Successful treatment of endometriosis requires differentiating its symptoms from other sources of pain. Not all pelvic pain is due to endometriosis. If non-endometriotic sources of pain are ascribed to endometriosis, the results of treatment will seem confusing. While the interpretation of symptoms is part of the art of medicine, endometriosis is usually a disease that produces predictable symptoms.

Symptoms of endometriosis

A very common pattern of emergence of symptoms due to endometriosis is the occurrence of increasingly severe pain with menstruation beginning with menarche. Family and friends may offer the faint encouragement that such pain is 'normal' or 'part of being a woman'. However, it should be kept in mind that the youngest patient ever reported with endometriosis was 10.5 years of age and was diagnosed surgically five months after menarche [1]. Over-the-counter pain pills, heating pads and warm baths are commonly used in this age group. With the increase of symptoms comes absence from school and occasional trips to the emergency room where examination and tests are frequently non-conclusive, and the opinion is given to the patient that she may be overreacting. The young woman begins to doubt herself, thinking that perhaps it is 'all in her head' and tries to keep her pain to herself as best she can. The pain, however, may begin to occur earlier and earlier before the menstrual flow and become impossible to ignore. Many patients are bothered between ovulation and menses by increasingly severe pain which can be described with adjectives, such as 'sharp', 'shooting', 'burning' or 'knife-like'. With menses, severe uterine cramping may be superimposed on the endometriosis pain which has increased during the luteal phase of the ovarian hormonal cycle. Some patients will clench their fists as if holding a knife and make phantom thrusts of the weapon into their pelvis or display a twisting motion of their fists as they thrust toward the physician. Sports activities may be dropped because the pain can be aggravated by strenuous, jarring activities. Sometimes, the pain may be aggravated by a bumpy automobile ride [2]. Invasive endometriosis with obliteration of the cul-de-sac may occasionally produce mild systemic flu-like symptoms and mild temperature elevation. Invasive disease of the uterosacral ligaments can sometimes radiate pain down the back of the leg into the thigh. If the doctor listens attentively to the patient, common patterns of descriptions will be heard which can frequently predict where endometriosis is located, or perhaps suggest that endometriosis is not a cause of pain.

Pelvic endometriosis

Endometriosis is a disease that occurs in recurring and predictable patterns, and specific symptoms can almost be predicted from the pattern of pelvic involvement. The cul-de-sac, uterosacral ligaments and medial broad ligaments are the sites most commonly involved by endometriosis. Disease in these areas can be affected by the function of adjacent structures including the rectum and the vaginal apex. Thus, with 'garden variety' peritoneal involvement of the cul-de-sac or uterosacral ligaments, bowel movements may be painful, especially during menses, as opposed to patients with obliteration of the cul-de-sac and rectal wall involvement who may complain of rectal pain with each bowel movement during the month, or rectal pain with flatus or sitting. Deep dyspareunia is extremely common in patients with superficial or deep disease of the cul-de-sac or uterosacral ligaments, and patients may complain that 'something is being hit' during intercourse. With obliteration of the cul-de-sac, deep penetration during intercourse may radiate pain into the rectum. Because of increasingly severe dyspareunia, intercourse is increasingly avoided, and the male partner becomes acutely aware that what should be an act of pleasure has become an act of pain and endurance for the woman.

Endometrioma cysts can cause pain in the area of the involved ovary especially if periovarian adhesions are put on stretch, although pain from adjacent pelvic areas may be present and confuse the issue. Occasionally, an ovarian cyst which is adherent to the pelvic sidewall may put pressure on an underlying ureter, which itself can further be entrapped in retroperitoneal fibrosis related to inflammation from the overlying cyst, or from invasive disease of a nearby uterosacral ligament which can send tendrils of fibrotic scarring around the ureter. Rarely, the ureter will be invaded by endometriosis spreading from invasive disease of the uterosacral ligament, with resultant hematuria or obstruction. Pressure on or invasion of the ureter by these processes may result in cyclic stricture of the ureter during menses with resultant hydroureteronephrosis that can cause renal flank pain. With slow stricture of the ureter, the kidney on the affected side may die a silent death.

Intestinal endometriosis

Isolated nodules of endometriosis of the sigmoid colon may be asymptomatic or cause vague left lower quadrant pain prior to a bowel movement. Endometriosis of the appendix and cecum is usually asymptomatic. Intestinal obstructive symptoms are rare when only the rectosigmoid colon or cecum is involved by endometriosis because of their relatively large diameters. When symptoms of bowel obstruction exist, they are almost always due to obstructive lesions of the distal ileum which obstructs more readily because of its smaller diameter.

Diaphragmatic endometriosis

Endometriosis of the diaphragm most commonly exists as a full-thickness lesion of the right hemidiaphragm [3]. Patients will complain of worsening right chest or shoulder pain with menses. Sometimes, the pain will radiate to the neck or upper arm and may be thought due to a muscular cause because it can possess an aching quality. Deep breathing may aggravate the pain and some patients must sleep upright during their episodes of diaphragmatic pain.

Umbilical endometriosis

This presents as a nodule in the inferior portion of the umbilicus which may grow slightly and become more painful with menses. Away from menses it can be smaller and less painful.

It is uncommon for a symptom of endometriosis to begin after the age of 30. In such a patient, some other gynecological source of pain, such as adenomyosis uteri, should be remembered.

Endometriosis vs other causes of pain

It is important to try to distinguish endometriosis pain from other causes of pain, especially pain coming from the uterus. This can sometimes be difficult as endometriosis pain and uterine pain can coexist, especially during menses. Trying to discriminate the source of pain is important since uterine pain may not respond to endometriosis surgery, and pain due to endometriosis may not respond to uterine surgery. This is part of the art of medicine. In plying this art, it is important to remember that uterine pain occurs primarily with menses, although some cases of adenomyosis may cause pain throughout the month but with more extreme menstrual aggravation. Uterine pain is often described as a cramping that may resemble labor contractions and which seems to emanate from a central suprapubic location. The uterus is notorious for radiating pain to the low back (through the uterosacral ligaments), anterior thighs (through the round ligaments) and to the umbilicus (through the obliterated umbilical arteries or urachus).

Physical examination

Endometriosis of the umbilicus is rare and will be identified as a painful swelling less than 1.5cm in diameter located within the umbilicus, usually in the inferior portion (Figure 1). There may be a bluish/black discoloration due to the accumulation of surrounding bloody material. Occasionally, there may be frank bleeding seen.

Visualization of the vulva or cervix will rarely show endometriosis. The posteror fornix will occasionally show endometriosis which is invading from an underlying nodule of a uterosacral ligament or rectum (Figure 2). Such a lesion may be missed unless the vaginal speculum is tilted posteriorly and opened so that the fornix can be displayed. Such vaginal endometriosis should bring to mind the high likelihood of obliteration of the cul-de-sac and rectal involvement.

Bimanual examination is performed to determine if enlargement of the uterus or ovaries exists. If the uterus is anterior in the pelvis, it can be palpated between the fingers of the internal and external hands. It is very important to note whether the uterus is tender when compressed and whether this reproduces any component of the patient's pain. Pain of a uterine origin may not respond to treatment of endometriosis but may respond to presacral neurectomy if the patient refuses hysterectomy.

Digital examination of the posterior vaginal fornix is the most important part of the search for evidence of endometriosis and is best performed after bimanual examination of the uterus and ovaries is completed. This part of the examination is performed only with the internal fingers. The external hand is placed behind the examiner's back or on the patient's leg. The cul-de-sac is lightly palpated with the fingertips while the patient's face is observed, followed by light palpation of each uterosacral ligament, again observing the patient's reaction. Endometriosis in these areas typically will be very tender, causing the patient to wince or cry out and even move up and away on the examining table. The presence and size of nodularity is also noted in these regions. On rectal examination, an effort is made to see whether the mucosa will slide across any nodules, since this may predict the depth of rectal resection required. The patient is asked to estimate how much of her pain is reproduced by examination as this will predict the success of surgery in relieving pain [4]. In some cases of endometriosis involving the bladder, a nodule may be felt anterior to the uterus, or the anterior vaginal wall may display nodularity and tenderness if the trigone of the bladder is involved. Some authors have recommended pelvic examination during menses when tenderness and nodularity may be increased [2, 5].

Endometriosis of the higher intestinal tract or diaphragm exists without identifiable signs on physical examination.

Laboratory tests and imaging

scans
There is no blood test that is specific for endometriosis. The CA125 level may be normal or elevated with ovarian or non-ovarian gynecologic cancer [6], benign ovarian masses, infections, molar pregnancy or fibroids [7], adhesions [8], adenomyosis, endometriosis [9], menstruation, or with a normal pelvis [10].

Transrectal ultrasound for the diagnosis of rectal nodules [11] or magnetic resonance imaging (MRI) studies for bladder nodules [12, 13] are used by some, but such scans appear to be unnecessary for the following reasons. Scans can frequently be negative. Scans do not treat the patient's pain. A negative scan in the presence of suggestive symptoms does not eliminate the need for surgical investigation with bowel preparation. The results of the scan do not alter how surgery is performed, since that is determined automatically by the surgical findings; if some type of bowel resection is seen to be necessary, it will be done regardless of the result of the scan. Scans add expense and no randomized controlled trial proves they are necessary or improve endometriosis surgical outcomes. Scans focus attention on themselves rather than on the patient. While scans may suggest involvement of an organ system which might alter the composition of the surgical team, endometriosis surgery by definition will require a multidisciplinary approach because of its ability to affect multiple organ systems. Thus, an endometriosis treatment program should be able to handle any surgical finding regardless of the presence or absence of preoperative scans.

Scans may be helpful in obese patients who are difficult to examine, or to determine if uterine fibroids or adenomyosis might be present, since the presence of a uterine abnormality may alter the surgical options presented to the patient. If the patient has uterine symptoms and an abnormal uterine scan, hysterectomy may be helpful.

Bowel preparation before surgery

Bowel preparation may be helpful for surgery on patients with known obliteration of the cul-de-sac since most of these patients will need some type of surgery on the rectum [4]. Patients with known intestinal involvement should receive a bowel prep. Patients with known or suspected presence of ovarian endometrioma cysts should have consideration of a bowel prep since the incidence of intestinal endometriosis is increased [14]. Bowel preparation should also be considered for patients with nodularity on examination which indicates more deeply invasive endometriosis possibly involving the rectum, as well as for patients with symptoms suggestive of intestinal involvement. Various osmotic or mechanical bowel prep regimens are available with or without the use of antibiotics and none is necessarily superior to others.

Bowel preparation is not without risks. At surgery, the colon will frequently be filled with fluid which may rush out and contaminate the pelvis and abdomen with a heavy bacterial innoculation (Figure 3) resulting in the risk of postoperative abscess formation. The combination of prophylactic antibiotics and bowel preparation may result in postoperative over-growth of Clostridium difficile which can cause low grade fever, intestinal cramping and watery diarrhea. This can be diagnosed with a Clostridium difficile toxin titer performed on a sample of the diarrhea, although false negatives are possible. Treatment is with oral metronidazole or vancomycin.

Surgical considerations

Uterine manipulation
It is virtually impossible to perform laparoscopic surgery for endometriosis without uterine manipulation. The patient's position is important for proper use of simple manipulators, such as a Hulka tenaculum. The patient's hips should be slightly off the end of the table so the handles of the manipulator can be pressed deeply into the crease of the buttocks, resulting in extreme anteversion of the uterus. The manipulator can then be pressed toward the ceiling, forcing the entire uterus anteriorly toward the abdominal wall and thus exposing the cul-de-sac for easy investigation and surgery. This is particularly vital during surgery for obliteration of the cul-de-sac, since such extreme uterine anteversion results in countertraction against the pull of the tissues posteriorly and will greatly help the dissection.

Patient positioning
The operating table should be able to tilt into extreme Trendelenburg position, which allows the bowel to fall superiorly out of the pelvis. Patients with endometriosis are typically young and healthy and can tolerate many hours in such a position without respiratory problems, although the anesthesiologist may notice a slight increase in required ventilatory pressures. Sub-cutaneous emphysema may occur and spread into the face (Figures 4 and 5), particularly in thin patients, but other than a slight increase in end-expiratory carbon dioxide (CO2) levels, this seems harmless.

The arms should be tucked to the sides and the elbows padded. Attention should be paid to maintaining patient temperature during surgery. The use of warm irrigation fluid during surgery, wrapping the legs and head in additional blankets and using warming devices for the upper torso and face or beneath the patient are all helpful. Leg stirrups should provide good padding for the popliteal and lower leg area, and the legs should be positioned so that the patient will not slide on the table toward the anesthesiologist since this will impair uterine manipulation.

Placement of trocars
Most surgeons use a laparoscope passed through a 10mm sheath in the umbilicus. The umbilicus is the thinnest part of the abdominal wall, and a vertical incision within the inferior umbilicus or across its center will simultaneously allow the most efficient entry and the best cosmetic result following surgery. The trocar and sheath can be placed directly with or without initial insufflation of the abdomen with a Verres needle. Direct trocar insertion may be safer than the use of Verres needle insufflation followed by trocar insertion. Needle insufflation before insertion of the main trocar means that two blind trocar insertions are required instead of one if the umbilical trocar is placed directly, thus increasing, arithmetically, the possibility of damage to underlying structures. The tip of reusable trocars remain slightly dull, especially compared to disposable trocars, so their use is safer since there is less potential for a sharp surgical injury. Indeed, studies have found that the risk of unintentional injury with direct insertion is 1/1838 (0.05%) [15], as compared with a risk of 19/470 (4%) [16], associated with prior Veress needle insufflation before insertion of the main umbilical trocar.

For direct trocar insertion, the abdominal wall around the umbilicus should be elevated with hands or towel clips (Figure 6). The tip of the trocar should be advanced toward the hollow of the pelvis. In obese patients, the bottom of the umbilicus may need to be grasped and everted with Allis clamps or hemostats since this area can be hidden beneath the enveloping fatty folds of the surrounding abdomen. An incision can be created across the center of the umbilicus and sometimes it is easiest to carry the incision into the peritoneal cavity for direct trocar placement. In obese patients, where the bottom of the umbilicus is visible, the trocar tip is advanced cautiously perpendicularly until it is felt biting into the fascia, then is directed toward the pelvis. Patients with endometriosis may have undergone many previous laparoscopies or laparotomies and alternate sites for insertion of an insufflating needle or initial trocar are sometimes recommended to avoid the theoretical complication of damage to bowel adherent around the umbilicus. However, the risk of damage to bowel with direct umbilical insertion has been found to be 1/2000 cases in the author's experience. It is more common to find light omental adhesions only around the umbilicus and these can be lysed either directly down the umbilical sheath or from an alternate direction through another port.

To perform laparoscopic surgery it is helpful to have other ports for graspers and suction-irrigators since this will decrease instrument changes and make surgery more efficient. A minimum of two additional ports is recommended. A triple-puncture technique consisting of insertion points at the umbilicus and a port lateral to each inferior epigastric vessel bundle is adequate for all pelvic surgery and most intestinal surgery for endometriosis. Insertion of these accessory trocars can be done safely by using the laparoscope which has been inserted down the umbilical sheath to view through the parietal peritoneum the course of the inferior epigastric vessels (Figures 7 to 9). The tip of the laparoscope can be advanced lateral to these vessels and a skin incision can be placed across the end of the laparoscope. Transillumination of the abdominal wall (Figure 10) will not reveal the inferior epigastric vessels in most patients, but may instead reveal superficial epigastric vessels which are in the subcutaneous fat. It is important to try to avoid both sets of vessels.

After the skin incision has been made lateral to the inferior epigastric vessels in a transilluminated area without obvious vascular markings, the distal shaft of the laparoscope can be used to push anteriorly against the inferior epigastric vessels while the accessory trocar is inserted horizontally immediately beneath the laparoscope (Figure 11). In this way, the trocar is never directed at the iliac vessels, but is instead directed into the middle of the pneumoperitoneum. The risk of bladder injury is decreased by such a port placement. Even if the accessory trocar does not injure the inferior epigastric vessels, vessels in the subcutaneous fat may be injured, particularly in obese patients where transillumination of the abdominal wall is useless. Blood dripping down the outside of the trocar (Figure 12) should alert the surgeon to the possibility of injury to the inferior or superficial epigastric vessels.

Some surgeons may place only a single additional 5mm port suprapubically in the midline. The bladder is the main organ at risk for such a port site. The placement of the skin incision for this port site is at the discretion of the surgeon based on training, experience, and the demands of the particular case. Depending on the direction of insertion of the trocar, the bladder could be injured whether the skin incision is placed 1cm or 10cm above the pubic symphysis. The outline of the bladder cannot always be seen when the bladder is empty because of retroperitoneal fat and areolar tissue. While the balloon of a retention catheter may be visible as a round mass, this is usually positioned at the internal urethral meatus well away from the dome of the bladder. Because of this, insertion of the trocar under direct visualization will not guarantee that bladder injury will be avoided. If the bladder is injured by the suprapubic trocar, such injury may not be apparent during surgery. No bleeding may be seen laparoscopically or in the catheter bag, and any urine leaking down the trocar sheath will simply be lost in the irrigation fluid used during surgery. The catheter bag may not fill with CO2 because the trocar sheath acts as a tamponade against CO2 entering the bladder. Thus, it is entirely possible that a bladder perforation may go undetected during surgery.

Placement of incision for laparotomy
In patients with such overwhelming disease that requires laparotomy, a low transverse incision is adequate. Division of the rectus muscles is not necessary. This incision can be used to treat virtually any case of pelvic or intestinal endometriosis, and a vertical subumbilical incision is rarely called for. Endometriosis of the diaphragm requires upper abdominal laparotomy, either through a subcostal incision for unilateral disease, or a short vertical incision for bilateral disease [3]. If laparoscopy is abandoned for laparotomy, consideration should be given to taking the patient's legs out of the stirrups and placement in the supine position if a lengthy laparotomy is anticipated. Disease of the ileum is difficult to reach with typical laparoscopic port placements, and it is far easier to extend the right lateral port incision minimally and deliver the ileum onto the abdominal wall for surgery, or to extend the umbilical incision slightly if the sigmoid, ileum and cecum all require surgery (Figure 13).

Visual identification of endometriosis
The visual appearance of endometriosis is important because every intellectual and therapeutic process concerning the disease starts with a surgeon identifying the disease in the pelvis. Protean visual appearances are possible with endometriosis. Young patients may have colorless lesions which can be identified as individual glands on close inspection (Figure 14). With the passage of time, slight fibrosis may occur around these glands due to the irritating effect of their secretions (Figure 15). Peritoneal blood painting [17] may be helpful in identifying subtle texture irregularities of the peritoneum which can be due to endometriosis (Figure 16). Peritoneal pockets, sometimes with multiple underlying subpockets, may be present on either side of the rectum (Figure 17) or occasionally on the pelvic sidewalls. These developmental peritoneal defects may contain endometriosis around their rims or in their depths.

Carbon deposition from previous laser vaporization may masquerade as hemorrhagic endometriosis, although it should be remembered that most patients will have lesions that lack hemorrhagic coloration or neovascularity [18]. Laser vaporization may not burn deeply enough to destroy even superficial endometriosis, and persistent disease can be present beneath the adhesions and carbon which laser leaves behind (Figure 18). Carbon within the tissue may incite a foreign body giant cell reaction which can be an iatrogenic cause of pain (Figure 19).

Other forms of endometriosis can become increasingly obvious as the effects of the disease are manifest over time. These forms are amply illustrated in the chapters on surgical treatment. Dense yellowish or whitish fibrosis is a common finding that can hide underlying disease. While many surgeons may consider such an appearance to be 'burned out' disease, this morphology of endometriosis represents a biologically active form of the disease which is 'burned in'. The endometriosis surgeon's caveat must be: any peritoneal abnormality, no matter how subtle, must be considered possible endometriosis until proven otherwise by biopsy. Only after the disease has been completely identified can it be completely removed.

Once surgery has been completed, the umbilical incision should be closed with absorbable suture placed in the umbilical fascia. Otherwise, omental prolapse may occur, especially in very thin patients. Smaller incisions can be closed with tape. Lengthy surgeries result in a larger volume of irrigation fluid left behind which will drain out of one or more of the laparoscopic puncture wounds. A drain left in the cul-de-sac will help keep bed-linen cleaner and also help detect postoperative bleeding.

Modern laparoscopic surgery can be very extensive and is not truly outpatient surgery. It is associated with a high incidence of postoperative nausea and vomiting. Most often, this is due to anesthesia, direct effects of surgery, or pain pills. Since postoperative vomiting can be a sign of a complication, it is unreasonable to send a patient home from the hospital while vomiting. Minimum criteria for discharge should include the ability to tolerate liquids and pain pills by mouth, ability to walk to the bathroom, and stable vital signs.

References

1. Goldstein DP, DeCholnoky C, Emans SJ. Adolescent endometriosis. J Adolesc Health Care 1980;1:37-41.
2. Counseller VS. Endometriosis: A clinical and surgical review. Am J Obstet Gynecol 1938;36: 877-88.
3. Redwine DB. Diaphragmatic endometriosis: diagnosis, surgical management, and long-term results of treatment. Fertil Steril 2002;77:288-96.
4. Redwine DB, Wright J. Laparoscopic treatment of obliteration of the cul-de-sac in endometriosis: Long term followup. Fertil Steril 2001;76: 358-65.
5. Fallon J. Endometriosis. Two hundred cases considered from the viewpoint of the practitioner. N Eng J Med 1946;235:669-73.
6. Niloff JM, Klug TL, Schaetzl E, Zurawski VR, Knapp RC, Bast RC. Elevation of serum CA125 in carcinomas of the fallopian tube, endometrium, and endocervix. Am J Obstet Gynecol 1984;148:1057-8.
7. Di-Xia C, Schwartz PE, Xinguo L, Zhan Y. Evaluation of CA 125 levels in differentiating malignant from benign tumors in patients with pelvic masses. Obstet Gynecol 1988;72:23-7.
8. Cheng Y-M, Wang S-T, Chou C-Y. Serum CA- 125 in preoperative patients at high risk for endometriosis. Obstet Gynecol 2002;99:375-80.
9. Koninckx PR, Riitinen L, Seppala M, Cornillie FJ. CA-125 and placental protein 14 concentrations in plasma and peritoneal fluid of women with deeply infiltrating pelvic endometriosis. Fertil Steril 1992;57:523-30.
10. Molo MW, Kelly M, Radwanska E, Binor Z. Preoperative serum CA-125 and CA-72 in predicting endometriosis in infertility patients. J Reprod Med 1994;39:964-6.
11. Fedele L, Bianchi S, Zanconato G, Tozzi L, Raffaelli R. Gonadotropin-releasing hormone agonist treatment for endometriosis of the recto- vaginal septum. Am J Obstet Gynecol 2000;183: 1462-7.
12. Donnez J, Spada F, Squifflet J, Nisolle M. Blad- der endometriosis must be considered as bladder adenomyosis. Fertil Steril 2000, 74:1175-81.
13. Vercellini P, Frontino G, Pisacreta A, De Giorgi O, Cattaneo M, Crosignani PG. The pathogene- sis of bladder detrusor endometriosis. Am J Obstet Gynecol 2002; 187:538-42.
14. Redwine DB. Ovarian endometriosis: A marker for more severe pelvic and intestinal disease. Fertil Steril 1999;73:310-15.
15. Kaloo P, Cooper M, Reid G. A prospective multicentre study of laparoscopic complications related to the direct-entry technique. Gynaecol Endosc 2002;11:67-70.
16. Jones KD, Fan A, Sutton C. Safe entry during laparoscopy: a prospective audit in a district general hospital. Gynaecol Endosc 2002;11: 85-9.
17. Redwine DB. Peritoneal blood painting: an aid in the diagnosis of endometriosis. Am J Obstet Gynecol 1989; 161:865-6.
18. Redwine DB. Age related evolution in color appearance of endometriosis. Fertil Steril 1987;48:1062-3.