Color Atlas of Gross Placental Pathology

lesions (Figure 8.2), parameters which tend to be masked by the pneumoperitoneum at standard laparoscopy. Figure 8.3 shows the same implants of red, active peritoneal endometriosis, as seen by standard laparoscopy (Figure 8.3a) and as seen by transvaginal hydrolaparoscopy (Figure 8.3b). An inexperienced eye may easily miss the implants on the image in Figure 8.3a, but will never miss them on Figure 8.3b. The classical ‘gunshot’ or ‘powder burn’ lesions of endometriosis on the surface of the peritoneum and of the

Figure 8.2

Trophic centripetal angiogenesis, often with tortuous blood vessels towards the lesion.

ovary, look exactly the same in a CO2 pneumoperitoneum as under water (Figures 8.4a and 8.4b).

The adhesions

The systematic inspection of the tubo-ovarian structures under fluid makes it easy to identify filmy, connecting, and

non-connecting endometriotic adhesions on the surface of the ovary and peritoneum. The use of saline as the distention medium also provides a remarkable delineation between the ovarian surface and the adhesions.

Unexplained ovarian adhesions were described by Jansen and Russell3 as subtle endometriosis, but subsequently received little or no attention. Ovarian endometriotic adhesions differ substantially from post-surgical and post-infectious adhesions4,5 and are characterized by prominent microvascularization (Figure 8.5) and the presence of inflammatory, hemosiderin-laden, endometriallike cells and fibrin deposits. Subtle endometriotic adhesions are found in 70% of infertile patients with minimal and mild endometriosis at transvaginal hydrolaparoscopy but also in 45% of patients with so-called

Figure 8.4a

So-called typical ‘gunshot’ or ‘powderburn’ lesions of endometriosis, on the surface of the peritoneum and of the ovary. These ovarian lesions can be the ‘tip of the iceberg’ of a larger and deeper endometrioma, especially when areas of retraction of the ovarian cortex can be seen at the same time. These findings warrant a surgical exploration and, if necessary, extensive bipolar coagulation following the selection of one or more biopsies.

Figure 8.4b

Typical late lesions are the so-called black-blue lesions, which ultimately heal by increasing fibrosis. (Reproduced from Brosens et al. Hum Reprod 1994; 9: 770–12 with permission from Oxford University Press.)

‘unexplained’ infertility. Non-connecting adhesions which may interfere with ovum capture and retrieval by the fimbriae,6 are seen in 55% of the ovaries in patients with minimal and mild endometriosis and in 36% of the ovaries in patients with unexplained infertility.7 Comparison of the techniques used to inspect the ovaries showed that standard laparoscopy required the systematic use of an additional instrument to manipulate and expose the inferior or anterior–lateral side of the ovary and the opposing posterior leaf of the parametrium. Traumatic bleeding occurred during that manipulation at standard laparoscopy in 7% of the 43 patients. In contrast, transvaginal hydrolaparoscopy allows for the inspection of the ovaries without any manipulation and detects three times more adhesions than standard laparoscopy. The clinical significance of

non-connecting adhesions, that are definitely not pathognomonic for endometriosis (other pathologies may induce these subtle inflammatory changes as well), must be clarified by further studies. Sampson8 and others9,10 have suggested a primary role of adhesions in the pathogenesis of the ovarian endometrioma (Figures 8.6a and 8.6b).

The endometrioma

The typical features of an endometrioma include:

1.Dense adhesions between the anterior side of the ovary near the hilus and the posterior leaf of the broad liga-

ment or pelvic wall (i.e. what one could call the ceiling of the ovarian fossa).

2.Retraction, scarification, and invagination of the ovarian cortex at the site of the adhesions.

3.The presence of endometriotic lesions and non-con- necting adhesions on the surface of the ovary.

4.The entrapment of a variable amount of chocolate.

5.The inner lining of the invaginated cortex inside the endometrioma with mucosa-like implants of red hemorrhagic lesions, the histology of which represents endometrial tissue as a surface epithelium with or without stroma.

These mucosa-type implants (Figures 8.6a, 8.6b and 8.7) are found most frequently at the site of the retraction and adhesion with the parametrium, i.e. endoscopically by the

introduction of an optic into the endometrioma.10 These ‘ovarioscopies’ and the result of the in-situ biopsies are in agreement with the hypothesis of Hughesdon9 that 93% of the endometrial cysts originate from invagination of the ovarian cortex (Figures 8.6a and 8.6b). Operative transvaginal hydrolaparoscopy allows for a progressive adhesiolysis/ovariolysis towards the ceiling of the fossa, followed by an incision of the chocolate cyst itself at the site of retraction. The chocolate content will be washed out by the continuous irrigation of saline, and once the vision becomes clear again, the red endometriotic implants can readily be peeled off their pearl-white background for histology and can subsequently be coagulated with the bipolar probe (see Figure 8.7). Currently, the technique is no

longer experimental, and endometrial cysts with a diameter ≤4–5 cm can be treated in this way by operative TVL.

The small endometrioma

In an ongoing clinical trial, TVL has proven to be superior to transvaginal ultrasound for the detection of ovarian endometriomas of ≤15 mm in diameter.11 A total of 564 consecutive women attending the infertility clinic were first investigated with transvaginal ultrasound. Patients without major pelvic pathology were investigated by transvaginal endoscopy that same day. Suspected endometriomas were explored by bipolar dissection and selective biopsies were taken whenever possible. Pelvic endometriosis was detected in 169 patients (29%) at transvaginal hydrolaparoscopy. Cystic ovarian endometriosis was present in 22 of these cases (13%). Transvaginal ultrasound only detected 5 out of 11 endometriomas measuring ≤15 mm in diameter, compared with the detection of 11 endometriomas out of 11 measuring >15 mm in diameter.

There is increasing evidence that the diagnosis of endometriosis, whatever its stage, is important in patients with infertility. The clinical significance of the detection and surgical treatment of small endometriomas in patients with infertility requires further investigation; however, their detection at such an early stage of the infertility exploration and their immediate conservative treatment as a prevention (of further development and growth towards the more severe stages of the disease, with the risk of laparoscopic cystectomy and subsequent premature ovarian failure) definitely deserves our full attention.

Disadvantages of transvaginal laparoscopy with regard to endometriosis

A first and obvious disadvantage of TVL is that not all the areas of the female pelvis are accessible or

approachable/reachable by transvaginal hydrolaparoscopy. Endometriotic lesions on the peritoneum of the vesicouterine fold will surely be missed. In a study of 716 women with endometriosis, Koninckx et al12 found the following anatomic distributions of endometriosis: cul-de-sac and uterosacral ligaments 69%, ovaries 45%, fossa ovarica 33%, and vesicouterine fold 24%. However, isolated peritoneal endometriosis in the vesicouterine fold is very rare and reported to be <4% of cases.13 So, if one cannot detect any endometriosis by transvaginal hydrolaparoscopy, then the vesicouterine area is probably free of endometriosis as well. Implants within the bladder, however, are within reach of the TVL set of instruments (Figure 8.8).

Figure 8.8

An irregularity in the bladder wall near the trigonum at vaginal ultrasound was explored endoscopically (‘cystoscopically’) using the same optic as the one that is used for TVL. This exploration revealed an endometriotic lesion lifting the bladder mucosa that shows a ‘cobblestone’ aspect. With the optic locked in the operating sheath (the one we normally use for operative hysteroscopy), a biopsy was taken and the lesion coagulated. Pathologic examination of the biopsy clearly indicated the presence of endometriosis.

Figure 8.9

When a rectovaginal nodule is felt at clinical examination, this constitutes a major contraindication to perform a transvaginal endoscopy. Since one can never predict if and how the anterior wall of the rectum is involved in this process, the risk of bowel perforation at the insertion of the TVL needle is too high. Thus, deep rectovaginal endometriosis is one chapter of the disease where a TVL cannot be used.

Endometriotic implants on the serosal surface of the bowel may also be missed, since a systematic exploration of bowel loops, e.g. during a laparotomy, is impossible by TVL, as all bowel loops look alike, are floating against each other, and simply cannot be checked over 360° and such a long distance. Endometriotic lesions on rare locations like the diaphragm will also be missed, since TVL can only reach as far as the umbilical region. On the other hand, endometriosis of the appendix has been documented by TVL on a number of occasions, except when adhesions did not allow for a visualization of the appendix itself.

A second disadvantage with regard to the detection and exploration of endometriosis is that deep rectovaginal indurations and nodules constitute a major contraindication to perform a transvaginal hydrolaparoscopy altogether. In such a condition – and that is the reason why a TVL is always preceded not only by a vaginal ultrasound but also by a clinical bimanual examination and even by a rectovaginal examination if in doubt – one can never exclude an abnormally high position of the anterior wall of

the rectum in the cul-de-sac, due to dense adhesions with and even involvement of that anterior wall by the process of rectovaginal endometriosis (Figure 8.9). Therefore, the risk of bowel perforation is too high when shooting the needle and inserting the instruments.

Conclusion

Transvaginal endoscopy, performed at an early stage of the infertility exploration, offers a powerful diagnostic tool that provides accurate and reliable information with regard to the detection and staging of endometriosis in the infertile patient without obvious pelvic pathology or with unexplained infertility. Moreover, it also enables us to treat the detected endometriosis efficiently in the majority of cases during that same session.

References

1.Laufer MR. Identification of clear vesicular lesions of atypical endometriosis: a new technique. Fertil Steril 1997; 68: 739–40.

2.Brosens IA, Puttemans P, Deprest J et al. The endometriosis cycle and its derailments. Hum Reprod 1994; 9(5): 770–1.

3.Jansen RPS, Russell P. Nonpigmented endometriosis: clinical, laparoscopic, and pathologic definition. Am J Obstet Gynecol 1986; 155: 1154–9.

4.Jirasek JE, Henzl M, Uher J. Periovarian peritoneal adhesions in women with endometriosis. J Reprod Med 1998; 43: 276–80.

5.Tulandi T, Chen MF, Al-Took S et al. A study of nerve fibers and histopathology of postsurgical, postinfectious, and endometriosisrelated adhesions. Obstet Gynecol 1998; 5: 766–8.

6.Gordts S, Campo R, Rombauts L, Brosens I. Endoscopic visualization of the process of fimbrial ovum retrieval in the human. Hum Reprod 1998; 13: 1425–8.

7.Brosens I, Gordts S, Campo R. Transvaginal hydrolaparoscopy but not standard laparoscopy reveals subtle endometriotic adhesions of the ovary. Fertil Steril 2001; 75(5): 1009–12.

8.Sampson JA. Peritoneal endometriosis due to the menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol 1927; 14: 422–69.

9.Hughesdon PE. The structure of the endometrial cysts of the ovary. J Obstet Gynaecol Br Emp 1957; 44: 481–7.

10.Brosens IA, Puttemans PJ, Deprest J. The endoscopic localization of endometrial implants in the ovarian chocolate cyst. Fertil Steril 1994; 61: 1034–8.

11.Gordts S, Puttemans P, Brosens I. Transvaginal hydrolaparoscopy but not transvaginal ultrasound allows detection of small ovarian endometriomas in patients with infertility. Fertil Steril 2004; 82: S338.

12.Koninckx PR, D’Hooghe TD, Oosterlynck D. Response to letter to editor. Fertil Steril 1991; 56: 590.

13.Jenkins S, Olive DL, Haney AF. Endometriosis: pathogenetic implications of the anatomic distribution. Obstet Gynecol 1986; 67: 335–8.

Introduction

Assessment of the Fallopian tube represents an integral part of the evaluation of the infertile couple. Because the findings obtained at hysterosalpingography (HSG), laparoscopy, or laparotomy are indirect, salpingoscopy has been introduced as an endoscopic examination that can directly evaluate the ampullary tubal mucosa. The standard procedure is transfimbrial salpingoscopy performed at the time of laparoscopy.1 It is a microendoscopic approach for directly visualizing the tubal mucosa from the ampullary–isthmic junction to the fimbria. Clinical and morphologic studies have shown a high correlation between the appearance of the tubal mucosa and the ultimate outcome in terms of pregnancies.1 Recently, it was suggested that performing salpingoscopy with laparoscopy could significantly increase accuracy in predicting shortterm fertility outcome.2 However, such salpingoscopy under transabdominal laparoscopy requires hospitalization and general anesthesia.

Transvaginal laparoscopy (TVL) was introduced by Gordts et al to explore the tubo-ovarian structures3 as well as the ampullary tubal mucosa4 in infertile patients. We have also been performing TVL and showed its usefulness in investigating infertile women.5–8 Two advantages of TVL have been shown: it is a less traumatic and more suitable outpatient procedure than transabdominal laparoscopy.

For these reasons, we have been performing TVL for the following five indications: diagnostic TVL5–8 for (1) tubal obstruction and/or peritubal adhesion are suggested by HSG, (2) serum antibody against Chlamydia trachomatis is positive, (3) diagnosis of early-stage endometriosis, and (4) unexplained infertility; operative TVL9,10 for (5) ovarian drilling using Nd:YAG or holmium laser in infertile women with polycystic ovary syndrome (PCOS).

In some women, endoluminal examination by salpingoscopy can be simultaneously performed under TVL. The transvaginal salpingoscopy under TVL is less invasive to infertile women because it does not require hospitalization or general anesthesia. The procedure was first described by Gordts et al.4 They reported that the fimbriae were visual-

ized in all patients with no obvious pelvic pathology, and cannulation of the distal tubal segment was achieved without manipulation of the tube in 20% before ovulation and 55% in the early luteal phase. Later, Watrelot et al reported that salpingoscopies were possible in 19% of women with post-PID (pelvic inflammatory disease) lesions without the need to stabilize the tubes. However, salpingoscopies were possible in 41% of those women using a grasp forceps introduced in the operative channel.11

We have also been performing salpingoscopy under TVL. Initially, transvaginal salpingoscopy was possible in 13 (18%) of 71 adnexa that were completely visualized.5 Suzuki et al recently reported the clinical significance of endoluminal assessment by transvaginal salpingoscopy in 130 infertile women.12 In this chapter, the authors would like to show some findings obtained by transvaginal salpingoscopy using data described in the manuscript.

Transvaginal hydrolaparoscopy and salpingoscopy

Consecutive series of 130 infertile women diagnosed tuboovarian structures and tubal passage using TVL (Circon ACMI, Stamford, CT, USA) were retrospectively analyzed between May 1999 and November 2003. The mean age of the subjects was 31.8 years (range: 22–43 years) and the average infertility period was 3.5 years (range: 0.8–15.0 years).

TVL was performed in the lithotomy position as we described previously.5–10 After disinfection, a Hys-cath (Sumitomo Bakelite Co. Ltd, Tokyo, Japan) was inserted into the uterine cavity for the use of chromotubation. The uterine cervix was lifted with a tenaculum placed on the posterior lip and the central part of the posterior fornix was infiltrated with 2 ml of 1% lidocaine. A Veress needle was introduced 1.5 cm below the cervix and inserted into the pelvic cavity. Approximately 100 ml of saline was instilled in the pouch of Douglas. A 3 mm blunt trocar was inserted

The posterior of the uterus and the bilateral tubo-ovarian structures were carefully observed by TVL. The ovary and fimbriae are clearly visible.

A view of the normal-looking fimbriae by TVL.

Tubal passage was confirmed using indigo carmine. A Hyscath was inserted into the uterine cavity for the use of chromotubation.

Chromotubation was useful to the operator to identify the fimbria.

Figure 9.5

The endoscope was inserted from the fimbria by the guidance of chromotubation.

by a stab incision in the posterior fornix and then a 2.7 mm diameter semirigid endoscope was used at an optical angle of 30° and a flow channel attached to a 3-CCD digital video camera. The saline irrigation was continued throughout the procedure to keep the bowel and tubo-ovarian structures afloat. The posterior of the uterus and the bilateral tubo-ovarian structures were carefully observed (Figures 9.1 and 9.2), and tubal passage was confirmed using indigo carmine (Figures 9.3 and 9.4).

Salpingoscopic findings

TVL was carried out in 130 infertile women in the study period. Access to the pouch of Douglas was achieved in 123 (94.6%) of 130 patients. Twenty-six tubes could not be visualized because of extensive adhesion. Two patients with a history of unilateral salpingectomy and a patient with unicorn uterus were excluded. Therefore, 217 adnexa were clinically evaluated.

In 89 (41.0%) of 217 tubes, a salpingoscopy could be performed. The endoscope was inserted from the fimbria (Figures 9.5 and 9.6) by the guidance of chromotubation (Figure 9.7) and the distal part of the tubal mucosa could be observed (Figures 9.8–9.14). This result is similar to the first report of transvaginal salpingoscopy by Gordts et al,4 who reported that the distal tubal segment was cannulated

Figure 9.6

The endoscopic finding of the normal-looking fimbria.

successfully without additional instrumentation in 47%. To increase the success of salpingoscopy rates, it may be necessary to stabilize the tubes using grasp forceps to introduce an operative channel demonstrated by Watrelot et al.11

Each tubal finding under TVL was assigned to one of three classes, comprising regular, convoluted, and hydrosalpinx, as shown in Table 9.1.13 Tubal obstruction was diagnosed in 36 (16.6%) of the 217 tubes, whereas convoluted tubes or hydrosalpinges were diagnosed in 102 (47.0%) of the 217 tubes. A salpingoscopy could be performed in 87 (48.1%) of 181 patent tubes, significantly higher than that of 2 (5.6%) in 36 obstructed tubes (p <0.0001). A salpingoscopy was easily performed in patent tubes because chromotubation allowed the operator to identify the fimbrial ostium easily. It was also successfully

Table 9.1 Classification of laparoscopic findings

Reproduced from Marchino et al,13 with permission.