Joseph C. Gambone
Endometriosis and adenomyosis often present difficult diagnostic challenges. In the case of endometriosis, few gynecologic conditions require such difficult surgical dissections.
Endometriosis is a benign condition in which endometrial glands and stroma are present outside the uterine cavity and walls. Endometriosis is important in gynecology because of its frequency, distressing symptomatology, association with infertility, and potential for invasion of adjacent organ systems, such as the gastrointestinal or urinary tracts.
The prevalence of endometriosis in the general population is not known, but it is estimated that 5% to 15% of women have some degree of the disease. At least one third of women with chronic pelvic pain have visualized endometriosis, as do a significant number of infertile women. Interestingly, endometriosis is noted in 5% to 15% of women undergoing gynecologic laparotomies, and it is an unexpected finding in about half of these cases.
The typical patient with endometriosis is in her 30s, nulliparous, and infertile. However, in practice, many women with endometriosis do not fit the classic picture. Occasionally, endometriosis may occur in infancy, childhood, or adolescence, but at these early ages, it is usually associated with obstructive genital anomalies such as a uterine or vaginal septum. Although endometriosis should regress after menopause unless estrogens are prescribed, 5% of new cases develop in that age group. In addition, the scarifying involution from preexisting lesions may result in obstructive problems, especially in the gastrointestinal and urinary tracts.
The pathogenesis of endometriosis is not completely understood. Genetic predisposition clearly plays a role. The following three hypotheses have been used to explain the various manifestations of endometriosis and the different locations in which endometriotic implants may be found:
1. The retrograde menstruation theory of Sampson proposes that endometrial fragments transported through the fallopian tubes at the time of menstruation implant and grow in various intraabdominal sites. Endometrial tissue, which is normally shed at the time of menstruation, is viable and capable of growth in vivo or in vitro. To explain some rare examples of endometriosis in distant sites, such as the lung, forehead, or axilla, it is necessary to postulate hematogenous spread.
2. The müllerian metaplasia theory of Meyer proposes that endometriosis results from the metaplastic transformation of peritoneal mesothelium into endometrium under the influence of certain generally unidentified stimuli.
3. The lymphatic spread theory of Halban suggests that endometrial tissues are taken up into the lymphatics draining the uterus and are transported to the various pelvic sites where the tissue grows ectopically. Endometrial tissue has been found in the pelvic lymphatics of up to 20% of patients with the disease.
Most authorities today believe that several factors are involved in the initiation and spread of endometriosis, including retrograde menstruation, coelomic metaplasia, immunologic changes, and genetic predisposition. A fundamental question is why all menstruating women do not develop endometriosis given that most if not all women have retrograde flow into the pelvis during menstruation. The amount of exposure to retrograde flow and the woman’s immunologic response appear to be critical. Researchers have identified differences in the chemical composition and biologic pathways of endometrial cells from women with endometriosis compared with unaffected women. They have also found significant differences in the inflammatory factors and growth factors in the peritoneal fluid of affected women. A clearer understanding of the pathophysiology of endometriosis would provide insights into more effective strategies for prevention and treatment.
SITES OF OCCURRENCE
Endometriosis occurs most commonly in the dependent portions of the pelvis. Specifically, implants can be found on the ovaries, the broad ligament, the peritoneal surfaces of the cul-de-sac (including the uterosacral ligaments and posterior cervix), and the rectovaginal septum (Figure 25-1). Quite frequently, the rectosigmoid colon is involved, as is the appendix and the vesicouterine fold of peritoneum. Endometriosis is occasionally seen in laparotomy scars, developing especially after a cesarean delivery or myomectomy when the endometrial cavity has been entered. It is probable that endometrial tissue is seeded into the surgical incision. Two of three women with endometriosis have ovarian involvement.
FIGURE 25-1 Common sites of endometriosis in decreasing order of frequency: (1) ovary, (2) cul-de-sac, (3) uterosacral ligaments, (4) broad ligaments, (5) fallopian tubes, (6) uterovesical fold, (7) round ligaments, (8) vermiform appendix, (9) vagina, (10) rectovaginal septum, (11) rectosigmoid colon, (12) cecum, (13) ileum, (14) inguinal canals, (15) abdominal scars, (16) ureters, (17) urinary bladder, (18) umbilicus, (19) vulva, and (20) peripheral sites.
Islands of endometriosis respond cyclically to ovarian steroidal hormone production. The implants proliferate under estrogenic stimulation and slough when support from estrogen and progesterone is removed with involution of the corpus luteum. The sloughed material induces a profound inflammatory response resulting immediately in pain and fibrosis in the long term. The macroscopic appearance of endometriosis depends on the site of the implant, activity of the lesion, day of the menstrual cycle, and time since implantation.
Lesions may be raised and flat with red, black, or brown coloration; fibrotic scarred areas that are yellow or white in hue; or vesicles that are pink, clear, or red (Figure 25-2). The color of the implant is generally determined by its vascularity, the size of the lesion, and the amount of residual sloughed material. Newer implants tend to be red, blood-filled active lesions. Older lesions tend to be much less active hormonally, scarred and blue-gray in color with a puckered appearance. These older inactive lesions have been called the “tattooing of endometriosis.”
FIGURE 25-2 A to D: Appearance of old endometriosis with “tattooing” (blue-gray lesions), and red, brown, and black raised lesions of active endometriosis at the time of laparoscopy.
Endometriomas of the ovary are cysts filled with thick, chocolate-colored fluid that sometimes has the black color and tarry consistency of crankcase oil. This characteristic fluid represents aged, hemolyzed blood and desquamated endometrium. Usually, endometrial glands and stroma are present in the cyst wall. Sometimes, however, the pressure of the enclosed fluid destroys the endometrial lining of the endometrioma, leaving only a fibrotic cyst wall infiltrated with large numbers of hemosiderin-laden macrophages. Generally, ovarian implants are associated with significant scarring of the ovary to the pelvic side wall or broad ligament. Histologically, two of four characteristics must be found in the endometrioma specimen to confirm the diagnosis—endometrial epithelium, endometrial glands, endometrial stroma, and hemosiderin-laden macrophages.
Although endometriosis is a benign process, it shares many characteristics with malignancy. It is locally infiltrative, invasive, and widely disseminated. It is also curious that cyclic hormones tend to induce growth, whereas continuous hormonal exposure, especially in high doses, generally induces significant regression.
The American Society of Reproductive Medicine employs a staging protocol in an attempt to correlate fertility potential with a quantified stage of endometriosis. This staging, which was initially based on the allocation of points depending on the sites involved and extent of visualized disease (Figure 25-3), was modified to include a description of the color of the lesions and the percentage of surface involved in each lesion type, as well as a more detailed description of any endometrioma.
FIGURE 25-3 Modified from the revised American Fertility Society classification of endometriosis.
(Reprinted with permission from the American Society for Reproductive Medicine. Fertil Steril 67:819–820, 1996.)
The characteristic triad of symptoms associated with endometriosis is dysmenorrhea, dyspareunia, and dyschezia. The pain that women suffer with endometriosis varies with the time since initiation. Early in the clinical course, women tend to have cyclic pelvic pain, which starts 1 to 2 days before the menstrual flow and resolves at the end of the menses. This secondary dysmenorrhea is thought to be related to the premenstrual swelling and extravasation of blood and menstrual debris, which induces an intense inflammatory reaction in the surrounding tissue mediated by prostaglandins and cytokines that are more directly responsible for triggering the pain sensation. Deep, infiltrating implants, especially those in the retroperitoneal space, are associated with more pain than are superficial lesions. Over time, the pain may become more chronic, with exacerbations at the time of the menses. Interestingly, there is no clear relationship between the stage of endometriosis and the frequency and severity of pain symptoms.
Dyspareunia is generally associated with deep-thrust penetration during intercourse and occurs mainly when the cul-de-sac, uterosacral ligaments, and portions of the posterior vaginal fornix are involved. Deep-thrust dyspareunia can also result from uterine immobility due to significant internal scarring caused by endometriosis. Endometriomas in these sites are usually exquisitely tender to palpation.
Dyschezia is experienced with uterosacral, cul-de-sac, and rectosigmoid colon involvement. As the stool passes between the uterosacral ligaments, the characteristic dyschezia is experienced. Premenstrual and postmenstrual spotting is a characteristic symptom of endometriosis. Menorrhagia is uncommon, the amount of menstrual flow usually diminishing with endometriosis. If the ovarian capsule is involved with endometriosis, ovulatory pain and midcycle vaginal bleeding often occur. Rarely, as other organ systems are involved, menstrual hematochezia, hematuria, and other forms of endometriotic sloughing become evident.
The association between mild to moderate endometriosis and infertility is not clear. When endometriosis distorts the pelvic structures, its role in infertility is more predictable.
Endometriosis presents with a wide variety of signs varying from the presence of a small, exquisitely tender nodule in the cul-de-sac or on the uterosacral ligaments to a huge, relatively nontender, cystic abdominal mass. Occasionally, a small, tender mulberry-like spot may be seen in the posterior fornix of the vagina. Characteristically, a tender, fixed adnexal mass is appreciated on bimanual examination. The uterus is fixed and retroverted in a substantial number of women with endometriosis. Occasionally, no signs at all are appreciated on physical examination.
The main differential diagnoses in the acute phase of endometriosis are (1) chronic pelvic inflammatory disease or recurrent acute salpingitis, (2) hemorrhagic corpus luteum, (3) benign or malignant ovarian neoplasm, and occasionally, (4) ectopic pregnancy.
The diagnosis of endometriosis should be suspected in an afebrile patient with the characteristic triad of pelvic pain, a firm, fixed, tender adnexal mass, and tender nodularity in the cul-de-sac and uterosacral ligaments. The characteristic sharp, firm, exquisitely tender “barb” (from barbed wire) felt in the uterosacral ligament is the diagnostic sine qua non of endometriosis, but this finding is generally present only in severe cases. An ultrasonic evaluation may indicate an adnexal mass of complex echogenicity, with internal echoes consistent with old blood. Serum levels of the cancer antigen CA 125 are frequently elevated in women with endometriosis. However, the positive predictive value of CA 125 for detecting endometriosis is low (about 20%), and this test should not used to diagnose endometriosis.
The definitive diagnosis is generally made by the characteristic gross and histologic findings obtained at laparoscopy or laparotomy. Unfortunately, even the most experienced surgeon may fail to identify endometriotic implants visually because the older implants may have a very subtle appearance and the deeper infiltrating lesions may not be visible at the surface. Biopsy of any suspicious lesions improves diagnostic accuracy.
The management of endometriosis depends on certain key considerations: (1) the certainty of the diagnosis, (2) the severity of the symptoms, (3) the extent of the disease, (4) the desire for future fertility, (5) the age of the patient, and (6) the threat to the gastrointestinal or urinary tract or both.
Treatment is indicated for endometriosis associated pelvic pain, dysmenorrhea, dyspareunia, abnormal bleeding, ovarian cysts, and infertility due to gross distortion of tubal and ovarian anatomy. Surgical intervention is required for an endometrioma larger than 3 cm, gross distortion of pelvic anatomy, involvement of bowel or bladder, and adhesive disease. Surgery may improve fertility for women with severe endometriosis. Medical therapy is generally the first line to treat other symptomatic women. There is no convincing evidence that treatment significantly improves fertility in women with mild endometriosis.
The most comprehensive surgery includes total abdominal hysterectomy, bilateral salpingo-oophorectomy with destruction of all peritoneal implants, and dissection of all adhesions. Usually, an appendectomy is also performed. Because of extensive adhesions, surgery for endometriosis is often technically challenging. If endometriosis involves the cul-de-sac or uterosacral ligaments, the proximity to the ureter, bladder, and sigmoid colon must be considered. If endometriosis obstructs the ureter, resection and ureteroplasty may be necessary to preserve renal function. Nearly 25% of kidneys are lost when endometriosis blocks the ureter. Obstruction of the rectosigmoid and even obstruction of the small intestine may require resection of the involved intestinal segment. The surgical risks must be carefully explained to the woman, as well as her subsequent need for treatment for loss of ovarian steroids. She also needs to understand that postoperatively, there is a 20% recurrence rate for endometriosis, usually involving the bowel.
Often the desire for future fertility precludes this surgical option. In this situation, laparoscopic or open surgery is designed to destroy all endometriotic implants and remove all adhesive disease. This usually involves excision (not lysis) of all adhesions and laser ablation or electrocautery of suspected implants. Large endometriomas (>3 cm) are amenable only to surgical resection. Because of extensive adhesive disease that generally surrounds these cysts, cystectomy is not always possible; oophorectomy may be necessary. Extensive tubal disease with or without ovarian involvement may be treated with removal of the affected organs but with uterine preservation for in vitro fertilization. Preoperative treatment with medical agents, such as gonadotropin-releasing hormone (GnRH) agonists for 3 to 6 months can improve surgical success.
The role of medical therapy postoperatively remains controversial, although it is indicated to treat women who have known residual disease diagnosed at surgery. There is a risk for recurrence of endometriosis throughout a woman’s life, so measures should be taken to reduce her risk for retrograde menstruation or cyclic ovarian sex steroid production. Depot medroxyprogesterone acetate (DMPA), continuous oral contraceptives, and the levonorgestrel-releasing intrauterine device (IUD) are all attractive long-term options.
Therapy should be targeted toward relieving the patient’s individual complaints and toward reducing the risk for disease progression. Asymptomatic women found incidentally to have endometriosis may not require any therapy. Dysmenorrhea due to endometriosis can be approached as outlined in Chapter 21, using nonsteroidal antiinflammatory drugs (NSAIDs) and reduction of menstrual flow with hormonal regimens such as low-dose oral contraceptives.
For relief of non-cyclic pelvic pain, short-term medical treatment may be used. NSAIDs, oral contraceptives, and progestins (e.g., medroxyprogesterone acetate) should be considered the appropriate first-line medical treatments for symptomatic endometriosis. When an inadequate response occurs, second-line medical treatment with either a GnRH agonist, higher-dose progestins, or danazol appears to be equally effective.Cost, individual response, and potential side effects generally guide selection of one agent or the other.
Danazol is an androgenic derivative that may be used in a “pseudomenopause” regimen to suppress symptoms of endometriosis if fertility is not a present concern. It is given over a period of 6 to 9 months, and doses of 600 to 800 mg daily are generally necessary to suppress menstruation. Through its weak androgenic properties, danazol decreases the plasma levels of sex hormone–binding globulin. The resulting increase of free testosterone may cause hirsutism and acne. Three years after cessation of danazol, 40% of patients have a recurrence of endometriosis. After a full course of danazol therapy, use of a cyclic oral contraceptive may help to delay or prevent such recurrence.
GnRH agonists cause a temporary medical castration, thereby bringing about a marked, albeit temporary, regression of endometriosis. Treatment of women with endometriosis with GnRH agonists usually produces relief of pain and involution of implants. The disadvantages of these agonists are related to cost, hot flashes, and side effects, including vaginal dryness. They also cause calcium loss from bone and an unfavorable lipid profile. If treatment with a GnRH agonist is effective in relieving chronic pelvic pain and surgery is not indicated, low-dose estrogen-progestin add-back therapy can permit longer-term use of GnRH agonists by mitigating the adverse impact of estrogen deficiency without reducing the efficacy of GnRH agonists.
Oral contraceptives and oral medroxyprogesterone acetate are more effective in treating endometriosis-associated pelvic pain than placebos. The levonorgestrel-releasing intrauterine system reduces dysmenorrheaand may be helpful for inducing regression of cul-de-sac implants without diminishing circulating estrogen levels in the serum.
Surgical and medical treatment options for endometriosis are summarized in Box 25-1.
BOX 25-1 Treatment Options for Endometriosis
Most definitive: Total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH-BSO) with destruction and removal of all peritoneal endometriotic implants and adhesions.
NOTE: As is the case with all treatments for endometriosis, there is always a risk for recurrence, even with “definitive” treatment.
Fertility preserving: Laparoscopic or open surgery (laparotomy) with destruction and removal of all peritoneal endometriotic implants and adhesions.
NOTE: Any endometriomas >3 cm in diameter should be removed surgically. Preoperative treatment for 3 to 6 months can improve surgical success.
First-line: Nonsteroidal antiinflammatory drugs, low-dose oral contraceptives, or progestins (e.g., medroxyprogesterone acetate)
NOTE: This treatment should be given an adequate trial of 3 to 6 months before initiating second-line therapy.
Second-line: Higher doses of progestins (e.g., medroxyprogesterone acetate, or Megace), danazol, or gonadotropin-releasing hormone analogues appear to be equally effective.
NOTE: Laparoscopic confirmation of the diagnosis of endometriosis before initiation of second-line treatment may be performed but is not required.
PREVENTION OF ENDOMETRIOSIS
Whenever severe dysmenorrhea occurs in a young patient, the possibility of varying degrees of obstruction to the menstrual flow must be considered. The possibility of a blind uterine horn in a bicornuate uterus or an obstructing uterine or vaginal septum should be kept in mind. In more than half of patients who are noted to develop endometriosis during childhood and adolescence, varying degrees of genital tract obstruction may be found.Whenever a congenital abnormality of the urinary or intestinal tract is detected, the genital tract should be investigated for an obstructive lesion. Infants with genital tract obstruction have been noted to develop endometriosis even in the first year of life. In all women, minimization of menstrual flow and suppression of ovarian cycling can reduce the risk for endometriosis.
Adenomyosis is defined as the extension of endometrial glands and stroma into the uterine musculature more than 2.5 mm beneath the basalis layer. Often this is an incidental finding on pathologic examination when it is seen in up to 60% of women in their 40s. About 15% of patients with adenomyosis have associated endometriosis. Islands of adenomyosis do not participate in the proliferative and secretory cycles induced by the ovary.
Generally, the gross appearance of the uterus consists of diffuse enlargement with a thickened myometrium containing characteristic glandular irregularities, with implants containing both glandular tissue and stroma(Figure 25-4). The endometrial cavity is also enlarged. Occasionally, the adenomyosis may be confined to one portion of the myometrium and take the form of a fairly well-circumscribed adenomyoma. Contrary to the picture in a uterine myoma, no distinct capsular margin can be detected on cut section between the adenomyoma and the surrounding myometrium. The distinction between adenomyosis and uterine leiomyoma may not always be clear on ultrasonic examination. Figure 25-5 illustrates the typical gross appearance of an enlarged uterus with extensive adenomyosis.
FIGURE 25-4 Histologic illustration of adenomyosis causing the enlargement of the uterus. A hyperplastic nodule of myometrium is seen. Note the endometrial glands and stroma.
FIGURE 25-5 Enlarged uterus cut open to demonstrate homogeneous enlargement due to adenomyosis. A diagnosis of leiomyomas may be incorrectly made at the time of pelvic examination.
(Courtesy of Dr. Sathima Natarajan, Ronald Reagan–UCLA Medical Center.)
Although many women are asymptomatic, those who suffer from this condition typically complain of severe secondary dysmenorrhea and menorrhagia. Even though the islands do not cycle in response to ovarian hormonal stimulation, there is still prostaglandin release and local inflammatory changes that can induce pain and tenderness and may disrupt the vasoconstriction of the arterial arcade supplying the endometrium. Deep-thrust dyspareunia, especially premenstrually, can be caused by adenomyosis.
On pelvic examination, the uterus is generally symmetrically enlarged and somewhat boggy and tender if the examination is conducted premenstrually. Occasionally, it may enlarge asymmetrically, which makes it very difficult to distinguish adenomyosis from a myomatous uterus. The consistency of the enlarged adenomyomatous uterus is generally softer than that of a uterine myoma.
The treatment of adenomyosis depends entirely on the symptoms and the possibility of other diagnoses. Any history of new onset or worsening menorrhagia, particularly in a woman with risk factors for endometrial cancer, should be investigated by endometrial biopsy or fractional dilation and curettage with or without hysteroscopy to rule out malignancy. Conservative management with NSAIDs and hormonal control of the endometrium are mainstays of therapy. Combination oral contraceptives or hormone-containing patches and vaginal rings may be used to reduce cyclic blood loss and menstrual pain. DMPA, levonorgestrel IUD, and continuous oral contraceptive pills can be used to try to achieve amenorrhea. If the woman is not a candidate for any of these medical interventions or if medical treatments do not sufficiently control her symptoms, hysterectomy may be indicated. Endometrial ablation to control the bleeding is another option.
D’Hooghe T.M., Hill J.A. Endometriosis. In: Berek J.S., editor. Berek & Novak’s Gynecology. 14th ed. Philadelphia: Lippincott Williams & Wilkins; 2007:1137-1184.
Gambone J.C., Mittman B.S., Munro M.G., et al. Consensus statement for the management of chronic pelvic pain and endometriosis: Proceedings of an expert-panel consensus process. Fertil Steril. 2002;78:961-972.
Stratton P., Sinaii N., Segars J., et al. Return of chronic pelvic pain from endometriosis after raloxifene treatment: A randomized controlled trial. Obstet Gynecol. 2008;111:88-96.
Wen D., Guo S.W. The search for genetic variants predisposing women to endometriosis. Curr Opin Obstet Gynecol. 2007;19:395-401.