The characterization of endometriosis has not changed much since Sampson first formally described it. It is the presence of endometrial glands and stroma outside the endometrial cavity and uterine musculature.

Since endometriosis is dependent on ovarian steroids for its existence and proliferation, its occurrence and clinical importance are confined generally to the reproductive years. Although endometriosis has been reported in premenarchal and postmenopausal women not using exogenous hormones, the mean age at diagnosis is 25–29 years [1].

The prevalence of endometriosis is difficult to establish. Different authors have suggested a prevalence of 3–10% in reproductive-aged women. A precise determination of prevalence is hard because definitive diagnosis requires laparoscopy with possible biopsies, and even the presumptive diagnosis may be skewed by symptoms, access to care, and cultural attitudes towards pain. However, studies have shown an association between endometriosis and delayed childbearing, lower BMI, and Caucasian race [2]. The prevalence of documented endometriosis at time of surgery ranges from 1% found in all gynecologic procedures to higher values (9–50%) when the surgery is performed to evaluate pelvic pain or infertility [3].

While the exact etiology of endometriosis is still unclear, there are several proposed theories.

The coelomic (peritoneal) cavity has undifferentiated cells that may undergo metaplastic transformation to endometriosis. Support for this hypothesis resides in cases of endometriosis in men, premenarchal girls, and those with congenital absence of the uterus. However, this theory remains problematic. Although the endometriosis cases mentioned above do occur, they represent a rare presentation of the disorder. Also, even though the coelomic membrane overlies the thoracic and abdominal cavities, the vast majority of endometriosis resides in the pelvis. Furthermore, metaplasia in general increases with older age, but endometriosis first presents at a relatively young age [4].

The transplantation theory holds that the endometrium is actively disseminated to an ectopic location. This can occur via mechanical, lymphatic or vascular mechanisms. Retrograde menstruation (mechanical transport) is widely accepted as a causative factor in endometriosis. Lending credence to this are several studies demonstrating lower endometriosis rates with later menarche, irregular menstrual cycles, more deliveries, and longer oral contraceptive use [5]. The fact that endometriosis is primarily a disease of menstruating women, especially those with regular menses, also sustains this theory.

The induction theory combines the above two hypotheses. It suggests that shed endometrium secretes a biochemical substance that causes differentiation of peritoneal cells.

Altered immunity, both humoral and cell mediated, appears to play a role in endometriosis as well. First, impaired immune detection of ectopic endometrial nests has been demonstrated. In women with endometriosis, there may be alterations in their human leukocyte antigen (HLA) class I expression, which affects immune recognition [6]. Second, a decreased immune response contributes to the persistence of the ectopic endometrium. Decreased natural killer cell activity in women with endometriosis reduces cytotoxicity to the endometriosis cells [7]. Apoptosis may be inhibited, both by an increased amount of antiapoptotic factors and by resistance to interferon-γ-induced apoptosis [8]. Compared to a nonendometriosis group, a survey demonstrated a higher association between women with endometriosis and other autoimmune inflammatory diseases, such as hyperthyroidism, fibromyalgia or chronic fatigue syndrome [9].

Recent research has focused on the adhesion and proliferation of endometriosis implants. One theory ties together altered immune response with growth factors that appear to be vital in the development of endometriosis, suggesting that increased peritoneal leukocytes and macrophages, together with endometriotic nests, secrete cytokines and growth factors. The cytokines, such as specific interleukins and tumor necrosis factor, stimulate implant growth, angiogenesis, and a greater inflammatory milieu [10].

Hormones may affect endometriosis. There is increased aromatase activity in women with endometriosis, which increases estrogen levels in the implants.

In some women, a genetic component of endometriosis has been established, probably via a multifactorial, polygenic inheritance pattern. First-degree relatives of women with endometriosis have a 7% chance of developing endometriosis compared to 1% chance in controls, i.e. unrelated persons [11]. A combination of different theories, plus alterations in immunity, expression of different growth factors, and possible genetic influences likely represent the complex etiology of endometriosis.

The most frequent pelvic locations for endometriosis are the ovaries (involved in 66% of cases), uterine ligaments (round, broad, uterosacral), pelvic peritoneum, and the rectovaginal septum. Other sites include the umbilicus, laparoscopy/laparotomy scars, hernial sacs, liver, appendix (involved in 2–4% of cases), small intestine, sigmoid, rectum, bladder, ureters, vulva/vagina/cervix, lymph nodes, extremities, pleural cavity, and lung. The multiplicity and widespread distribution of these sites make acceptance of one histogenetic theory difficult.

Like everything else connected with endometriosis, the gross pathology is characterized by variability. Endometriotic lesions classically appear as “powder-burn,” black or blue-ish areas. However, they may be clear, red, yellow or white. The different manifestations of endometriosis may depict the various phases of the disease, with clear or reddish lesions signaling active proliferation and darker lesions or fibrotic areas indicative of quiescence. At times, the only evidence of endometriosis is a peritoneal window. With diagnostic uncertainty, including areas of suspected nonpigmented endometriosis, laparoscopic biopsy will help establish the histologic diagnosis. When biopsy specimens of grossly normal peritoneum are taken, 25% of patients demonstrate microscopic endometriotic foci.

Endometrial islands may occur anywhere on the pelvic peritoneum, initially involving just the serosal surface but, with progression, possibly infiltrating deeply (>5–6 mm) into the retroperitoneum. Women presenting with endometriosis and pelvic pain usually have this type of deep penetration. Occasionally, invasion and penetration occur in the sigmoid. Progressive submucosal scarring can result in luminal constriction. However, mucosal involvement with associated rectal bleeding is a late phenomenon.

More significant ovarian involvement means the formation of unilateral or, commonly, bilateral endometrial cysts (endometriomas, “chocolate” cysts). Even when quite small, the cysts show a strong tendency to perforate with escape of menstrual blood and subsequent ovarian adherence to any adjacent structure, usually the posterior surface of the broad ligament or uterus.

Definitive diagnosis requires microscopic demonstration of endometrial tissue, preferably both glands and stroma. However, a wide range of patterns may occur, from normal-appearing to completely denuded endometrium, secondary to repeated menstrual bleeding and desquamation, with hemorrhage and pigment-laden macrophages the only microscopic clues. In one-third of clinically typical endometriotic cases, a specific histologic diagnosis cannot be made.