The Diagnosis and Treatment of Deep Infiltrating Endometriosis
Background: Endometriosis and adenomyosis uteri are the most common benign disorders affecting girls and women after uterine myomas (fibroids), with a prevalence of roughly 5% to 15%. There have been many advances in diagnostic assessment and in our understanding of the disease over the past decade. Steady improvements in treatment have been accompanied by heightened consciousness of the diagnosis among the affected women and the doctors who care for them.
Methods: A selective literature search was carried out in the Cochrane and PubMed databases using the key words “endometriosis,” “deep infiltrating endometriosis,” “endometriosis AND diagnostics,” “endometriosis AND surgical therapy,” “endometriosis AND endocrine treatment,” and others. The AWMF and ESHRE guidelines were also taken in account.
Results and Conclusion: The main manifestations are primary or secondary dysmenorrhea, bleeding disturbances, infertility, dysuria, pain on defecation (dyschezia), cycle-dependent or (later) cycle-independent pelvic pain, nonspecific cycle-associated gastrointestinal or urogenital symptoms. Cycle-associated problems of urination and/or defecation that are due to endometriosis are most common in young, premenopausal women. Whenever such manifestations are present, endometriosis should be considered in the differential diagnosis, and evidence for it should be sought in the clinical history, physical examination, and ultrasound findings. If endometriosis is histologically confirmed and is of the deeply infiltrating kind, the recommended management today is to refer the patient to an endometriosis center.
Endometriosis is a disease of the uterus in which tissue from the uterine cavity becomes implanted in the abdominal cavity and, rarely, metastasizes to organs at a distance from the uterus. Endometriosis tissue is biologically the same as basal endometrial tissue. Foci of endometriosis consist of glands, stroma cells, and smooth muscle; they are supplied by nerves (neurogenesis), lymphatic vessels, and blood vessels (angiogenesis) (1–5). Endometriosis cells express estrogen receptors (ER α/β) and progesterone receptors (PR A/B) and therefore respond to endocrine treatments (6, 7). Although endometriosis is considered a disease of women of child-bearing age, its occasional occurrence before the menarche has been histologically documented (4, 8). Postmenopausal endometriosis accounts for less than 3% of cases.
The German-language medical literature uses Latin terms to classify endometriosis by site: “endometriosis genitalis interna” is the name given to adenomyosis uteri, while “endometriosis genitalis externa” designates disease in the internal female genital tract and “endometriosis extragenitalis” designates disease elsewhere, e.g., in the appendix, the bowel, the urinary bladder, the ureter, the vagina, the lung, the liver, the umbilicus, and other, rare locations. Other classification schemes simply speak of peritoneal, ovarian, and deep infiltrating endometriosis. Any manifestation of endometriosis that is located other than in the superficial tissues of the rectovaginal septum and vaginal fornix, the pelvic wall, parametrium, bowel, uterus, or urinary bladder can be called deep infiltrating endometriosis. Endometriosis is currently staged according to the system of the American Society of Reproductive Medicine (ASRM) and the experimental ENZIAN classification.
The learning objectives of this article are
to describe the main manifestations of endometriosis, to present the diagnostic evaluation and classification of endometriosis, and to give an overview of the available curative and palliative treatment options.
The estimated prevalence of endometriosis is 5% to 15% among all women of child-bearing age; its prevalence is higher in some subgroups (2, 4, 6). 20% to 48% of women suffering from infertility have endometriosis. Among young women with chronic pelvic pain that does not respond to hormonal therapy or to treatment with nonsteroidal anti-inflammatory drugs (NSAID), the prevalence of endometriosis is roughly 70% (4). No robust data are available concerning the prevalence of deep infiltrating intestinal endometriosis, or of endometriosis of the urinary tract. Ureteric endometriosis, for example, has been described in 0.1% to 0.4% of all cases of endometriosis, while the overall prevalence of urogenital endometriosis is said to be 1% to 2% of the overall prevalence of endometriosis (9, 10).
According to the transplantation hypothesis, viable endometrial cells enter the abdominal cavity through retrograde menstruation and become implanted there (11). In what is called retrograde menstruation, the patient bleeds from the vagina during her menstrual period, but menstrual blood also simultaneously enters the abdominal cavity by way of an open Fallopian tube. This hypothesis fails to explain why endometriosis does not affect all women, because all women are thought to have retrograde menstruation. An answer is supplied by the currently favored etiological hypothesis for endometriosis, which is known as the tissue injury and repair (TIAR) concept (7). Embryologically speaking, the uterus is made up of the archimetra (the endometrial glands and stroma and the subvascular layer) and the neometra (the vascular and supravascular layers) (6, 7). The archimyometrium is seen on ultrasonography as a hypodense halo, and on magnetic resonance imaging (MRI) as a hypointense junctional zone, i.e., a layer that is distinct from the normal endometrium (2, 7). Microtrauma can occur at the interface between different layers of uterine tissue, e.g., in the region of the fundocornual raphe, as a consequence of the estrogen-driven increase in peristalsis (7). The repair mechanisms that then come into play are associated with local hyperestrogenism because of aromatase overexpression; they lead, therefore, to paracrine-estrogen-induced uterine hyper- and dysperistalsis, with desquamation and dislocation of the basal endometrium through the Fallopian tubes and out into the abdominal cavity (Figure 1 gif ppt) (4, 7). On the other hand, cells of the basal layer can also continually infiltrate the myometrium, giving rise to the fully developed clinical picture of uterine adenomyosis. Endometriosis and adenomyosis are two sides of the same coin (1, 6, 7). There are likely to be stem cells in the basal endometrial layer that play an essential role both in the cyclical regeneration of eutopic endometrial tissue and in endometriosis (12). At present, findings from the fields of genomics and proteinomics are shedding new light on the molecular biology of endometriosis (13).
The main manifestations of endometriosis include:
- Uterine hemorrhagic disturbances
- Primary or secondary sterility.
Women with endometriosis also complain, depending on the site of the disease, of gastrointestinal, urological, autonomic, and nonspecific manifestations, and of symptoms resembling chronic fatigue. We have noted depressive mood alterations and/or clinically relevant depression or anxiety disorders in more than 60% of the women with endometriosis whom we have treated. Thus, whenever advanced endometriosis is diagnosed, psychotherapy should always be considered as a potential component of treatment.
Endometriosis of the bladder
Tough, nodular adenomyofibrohyperplasia is found at the site of infiltration (Figure 2a jpg ppt) and can lead to painful, ineffective bladder contractions, as well as to microcirculatory disturbances in the urothelium, with resulting micro- or macrohematuria. The diagnostic assessment includes history-taking, vaginal palpation, vaginal ultrasonography with a full bladder, and magnetic resonance imaging (MRI) (Figure 2c). Invasive diagnostic techniques include cystoscopy and laparoscopy (10, 14) (eBox). Transurethral resection (TUR) is contraindicated in endometriosis, because endometriosis infiltrates transmurally from the outside in, i.e., toward the endothelium, and thus cannot be removed through the urethra (eFigure 1 a–f jpg ppt). Intravesical lesions can be biopsied through the cystoscope, and ureter stents can be inserted cystoscopically if necessary (10, 15).
Endometriosis of the ureter can be either intrinsic or extrinsic (15). These two types often cannot be reliably distinguished from each other before surgery. In the external type, which is more common, the ureter is compressed by the shrinkage of endometriosis tissue that encompasses it on the outside; this finding is typically described in cases of bilateral ovarian endometriosis (“kissing ovaries,” as named by Michel Mueller of Bern). The site of least resistance is the point where the sacrouterine ligament, the ureter, and the uterine artery cross (16) (eFigure 2 jpg ppt). Intrinsic ureteric endometriosis is rarer and infiltrates multiple layers of the ureter. It is present in less than 0.3% of all women with endometriosis (17). Its manifestations range from nonspecific pelvic pain to flank pain, renal obstruction (usually unilateral), and asymptomatic hydronephrosis, with loss of function of the affected kidney(s). Recommended studies for diagnostic evaluation include renal ultrasonography, an intravenous urogram, or MRI excretion urography, if available. Renal function should be assessed with laboratory tests (creatinine, blood urea nitrogen) and/or renal scintigraphy (eBox).
Infiltration of the adjacent intestine and of the sacrouterine ligaments, in addition to adhesion formation, can lead to partial or complete obliteration of Douglas’s pouch. This, combined with severe accompanying adenomyosis, causes infertility. Pelvic pain is often severe; intestinal manifestations are almost always present; and dyspareunia, sometimes leading to loss of libido, is typical. These problems are due to the location of the invasive foci and their innervation (18). In endometriosis of the bowel, transmural infiltration generally leads to stenosis or, rarely, occlusion of the intestinal lumen (Figure 2b). Abnormal microcirculation between the endometriosis nodule and the intestinal mucosa (less commonly, infiltration of the mucosa) causes cycle-dependent intestinal hemorrhage. Simultaneous infiltration of the parametria often involves the ureters as well (15, 19), so that bilateral ureterolysis and surgical exposure of the ureters may be necessary (9, 16, 19). For optimally precise preoperative diagnostic evaluation, history-taking and rectovaginal gynecological examination should be supplemented by transrectal ultrasonography, in combination with rectosigmoidoscopy, to determine whether the mucosa is involved. Invasive foci can be documented by MRI, which can also provide very clear evidence of uterine adenomyosis (Figure 2d). Tissue is obtained for histological diagnosis and staging either by laparoscopy or by laparotomy. Laparotomy is now performed less commonly than in the past (eBox).
The surgical treatment of endometriosis
Endometriosis is a chronic, hormone-dependent disease of the uterus, with a highly variable clinical course. Thus, the treatment should be designed according to the patient’s individual needs. This does not mean that it should be chosen arbitrarily (20). The physician should discuss with the patient whether the primary reason for treatment is acute or chronic endometriosis-related pain or an as yet unfulfilled desire to bear children (21).
The biology of endometriosis implies that the best way to treat symptomatic patients is with an individualized combination of surgery and endocrine (usually anti-estrogenic) pharmacotherapy, supported by complementary treatment approaches (eTable 1 gif ppt).
Laparoscopy is the gold standard for the surgical treatment of endometriosis. Robust evidence is currently unavailable for the surgical methods that are used to treat endometriosis or deep infiltrating endometriosis (as is the case for many other diseases as well).
Although the best treatment for endometriosis is generally surgery combined with pharmacotherapy, asymptomatic patients with stenosing ureteric endometriosis are an exception to this rule. For these patients, surgical treatment is clearly indicated (evidence level 1a). It remains unclear whether asymptomatic ovarian endometriomas require operative treatment, and surgery in this area can damage the adjacent normal ovarian tissue. The question is relevant for no more than a few women, as only about 5% to 10% of all cases of endometriosis are thought to be asymptomatic (evidence level 4). At any rate, it is just as wrong to neglect the clinical manifestations of endometriosis (eBox) as it is to pursue zealous medical and/or surgical overtreatment of the disease, while disregarding the psychosomatic, social, occupational, and conjugal problems that it causes. Cyclical symptoms and pain that make a woman feel ill, confine her to bed, or lead her to abuse medications require timely clinical and laparoscopic evaluation by a gynecologist. Endometriosis should be either excluded or confirmed before any treatment is initiated. The most serious complication of endometriosis treatment is chronification of the patient’s symptoms.
Endometriosis of the bladder
The primary treatment modality for symptomatic endometriosis of the bladder is surgery (10, 15). The surgeon’s experience determines whether this is best done via laparoscopy or laparotomy (10, 14). A number of innovative surgical techniques have been developed (10, 15, 17).
First, the infiltrated portion of the bladder should be dissected free of the body of the uterus or the cervicoisthmic junction until the macroscopically disease-free vesico-uterine space is reached (eFigure 1). Next, a whetstone- or orange-slice-shaped partial vesical resection is performed (10). The trigone of the bladder near the ostium, together with its neural structures, is the most vulnerable part of the bladder whenever partial vesical resection is performed, either openly or laparoscopically (10, 15). Nonetheless, an R0 resection should be the goal. The bladder is then closed with a seromuscular suture and tested for leak tightness by retrograde filling with methylene blue dye. Transurethral urinary drainage is recommended for six days after surgery. The most serious complication of this operation is a so-called neurogenic bladder: vesical denervation, caused either by endometriosis or its treatment, may necessitate either permanent catheterization or the implantation of a vesical neurostimulator in a young female patient. Adjuvant anti-endocrine therapy is given in accordance with the current national guidelines for deep infiltrating endometriosis (e4).
Rehabilitation measures or treatment in specialized facilities is indicated for many women who suffer from endometriosis as a chronic disease (evidence level 4).
Endometriosis of the ureter
Surgery for deep infiltrating endometriosis, or for pelvic adhesions secondary to endometriosis, carries an elevated risk of ureteric complications (9). These can include urinoma formation or uroperitoneum leading to infection or other adverse consequences for the involved kidney. Complex operations should be planned and carried out in an interdisciplinary collaboration (20). In extrinsic ureteric endometriosis, the goal of surgery is freeing (ureterolysis) and decompression of the ureter (15). In intrinsic ureteric endometriosis, an additional objective is partial resection of the ureter with end-to-end anastomosis or direct ureteric neoimplantation, e.g., with the psoas hitch technique (15). The ureter often must be freed of surrounding tissue all the way to its junction with the bladder to allow safe resection of the infiltrated parametria. At the same time, the retroperitoneal course of nerves lying in the operative field (such as the hypogastric, splanchnic, femoral, and obturator nerves) must be laparoscopically exposed, so that a neurogenic bladder-emptying disturbance can be avoided. After extensive surgery in the area of the ureters, it is recommended that ureteric stents should be left in place for four to six weeks.
Asymptomatic or oligosymptomatic deep infiltrating endometriosis is rare and can be treated conservatively, after thorough discussion with the patient, as long as evaluation reveals no stenosis, hemorrhage, or progressive disease needing treatment. Such patients should undergo annual rectovaginal examinations by an experienced gynecologist. Depending on the findings, further diagnostic studies may be indicated, e.g., MRI of the uterus and rectovaginal septum or transrectal ultrasonography, combined with rectosigmoidoscopy where appropriate (evidence level 2). Surgery is currently the treatment of choice for symptomatic rectovaginal endometriosis (20). Many operative techniques have been developed for this purpose, all of them with the goal of an R0 resection (eTable 1). Regardless of the surgical approach used, the infiltrated rectosigma or sigma must be mobilized away from normal and pathological adhesions (Case Illustration jpg ppt) and then resected, after which an end-to-end anastomosis is performed. The patient must, of course, be fully informed about the nature of the procedure before it is performed (16, 19). Women with deep infiltrating endometriosis are today best served by treatment in a specialized endometriosis center. This is the case because of the potential complications of treatment, which become more likely with increasing severity of the illness. Special attention must be paid to the problem of suture-line dehiscence leading to rectovaginal fistula formation and further problems (22–24). As far as the treatment of pain is concerned, the data from clinical trials published to date have not led to the widespread adoption of laparoscopic uterine nerve ablation (LUNA) or other nerve-ablation techniques. Any adjuvant and/or experimental treatments that might be proposed should be discussed with the patient individually, in the light of her particular clinical circumstances and living situation (25).
Once the diagnosis of endometriosis is histologically confirmed, endocrine pharmacotherapy can be used as a neo-adjuvant or adjuvant measure, as well as for the treatment of recurrences. Surgeons generally do not favor neo-adjuvant endocrine pharmacotherapy because of its unfavorable effect on tissue planes. In cases of extensive endometriosis, and particularly in deep infiltrating endometriosis, an R0 resection can only rarely be achieved. It therefore makes sense to give adjuvant endocrine pharmacotherapy with the goal of transient therapeutic amenorrhea. The following options are available at present:
b) oral contraceptives
c) GnRH analogues
d) pain therapy
e) combinations of the above
f) experimental treatment approaches.
Gestagens effect a secretory transformation of the endometrium after previous exposure to estrogens (Tables 1 and 2).
The potential side effects of gestagen supplementation are:
- breakthrough bleeding (40% to 80%)
- weight gain and fluid retention (40% to 50%)
acne, skin changes of endocrine origin, and other skin problems (20%)
- hot flashes, loss of libido (more than 30%)
- breast tenderness (10%)
- headache (10%)
vaginal dryness, transient alopecia, osteopenia, mood swings (10%)
Gestagens also have an effect on a variety of metabolic processes and indices:
atherogenic indices (rising LDL, falling HDL) carbohydrate metabolism (lower glucose tolerance, rising insulin resistance).
Gestagen supplementation has also been found to lead to a higher incidence of venous thromboembolism. Other long-term effects include a mild natriuretic effect, alteration of cervical secretions, and vaginal candidiasis (a common finding).
Oral contraceptives (when used off label for this indication) bring about a so-called pseudopregnancy regimen (Table 2 gif ppt, eTable 1). Their well-known side effects, which vary in frequency from one preparation to another, include breakthrough bleeding, nausea, headache, and an elevated risk of venous thromboembolism, as well as loss of libido, cutaneous reactions, sodium and fluid retention leading to weight gain, breast tenderness, and a rise in blood pressure. Generally, however, oral contraceptives are very well tolerated.
The goal of treatment is suppression of the menses (therapeutic amenorrhea). If breakthrough bleeding occurs, the patient can take one oral contraceptive tablet twice a day for as long as smear bleeding persists and for one day afterward, and then return to a single tablet per day. It is important for the patient to be properly informed.
GnRH analogues bring about a “functional oophorectomy,” i.e., a state of pharmacologically induced hypogonadotropic hypogonadism (Table 2, eTable 1). This, in turn, causes well-recognized side effects:
- hot flashes and perspiration (80% to 90%)
- sleep disturbance (60% to 90%)
- vaginal dryness (30%)
- headache (20% to 30%)
mood changes (depressive mood change because of estrogen withdrawal, more than 10%)
- osteopenia, loss of libido (more than 30%)
- weight gain (ca. 15%).
Combined treatment approaches (pharmacotherapy and pain therapy)
In addition to surgery and pharmacotherapy, complementary treatments can be used whose efficacy has not been documented by scientific evidence. Women whose quality of life is impaired by cyclical or chronic pain want treatment in order to achieve a pain-free state with a better quality of life and an improved ability to engage in productive activities (eTable 2 gif ppt). In a specialized endometriosis center, the patient’s individual situation can be stabilized or improved through a team effort, with the active participation of the patient herself, her treating gynecologist, the surgeon, the pain specialist, and the psychotherapist. Before treatment, these women’s problems are often severe enough to cause the loss of a job or of a life partner.
Experimental treatment approaches
Endometriosis cells manifest properties such as invasiveness, migration, metastasis, angiogenesis, and neurogenesis that call to mind similar properties of malignant tumors. Their responsiveness to cytokines, tumor necrosis factor (TNF-α), cyclooxygenase-2 (COX-2), oxytocin, and aromatase is currently being exploited in the attempt to devise new methods for diagnosis and treatment (3, 4, 13, 25). Although a combination of aromatase inhibitors with gestagens or GnRH analogues has been proven effective, the practicality of this form of treatment is currently limited both by its side effects and by its cost (4).
Therapeutic nihilism is not justified, even for women with extensive endometriosis. Germany is currently in the European vanguard with respect to the quality of care for patients with endometriosis, the available institutional structures for patient care and research, physician training, continuing medical education, and the certification of endometriosis centers at various levels. All of these activities are carried out with the aid of the German Foundation for Endometriosis Research (Stiftung Endometriose-Forschung [SEF]), the European Endometriosis League (EEL), and the German Endometriosis Association (Endometriose-Vereinigung Deutschland e.V.) (an umbrella organization of self-help groups).
We sincerely thank Prof. Dr. G. Leyendecker (Darmstadt) for his critical remarks and suggestions, as well as Drs. Cordula von Kleinsorgen, Raida Dakkak, and Gabriela Rosenow for their advice and support. We also extend our thanks to our clinical partners, PD Dr. K. Krüger (Department of Radiology), PD Dr. R.-M. Liehr (Department of Internal Medicine/Gastroenterology), Dr. J. Haßelmann (Urology Department) and Prof. Dr. U. Adam (Visceral Surgery) at the Vivantes Humboldt-Klinikum, Berlin.
Conflict of Interest Statement
Dr. Gülden Halis has received study support and congress support from Takeda Pharma Germany and Bayer-Schering Pharma, as well as fees for lecturing or consulting contracts from Bayer-Schering-Pharma. Dr. Sylvia Mechsner has received study support from Bayer-Schering Pharma and lecture fees from Wyeth, Bayer Schering Pharma, and Takeda Pharma Germany. Prof. Andreas D. Ebert has received research and congress support from Takeda Pharma Germany, Novartis, and Bayer-Schering Pharma, as well as fees for lecturing or consulting contracts from Bayer-Schering Pharma, Wyeth, and Ethicon Endosurgery.
Manuscript submitted on 5 March 2010; revised version accepted on 12 May 2010.
Translated from the original German by Ethan Taub, M.D.
Prof. Dr. med. Dr. phil. Dr. h. c. Andreas D. Ebert
Deutsches Endometriosiszentrum Berlin
Klinik für Gynäkologie und Geburtsmedizin
Am Nordgraben 2
13509 Berlin, Germany
@For e-references please refer to:
peritoneal endometriotic lesions and ovarian endometriotic cysts. Fertil Steril. 2005; 83 (Suppl 1): 1220–31.
Invest 2009; 67: 158–61.
Invasive Gynecol 2009; 16: 713–9.
Praxis für Fertilität; Kinderwunsch- und Endometriosezentrum Berlin: Dr. med. Halis
Endometriosezentrum Charité, Klinik für Gynäkologie, Charité-Universitätsmedizin, Campus Benjamin Franklin, Berlin: Dr. med. Mechsner