach is not effective, then IVF is typically recommended (Table 1). This approach to the treatment of infertility associated with endometriosis is described in more detail below.

A large clinical trial demonstrated that surgical treatment (excision or ablation) of endometriosis improves fertility in infertile women with endometriosis. In this study, 341 infertile women with Stage I or II endometriosis were randomized to have a diagnostic laparoscopy or a diagnostic laparoscopy combined with surgical resection or ablation of endometriosis lesions. During 36 weeks of postoperative follow-up, 18% of the women in the diagnostic laparoscopy group became pregnant and 31% of the women in the diagnostic laparoscopy plus excision or ablation of endometriosis lesions group became pregnant (p<0.006). This well designed clinical trial suggests that surgical resection of endometriosis lesions can improve fertility in infertile women with endometriosis. However, pelvic surgery can cause pelvic adhesions, which can reduce fertility potential. Therefore, in the opinion of this author, infertile women should not have multiple operations to treat endometriosis lesions that are believed to be causing infertility. If the first adequate and complete operation is not followed by pregnancy during the next 12 months, then empirical induction of multi-follicular ovulation with clomiphene or gonadotropins in combination with intrauterine insemination is commonly recommended.

Clinical trials suggest that treatment of infertility with induction of multi-follicular ovulation with clomiphene or gonadotropins in combination with intrauterine insemination is effective in treating infertility associated with endometriosis. Fedele and colleagues randomized 40 women with Stage I or Stage II endometriosis and infertility to 3 cycles of human menopausal gonadotropin (LH and FSH-Pergonal, Repronex) injections plus intrauterine insemination (IUI) or to no treatment. The LH and FSH injections induce multi-follicular ovulation. The pregnancy rate per cycle was 4.5% in the no treatment group and 15% in the LH-FSH-IUI group (p<0.05). In a large randomized clinical trial, 932 women with infertility, many of whom had endometriosis, were randomized to receive intracervical sperm insemination (ICI) alone, intrauterine insemination (IUI), FSH (Gonal-F, Follistim) injections plus intracervical sperm insemination or FSH injections plus intrauterine insemination. Intracervical sperm insemination (ICI) was designed as a "control" that would resemble the effects of natural intercourse. Intrauterine insemination (IUI) is a commonly used fertility treatment where an ejaculated semen specimen is treated in the laboratory to concentrate the sperm in a small volume of buffer while removing all the semen. The concentrated sperm pellet is then delivered to the upper portion of the uterine cavity with a small catheter passed through the cervix. ICI and IUI are timed to occur just before ovulation with the use of urine LH detection kits. FSH injections are given to stimulate multi-follicular ovulation. After 4 treatment cycles, the pregnancy rate in each group was: FSH plus IUI-33%, FSH plus the control ICI treatment- 19%, IUI alone 18%, the control ICI treatment alone 10% (p<0.01). This study demonstrates that FSH plus IUI or FSH plus ICI, or IUI alone improves pregnancy rate over an intervention designed to simulate intercourse alone (ICI). A major complication associated with the FSH injections was twin and triplet pregnancy. The maternal and fetal risks associated with triplet pregnancy may diminish the clinical utility of FSH injections in this infertility setting. Compared to FSH, clomiphene (Serophene, Clomid) ovulation induction is associated with fewer twin and triplet pregnancies. Clomiphene plus IUI therapy may result in a better balance of benefits (pregnancy) and risks (triplet pregnancy) than FSH plus IUI therapy.

The infertility treatment, which has the greatest efficacy in women with endometriosis, is in vitro fertilization and embryo transfer (IVF). IVF is associated with a 30% per cycle pregnancy rate in the treatment of infertility caused by endometriosis. In many centers, if surgery, followed by IUI alone, followed by multi-follicular ovulation induction plus IUI fail to result in a pregnancy, IVF is recommended. In some centers, IVF is offered on a "fast clinical track" as the first intervention in the setting of endometriosis and infertility, especially if the infertile female is more than 37 years of age. In the practice of the author, infertile women < 32 years of age have the time to progress in a deliberate manner through all the standard treatment steps: surgery, IUI alone, clomiphene plus IUI, FSH injections plus IUI and finally IVF. For infertile women > 37 years of age, rapid progression to IVF is indicated to avoid the possibility that all the "healthy" follicles will be depleted before a pregnancy is achieved.

Pelvic Pain - The Traditional Approach:

Many women with endometriosis present with pelvic pain, including severe dysmenorrhea, dyspareunia, dyschezia and lower abdominal pain not related to menses. The standard approach to a woman with severe pelvic pain, who has not had significant pain relief when treated with NSAIDs and oral contraceptives, is to perform laparoscopy to determine the cause of the pain and to see if endometriosis is present. The advantage of placing surgery in a central position in the diagnosis and treatment of women with pelvic pain is that a definitive diagnosis can often be determined and if endometriosis is demonstrated, surgical resection of lesions can produce significant pain relief for 12 months or more. In one randomized controlled clinical trial, surgical resection of endometriosis lesions was demonstrated to produce pain relief for up to 12 months post-operatively. In this study, the "active" surgical procedure consisted of diagnostic laparoscopy combined with laser ablation of endometriosis lesions plus uterine nerve ablation. The "control" surgical procedure was a diagnostic laparoscopy plus aspiration of pelvic peritoneal fluid. The physicians who were evaluating the response of the patients and the patients were not informed as to whether the "active" or "control" operation had been performed. In this randomized study, the active surgical procedure resulted in improvement in pain in 63% of women and the control procedure resulted in improvement in pain in 23% of the women at 6 months follow-up (p<0.01). This study suggests that surgical treatment of endometriosis can be effective in the treatment of pain.

Recent clinical trials suggest that after surgical treatment of endometriosis, hormone therapy for 6 months with a GnRH agonist delays the recurrence of pelvic pain by 12 months. In one clinical trial, 109 women with laparoscopically proven endometriosis, who had undergone laparoscopic treatment of endometriosis lesions, were randomized to receive either a GnRH-agonist analogue, nafarelin (Synarel) 200 ug nasal insufflation twice daily or a placebo nasal insufflation twice daily for 6 months post-operatively. A major outcome variable was the median time after treatment to initiation of an alternative therapy, which reflects the recurrence of severe pelvic pain. The women who were treated with nafarelin post-operatively had a median time to initiation of an alternative therapy >24 months. In the group treated with placebo the median time to initiation of an alternative therapy was 12 months (p<0.001). This study suggests that the combination of surgery plus post-operative hormone therapy may be helpful in extending the time to recurrence of pelvic pain in women with endometriosis. The author offers all women with endometriosis post-operative hormone therapy with either a combination estrogen-progestin oral contraceptive or a GnRH agonist analogue.

Pelvic Pain - The Use of the "Clinical Diagnosis" of Endometriosis:

The current "gold standard" for the diagnosis of endometriosis is surgical documentation of endometriosis lesions by visual detection. The advantages of this approach are that it is very unlikely that a serious condition, such as ovarian cancer, will be missed or misdiagnosed, and that treatment of the endometriosis can take place at the time of surgical visualization. There are a number of clinical disadvantages to using surgical visualization of lesions as the "gold standard" for the diagnosis of endometriosis. These disadvantages include: the requirement for surgery and anesthesia, the possibility that a lesion thought to be endometriosis by visual inspection might not prove to be endometriosis on histological analysis and the possibility that some deep endometriosis lesions might be difficult to visualize and consequently not detected. An alternative to surgery is the use of "clinical diagnosis", based on history, physical examination and non-invasive laboratory tests, to make the diagnosis of endometriosis. A recently published study supports the concept of making a "clinical diagnosis" of endometriosis in certain clinical situations, and using "empirical" hormone treatment for the pelvic pain presumed to be due to endometriosis.

A recent study focused on 100 women with chronic pelvic pain who met 13 inclusion criteria: 1) moderate to severe chronic pelvic pain, 2) age 18 to 45 years, 3) regular menstrual cycles, 4) no previous diagnosis of endometriosis by a surgical procedure, 5) no hormone treatment in past 3 months, 6) no evidence for a primary disease of the urinary tract or bowel as the cause of the pain, 7) no history of excessive use of alcohol, tranquilizer drugs or illicit drugs, 8) normal pelvic ultrasound, 9) normal complete blood count, 10) normal urinalysis, 11) absence of gonorrhea and chlamydia on tests of an endocervical specimen, 12) negative pregnancy test, and 13) failure of non-steroidal anti-inflammatory medications and doxycylcine treatment to improve the pain symptoms.

The women were randomized to receive treatment with depot leuprolide (Lupron) 3.75 mg intramuscular injections every 4 weeks for 12 weeks or to receive a placebo injection for 12 weeks. Pain was measured with the 4-point Biberoglu and Behrman scale. At the end of 12 weeks treatment, all the women underwent laparoscopy to assess the accuracy of the "clinical diagnosis" of endometriosis. At the end of the study, at laparoscopy, it was determined by surgical visualization of lesions, that 87% of the women in the placebo group and 78% of the women in the leuprolide group did have endometriosis. This suggests that the 13 point inclusion criteria listed above is relatively effective in identifying women with chronic pelvic pain who have endometriosis as the cause of their pain (at least in this study population). In addition, the study showed that 82% of the women treated with empirical leuprolide had pain relief compared to 39% of the women treated with placebo. This result indicates that in women with a clinical diagnosis of endometriosis, "empirical" hormone treatment with leuprolide acetate depot 3.75 mg intramuscular injection every 4 weeks (Lupron) can be effective in relieving pelvic pain.

This study, and others, supports the concept of making the clinical diagnosis of endometriosis based on history, physical findings and limited laboratory testing. Women clinically diagnosed with endometriosis can then be offered treatment with a GnRH analogue, danazol or a progestin for their pelvic pain. A recent American College of Obstetricians and Gynecologists technical bulletin, "Medical Management of Endometriosis", discussed this evolving clinical paradigm and concluded that clinical diagnosis of endometriosis and empirical hormone treatment can be efficacious.

Hysterectomy and Bilateral Oophorectomy:

Hysterectomy and bilateral oophorectomy (TAH_BSO) have been reported to provide long-term pain relief in approximately 85% of women with endometriosis. TAH-BSO treatment is only available to women who have definitely completed their family. Following TAH-BSO, hormone replacement therapy with standard regimens can treat vasomotor symptoms without increasing the risk of recurrence of endometriosis and pelvic pain. Large-scale follow-up trials indicate that women with endometriosis and severe pelvic pain who were treated with TAH-BSO have long-term improvement in their pain symptoms and are satisfied with their choice of treatment.

Medical Treatment of Pelvic Pain Caused by Endometriosis:

Women with endometriosis and pelvic pain can be successfully treated with hormone therapy. The practice of the author is to start with hormone treatments with the fewest side effects (non-steroidal anti-inflammatory agents used at near maximal doses, oral contraceptives used in a cyclic or continuous manner) and then to progress sequentially to use the more powerful hormone treatments that are associated with significant side effects (GnRH analogues such as nafarelin and leuprolide, danazol). This progression is outlined in Table 2.

The standard combination estrogen-progestin oral contraceptive appears to be effective in the treatment of pelvic pain caused by endometriosis. The standard oral contraceptive pill contains a progestin in every pill. The progestin blocks growth in many endometriosis lesions, in part, by blocking the growth promoting properties of estrogen.

An alternative to the standard schedule (21 hormone pills, followed by 7 days off the pill) for administering combination estrogen-progestin oral contraceptives is to prescribe 63 active hormone pills in a row (3 packs of 21 active pills) followed by 7 days off the pill, followed by 63 active pills in a row. This approach has been described by the author as a "mini-pseudopregnancy" because it often results in menses every 10 weeks rather than every 4 weeks. An even longer cycle can be prescribed by recommending the use of 105 (5 packs of 21 active pills) in a row, followed by 7 days off medication, followed by 105 active pills in a row. In one study, women who were allowed to extend the number of consecutive active pills and avoid the pill free week reported a high degree of satisfaction with the extended pill regimen. The major disadvantage to the use of "long cycles" is that many women develop "break-through bleeding" and have a small of amount of uterine bleeding, often reported to be brown in color, many days per week. Break through bleeding is the main reason women using a "mini-pseudopregnancy" discontinue the medication.

If oral contraceptives are not effective then the patient can be treated with nafarelin (Synarel) 200 ug twice daily by nasal insufflation or leuprolide acetate depot (Lupron), 3.75 mg intramuscular injection every 4 weeks. A standard course of GnRH agonist treatment is 24 weeks, but treatment regimens as short as 12 weeks have been shown to be effective. The GnRH agonist analogues, nafarelin and leuprolide, initially stimulate pituitary release of LH and FSH, but with chronic use, they paradoxically down-regulate LH and FSH secretion, resulting in cessation of ovarian follicular growth and suppression of ovarian production of estrogen and progestin. The absence of ovarian production of estrogen and progestin, results in amenorrhea and hypoestrogenic side effects such as hot flashes and increased rate of bone resorption. The hypoestrogenism induced by the GnRH agonists is associated with relief of pelvic pain in approximately 85% of women with endometriosis. This strong beneficial effect needs to be balanced against the side effects of hot flashes and bone loss. For some women, the pain relief is so impressive and the side effects are so minimal that they want to stay on the treatment for longer than 6 months. This is safe, if the patient is carefully monitored for bone loss. Low-dose steroid "add back" can be added to the GnRH agonist treatment to reduce the rate of bone loss and to treat hot flashes. Add-back regimens, which have been demonstrated, to be effective are listed in Table 3. After discontinuation of treatment, pelvic pain tends to recur in most women over the next 12 months.

Danazol (Danocrine), an orally active derivative of testosterone, is as effective as nafarelin and leuprolide in the treatment of pelvic pain caused by endometriosis. Danazol is typically prescribed at doses of 200 mg twice daily for up to 6 months. If this dose is not effective in relieving pelvic pain, then the dose can be increased to 200 mg three or four times daily. Danazol is no longer widely used for the treatment of endometriosis because it is associated with significant weight gain (average-4 kg), deepening of the voice, hirsutism and musculoskeletal pain. The current use of danazol tends to be as a "last resort" after NSAIDs, oral contraceptives and the GnRH agonist analogues have failed to adequately treat pelvic pain caused by endometriosis.

Multi-Specialty Pain Practice:

A major challenge in pain treatment is to effectively integrate the psychosocial-mind-brain-body paradigms. The psychosocial paradigm emphasizes the importance of social status and experiences in the family and society on health outcomes. The mind paradigm emphasizes the importance of psychological factors on health. The brain paradigm focuses on neural genes and neurochemistry in the etiology of pain. The body paradigm focuses on the diseased organs that trigger the pain stimulus. There is ample evidence that in women with endometriosis and pelvic pain, factors such as depression, somatization, past history of sexual or psychological abuse play important roles in modulating the patient's perception of her pain. For patients with pelvic pain who have not had significant relief from hormonal or surgical treatment, the multispecialty pain practice may prove to be effective.

There are few randomized studies of the efficacy of psychological or psychosocial interventions in the treatment of pelvic pain. In one clinical trial, the effectiveness of a focused organic (surgical) approach to pelvic pain was compared to a multidisciplinary approach to pelvic pain that included surgical, psychological, dietary and social interventions. The investigators reported that the multidisciplinary approach to pelvic pain resulted in greater, long-term improvement in self-reported pain scores than the approach that focused on treating organic disease with surgery. Most large medical centers have an organized pain treatment center that can develop a multidisciplinary treatment plan for the patient with chronic pelvic pain that has failed to respond to surgical intervention and hormone treatment.

Rapid diagnosis is critical to the proper treatment of disease. Two independent studies reported that the average time between onset of the symptoms of endometriosis (pelvic pain) and the definitive diagnosis was over 6 years. The delay between onset of symptoms and definitive diagnosis results in significant frustration for both patients and clinicians. Delay in diagnosis of endometriosis can be reduced if clinicians actively consider the diagnosis in all women with pelvic pain or infertility.

Table 1: Sequential treatment approach to the woman with infertility and endometriosis.
Step1: Surgical diagnosis and treatment. Follow-up for 12 months to see if pregnancy occurs.
Step2: Intrauterine insemination.
Step 3: Clomiphene induction of multi-follicular ovulation plus intrauterine insemination.
Step 4: FSH injections for induction of multi-follicular ovulation plus intrauterine insemination.
Step 5: In vitro fertilization and embryo transfer (IVF)

Table 2: Hormonal treatments for pelvic pain caused by endometriosis. The Step 1,2 and 3 interventions are associated with less severe side effects than Step 4 and 5 treatments.
Step 1: Non-steroidal anti-inflammatory agents at maximal doses.
Step 2: Oral contraceptives used in a cyclic manner.
Step 3: Oral contraceptives used in "long-cycles".
Step 4: GnRH agonist analogue treatment: nafarelin (Synarel- 200ug by nasal insufflation bid) or leuprolide (Lupron Depot, 3.75 mg IM q 4 wks)
Step 5: Danazol (Danocrine-200mg po. bid or tid or qid, daily)

Table 3: Steroidal add-back regimens that can reduce the rate of bone loss and hot flashes associated with GnRH agonist treatment (nafarelin or leuprolide).
Add-Back MedicationDose
Norethindrone Acetate (Aygestin)5 mg po daily
17beta-estradiol, transdermal patch plus
Medroxyprogesterone acetate (Provera)
25 ug patch
5 mg daily
17beta-estradiol, micronized (Estrace)
Norethindrone Acetate (Aygestin)
2 mg po daily
1 mg po daily
Conjugated Equine Estrogen (Premarin)
Medroxyprogesterone Acetate (Provera)
0.625 mg po daily
5 mg po daily
Conjugated Equine Estrogen (Premarin)
Medroxyprogesterone Acetate (Provera)
0.3 mg po daily
5 mg po daily

Editor's Note:
Gratitude is expressed to Dr. Robert L. Barbieri for his priceless encouragement and support. His pioneer work in reproductive health has helped millions of women worldwide. We are also thankful to Con-Therapy for giving us permission to print it. Readers are encouraged to visit the web site of WHO on Reproductive Health:

© Women's Health and Education Center (WHEC)