PATHOGENESIS AND TREATMENT OF DEEP ENDOMETRIOSIS
by M Nisolle and J Donnez
Cliniques Universitaires St Luc, Department of Gynaecology, Avenue Hippocrate 10 -1200, Brussels, Belgium

Introduction

It is generally believed that endometriosis is caused either by the implantation of retrograde menstrual endometrial cells, or by metaplasia.

In the pelvis, three different forms of endometriosis must be considered (1): 1) peritoneal endometriosis, 2) ovarian endometriosis, 3) rectovaginal septum endometriosis.

Recently, Nisolle (2) has demonstrated that these three entities are distinct and have a different histopathogenesis.

The deep-infiltrating endometriotic nodule in the recto-vaginal septum was studied by Koninckx (3). These authors described three subtypes of deep endometriosis depending on the depth of infiltration. The recto-vaginal septum endometriotic nodule was considered as the deepest form of endometriosis and a result of the natural evolution of peritoneal endometriosis in some women. Other authors (4-6) have suggested that the recto-vaginal endometriotic nodule is an adenomyotic nodule whose histopathogenesis is probably not related to the implantation of regurgitated endometrial cells but to the metaplasia of MŸllerian remnants.

Treatment options for pain or infertility secondary to cul-de-sac obliteration include ovarian suppression therapy with Danazol or gonadotropin-releasing hormone agonists, or surgery (3,7,8). During the last decade, some gynaecologists (7,9,11) have developed an endoscopic technique.

Histology of deep endometriosis

Deep vaginal endometriosis associated with pelvic endometriosis can take the form of nodular or polypoid masses involving the posterior vaginal fornix. It has been called an "adenomyotic nodule of the recto-vaginal septum" (12).

Histologically, scanty endometrial-type stroma and glandular epithelium is disseminated in muscular tissue proliferation. Indeed, in all cases, hyperplastic smooth muscle was observed surrounding the endometriotic tissue. This smooth muscle proliferation is probably a reaction to the presence of endometriotic foci in the recto-vaginal septum.

A circumscribed nodular aggregate of smooth muscle, endometrial glands, and usually, endometrial stroma thus represents the typical histological confirmation of a recto-vaginal endometriotic nodule. This histological description, similar to the histological description of adenomyosis, led us to suggest that the so-called endometriotic nodule of the recto-vaginal septum is, in fact, an adenomyoma or an adenomyotic nodule.

Endometriotic glands and stroma were often discovered by serial section up to the vaginal mucosa, which was sometimes replaced by endometrial epithelium. It is obvious that the invasion process of the smooth muscle by glandular epithelium does not require the presence of stroma. Indeed, glandular epithelium situated deep in the muscle was sometimes observed without any surrounding stroma.

Hypothesis of pathogenesis

Koninckx and Martin (3) suggested that the "endometriotic" nodule is the consequence of deep infiltration by active disease and that three types of "deep-infiltrating endometriosis" can be distinguished according to the depth of infiltration. According to their histological findings, the presence of loose connective tissue permits infiltration.

As the so called endometriotic nodule of the recto-vaginal septum, which consists essentially of smooth muscle with active glandular epilithum and scanty stroma, has a histological description similar to that of uterine adenomyosis, we have suggested that it is in fact a adenomyotic nodule which develops from MŸllarian rests. (5,6,12).

It has been proved (13) that Müllerian rest cells are able, after a long period of quiescence, to proliferate and differentiate, explaining the development of uterine adenomyosis. There is no reason why this hypothesis cannot be extended to explain the development of the recto-vaginal adenomyotic nodule.

But why do such Müllerian rest cells begin to proliferate and differentiate? A possible causal relationship with dioxin pollution has been proposed by Koninckx et al (14). Eskenazi and Kimmel (15) pointed out that in animal models, postnatal exposure to dioxin or dioxin-like compounds has been associated with endometriosis in females. Further studies in humans are needed to confirm these animal models and to explain the mechanism by which dioxin-like compounds could cause reproductive effects.

Smooth muscle proliferation and fibrosis are responsible for the nodular aspect of endometriosis located in the recto-vaginal septum. The clinical diagnosis is only made when smooth muscle proliferation is sufficient to be felt by vaginal examination. In our opinion (2), at least two hypotheses could explain this smooth muscle proliferation:

  1. Endometriotic foci involving smooth muscle are typically associated with striking proliferation of the smooth muscle, creating an adenomyomatous appearance similar to that of adenomyosis in the endometrium (16);
  2. The endometriotic stroma may exhibit smooth muscle metaplasia, as has been demonstrated within the wall of ovarian endometriotic cysts (17,18).

The first hypothesis is probably the most accurate. When endometriotic glands, whose histogenesis is probably metaplasia, affect the recto-vaginal septum which contains smooth muscle, smooth muscle proliferation can take place and the nodule thus develops. (2).

Hyperplasia of the smooth muscle in the septum provokes a perivisceritis through an inflammatory process followed by secondary retraction. This perivisceritis is not an endometriotic rectal lesion or an invasion of the rectal wall by the endometriotic process, as has been suggested by many authors (3,7,9) but in our opinion, (6) only the consequence of serosal retraction due to the inflammatory process and fibrosis on the anterior wall of the rectum.

Laparoscopic treatment of deep endometriosis

Our series of 480 cases of recto-vaginal endometriosis is presented here. In the majority of cases, the main presenting symptom was severe pelvic pain. Twenty-five percent of patients suffered pelvic pain and infertility.

Preoperative radiography of the colon was carried out in order to detect any involvement of the rectal surface. Profile radiography of an air contrast barium enema offers the best evaluation of the infiltration of the anterior wall of the rectum.

The surgical method involved the laparoscopic separation of the anterior surface of the rectum from the posterior vagina and the excision or ablation of endometriosis in that area. A mechanical bowel preparation was administered orally on the afternoon before surgery to induce brisk, self-limiting diarrhoea that rapidly cleansed the bowel without disrupting the electrolyte balance. All the laparoscopic procedures were performed using general anaesthesia. A 12-mm operative laparoscope was inserted through a vertical intraumbilical incision. Three other puncture sites were made 2-3 cm above the pubis, in the midline and in the areas adjacent to both the deep inferior epigastric vessels, which were identified before the cannulae were inserted.

Deep fibrotic nodular endometriosis involving the cul-de-sac required an excision of the nodular fibrotic tissue from the posterior vagina, rectum, posterior cervix and utero-sacral ligaments. As described by Reich et al (7), attention was first directed towards a complete dissection of the anterior rectum throughout its area of involvement until the loose tissue of the recto-vaginal space was reached.

A sponge on ring forceps was inserted into the posterior vaginal fornix and a dilatator (Hegar 25) was placed in the rectum. A cannula was inserted into the endometrial cavity to antevert the uterus. The peritoneum covering the cul-de-sac of Douglas was opened between the "adenomyotic" lesion and the rectum.

We first freed the anterior rectum from the loose areolar tissue of the recto-vaginal septum before excising or vaporising visible and palpable deep fibrotic endometriosis. This approach was possible even when anterior rectal muscular infiltration was present. Careful dissection was then carried out using the aquadissector for blunt dissection and the CO2 laser for sharp dissection until the rectum was completely freed and lying below the lesion. Excision of the fibrotic tissue on the side of the rectum was attempted only after the rectal dissection was complete. A partial rectal resection was never performed in our series. In cases of deep-infiltrating lesions, where the vaginal wall is more or less penetrated by the adenomyosis and excision of a part of the vagina is essential.

Dissection was performed not only with the removal of all visible endometriotic lesions, but also the vaginal mucosa with at least a 0.5 cm disease-free margin. Lesions extending totally through the vagina were treated with en bloc laparoscopic resection from the cul-de-sac to the posterior vaginal wall; the pneumoperitoneum was maintained and the posterior vaginal wall was closed vaginally.

In our series of 480 cases, laparoscopic rectal perforation occurred on four occasions. Perforation was diagnosed at the time of laparoscopy. In two cases, the rectum was repaired by laparotomy and in the other two cases by colpotomy.

Laparoscopic dissection was successfully performed in all cases, even when the radiography of the colon showed bowel involvement. During the same period, five cases of rectal endometriosis and eight cases of sigmoid colon endometriosis were diagnosed. None of the 13 patients had nodules in the septum, but all had bowel wall endometriosis, which caused menstrual rectorrhagia. In these 13 cases, laparotomy and bowel resection were performed. Histology proved the invasion of the mucosa by endometriotic tissue.

The absence of continuing development of the "rectal lesion" after removal of the nodule supports our hypothesis concerning its recto-vaginal septum origin, and strongly suggests that it is not necessary to excise the anterior wall of the rectum for successful treatment (12,19). While we recognise that bowel resection may be necessary in rare cases, it seems prudent to be conservative, rather than encourage, the widespread use of an aggressive surgery.

In conclusion, deep-infiltrating endometriosis should be considered as a specific disease, different from mild or minimal endometriosis and ovarian cystic endometriosis. The lesion originates from the recto-vaginal septum tissue and consists essentially of smooth muscle with active glandular epithelium and scanty stroma. We suggest calling this disease: "recto-vaginal adenomyosis".

REFERENCES
  1. Donnez J., Nisolle M., Casanas-Roux F. (1992) Three-dimensional architectures of peritoneal endometriosis. Fertil. Steril., 57, 980-983.
  2. Nisolle M (1996). Peritoneal, ovarian and recto-vaginal endometriosis are three distinct entities. Thse d'AgrŽgation, UniversitŽ Catholique de Louvain.
  3. Koninckx P. (1993) Deeply infiltrating endometriosis. In Brosens I. and Donnez J. (eds) Endometriosis: Research and Management. Parthenon Publishing, p. 437-446.
  4. Brosens IA (1994). Is mild endometriosis a progressive disease? Hum. Reprod., 9, 22092211.
  5. Donnez J., Nisolle M., Casanas-Roux F., Anaf V., Smets M. (1994) Laparoscopic treatment of recto-vaginal septum endometriosis in Donnez J. and Nisolle M. (eds), An Atlas of Laser Operative Laparoscopy and Hysteroscopy, Parthenon Publishing, pp 75-85.
  6. Donnez J, Nisolle M (1995). L'endomŽtriose pŽritonŽale, le kyste endomŽtriotique ovarien et le nodule de la lame recto-vaginale sont trois pathologies diffŽrentes. Editorial. RŽfŽrences en GynŽcologie-ObstŽtrique, 3, 121-123.
  7. Reich H, McGlynn F, Salvat J (1991). Laparoscopic treatment of cul-de-sac obliteration secondary to retrocervical deep fibrotic endometriosis. Reprod Med, 36, 516.
  8. Donnez J, Nisolle M, Casanas-Roux F (1990). Endometriosis - associated infertility: Evaluation of preoperative use of Danazol, Gestrinone and Buserelin. Int. J. Fertil., 35, 297301.
  9. Nezhat C, Nezhat F, Pennington E (1992) Laparoscopic treatment of lower colorectal and infiltrative recto-vaginal septum endometriosis by the technique of video laparoscopy. Br J Obstet Gynaecol, 99, 664-667.
  10. Canis M, Wattiez A, Pouly JL, Bassil S, Bouquet de Jolinire J, Chapron C, Manhes H, Mage G, Bruhat MA (1993). Laparoscopic treatment of endometriosis. In Brosens I and Donnez J (eds), Endometriosis: Research and Management. Parthenon Publishing, p. 407-417.
  11. Donnez J., Nisolle M., Casanas-Roux F., Clerckx F. (1993) Endometriosis: rationale for surgery. In Brosens I. and Donnez J. (eds) Endometriosis: Research and Management. Parthenon Publishing, pp 385-395.
  12. Donnez J, Nisolle M, Casanas-Roux F, Bassil S, Anaf V (1995). Recto-vaginal septum endometriosis or adenomyosis: laparoscopic management in a series of 231 patients. Hum. Reprod., 10, 630-635.
  13. Minh HN, Smajda A, Orcel L (1988) Etude morphologique comparŽe du mŽsothŽlium pŽritonŽal de l'ŽpithŽlium germinatif de lĠovaire. DŽduction histogŽnŽtique sur lĠendomŽtriose. J. Gynecol Obstet Biol Reprod, 17, 479-484.
  14. Koninckx PR, Braet P, Kennedy SH, Barlow DH (1994) Dioxin pollution and endometriosis in Belgium. Hum. Reprod. 9, 1001-1002.
  15. Eskenazi B, Kimmel G (1995). Workshop on perinatal exposure to dioxin-like compounds. II. Reproductive effects. Environ Health Perspet, 103, 143-145.
  16. Scully RE, Richardson GS, Barlow JF (1966). The development of malignancy in endometriosis. Clin. Obstet. Gynecol., 9, 384-411.
  17. Scully RE (1981). Smooth-muscle differentiation in genital tract disorders (Editorial). Arch. Pathol. Lab. Med., 105, 505-507.
  18. Cozzuto C (1981). Uterus-like mass replacing ovary. Report of a new entity. Arch. Pathol. Lab. Med., 105, 508-511.
  19. Donnez J, Nisolle M (1995). Advanced laparoscopic surgery for the removal of recto-vaginal septum endometriotic or adenomyotic nodules. Baillire's Clinical Obstet Gynecol, 9, 769-774.

Return to the Table of Contents

 {FILE "e:\platform\iis\wwwroot\oblive\eago\_footers\footer.txt"}