Endometriosis

INTRODUCTION

Background: Endometriosis, the presence of endometriumlike glands and stroma outside the uterus, is a common, poorly understood, and extremely debilitating benign gynecological condition. The psychological impact of the severe pain experienced by the patient is compounded by the negative impact of the disease on fertility. Our understanding of the etiology and pathophysiology of endometriosis is limited by the lack of a suitable animal model on which to study the anatomic correlates and natural history of disease (ACOG, 1999). No cure for the disease exists, and treatment is directed toward medical suppression, surgical excision, and symptom alleviation.

Adenomyosis is the invasion of the myometrium by endometrial tissue.

Frequency:

• In the US:

  Endometriosis occurs in 7-10% of women in the general population (Wheeler, 1989). It is an estrogen-dependent disease and, thus, usually affects women of reproductive age. Endometriosis has a prevalence of 20-50% in infertile women (Rawson, 1991; Strathy, 1982; Verkauf, 1987) and as high as 80% in women with chronic pelvic pain (Carter, 1994).

  Evidence of endometriosis was found on laparoscopy in 20-50% of asymptomatic women (Williams, 1972). About 4 per 1000 women are hospitalized with endometriosis each year.

  A familial association exists, with a 10-fold increased incidence in women with an affected first-degree relative (Cramer, 1987). Monozygotic twins are markedly concordant for endometriosis (Hadfield, 1997).

CLINICAL

History: A significant number of women with endometriosis remain asymptomatic.

• Cyclic pain: Cyclic pain is pain that accompanies bleeding at the time of menstruation. This could involve the bladder (hematuria), bowel (hematochezia and painful defecation) or, rarely, bleeding at uncommon sites such as the umbilicus, abdominal wall, or perineum.

• Chronic pain: The most important point to remember is that the degree of visible endometriosis has no correlation with the degree of pain or other symptomatic impairment (Demco, 1998). Pain, does, however, correlate with the depth of tissue infiltration (Konickx, 1992; Konickx, 1994). Midline disease generally is believed to be more painful than lateral disease.

• Acute exacerbations are believed to be caused by chemical peritonitis due to leakage of old blood from an endometriotic cyst. Recently, with conscious laparoscopic pain mapping, the discovery was made that painful lesions involve peripheral spinal nerves rather than autonomic nerves (Demco, 1998).

• Dysmenorrhea: Secondary dysmenorrhea occurs twice as often in women with endometriosis as in controls (Williams, 1977). Pain frequently commences prior to menses. In a patient presenting with significant dysmenorrhea, endometriosis should be suspected, and the patient should be started on empiric therapy.

• Dyspareunia: Deep dyspareunia may be due to scarring of the uterosacral ligaments, nodularity of the rectovaginal septum, cul-de-sac obliteration, and/or uterine retroversion. All of these also may lead to chronic backache. Both of these symptoms are exaggerated during menses.

Physical:

• Pelvic examination: Tenderness on examination is best detected at the time of menses. Nodularity of the uterosacral ligaments and the cul-de-sac may be found. The uterus may be fixed in retroversion owing to adhesions. Occasionally, a bluish nodule may be seen in the vagina due to infiltration from the posterior vaginal wall.

Causes: Theories of causation include the following:

• Retrograde menstruation: Sampson described this theory in his classic paper published in 1927. Retrograde flow of endometrial tissue through the fallopian tubes into the peritoneal cavity is believed to cause endometriosis. Various animal experiments and clinical observations support this theory (Liu, 1986; Kruitwagen, 1991; Scott, 1953; Hooghe, 1996).

• Lymphatic and vascular spread: These pathways may explain the occurrence of endometriosis at distant, noncontiguous sites. Ovarian endometriosis also is believed to be caused by lymphatic spread (Ueki, 1991), although superficial ovarian endometriosis may also be due to implantation via retrograde menstruation.

• Coelomic metaplasia: Transformation of coelomic epithelium into endometrial type glands in response to as yet unknown stimuli could explain endometriosis in unusual sites (Suginami, 1991). Coelomic metaplasia also is believed to explain the occurrence of endometriosis in women who have undergone total hysterectomy and are not taking estrogen replacement (Metzger, 1991). Endometriosis also may occur in men on high-dose estrogen therapy (Schrodt, 1980).

• Immunogenetic defects: These are believed to increase the susceptibility of a woman to endometriosis. Humoral antibodies to endometrial tissue also have been found in sera of women with endometriosis (Mathur, 1982). Recent work has focused on studying the differences between eutopic endometrium and endometriosis. In endometriosis, an aberrantly expressed factor SF-1 activates the expression of the enzyme aromatase, which converts C19 steroids to estrogens. Estrogen , consequently, increases the synthesis of PGE2, which exerts a positive feedback effect resulting in increased aromatase activity. Additionally, endometriotic tissue is deficient in the enzyme 17-beta hydroxy steroid dehydrogenase type 2, which converts E2 in eutopic endometrium to the less potent E1 under the direction of progestins. A case report has been published describing treatment with the aromatase inhibitor Anastrozole in a women with severe postmenopausal endometriosis, resulting in dramatic symptom resolution (Zeitoun 1999). Additional data are needed before recommending this as treatment.

• Anatomic spread

• Ovary: This is the most common site for endometriosis. Spread to the ovary is believed to be lymphatic (Ueki, 1991) although superficial implants maybe due to retrograde menstrual flow, since the ovaries are in a dependent part of the pelvis. Lesions can vary in size from spots to large endometriomas. The classic lesion is a chocolate cyst of the ovary that contains old blood that has undergone hemolysis. Once intracystic pressure rises, the cyst perforates, spilling its contents within the peritoneal cavity. This can cause severe abdominal pain that is associated with the exacerbations typically seen with endometriosis. The inflammatory response causes adhesions that further increase the morbidity of the disease.

• Uterine serosa: Vesicular lesions may provoke an inflammatory response and scarring that cause the bladder to adhere anteriorly. Posteriorly, the disease may cause obliteration of the cul-de-sac and form dense adhesions between the posterior vaginal wall or cervix and the anterior rectum. Severe dyspareunia, dyschezia, and alteration of bowel habits are the clinical sequelae of this common spread.

• Uterosacral ligaments: Deep peritoneal disease is caused by infiltration of the uterosacral ligaments and rectovaginal septum by endometriotic nodules. Tethering of the uterus can lead to fixed retroversion. Dyspareunia is an important feature.

• Bowel: Through contiguous spreading, endometriosis may invade the rectovaginal septum and the anterior rectal wall. It also may involve the upper rectum and sigmoid colon, infiltrating the muscularis. Cyclical rectal bleeding (hematochezia) is pathognomonic of endometriosis. Transmural bowel involvement by endometriosis, however remains a rarity. The ileum, appendix, and cecum also may be involved, leading to intestinal obstruction. Cicatrization as a consequence of endometriosis may lead to symptoms of obstruction even in the postmenopausal woman.

• Genitourinary tract: Although uncommon, interference in the genitourinary tract by endometriosis can affect the bladder, ureters, and kidneys by invasion, compression, or scarring. Medical therapy has less than satisfactory results and surgical intervention often is required.

• Uncommon sites: Incisional scars, the umbilicus, and the thoracic cavity are also sites for endometriosis. Catamenial or cyclic pneumothorax can cause hemoptysis. Remember that ectopic endometrial tissue theoretically can undergo malignant transformation; histologic evaluation may be necessary.

• Postmenopausal endometriosis: This may be encountered in women who are on estrogen replacement therapy. Occasionally, endometriosis can be stimulated in an ovarian remnant after total abdominal hysterectomy if estrogen replacement therapy is being given. Extrapelvic endometriosis is believed to be hormone resistant when it occurs after surgical castration (Metzger, 1991). Transplantation of endometrial implants during the original surgery is believed to explain this occurrence. Coelomic metaplasia also has been advanced as a possibility.

WORKUP

Lab Studies:

• Serum CA-125: Serial measurements have a low sensitivity in detecting endometriosis (Barbati, 1994), but are useful as prognosticators of treatment outcome (Chen, 1998). However, normal posttreatment values do not mean that endometriosis is absent.

Imaging Studies:
　

• Ultrasonography: Transvaginal sonography is a useful method of identifying the classic chocolate cyst of the ovary. The typical appearance is that of a cyst containing low-level homogenous internal echoes consistent with old blood.

• MRI: MRI has a low sensitivity in detecting rectal involvement (Kinkel, 1999) but may detect deeply infiltrating endometriosis that involves the uterosacral ligaments and cul-de-sac. The cost-effectiveness of this imaging modality for endometriosis has yet to be justified in view of the low diagnostic yield.

Other Tests:
　

• Laparoscopy: Laparoscopy is considered the primary diagnostic modality for endometriosis. Endometriosis has been described as protean in appearance. The classic lesions are blue-black or have a powder-burned appearance. However, the lesions can be red, white, or nonpigmented. Peritoneal defects and adhesions are also indicative. Bear in mind that microscopic evidence of endometriosis may be found in normal-appearing peritoneum.

• Recently, conscious pain mapping with the patient awake has been used to locate the specific areas that cause pain (Demco, 1998). Subsequently, the patient is placed under anesthesia and those deposits are ablated.

TREATMENT

Medical Care:

• Endometriosis and subfertility

• Peritubal and periovarian adhesions can interfere mechanically with ovum transport and contribute to subfertility. Peritoneal endometriosis has been postulated to contribute to subfertility by interfering with tubal motility, folliculogenesis, and corpus luteum function. Studies concerning the role of prostaglandins have proved inconclusive.

• Medical treatment of minimal or mild endometriosis has not been shown to increase pregnancy rates (Badawy, 1988). Moderate-to-severe endometriosis should be treated surgically (Marcoux, 1997).

• Other options for achieving pregnancy include intrauterine insemination, superovulation, and in vitro fertilization. In a case-controlled study, pregnancy rates with intracytoplasmic sperm injection were not affected by the presence or extent of endometriosis (Bukulmez, 2001).Further, other analyses have shown improvement in IVF pregnancy rates with pre-treatment of Stage 3 and 4 endometriosis with GnRH agonists.

• Interval treatment: Some authorities believe that endometriosis should be suppressed prophylactically by continuous combined oral contraceptives, gonadotropin-releasing hormone (GnRH) analogs, medroxyprogesterone, or danazol in order to cause regression of asymptomatic disease and enhance subsequent fertility. Surgical ablation of asymptomatic endometriosis also has been shown to improve fecundity rates on a 3-year follow-up (Marcoux, 1997).

  　

• Recurrent pregnancy loss: On the basis of controlled prospective studies, no evidence indicates that endometriosis is associated with recurrent pregnancy loss, or that medical or surgical treatment of endometriosis reduces the spontaneous abortion rate (Vercammen, 2000).

  　

• Medical therapy: Combination oral contraceptives (COCP), danazol, progestational agents, and GnRH analogs form the mainstay of medical therapy. All these therapies have similar clinical efficacy in terms of reduction in pain-related symptoms and duration of relief.

• COCP: COCPs act by ovarian suppression and continuous progestin administration.

  　

  • Initially, a trial of continuous or cyclic COCP should be given for 3 months. If pain is relieved, this treatment is continued for 6-12 months. Subsequent pregnancy rates upon discontinuation of the pill are 40-50%. This applies to a population unselected for stage and fertility status of the disease.

    　

  • Although there is little to choose from among individual formulations, note that the long-term efficacy of multiphasic preparations is yet unproven.

    　

  • Continuous noncyclical administration of COCPs, omitting the placebo menstrual tablets, for 3-4 months helps avoid any menstruation and associated pain.

  　

• Progestational agents: All progestational agents act by decidualization and atrophy of the endometrium.

  　

  • Medroxyprogesterone acetate has proven efficacy in pain suppression in both the oral and injectable depot preparations (Hull, 1987; Kaupilla, 1993). Oral doses of 10-20 mg per day can be given continuously. The time to resumption of ovulation is longer and variable with depot preparations. Adverse effects include weight gain, fluid retention, depression, and breakthrough bleeding.

    　

  • Megestrol acetate has been used in doses of 40 mg with similarly good results (Schlaff, 1990).

  　

• GnRH analogs: The agents produce a hypogonadotrophic, hypogonadic state by down-regulation of the pituitary gland. Currently, goserelin and leuprolide acetate are the commonly used agonists.

  　

  • Once again, the efficacy is limited to pain suppression, and fertility rates may show no improvement (Hughes, 2001).

    　

  • Treatment usually is restricted to monthly injections for 6 months.

    　

  • Loss of trabecular bone density caused by GnRHis restored by 2 years after cessation of therapy (Paoletti, 1996). Other prominent adverse effects include hot flushes and vaginal dryness.

    　

  • Much interest has been shown in the question of whether add-back therapy should be instituted to prevent osteoporosis and hypoestrogenic symptoms. Hormone replacement therapy (HRT) preparations, progestins, tibolone maleate, and bisphosphonates all have been shown to be effective (Friedman, 1993; Taskin, 1997; Surrey, 1995; Lane, 1991). Add-back therapy has been shown to prevent loss in bone density and to relieve vasomotor symptoms without reducing the efficacy of GnRH regimes. GnRH agonists have been used for 12 months with norethindrone add-back therapy with good results (Surrey, 1999).

    　

  • A clinical trial has shown that a 3-month empiric course of GnRH, based on a diagnostic algorithm without definitive laparoscopic diagnosis, is efficacious (Ling, 1999). However, long-term effects of GnRH analogs on bone density in this population remain unproven. Therefore, " add-back" therapy remains the standard of care whilst on GnRH treatment. Also, empiric treatment without a firm diagnosis could result in unnecessary treatment in patients with chronic pelvic pain, whose condition could be due to other causes. In Ling's study, 13% of subjects were shown not to have endometriosis.

    　

  • GnRH therapy also has been proved to relieve the pain and bleeding associated with extrapelvic distant endometriosis (Espaulella, 1991).

  　

• Danazol: Danazol acts by inhibiting the midcycle follicle-stimulating hormone (FSH) and luteinizing hormone (LH) surges and preventing steroidogenesis in the corpus luteum. It is the most extensively studied agent for endometriosis.

  　

  • Danazol has been shown to be as effective as any of the newer agents, but with a higher adverse effect profile.

    　

  • Its androgenic manifestations include oily skin, acne, weight gain, deepening of voice, and facial hirsutism. Hypoestrogenic features due to danazol include emotional lability, hot flashes, vaginal dryness, and reversible breast atrophy.

    　

  • The recommended dose is 600-800 mg per day. However, smaller doses have been used with success (Telimaa, 1987; Dmowski, 1982).

    　

  • Because of the possibility of virilizing changes in a female fetus, additional barrier contraception must be used while on danazol therapy.

Surgical Care: Surgical care can be classified broadly as conservative when reproductive potential is retained, semiconservative when reproductive ability is eliminated but ovarian function is retained, and radical when the uterus and ovaries are removed. Age, desire for further childbearing, and deterioration of quality of life are the main considerations when deciding on the extent of surgery.

• Conservative surgery

• With conservative surgery, the aim is to destroy visible endometriotic implants and lyse peritubal and periovarian adhesions that are a source of pain and may interfere with ovum transport. Laparoscopic approach is the method of choice in treating endometriosis conservatively (Cook, 1991; Saidi, 1996). Ablation can be performed with laser or electrodiathermy. Overall, the recurrence rate is 19% and is similar for all techniques (Revelli, 1995). Ovarian endometriomas can be treated by drainage or cystectomy. Laparoscopic cystectomy was found to yield better pain relief and pregnancy rates than drainage (Beretta, 1998). Medical therapy with GnRH agonists reduces the size of the cyst but does not influence pain relief (Donnez, 1990).

• Presacral neurectomy is an operation used to relieve severe dysmenorrhea. The nerve bundles are transected at the level of the third sacral vertebra and the distal ends ligated. Vascular injury to the middle sacral artery and vein is a potential complication, and some authors advocate prophylactic ligation. Also, constipation is a long-term side effect (94%) of this procedure and should be factored in while deciding upon this procedure.

• Nodularity of the uterosacral ligaments may contribute to dyspareunia and low back pain. The transmission of neural pathways is via the Lee-Frankenh酳ser plexus. Laparoscopic uterine nerve ablation (LUNA) is performed to interrupt the pain fibers. Potential complications of this procedure included uterine prolapse and pelvic denervation. A systematic review of trials of LUNA found no advantage in terms of pain relief when compared to placebo (Wilson, 2001). However, when combined with laparoscopic ablation, LUNA significantly reduced pain attributed to endometriosis (Sutton, 1997). In patients with subfertility, tissue ablation significantly increased the cumulative pregnancy rate (Marcoux, 1997).

• For patients with mild disease, postoperative adjunctive hormonal treatment has been shown effective in reducing pain but has no impact on fertility. GnRH analogs, danazol, and medroxyprogesterone all have been found to be useful for this indication (Prentice, Deary, and Bland, 2001; Prentice, Deary, Goldbeck-Wood, Farquhar, and Smith 2001; Selak, 2001; Moore, 2001). However, for severe endometriosis, the efficacy of preoperative or postoperative hormonal treatment has yet to be established.

• Semiconservative surgery

• The indication for this type of surgery is mainly in women who have completed their childbearing, are too young to undergo premature menopause, and are debilitated by the symptoms. Such surgery involves hysterectomy and cytoreduction of pelvic endometriosis. Ovarian endometriosis can be removed surgically, as one tenth of functioning ovarian tissue is all that is needed for hormone production. Patients who undergo hysterectomy with ovarian conservation have a 6-fold higher rate of recurrence than women who undergo oophorectomy. (Namnoum, 1995).

  　

• Medical therapy in women who have completed childbearing is equally efficacious for symptom suppression. (Vercillini, 1997; Waller, 1993; Canadian Consensus Conference on Endometriosis, 1999)

• Radical surgery

• This involves total hysterectomy with bilateral oophorectomy and cytoreduction of visible endometriosis. Adhesiolysis is performed to restore mobility and normal intrapelvic organ relationships.
• Ureteric obstruction may warrant surgical release or excision of a damaged segment. Bowel obstruction may need a resection anastomosis or a wedge resection if the obstruction is confined to the anterior rectosigmoid.
• Endometriosis may recur in 15% of women after extirpative surgery, irrespective of whether ERT is given postoperatively (Redwine, 1994). ERT can be instituted safely immediately after surgery, especially in younger women who face the prospect of accelerated bone loss and vasomotor symptoms (Redwine, 1994; Hickman, 1998). There are no trials reporting the use of estrogen + progestin compared to estrogen alone therapy postoperatively. Theoretically, however, the addition of a continuous progestin could decrease the regrowth of endometriosis.

• Comparison of medical and surgical therapy: In women who wish to preserve their reproductive potential, the rates of recurrence of pain symptoms are 44% with surgical management and 53% with medical management (Sutton, 1997; Waller, 1993).

MEDICATION

The goals of pharmacotherapy are to reduce morbidity and prevent complications.

Drug Category: Oral contraceptives -- COCPs act by ovarian suppression and continuous progestin administration. Initially, a trial of continuous or cyclic COCP should be given for 3 months. If pain is relieved, this treatment is continued for 6-12 months. Subsequent pregnancy rates are 40-50% upon discontinuation of the contraceptive pill. Although individual formulations offer few variations, note that the long-term efficacy of multiphasic preparations is yet unproven. In addition, continuous noncyclical administration of COCPs omitting the placebo menstrual tablets for 3-4 months helps avoid any menstruation and associated pain.

Drug Category: Progestational agents -- All act by decidualization and atrophy of the endometrium.

Drug Category: GnRH analogs -- GnRH analogs produce a hypogonadotrophic, hypogonadic state by down-regulation of the pituitary gland. Currently, goserelin and leuprolide acetate are the commonly used agonists.

Drug Category: Antigonadotropic agent -- Acts by inhibiting the midcycle FSH and LH surge and preventing steroidogenesis in the corpus luteum. It is the most extensively studied agent for endometriosis. Danazol has been shown to be as effective as any of the newer agents, but with a higher adverse effect profile.

FOLLOW-UP

Prognosis:
　

• Endometriosis was found to resolve spontaneously in one third of women who were not actively treated (Harrison, 2000).