In endometriosis typical symptoms include dysmenorrhoea, deep dyspareunia (pain on deep penetration), dyschezia (pelvic pain with defecation), dysuria (pain with micturition) although the association between endometriosis stage and severity of pelvic symptoms has been demonstrated to be marginal and inconsistent [11]. Symptoms are usually synchronous with menstruation.

Pelvic pain is an important issue in the health care of women contributing to 10 % of all outpatient gynaecological visits, 40 % of laparoscopies and is the indication for 10–15 % of hysterectomies [12]. The existence of a relationship between chronic pelvic pain symptoms and endometriosis is widely accepted, but various other painful pelvic symptoms are also normally present in the general population.

Women presenting with rectal endometriosis are more likely to report an increase in intensity and duration of dysmenorrhoea, while deep dyspareunia appeared to be more severe in women with superficial endometriosis. Women presenting with rectal endometriosis are more likely to present cyclic defecation pain (67.9 %), cyclic constipation (54.7 %) and a significantly longer stool evacuation time, although these complaints are also frequent in women with Stage 1 endometriosis and in women with deep endometriosis without digestive involvement. No independent clinical factor has been found to be related to infiltration of the rectum by deep endometriosis and few women with rectal endometriosis present with rectal stenosis, however these women are significantly more likely to report constipation, defecation pain, appetite disorders, longer evacuation time and increased stool consistency without laxatives. Various digestive symptoms seem to be more related to cyclic inflammation than to rectal infiltration by the nodule, as they occur in women free of rectal involvement [13].

Ileocecal endometriosis may mimic appendicitis while other intestinal locations may cause bloating and discomfort related to narrowing, however intestinal occlusion is exceptional.

Transvaginal sonography (TVS) performed after bowel preparation should be the first-line imaging examination and is the best imaging modality for identifying intestinal lesions, determining the depth of bowel wall invasion and the circumference of involved bowel. This method allows dynamic evaluation from the anal verge to the sigmoid with high spatial resolution and minimal patient discomfort. The proximity between the transducer and the targeted structure provides superior contrast resolution, which is important for visualizing small and laterally located lesions. A useful tool for preoperative mapping of endometriosis by TVS (the Endometriosis Surgical-Ultrasonographic System) has recently been developed with the specific aim of creating a common language so that physicians who are dedicated to the diagnosis and treatment of patients with severe endometriosis can accurately share clinical data. It gives clinicians the opportunity to decide on the best surgical approach, to evaluate the potential need to involve other surgical specialists (general surgeon or urologist), to establish a tailored management of the disease, and to properly inform patients of the extent of the disease and therapeutic options [14].

TVS is as accurate as transrectal US for diagnosing intestinal lesions and identifying the bowel layers affected, and it yields better results than magnetic resonance (MR) imaging for the assessment of deeply infiltrating endometrial implants in other pelvic locations, especially small (<1.5-cm-diameter) lesions of the uterosacral ligament and bladder [15]. MR imaging is an excellent method for identifying old hemorrhagic content that characterizes endometriomas and for mapping multiple DIE implants, given its large field of view, multiplanar capabilities, and outstanding contrast resolution. Extensive pelvic adhesions and ureteral involvement are two important indications for MR imaging [16]. In DIE lesions are mostly hypoechoic in comparison with the myometrium: on MR images, they have signal intensity similar to that of smooth muscle, with low signal intensity on T2-weighted images, intermediate signal intensity on T1-weighted images, and minimal enhancement after the intravenous injection of contrast material. Cystic areas may be present, with or without hemorrhagic content.

MR imaging is less sensitive than endorectal ultrasound (ERUS) for DIE in the rectal wall but is more specific, up to 95 % [3]. The MR sensitivity may be improved with by use of endocavity probes [16] or with intra-rectal contrast [17] (Fig. 20.4). Since intestinal involvement is often associated with multifocal disease, MR imaging is essential for complete staging of disease. ERUS is the standard technique to evaluate DIE of the rectal muscular layer with 97 % sensibility and 85 % specificity [18]. TVS may have similar sensitivity and specificity for rectal endometriosis with less patient discomfort and in expert hands could replace ERUS [7]. CT scan is of less value and is used when MRI is impracticable or for specific sites (e.g. urethral involvement) [17].

Endometriosis is an estrogen-dependent disease that tends to disappear with menopause. Medical treatments, based on the concept that the eutopic and ectopic endometrium respond similarly to sex hormones, are hormonal and based on blocking ovarian function. Several therapeutic classes are available (combined oral oestroprogestative, GnRH analogues etc.) and offer free intervals from pain causing atrophy of endometriotic implants. Their effectiveness is similar, so the choice is based on cost and balance of side effects [20]; in this context, combined oral contraception is often used initially [11]. In infertile patients, in vitro fertilization may be considered either before or following surgery for DIE [21].

Bowel surgery should only proceed on the basis of shared decision-making after thorough consideration of risks versus benefits, ideally following multi-disciplinary consultations and full information to the patient who shares the final decision. The role of a purely diagnostic laparoscopy has been questioned and, ideally, there should always be the option of continuing to surgical removal of endometriosis, within the limitations of the surgeon’s expertise. It is also important, particularly in cases of more severe endometriosis, that surgeons consider the option of limiting surgical excision at an initial operation in order to refer to a surgeon better equipped to deal with endometriosis, as a single radical surgery has been shown to deliver the greatest benefit [22]. Surgery in the follicular phase of the menstrual cycle avoids the presence of a hemorrhagic corpus luteum and one study suggested an increased recurrence rate for surgery undertaken in the luteal phase, possibly due to re-implantation through retrograde loss of endometrial tissue at subsequent menses while the sites of surgically removed lesions are healing [23].

The experience of the surgeon is critical in a decision for local removal of a nodule infiltrating the intestinal wall as incomplete resection may not relieve symptoms [24], while radical intervention increases the risk of major complications such as ureteric and rectal injuries.

First operations tend to produce a better response than subsequent surgical procedures, with pain improvements at 6 months in the region of 83 % for first excisional procedures versus 53 % for second procedures [22, 25]. Excessive numbers of repeated laparoscopic procedures should therefore be avoided.

The surgical options for treating DIE in the bowel include peeling, disc excision or segmental excision and re-anastomosis. The peeling technique has the advantage that it avoids opening the bowel but it carries a risk of incomplete excision [6] and bowel microperforation that may not be recognized potentially leading to post-operative pelvic peritonitis. The hydropneumatic test (visualization of bubbles in the pelvis filled with water after air insufflation of the rectum) may be used to identify microperforation but is not always diagnostic and resection may be considered as safer in some cases.

Nezhat et al. [26] were the first to report a case series of eight patients treated with laparoscopic disk excision for endometriosis affecting the anterior colonic wall. Anterior rectal wall excision using a circular stapler was first proposed by Gordon et al. [27] to avoid the risks of a low extraperitoneal anastomosis in cases of nodular endometriosis invading the rectovaginal septum. Others [28, 29] have confirmed the feasibility of laparoscopic full thickness disk excision of endometriosis. The upper limit of the size of the lesion that can be removed is questionable and subjective because it is assessed visually, but it is generally agreed that lesions excised in this way should not exceed 2–3 cm in diameter and should not involve more than one-half of the circumference of the rectum. Disk excision may result in incomplete excision as endometriosis may infiltrate the large bowel wall preferentially along the myenteric plexus up to 3 cm from the palpated lesion [30].

The decision to carry out segmental bowel resection should be individualized. A decision to resect is supported by the depth of nodular infiltration, size and multicentricity of nodules, and the risk of incomplete excision [31]. Currently, only a complete surgical resection based on removal of all endometriosis lesions is considered adequate to controls the symptoms [32, 33] and prevent recurrence [34, 35]. The issue of how to treat lymph node involvement is unresolved as the clinical importance of lymph node involvement is not yet clarified [36].

The variety and depth of infiltration into other organs is testified by the number of previous surgical interventions, up to 82 % in the series reported by Dousset et al. [37] and by the number of synchronous procedures on the reproductive and urinary organs. Ileocolic locations should be searched carefully intraoperatively because preoperative imaging fails to identify these in over 50 % of cases. These may be treated by appendectomy, caecal, ileal or ileocaecal resection depending on the location.

Patients with DIE and rectal involvement may require a sub-total or total proctectomy depending on the level of lesion in relation to the anal sphincter with subsequent low or ultralow coloanal anastomosis. The resection may be extended “en block” to posterior vaginal cul-de-sac, uterosacral ligaments or sigmoid when involved; this is extensive surgery and requires in most cases a protective temporary ileostomy to prevent the sequele of an anastomotic leak or a rectovaginal fistula.

The ureters should always be identified and followed distally to ensure the absence of infiltration by endometriotic nodule. Extrinsic involvement is treated with a “decompression” protected by a double J stent. Intrinsic ureteral involvement is treated by resection followed by ureteral-bladder reimplantation or nephrectomy in case of chronic ureterohydronephrosis. During mobilization, the hypogastric plexus should be identified and preserved to avoid sexual or urinary functional alteration such as peripheral neurogenic bladder. This may be achieved by preserving the superior haemorrhoidal artery. Post operative neurogenic bladder is more common following proctectomy with coloanal anastomosis, total hysterectomy and presence of up to four DIE pelvic locations [37].