Salpingitis

Most salpingitis is the result of an ascending infection from the lower genital tract, caused primarily by Chlamydia trachomatis and Neisseria gonorrhoeae (see Chapter 61, Pelvic inflammatory disease, and Chapter 62, Sexually transmitted infections, for more information). These micro-organisms cause mucopurulent endocervicitis, and break down major cervical barriers, such as the mucus plug that protects the endometrium and upper genital tract, and permits ascending infection. Other pathogens such as Streptococcus sp., Escherichia coli and anaerobes ascend through the breached barrier and reach the upper genital tract (Rice and Schachter 1991). In a significant proportion of cases, no aetiological agent is identified; this suggests that a range of other aetiological agents have a role, but the identification of novel organisms is difficult to ascertain because of the limited availability of diagnostic tests. Serological evidence has associated Mycoplasma genitalium with pelvic inflammatory disease (Simms and Stephenson 2009).

In the majority of patients, salpingitis is due to a sexually transmitted infection. In other cases, iatrogenic manoeuvres such as insertion of an intrauterine device, termination of pregnancy, hysterosalpingography or curettage can spread a cervical infection into the uterus and tubes. In a small proportion of women, pelvic infection may be secondary to gastrointestinal infections such as perforated appendicitis. Mycobacterium tuberculosis is still seen in developing countries; however, migration to developed countries has resulted in a significant increase in the number of tuberculosis notifications since the 1980s (Rose et al 2001).

Acute salpingitis causes loss of ciliated cells and distal occlusion of the fallopian tubes. Some of the deciliation may be recovered later, but subsequent immune response following the acute phase causes permanent scarring, loss of mucosal folds and flattening of the mucosa (Lyons et al 2006). Distal occlusion of the tubes together with chronic inflammatory process could result in fluid collection in the lumen of the fallopian tube, leading to hydrosalpinges. Pelvic peritonitis during the acute phase may be followed by adhesions around the fallopian tubes and ovaries, distorting the pelvic anatomy. Perihepatic inflammation may cause ‘violin-string’ adhesions known as ‘Fitz–Hugh–Curtis syndrome’ (Figures 24.1–24.3).

Figure 24.1 Laparoscopic view of a pelvis after pelvic inflammatory disease. There is a ‘curtain’ of adhesions covering the pelvic organs.

Figure 24.2 Laparoscopic photograph of perihepatic adhesions: Fitz–Hugh–Curtis syndrome.

Figure 24.3 Laparoscopic appearance of bilateral hydrosalpinges as a result of previous inflammatory disease.

Other causes

Endometriosis, fibroids, previous pelvic/tubal surgery, salpingitis isthmica nodosa, endosalpingiosis and cornual polyps can be the cause of cornual obstruction or tubal damage. Endometriosis is reviewed in Chapter 33. In some patients, tubal damage is secondary to previous tubal or pelvic surgery such as salpingotomy for ectopic pregnancy, ovarian cystectomy, myomectomy, ovarian wedge resection and shortening of round ligaments. Salpingitis isthmica nodosa was described by Chiari (1887) as nodular thickening of the proximal part of the fallopian tube. The aetiology of this entity is unknown but it is probably due to a non-inflammatory process.

Selective transcervical salpingography

Selective transcervical salpingography and tubal catheterization can be used for the treatment of proximal tubal obstruction in selected cases. A review of the literature on fluoroscopic tubal cannulation showed pregnancy rates of 22.7% and ectopic pregnancy rates of 3.2% (Honore et al 1999).

Reconstructive tubal surgery

With the advent of IVF-ET, the role of reconstructive tubal surgery has diminished. However, this type of surgery is still appropriate and effective for properly selected individuals (National Institute for Clinical Excellence 2004). Initial methods for tubal surgery were based on microsurgical techniques. The use of magnification and microsurgical techniques has been the traditional method. However, with advances in laparoscopic surgery, most of these methods are now possible to perform at laparoscopy.

The criteria for patient selection for treatment depend very much on the surgeon’s experience and on the possibility of being able to offer alternative treatment. Other points to be considered when making the decision regarding treatment are the patient’s wishes about the type of treatment that she prefers and the problem of pain. Large numbers of patients with tubal disease have pelvic pain. In some of them, the surgical freeing of the ovaries and tubes from adhesions can give symptomatic relief.

Microsurgical treatment

The microscope was first used for tubal surgery by Walz (1959). Whilst others used delicate electrosurgery and magnification (loupes) in the treatment of hydrosalpinges, Paterson and Wood (1974) in Australia and Winston and McClure-Browne (1974) in England adapted their experience from animal experiments to humans and operated on infertile women, under high magnification, using an operating microscope.

Microsurgery is not only the use of a microscope; it is based on gentle tissue handling, reperitonealization of raw areas, the use of non-resorbable suture materials, and irrigation of the tissues using Ringer’s lactate solution as this tends to prevent adhesion formation.

The use of the microscope was originally much debated, but microsurgery is now well established and is a routine procedure for tubal infertility.

Cornual occlusion

Cornual occlusion due to inflammatory causes was treated by uterotubal implantation with very poor results. Ehrler (1963) described his technique and suggested that the intramural portion of the tube could be spared in most patients. Microsurgical methods described by Winston (1977) and Gomel (1977) became the surgical technique of choice. Cornual implantation is now rarely used. Tubocornual implantation is avoided whenever possible. Destroying a possible sphincter at the uterotubal junction is associated with excessive bleeding and damage to the tubal blood supply; it shortens the tube and may increase the risk of rupture of the uterus in the event of subsequent pregnancy, which means that patients must be delivered by caesarean section.

The use of magnification allows better identification of the intramural portion of the tube by careful shaving of the cornua until healthy tissue is found, permitting a more accurate tissue apposition with a watertight anastomosis between healthy tubal tissues. Once the ends are well defined, the anastomosis is done in two layers, using 8/0 nylon as a suture. The suture material should not penetrate the mucosa, only the muscularis (Seki et al 1977). In some cases, the use of a temporary splint gives considerable help, especially in deep cornual tubal anastomosis (Figure 24.7), but this should be removed at the end of the surgical procedure; if left in situ, the splint should not remain for more than 48 h as longer periods of time cause mucosal damage. Tension between the anastomosed ends must be avoided. A stitch of 6/0 Prolene between both ends of the mesosalpinx should be applied as a stay suture. More details about the technique have been given elsewhere (Winston 1977, Margara 1982).

Figure 24.7 Cornual block diagram showing a cornual anastomosis. (A) Superficial cornual block. (B) Opening of the cornua to expose the intramural part of the tube. (C) Shaving of the tube to find healthy mucosa for anastomosis. (D) Cornual anastomosis completed.

Tubal anastomosis

Tubal anastomosis for reversal of sterilization is the most successful technique in microsurgery for the following reasons: healthy tissues are anastomosed, the localized damage is removed completely (Figure 24.8) and these patients are generally fertile.

Figure 24.8 Anastomotic procedure. (A) Both tubal lumens are exposed (healthy mucosa). (B) The first stitch is at 6 o’clock and must be extramucosal. (C) The anastomosis is completed in two layers. The suture material used is 8/0 nylon.

Success depends on the length of the remaining tube. The minimum length of tube necessary to maintain fertility in women is not known. In animal experiments, fertility diminished in a linear fashion depending on the length of the ampulla resected; when more than 70% was missing, none of the animals became pregnant. Resection of the ampullary–isthmic junction did not appear to alter fertility (Winston et al 1977). The rabbit is far from an ideal model for human tubal physiology, but these results emphasize the fact that the ampulla seems to be important in maintaining fertility.

The length of time between sterilization and reversal is important and has a prognostic value. Vasquez et al (1980) demonstrated that after 5 years of sterilization, the proximal portion of the tube had a severely damaged mucosa with flattening of the epithelium and polyp formation. The surgical technique of reversal of sterilization has been described by Winston (1977) and Margara (1982).

Hydrosalpinges

As well as being a cause of infertility, hydrosalpinges also have a detrimental effect on the outcome of IVF-ET treatment, with reduced pregnancy and implantation rates and increased miscarriage rates. The hydrosalpinx fluid is known to be embryotoxic and may also reduce endometrial receptivity. Removal of hydrosalpinges prior to IVF-ET seems to lead to better results (Nackley and Muasher 1998, Zeyneloglu et al 1998). However, some groups believe that salpingectomy should only be performed when there is severe tubal pathology, and other patients should be given the chance of tubal surgery (Puttemans and Brosens 1996). A Cochrane review concluded that further randomized trials are required to assess other surgical treatments for hydrosalpinx, such as salpingostomy, tubal occlusion or needle drainage of a hydrosalpinx at oocyte retrieval (Johnson et al 2004).

The value of microsurgery varies in the treatment of hydrosalpinges. In some cases, where the tube is completely free of adhesions, the use of loupes is sufficient. Where complex adhesions are present, the use of the microscope is beneficial. When performing a salpingostomy, the following points should be borne in mind. Before starting the salpingostomy, mobilization of the tube must be completed. Division of adhesions between tube and ovary or other pelvic organs is very important in order to leave the tube fully mobile, with the possibility of the new ostium being able to cover the whole ovarian surface and making egg pick-up more likely.

While dividing adhesions, special care must be taken to avoid damage to the fimbrial blood supply. These vessels are in the area of the connecting ligament between the ovary and the tube at the outer margin of the mesosalpinx. The hydrosalpinx must be opened at the most terminal part: the ‘pucker point’. This is where the fimbrial end has closed; it is clearly seen under the microscope as a thin fibrous line, often with an H-shaped configuration, and is not always the thinnest part of the tube. Linear salpingostomy has a high chance of healing over. Using fine diathermy, the tube is opened and a glass probe is introduced, following the fibrous tracts parallel to the blood vessels and ensuring that the mucosal folds are not cut. Using small incisions, the new tubal ostium is completed and then the mucosa can be everted (Figure 24.9).

Figure 24.9 Salpingostomy. (A) Identification of the terminal part of the hydrosalpinx where the incision must be made. (B) An incision is made using diathermy, and the tip of the glass probe is inserted into the hole in the tube. (C) The salpingostomy is enlarged using the diathermy needle and glass rods. (D) The salpingostomy is completed.

Two or three stitches of 8/0 nylon are used to secure the mucosal eversion. If the ovarian surface is damaged during the division of adhesions, the raw area is repaired using fine non-absorbable suture material to avoid recurrence of adhesions.

Adhesions

Omental adhesions are not infrequent and when they are more than minimal, a partial omentectomy is performed. It is best done at the beginning of the operation. Fine 2/0 linen is used to secure the pedicles. It is not used as a routine procedure, but seems to be a very effective way of avoiding recurrent adhesions in the pelvis.

The most frequent adhesions are between the ampulla, the ovary and the mesosalpinx. It is easy to work from the isthmus towards the fimbrial end. Using a glass probe, the adhesions are hooked and incised with monopolar diathermy. Care should be taken not to damage the tubal peritoneum. The use of the microscope simplifies the process because the peritoneal edges can be seen easily. If the tubal peritoneum is incised, it must be repaired using 8/0 nylon as the suture material.

Ovarian adhesions should be removed from the ovarian capsule using diathermy or scissors, leaving the ovarian capsule as free as possible. Ovulation and egg release seem to improve after careful ovariolysis and may depend on the amount of the ovarian surface left free (Figure 24.10).

Figure 24.10 Laparoscopic appearance in a patient with severe periovarian adhesions after pelvic inflammatory disease.

Special attention must be paid to raw areas. The uterine surface and the surrounding peritoneum must be inspected carefully. Peritonealization is very important and all raw areas must be covered. If the ovarian fossa has been damaged in order to free or liberate a firmly adherent ovary, the raw area should be closed using a linear suture of 4/0 Prolene. If the raw area cannot be peritonealized using the surrounding peritoneum, a peritoneal graft can be applied. The peritoneum should be thin and without fatty tissue. It is attached to the raw area using 8/0 nylon or 6/0 Prolene. The donor areas can be the peritoneal layer of the anterior abdominal wall, the peritoneal space between the round ligament and the bladder and, in some cases, the peritoneum of the mesentery of the small or large bowel. This technique has proved very effective in experimental animals, and the results with humans are very encouraging. Synthetic materials are available to replace peritoneum (Hunter et al 1988, Haney and Dotty 1992). They can be applied during open surgery as well as laparoscopic procedures, and can remain in situ, apparently without side-effects; if removal is necessary, this can be done through a laparoscope.

It must be emphasized that the treatment of tubal damage due to tuberculosis is always medical and the tubal damage cannot be repaired by surgery. In this group of patients, if the uterine cavity remains unaffected or without damage, IVF is the only option.

Laparoscopic surgery

With the development of endoscopic techniques in the last three decades, gynaecologists, using a laparoscope, have been able to perform numerous operations in the field of infertility as well as in general gynaecology.

Since the first diagnostic use of the endoscope in the 1800s, developments in optics and related technology have revolutionized the use of the laparoscope. In the 1940s, Palmer in France promoted the laparoscope as a diagnostic instrument. In the 1970s, Gomel published the first results of laparoscopic surgery. He performed salpingo-ovariolysis and fimbrioplasties using this approach, with results comparable to those he achieved using magnification in open procedures. At the same time, Semm in Germany also developed the technique and a large number of new instruments became available for laparoscopic surgery.

The development of laser beams for medical use also had an impact in laparoscopic surgery. The use of video-laparoscopy has become a standard part of laparoscopic surgery. It is less tiring, and the assistant has a more active role during surgical procedures, thus it facilitates training and teaching. Different methods of dissection using laser diathermy (Donnez and Nisolle 1989) or cold cutting (Serour et al 1989) have been discussed. There is no evidence that one is better than the other.

After the initial experiences, procedures that can now be performed safely using operative laparoscopy include salpingo-ovariolysis, fimbrioplasty, salpingostomies, tubal reanastomosis, and management of ectopic pregnancy and endometriosis.

One of the most important attributes of this method is that it probably has a low rate of adhesion formation. There is no doubt that laparoscopic adhesiolysis is very effective for pain, and open surgery may cause more adhesion formation (Lundford et al 1991). The question still remains whether patients with severe tubal damage, not suitable for open procedures, should be operated upon using the laparoscopic approach when the prognosis is very poor and alternative methods are available to them.

Great financial advantages are offered by this method. This procedure can be done as a day case or overnight stay, and patients resume their normal activities in a very short period of time. Laparoscopic surgery, as with any other surgical procedure, has its risks and therefore intensive training is fundamental in order to diminish complications and achieve good results.

Proximal tubal disease

Selective salpingography and tubal catheterization can achieve high patency rates and pregnancy rates of 49% (Honore et al 1999). With the exception of reversal of sterilization, laparoscopic surgery has a limited place in the treatment of proximal tubal disease. Microsurgical cornual anastomosis offers very good results in selected groups of patients. Postoperative laparoscopies show a high patency rate amongst those patients who have not conceived. The major limiting factors seem to be the recrudescence of the disease or extension of the original inflammation into the anastomotic site, rather than lack of patency. Gomel (1980) reported that 53% of patients conceived and had at least one term pregnancy.

Following tubocornual anastomosis by microsurgery, the reported term pregnancy rates in the literature vary between 22% and 57%, and ectopic pregnancy rates vary between 2% and 12%. These results are similar to cumulative delivery rates following five cycles of IVF-ET (Posaci et al 1999). In contrast, the outcome of tubouterine anastomosis is much poorer, and this procedure has mostly been replaced by tubocornual anastomosis.

The surgical treatment of salpingitis isthmica nodosa in the infertile patient is similar to that of cornual block. In nearly half of patients, the tubes are open and the diagnosis is made on the basis of the typical images of diverticulae on the hysterosalpingogram and at laparoscopy. The anastomotic procedure is generally much easier, but the length of isthmic portion that must be removed can be difficult to assess. The prognosis depends on the length of tube removed, but if insufficient tissue is removed, the surgical procedure is more likely to fail. The use of the microscope and the experience of the surgeon are very important in evaluation of the amount or length of tissue that must be resected. In general, this condition involves most of the isthmic portion of the tube, so the anastomosis is between the cornu and the ampullary–isthmic junction.

Where the intramural portion of the tube is extensively involved in the process, the patient should probably not be treated surgically. The surgical procedure itself does not vary from that of tubocornual anastomosis. When the whole of the isthmus is removed, the problem of tension at the anastomotic level can be solved using stay sutures of 6/0 Prolene between the uterus and the mesosalpinx in order to approximate the tubal ends.

Reversal of sterilization

Reversal of tubal sterilization is the most successful procedure in tubal surgery. Both open microsurgical and laparoscopic reversals offer high pregnancy rates. The pregnancy rates in the literature following reversal of sterilization with microsurgery vary between 33% and 86%, with ectopic pregnancy rates of 1–14%. Pregnancy rates of up to 87.1% have been reported following laparoscopic reversal of sterilization (Yoon et al 1999). The success rates are lower after sterilization with cautery or Pomeroy’s technique compared with clip or ring sterilization (Posaci et al 1999).

Salpingostomy

In general, it is difficult to assess the results of salpingostomy because a very heterogeneous group of patients is involved. Patient selection varies widely between units and there is no agreement regarding classification, especially where salpingostomy and fimbrioplasty are concerned.

Many gynaecologists think that salpingostomy is obsolete because IVF techniques offer comparable or better results in some cases. As in any other surgical procedure, good results will be achieved in well-selected patients.

Boer-Meisel et al (1986), in a prospective study, classified hydrosalpinges as grades I, II and III, based on the nature and extent of the adhesions, the microscopic aspect of the endosalpinx, the thickness of the tubal wall and the diameter of the hydrosalpinx. In their series, 77% of patients with grade I hydrosalpinx had the possibility of conception, compared with 21% with grade II and only 3% with grade III. Thus, surgery is the obvious treatment for patients with grade I hydrosalpinges. Patients with grade III hydrosalpinges should avoid salpingostomy and should be treated using IVF techniques after salpingectomy. The difficult group is that of grade II hydrosalpinx, where treatments with IVF or tubal surgery have the same prognosis. Age, social and religious background, the possibility of alternative treatment and the patient’s wishes must be considered very carefully.

The success of tubal microsurgery for distal tubal lesions depends on several factors. Patients with extensive pelvic adhesions, a thick tubal wall and abnormal tubal mucosa at salpingoscopy have poor results. Type of surgery also affects the outcome. Pregnancy rates as high as 59–60% can be achieved with fimbrioplasty, but the results after salpingostomy may not be as good (Posaci et al 1999). Patients with poor prognostic factors will have a higher chance of achieving pregnancy with IVF, whereas in the absence of poor prognostic factors, tubal microsurgery should be the first choice.

Laparoscopic surgery

Laparoscopic adhesiolysis for pelvic adhesions results in very good results, with term pregnancy rates of 47–62% and ectopic pregnancy rates of 4–8%. These results are similar to those achieved after microsurgery, so laparoscopic surgery should be the first option for suitable patients (Posaci et al 1999).

Success rates of laparoscopic surgery for distal tubal lesions also depend on several prognostic factors, including the presence of extensive pelvic adhesions, a thick tubal wall and severe endotubal damage. The reported intrauterine pregnancy rates following laparoscopic surgery for distal tubal lesions vary between 13% and 51%, with ectopic pregnancy rates of 3–23% (Posaci et al 1999). It appears that laparoscopic surgery offers similar success rates compared with microsurgery for lesser degrees of distal tubal damage. However, open microsurgery may be slightly better for more severe distal tubal disease. For the treatment of moderate to severe distal tubal disease, postoperative pregnancy rates are low, even after microsurgery, despite long periods of observation (>2 years), with remarkably high ectopic pregnancy rates. For this reason, IVF is preferable in these situations (Ozturk and Saridogan 2009).