Redefining endometriosis in the modern era

Introduction

Endometriosis is tissue which somewhat resembles the endometrium lining the inside of the uterus, but which is found outside the uterus where it doesn't belong. The chief symptom is pain. Symptomatic endometriosis affects approximately 5% - 10% of women, although up to 40% of asymptomatic women may harbor the disease, regardless of the indication for surgery [1]. This makes endometriosis one of the most common human diseases on the face of the earth. Although endometriosis has been a subject of the medical literature since the 19th century, confusion has raged for decades regarding its origin and proper treatment. Finally, at the dawn of the 21st century, opinion is being replaced by fact, which is allowing better diagnosis and more rational treatment of the disease.

A brief history of the study of endometriosis

In the historical evolution of knowledge of any disease, the most severe and therefore the most clinically obvious manifestation may be identified earliest. This severe form of disease may not be a good representation of the true nature of the disease. A disease requiring surgery for diagnosis and which is frequently misdiagnosed at surgery will undergo two levels of diagnostic selection bias, further obscuring the true nature of the disease. A disease whose symptoms may be mimicked by other diseases will be inaccurately judged on its own true merit. All of these factors together will lead to misunderstanding of the disease. As a result, myths arise to explain bewilderment, futile attempts at treatment are inevitable, and a confused medical profession will follow false prophets into an intellectual wilderness. This is exactly what has happened for the last 85 years with endometriosis.

The earliest manifestations of endometriosis described in individual case reports in the literature have included severe nodular rectovaginal disease [2,3,4] and large ovarian "chocolate" cysts [5]. Nodules of endometriosis involving parenchymal structures, such as the uterosacral ligaments or bowel wall, were noted to histologically resemble the adenomyomas which were known to affect the uterine myometrium. Thus, early manifestations of severe endometriotic disease were termed "adenomyomas" and were identified in the vagina, rectum, and the round ligament of the uterus as it traversed the inguinal canal.

...although endometriosis has been a subject of the medical literature since the 19th century, confusion has raged for decades regarding its origin and proper treatment. Finally, at the dawn of the 21st century, opinion is being replaced by fact, which is allowing better diagnosis and more rational treatment of the disease...

In an early influential paper, ovarian "chocolate" cysts in 23 women were described, thus giving the impression that the ovaries were the most common site of pelvic involvement by endometriosis, even though most of the chocolate cysts described in that paper were corpora lutea. Because no teenager or menopausal woman was among those 23 women with chocolate cysts of the ovary, the notion was born that endometriosis was a disease of the reproductive years and that menopause was protective against it. In that intellectually naive era, natural fertility was assumed to approach 100%. Since about 60% of married women with endometriosis in that study had been pregnant, it was assumed that pregnancy was protective against the disease and that endometriosis severely decreased fertility. Due to the fact that some endometriosis histologically resembled endometrium, the simplistic idea of reflux menstruation was born as the mechanism of origin of the disease and has reigned as the most widely-held theory of origin ever since. Sampson's theory of reflux menstruation [6] alleged that endometriosis arises because particles of endometrium, which are shed during menses travel from the uterine cavity in a retrograde fashion out of the fimbriated end of the fallopian tubes. These cells then attach to peritoneal surfaces, proliferate, and invade to become the disease called endometriosis. Endometriosis was long regarded as an incurable disease since the pelvis would be re-seeded with disease each month during menses. Since endometriosis was declared to be an autotransplant identical to endometrium, it was thought to also bleed each month during menses. This would give lesions a largely hemorrhagic visual manifestation and led to the notion of the "black powder-burn" lesion as the most common visual manifestation of disease. It was not until the mid-1980s that clinical researchers, aided in large part by the magnification of laparoscopy, documented the frequency of early, subtle, non-hemorrhagic disease, thus completing the morphologic spectrum of endometriosis [7,8]. The visual manifestation of endometriosis, which had been long accepted as the most common was found to be one of the least common, immediately calling into question anything published before that time.

A paradigm collapse is desperately overdue. The medical community needs to acknowledge that early gynecological authors didn't know what endometriosis looked like, and that endometriosis is curable by conservative surgery, and that endometriosis differs in dozens of profound ways from endometrium, and that evidence of initial attachment and secondary invasion has been sought but never found [9], and that supporters of the theory of reflux menstruation steadfastly refuse to supply what should be easily-obtainable photo-documentation of initial attachment and secondary proliferation and invasion by refluxed endometrial cells (which allegedly occurs by the billions), so that finally, Sampson's theory will be discarded in favour of a theory based on science and not belief. The fact that this theory has survived so long with so many fatal contradictions and no confirming evidence is remarkable and speaks poorly of the gynecological profession, which has used it for decades and continues to do so as a convenient explanation for all treatment failures instead of blaming the ineffective medical and surgical treatments which are still used to this day by the majority.

The origin of endometriosis

To be considered rational, any theory of origin of endometriosis must satisfy what is known about the disease, including the following facts. It occurs as early as age 10 [10] and as late as age 78 [11]. Evidence of possible endometriosis has been found in the cul-de-sac of a female infant dying of Sudden Infant Death Syndrome [12]. Early endometriosis can be very subtle in appearance and easily missed, but may become more obvious over time thus giving the incorrect impression new disease has appeared [7]. It occurs in elderly males with metastatic prostate cancer who undergo estrogen therapy [13,14,15]. Endometriosis is not an autotransplant since it differs so fundamentally from eutopic endometrium. These differences include morphologic, histologic, immunohistochemical, enzymatic, gene expression, and chromosomal makeup [16]. Autotransplants essentially remain identical to the native tissue of origin. Endometriosis is associated with an increased incidence of other abnormalities, including immunologic differences [17,18], genetic differences [19,20,21,22], and fundamental differences of the endometrium [23,24,25].Endometriosis does not spread geographically throughout the pelvis with advancing age [26,27] although local invasion or fibromuscular metaplasia may occur and give the impression of slight local spread [28]. Most untreated patients do not have progression of their disease [29,30,31,32]. Endometriosis is curable by excision in over 50% of patients by a single surgery [33,34] and in another 50% of patients by a second surgery (unpublished data).

Sampson did not have all the facts we have today, so his theory was based largely on speculation. Under those circumstances existing in that era, there would be a very low probability that his theory should be correct.

The facts above inevitably point toward an embryonic origin of the disease. At the moment of conception genetic, environmental, and chance factors combine to result in tracts of target tissue being laid down across the posterior coelomic cavity during pelvic organogenesis. Given the small size and plasticity of the embryo, and the variability which is the hallmark of biology, these tracts can sometimes be located outside the pelvis, such as on the diaphragm, in the brain, or in the lower extremities. These tracts are the result of abnormal differentiation and migration of Müllerian duct precursors and may contain rests of actual endometriosis or simply carry the potential to undergo metaplasia under the influence of estrogen production at puberty. At first, the rests are colorless and inconspicuous, or tracts of mesencyhmal target substrate may be undifferentiated and therefore unidentifiable. With rising estrogen levels, the glandular elements begin to secrete an unidentified paracrine product which can irritate tissue and cause pain. Nearby capillaries can be destabilized and bleed. Epithelial growth factors, chemokines, and cytokines associated with tissue repair in response to chronic injury result in angiogenesis, neovascularity, and overlying fibrosis. Mesenchymal tracts associated with parenchymal structures such as the uterosacral ligaments, or muscularis of the bowel or bladder may undergo fibromuscular metaplasia around small rests of endometriosis. Thus, lesions that were once colorless or clear may become reddish, then yellow or whitish due to fibrosis, then blackish when trapped blood degenerates. Parencymal structures may develop progressive nodularity and appropriate adhesion formation resulting from the chronicity of the irritative process. Not all of these embryologically-patterned tracts or rests will have the same inherent level of potential biologic activity, either in the same pelvis or from woman to woman. It is possible to see innocuous-appearing endometriosis in one part of the pelvis and angry, aggressive disease in another. Also, not all disease will change in appearance over time, and some women will have only superficial, colorless disease forever. By the time a woman is in her mid-twenties, she has probably formed most of the endometriosis she will ever form in her lifetime, leading to the possibility of cure following complete excision.

...Sampson did not have all the facts we have today, so his theory was based largely on speculation. Under those circumstances existing in that era, there would be a very low probability that his theory should be correct...

Symptoms

Endometriosis causes geographically precise pain but can sometimes produce more widespread pain. The cul-de-sac is the pelvic area most commonly involved, so disease in this site is impacted by intercourse, bowel movements, or sitting, resulting in pain with those actions, particularly just before or during menses. Endometrioma cysts of the ovary can cause ipsilateral pain, especially if periovarian adhesions are stretched by cyst growth. Leakage of cyst contents can cause acute severe pain lasting several hours, with a more widespread pelvic and lower abdominal discomfort lasting several days until the irritative liquid is reabsorbed. Intestinal endometriosis may be asymptomatic if it is superficial, while rectal nodules associated with complete obliteration of the cul-de-sac will cause painful bowel movements even away from menses. Invasive endometriosis of a uterosacral ligament may involve the adjacent ureter with surrounding fibrosis or rarely invade the ureter, resulting in symptoms of hydroureter and hydronephrosis. If ureteral stricture is gradual, silent permanent loss of kidney function may occur. Large nodules of the terminal ileum may cause symptoms of partial bowel obstruction. Right chest and shoulder pain associated with menses may result from diaphragmatic endometriosis (figure 1).

Figure 1.

Symptomatic endometriosis of the diaphragm (arrows) is usually located on the posterior portion of the right hemidiaphragm. This area cannot be seen easily with a laparoscope placed through the umbilicus, but is always visible through a 5 mm laparoscope advanced through a sheath placed beneath the right costal margin, as seen in this frame.

It is important to try to distinguish between pelvic pain which may or may not be due to endometriosis, otherwise diagnostic confusion will result and outcome following treatment will be confounded. As one important example, uterine cramping with menses is a symptom which may not be due to endometriosis but which can be caused by adenomyosis, uterine leiomyomata, or primary dysmenorrhea.

Physical examination

Tenderness or nodularity of the cul-de-sac and uterosacral ligaments during gentle stroking of these areas is pathognomonic of endometriosis. Enlargement of an ovary may or may not be due to endometriosis. Occasionally a patient may have endometriosis of the posterior vaginal fornix (figure 2) eroding through from a nodule of the uterosacral ligament or rectum. This will be visible on office pelvic exam if the speculum is directed posteriorly. Endometriosis of a cesarean section scar or of the round ligament as it exits the inguinal canal will manifest as a painful lump which may enlarge during menses.

Figure 2.

Endometriosis of the posterior vaginal fornix has caused epithelial piling (within the circle). Vaginal endometriosis may be associated with obliteration of the cul-de-sac and results from extension of invasive disease of the uterosacral ligaments or from a rectal nodule.

Diagnosis

Since endometriosis is the leading cause of pelvic pain in women of reproductive years, it should always be the first on the list of differential diagnoses. In most patients, a presumptive diagnosis of endometriosis can be made on the basis of a typical history of pain combined with findings on pelvic exam. Imaging tests are usually negative since most patients do not have ovarian cysts or severe intestinal involvement. Negative imaging tests in the face of significant symptoms will not eliminate the need for surgical evaluation. When imaging tests are positive, the patient usually has flagrant findings on pelvic exam, such as tenderness and nodularity of the posterior pelvis or ovarian enlargement.

In obese patients or patients with uterine cramping, imaging tests may help augment findings on pelvic exam or suggest a non-endomtriotic uterine pathology, which would not respond to endometriosis surgery. Given the multi-organ system involvement possible with endometriosis, expert surgical management will always include the capability of treating pelvic, bowel, bladder, diaphragmatic or ureteral disease regardless of imaging test results. For this reason, not all experts in surgical treatment obtain any pre-operative imaging tests. Surgery is the most accurate diagnostic test for endometriosis, but the surgeon must be familiar with the all the possible appearances of endometriosis, ranging from the subtle to the extreme (figures 3-7).

Figure 3.

Subtle endometriosis in a teenager may be virtually colorless. Here on the left broad ligament, clear papules and whitish macules with an individual gland of endometriosis (arrow), visible beneath the peritoneal surface.

Figure 4.

With the passage of time, colorless glandular lesions (arrow) may secrete a paracrine substance which can destabilize adjacent capillaries, resulting in hemorrhage. Faint fibrosis of the peritoneum may also result, partially obscuring vessels running beneath the peritoneum. Early angiogenesis is visible near the middle of the hemorrhagic area, resulting from secretion of vascular epithelial growth factor.

Figure 5.

With sufficient biological activity and the passage of time, more fibrosis and angiogenesis can form and obscure the peritoneal endometriotic lesions. Other less biologically active glandular structures can still be seen (arrows). A small surgical staple used in a previous surgical procedure can be seen in the cul-de-sac.

Figure 6.

This frame shows the bladder peritoneum in the top half and the anterior uterine fundus below. In older patients, black "powderburn" lesions such as these are more prominent. These represent more scarring and entrapment of blood beneath the peritoneal surface.

Figure 7.

In this patient, the ovaries are enlarged and adherent to the posterior uterus. The cul-de-sac is hidden beneath and is completely obliterated. Many surgeons would interpret this surgical appearance as representing only ovarian endometriosis associated with adhesions, instead of the correct interpretation of invasive disease of the pelvic floor and rectum. These additional findings are apparent only with subsequent dissection.

Medical treatment of symptoms

The historical observations that pregnancy might be protective against endomeriosis and that the disease was rarely symptomatic after menopause led to the development of medicines to mimic these presumed beneficial hormonal states. Thus, birth control pill or progesterone therapy mimicked pregnancy, while danazol and gonadotropin-releasing hormone agonists mimicked menopause. When these drugs were introduced, it was firmly believed that endometriosis could be physically eradicated and cured by mimicking the naturally occurring hormonal states which were assumed to cure the disease.

Since no one had done the simple studies to prove that pregnancy or menopause cured endometriosis, these hopes were unfounded and were based on observations of symptom response rather than disease response, a basic epidemiologic error. The only indication for medical therapy of endometriosis is for temporary reduction of symptoms. Since medical therapy has been shown not to improve fertility, it is contraindicated for the treatment of infertility associated with endometriosis.

Initial medical treatment of symptoms includes pain pills or birth control pills. When the diagnosis of endometriosis is finally suspected or confirmed by surgery, medicines more specific for the treatment of endometriosis are begun. These include gonadotropin-releasing hormone agonists, danazol, or gestrinone. Many patients who do not respond to an initial round of medical therapy are subjected to repeated medical therapy, which is obviously nonsensical: if a therapy did not work the first or second time, why would it work on the third or fourth attempt?

Pharmaceutical companies have observed that most gynecologists do a poor job of treating endometriosis surgically and have promoted medical therapy as a treatment choice for those clinicians who are not expert at surgical treatment. Thus, medical therapy is the hallmark of those who are not experts in treating endometriosis.

...more good could be done more quickly for more women if excision were adopted by more practitioners...

Surgical treatment of endometriosis

Since medicines treat only symptoms, surgery is the only treatment for endometriosis. The question then becomes: Which form of surgery will most completely eradicate endometriosis from the body? Given the invasive nature of the disease, all experts are in agreement that excision is the treatment of choice for endometriosis. Excision is the only form of treatment which has been documented to be able cure endometriosis by findings at repeat surgery. Although thermal ablation by laser vaporization and electrocoagulation are frequently used to treat endometriosis, frequently these methods don't burn deeply enough to destroy all disease (figures 8-9), and surgeons will be understandably hesitant to burn over or on vital structures invaded by the disease. Neither of these thermal ablation treatments has documentation of efficacy as assessed by repeat operation, and electrocoagulation of endometriosis has not been described in sufficient detail in the literature to be used rationally in human females.

Figure 8.

These individual glands of superficial endometriosis of the left broad ligament were not destroyed by laser vaporization, although filmy adhesions were left following treatment. Microscopy is shown in Figure 9.

Figure 9.

A gland of endometriosis (asterisk) lies just beneath the peritoneum of the patient shown in Figure 8. Filmy adhesions (arrow) can be induced by laser injury.

Techniques for excision of endometriosis have been developed which can treat any form of the disease anywhere in the body, usually by laparoscopy (figures 10-11). Gynecological cancer is far less prevalent than endometriosis, and surgery for advanced endometriosis is universally acknowledged as being the most difficult in the gynecological repertoire.

Figure 10.

Biologically active endometriosis of the left broad ligament is associated with thick yellowish scarring and retroperitoneal trapped blood. The ureter runs beneath the lesion.

Figure 11.

Aggressive laparoscopic excision of the lesion shown in Figure 10 requires ureterolysis. The left ureter (arrow) was involved by mild retroperitoneal fibrosis extending from the fibrotic peritoneal nodule within the circle.

For this reason, three levels of endometriosis care are recognized. At the first level, symptoms may be treated medically by a general practitioner or gynecologist. At the second level, a general gynecologist diagnoses the disease surgically and treats by a single attempt at superficial thermal ablation which should not be repeated if unsuccessful. At the third level, an endometriosis expert treats the disease by aggressive and complete surgical excision, with the aid of urologists or general surgeons as needed (figure 12). Relief of symptoms caused by endometriosis is predictable and highly successful following aggressive excision of endometriosis. Symptoms caused by other gynecological conditions will not be affected by excision of endometriosis, and hysterectomy may be indicated in some patients for treatment of uterine pathology (figure 13).

Figure 12.

The rectum is slightly rounded (arrow) at the point of adherence to the right uterosacral ligament. This rounded shape indicates that the rectal wall is involved by endometriosis and some type of rectal surgery will be necessary for complete removal. The cul-de-sac to the left of the rectum shows superficial glandular lesions of endometriosis with surrounding stellate scarring.

Figure 13.

The posterior uterine fundus exhibits several darkly hemorrhagic spots representing either adenomyosis or endometeriosis, with adjacent neovascularity. Excision of endometriosis will not treat uterine symptoms caused by such findings. The cul-de-sac is completely obliterated, but the rectal wall is flat, indicating that there is little or no rectal involvement by endometriosis. The left ovary is cystic and hidden beneath peritoneal folds which are adherent to the left uterosacral ligament. Superficial glandular lesions of endometriosis with faint fibrosis are seen on the right broad ligament.

It is popular in some areas to treat endometriosis surgically by removing something else, such as the uterus, tubes, and ovaries. The hope is that endometriosis will wither away in the absence of estrogen. Since endometriosis occurs primarily on peritoneal surfaces away from the female pelvic organs, this surgical strategy will leave disease behind in most patients. Some 10% to 20% of women will remain symptomatic because of their retained disease. This may be explained in part by the fact that many lesions of endometriosis contain aromatase enzyme, which is able to convert circulating androstenedione into estrogen. So even in the absence of endogenous or exogenous estrogen, endometriosis can continue to produce its own estrogen and remain symptomatic.

The future

Research is being conducted into novel medical treatments directed against aromatase enzyme or against the angiogenesis which accompanies the development of some (but not all) lesions of endometriosis. Since the origin of endometriosis is so closely tied to embyrological and genetic factors, true prevention of the disease is likely to require some type of gene therapy.

While there is still much to learn about endometriosis, it is clear that throughout the world most women suffering from the disease do not receive the most effective current treatment, which is complete excision.

More good could be done more quickly for more women if excision were adopted by more practitioners, with difficult cases quickly referred to super-specialists in endometriosis surgery at select centers of excellence around the world.

References

  1. Rawson, J. M. (1991). Prevalence of endometriosis in asymptomatic women. Journal of Reprodive Medicine, 36, 513-515.
  2. Stevens, T. G. (1910). Adenomyoma of the vaginal wall. Proceedings of the Royal Society of Medicine, 3, 57-58.
  3. Lockyer, C. (1913). Adenomyoma in the recto-uterine and recto-vaginal septa. Proceedings of the Royal Society of Medicine, 4, 112-116.
  4. Cullen, T. S. (1914). Adenomyoma of the rectovaginal septum. Journal of the American Medical Association, 62, 835-839.
  5. Sampson, J. A. (1921). Perforating hemorrhagic (chocolate) cysts of the ovary. Archives of Surgery, 3, 245-323.
  6. Sampson, J. A. (1927). Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. American Journal of Obstetrics and Gynecology, 14, 422-469.
  7. Redwine, D. B. (1987). Age related evolution in color appearance of endometriosis. Fertility and Sterility, 48, 1062-1063.
  8. Stripling, M. C., Martin, D. C., Chatman, D. L., Vander Zwaag, R., & Poston, W. M. (1988). Subtle appearance of pelvic endometriosis. Fertility and Sterility, 49, 427-31.
  9. Redwine, D. B. (2003). "Invisible" Microscopic Endometriosis: A Review. Gynecologic and Obstetric Investigation, 55, 63-67.
  10. Goldstein, D. P., De Cholnoky, C., & Emans, S. J. (1980). Adolescent endometriosis. Journal of Adolescent Health Care, 1, 37-41.
  11. Haydon, G. B. (1942). A study of 569 cases of endometriosis. American Journal of Obstetrics and Gynecology, 43, 704-709.
  12. Redwine, D. B. (1988). Mülleriosis: the single best fit model of origin of endometriosis. Journal of Reproductive Medicine, 33, 915-920,
  13. Pinkert, T. C., Catlow, C. E., & Straus R. (1979). Endometriosis of the urinary bladder in a man with prostatic carcinoma. Cancer, 43, 1562-1567.
  14. Oliker, A. J., & Harris, A. E. (1971). Endometriosis of the bladder in a male patient. Journal of Urology, 106, 858-859.
  15. Beckman, E. N., Pintado, S. O., Leonard G. L., & Sternberg W. H. (1985). Endometriosis of the prostate. American Journal of Surgery, 9, 374-379.
  16. Redwine, D. B. (2002). Was Sampson wrong? Fertility and Sterility, 78, 686-693.
  17. Kennedy, S. H., Nunn B., Cederholm-Willimas, S. A., & Barlow, D. H. (1989). Cardiolipin antibody levels in endometriosis and systemic lupus erythematosus. Fertility and Sterility, 52, 1061-1062.
  18. Sinaii, N., Cleary, S. D., Ballweg, M. L., Nieman, L. K., & Stratton, P. (2002). High rates of autoimmune and endocrine disorders, fibromyalgia, chronic fatigue syndrome and atopic diseases among women with endometriosis: a survey analysis. Human Reproduction, 17, 2715-2724.
  19. Eyster, K. M., Boles, A. L., Brannian, J. D., & Hansen K. A. (2002). DNA microarray analysis of gene expression markers of endometriosis. Fertility and Sterility, 77, 38-42.
  20. Chang, C.-C., Hsieh, Y.-Y., Tsai, F.-J., Tsai, C.-H., Tsai, H.-D., & Lin, C.-C. (2002). The proline form of p53 codon 72 polymorphism is associated with endomtriosis. Fertility and Sterility, 77, 43-45.
  21. Wieser, F., Schneeberger, C., Tong, D., Tempfer, C., Humber, J.C., & Wnzi, R. (2002). PROGINS receptor gene polymorphism is associated with endometriosis. Fertility and Sterility, 77, 309-312.
  22. Treloar, S. A., O'Connor, D. T., O'Connor, V. M., & Martin, N. G. (1999). Genetic influences on endometriosis in an Australian twin sample. Fertility and Sterility, 71, 701-710.
  23. Gebel, H. M., Braun, D. P., Tambur, A., Frame, D., Rana, N., & Dmowski, W.P. (1998). Spontaneous apoptosis of endometrial tissue is imparied in women with endometriosis. Fertility and Sterility, 69, 1042-1047.
  24. Kim, S. H., Choi, Y. M., Chae, H. D., Kim, K. R., Kim, C. H., Kang, & B. M. (2001). Increased expression of endoglin in the eutopic endometrium of women with endometriosis. Fertility & Sterility, 76, 918-922.
  25. Lessey, B. A., Castelbaum, A. J., Sawin, S. W., Buck, C. A., Schinnar, R., Bilker, W., & Strom, B. L. (1994). Aberrant integrin expression in the endometrium of women with endometriosis. Journal of Clinical Endocrinological Metabolism, 79, 643-649.
  26. Redwine, D. B. (1987). The distribution of endometriosis in the pelvis by age groups and fertility. Fertility and Sterility, 47, 173-175.
  27. Marana, R., Muzii, L., Caruana, P., Dell'Acqua, S., & Mancuso, S. (1991). Evaluation of the correlation between endometriosis extent, age of the patients and associated symptomatology. Acta Europaea Fertilitatis, 22, 209-212.
  28. Koninckx, P. R., Meuleman, C., Demeyere, S., Lesaffre, E., & Cornillie, F. J. (1991). Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertility and Sterility, 55, 759-765.
  29. Mahmood, T. A., & Templeton, A. (1990). The impact of treatment on the natural history of endometriosis. Human Reproduction, 5, 965-970.
  30. Thomas, E.J., & Cooke, I. D. (1987). Impact of gestrinone on the course of asymptomatic endometriosis. British Medical Journal, 294, 272-274.
  31. Telimaa, S., Ronnberg, L., & Kauppila, A. (1987). Placebo controlled comparison of danazol and high-dose medroxyprogesterone acetate in the treatment of endometriosis after conservative surgery. Gynecological Endocrinology, 1, 363-371.
  32. Harrison, R. F., & Barry-Kinsella, C. (2000). Efficacy of medroxyprogesterone treatment in infertile women with endometriosis: a prospective, randomized, placebo-controlled study. Fertility and Sterility, 74, 24-30.
  33. Redwine, D. B. (1991). Conservative laparoscopic excision of endometriosis by sharp dissection: Life table analysis of reoperation and persistent or recurrent disease. Fertility and Sterility, 56, 628-634.
  34. Abbott et al. (unpublished)