Leuven Institute for Fertility and Embryology, Leuven, Belgium. e-mail: ivo.brosens{at}med.kuleuven.ac.be
Dear Sir,
What is now universally called endometriosis was initially described as adenomyosis externa, ovarian endometrioma and peritoneal endometriosis. In the past, it has taken 25 years to recognize that they all are phenotypes of the same pathological process (Benagiano and Brosens, 1991). In his debate article, Garry (2004
) proposes to revive the names of the old investigators and to differentiate between a Cullens syndrome and a Sampsons syndrome. Sampsons syndrome would include superficial lesions with little or no associated smooth muscle metaplasia, minor symptoms, less structural changes in the pelvis and substantially less risk of major complications, while Cullens syndrome is to include nodular lesions with fibromuscular hyperplasia, more severe symptoms and almost all lesions at risk of developing malignancy and all lesions at risk of lower bowel and urinary tract involvement. Although ovarian endometrioma was not described by Cullen, this lesion is assimilated into Cullens syndrome as a deeply invasive lesion.
It is, indeed a cruel trick of history that the great contribution of Cullen to the knowledge of the nosographic entity of endometriosis is usually all but ignored. By 1920, Cullen presented a comprehensive review of all his findings, reporting the heterotopic presence of lesions characterized by the presence of uterine glands, stroma and smooth muscle (adenomyoma) in the uterus, the rectovaginal space, the supporting uterine ligaments, the wall of the bowel and on the ovary. On the other hand, it is customary to consider Sampson as the discoverer of endometriosis, and indeed his work in 1927 provided the first theory on the pathogenesis of the disease. The main contribution of Sampson, however, was in 1921 when he described in women operated at the time of menstruation that the ectopic endometrial-like tissue present in haemorrhagic cysts of the ovary showed evidence of desquamation and bleeding like eutopic endometrium. This observation led to the discovery that some haemorrhagic (chocolate) cysts of the ovary were of endometrial origin.
In the same publication, Sampson described the complex pathology of the chocolate cyst in a series of 22 patients. The mean age of the women was 37.4 years and in two-thirds of the patients severe pelvic pain was the leading symptom. All cysts had a thickened, fibrotic wall and were adherent to the posterior side of the uterus, parametrium or ligaments, and showed at the site of adhesion a typical defect which he interpreted at that time as the site of perforation. In 78% of the cysts, he described the presence of adenomyotic nodules in the adherent fibromuscular or myometrial tissue. Clearly, the endometriotic cyst detected today by ultrasound in the asymptomatic young woman with infertility differs in the vast majority by the absence of a fibrotic wall and adenomyotic lesions in the adherent tissue (Brosens et al., 1994; Gordts et al., 2004
).
In 1927, when Sampson had abandonned his belief that the ovary was the source of peritoneal endometriosis, he speculated that the chocolate cyst could be explained by endometrial tissue developing on the surface of the ovary caused adhesions which fused the ovary to the uterus and an endometrial cavity developed in this situation. Subsequently, Hughesdon (1957) demonstrated in a detailed study of 29 ovaries with in situ endometriomas that the cavity is formed by invaginated cortex, and the ectopic endometrium is superficial and not ate its way into the ovary like insects into an apple. In young women, the endometriotic implants lining the wall of the pseudocavity are as superfical as on the outer cortex (Brosens et al., 1994
). If the ovarian endometrioma is described as a deep invasive lesion, the adenomyotic or invasive lesion is not in, but outside the ovary in the adherent fibromuscular or muscular tissue. Both Sampson and Hughesdon would be shaken to learn that the endometriotic ovarian cyst is now classified as a deeply invasive lesion.
The current terminology of invasive endometriosis for deep, nodular lesions may be misleading. There is indeed no evidence that all small adenomyotic lesions will necessarily progress to or acquire a destructive invasive phenotype. Although there is unequivocal evidence that endometriotic cells have invasive potential in in vitro invasion assays, in vivo, endometriosis does not invade the ovarian stroma or the fat tissue in the retroperitoneal space. A non-destructive infiltration is seen in loose connective tissue, with smooth muscle metaplasia and hyperplasia of the fibromuscular or muscular structures. It appears just as possible that the extent of invasion is determined primarily by the microanatomy or the implant, such as the presence of a surface epithelium, and the hormonal response with interstitial bleeding and formation of haemorrhagic micro-cavities and micro-endometriomas. In a pro-inflammatory environment, endometriosis remains basically a self-limiting disease with adhesion formation, fibrosis and retraction, and even this healing process may be associated with a risk of bowel and ureter obstruction.
While we have proposed (Brosens, 1993) the distinction between the dominantly haemorrhagic and dominantly adenomyotic lesions as a basis for clinical classification, what is suggested by Ray Garry is inferring that Sampson discovered peritoneal endometriosis and that the name of Cullen can be linked with the ovarian endometrioma. The use of the eponyms in this context would do great injustice to the pioneering work of these early investigators.
When the complex pathology of endometriosis as described in detail by these authors is not appreciated, surgery is likely to result in overtreatment and unnecessary surgery, or undertreatment with risk of recurrence. In addition, any surgery in the pro-inflammatory microenvironment associated with endometriosis carries an increased risk of a post-surgical adhesive disease. Therefore, the first surgical procedure for endometriosis in a young patient may well determine her reproductive outcome. Today, we can appreciate centres of excellence with supersurgeons, but there is obviously a greater need that all residents receive during their supervised years a full training programme in endoscopic surgery to learn how to operate, when to operate and when not to operate.
Ivo Brosens
Leuven Institute for Fertility and Embryology
Leuven
Belgium
E-mail: ivo.brosens{at}med.kuleuven.ac.be
References
Benagiano G and Brosens I (1991) The history of endometriosis: identifying the disease. Hum Reprod 6,963968.[Abstract]
Brosens I.A (1993) Classification of endometriosis revisited. Lancet 341,630.[Medline]
Brosens I.A, Puttemans PJ and Deprest J (1994) The endoscopic localization of endometrial implants in the ovarian chocolate cyst. Fertil Steril 61,10341038.[Medline]
Cullen TS (1920) The distribution of adenomyoma containing uterine mucosa. Arch Surg 1,215283.
Garry R (2004) The endometriosis syndromes: a clinical classification in the presence of aetiological confusion and therapeutic anarchy. Hum Reprod 19,760768.
Gordts SY, Gordts S, Campo R, Valkenburg RM, Puttemans P and Brosens I (2004) Transvaginal hydrolaparoscopy, but not transvaginal ultrasound allows detection of small ovarian endometriomas. (in press)
Hughesdon PE. (1957) The structure of the endometrial cysts of the ovary. J Obstet Gynaecol Br Emp 44,481487.
Sampson JA (1921) Perforating hemorrhagic (chocolate) cysts of the ovary. Arch Surg 3,245323.
Sampson JA (1927) Peritoneal endometriosis due to the menstrual dissemination of endometrial tissues into the peritoneal cavity. Am J Obstet Gynecol 14,422469.