1 II Department of Obstetrics and Gynecology, Clinica L.Mangiagalli and 2 Macedonio Melloni Hospital, University of Milan, Milan, Italy
3 To whom correspondence should be addressed at: II Department of Obstetrics/Gynecology, Via Commenda 12, 20122 Milano, Italy. e-mail: dadosomigliana{at}yahoo.it
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: adhesion/deep endometriosis/endometriosis/pathogenesis/superficial endometriosis
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In this study, we have hypothesized that if deep endometriosis and the other forms of the disease do not share a common pathogenetic mechanism, they should not be significantly associated. In other words, if a different and peculiar pathogenetic mechanism leading to deep nodules exists, the frequency of the presence of other forms of endometriosis in patients with deep endometriosis should be similar to that observed in the general population. Therefore, to gain insights into this debated topic, we have evaluated the frequency of non-deep forms of the disease such as superficial implants, endometriotic ovarian cysts and pelvic adhesions among patients affected by deep endometriosis.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Deep endometriosis was defined as the presence of histologically confirmed peritoneal endometriosis infiltrating to a depth of at least 5 mm beneath the peritoneal surface (Cornillie et al., 1990). Description of surgery was always performed by one of five different surgeons (M.B., M.C., M.I., M.V. and E.S.) who have been actively engaged for a long time in endometriosis clinics and research. Information regarding clinical and surgical data was extracted from detailed patients charts. According to the protocol used in our Department, descriptions of the anatomical pelvic situation and surgical procedures are extremely detailed. Specifically, the following items always have to be described: abdominal situation, uterus, right ovary, right tube, left ovary, left tube, Douglas cul-de-sac, vesico-uteral cul-de-sac, other peritoneal locations, tube patency assessed by salpingo-cromo-scopy and other remarks. Moreover, in our unit, a sketch of the anatomical situation is systematically drawn. Cases were discarded if written and graphic description did not coincide and/or if an infective origin of adhesions was suspected. Adhesions were considered of endometriotic nature if not attributable to another cause. The diameter of deep endometriotic nodules was recorded at the time of histological examination or at the time of surgery when lesions could not be entirely surgically removed. Endometriotic cysts were always entirely or partially excised and sent for microscopic evaluation to confirm the diagnosis. Superficial endometriotic implants were excised and histologically evaluated only if there were doubts regarding the endometriotic nature of the lesions. Otherwise, they were electro-coagulated at the time of surgery.
Ninety-five patients were recruited. Two cases were discarded since the written and graphic descriptions were found to be discordant, leaving 93 patients for data analysis. The median (range) diameter of deep endometriotic nodules was 1.5 (0.54.0) cm. Eighteen (19.3%) patients were found to have more than one nodule. The mean ± SD (range) age of patients was 30.7 ± 4.8 (1941) years. Pain symptoms were distributed as follows: 77.4% reported moderate to severe dysmenorrhoea, 46.2% deep dyspareunia, 51.6% chronic pelvic pain and 7.5% painful defecation. Fourteen percent did not report relevant pain symptoms at all. A total of 64.5% reported more than one pain symptom. Other complaints included lower urinary tract symptoms (5.4%) and gastrointestinal symptoms (11.8%). Infertility, defined as failure to achieve a pregnancy within 12 months, was reported by 37.6% of women. The median (range) duration of infertility was 36 (12120) months. A proportion of 15.1% women had previously delivered. Twenty-six patients had previously undergone pelvic surgery. For 21 of them, the main indication for prior intervention was endometriosis. The vast majority of patients were operated on using laparoscopic techniques. Laparotomy was required in 7.5% of cases. According to the ASRM classification (ASRM, 1997), 18.3% were stage I, 19.4% stage II, 38.7% stage III and 23.7% stage IV. Concomitant gynaecological pathological findings were observed in 13 cases; specifically, myomas were documented in seven cases, non-endometriotic ovarian cysts in three cases, uterine septum in two cases and endometrial polyps in one case.
A binomial probability distribution model was used to calculate the 95% confidence interval (95% CI) of the frequency of the presence of non-deep peritoneal forms of the disease (Pearson and Hartley, 1970).
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
There are at least three major concerns regarding our study design. First, we did not include controls. In this regard, given both the aims of our study and the impossibility of adequately detecting the presence of endometriosis without a direct visualization of the pelvis, only a surgical control group would have been accepted. A population-based group is thus advisable but also unrealizable due to the ethical concerns related to the necessity of an invasive diagnosis (Zondervan et al., 2002). On the other hand, considering the strong association observed in this study, a pure matter of chance appears extremely unlikely. Indeed, it is currently estimated that there is a 10% prevalence of endometriosis in the general population (Eskenazi and Warner, 1997
). This rate appears significantly lower when compared with the rates observed in our study (the 95% CI of the rate of association between deep endometriosis and other forms of endometriosis varied between 87.7 and 97.2%). A second concern with this study is related to its retrospective design. In this regard, it should be emphasized that a rigid and complete protocol for both clinical data recording and anatomical surgical description is systematically followed in our Department. The availability of both a written and graphic description of the anatomical situation allowed us both to control the accuracy of data and to define precisely the localization of the lesions. It is of note that only two cases had to be discarded due to discrepancies between the written and graphic description. Therefore, we believe that, despite the retrospective design of our study, available data should be considered accurate, thus supporting the reliability of our observations. Finally, it has to be noted that the presence of an association, although highly suggestive, does not definitively demonstrate that deep endometriotic nodules should not be considered a different form of endometriosis. Indeed, it cannot be totally excluded that deep endometriosis and the other forms of the disease are two distinct entities that only share some common risk factors. In our opinion, this interpretation is unlikely, however, considering the extremely high degree of association observed.
Although the present study supports a common pathogenetic mechanism for the different forms of endometriosis, it does not clarify why, from a histological point of view, deep endometriotic nodules but not the other forms of the disease resemble adenomyosis. In this regard, it has to be noted that Anaf et al. (2000b), using immunochemical techniques with a monoclonal antibody against muscle-specific actin, recently have demonstrated that a smooth muscle component is in fact present in all types of endometriotic lesions. On the other hand, they failed to observe this component in disease-free peritoneum. These authors thus hypothesize that the smooth muscle component may result from the totipotential capacity of the pelvic and lower abdomen mesothelium to differentiate. In other words, the implanted endometrium may cause a metaplastic response in the underlying tissue. This metaplastic response might differ from one location to the other, thus explaining histological differences among the different forms of endometriosis. In light of these findings, the definition of distinct endometriotic entities based on the difference in the tissue composition of the lesions (endometriotic lesions versus adenomyotic nodules) appears inconsistent. Furthermore, in refutation of the metaplasia theory, patients with deep endometriosis of the rectovaginal septum have about a one-third reduction in the depth of the Douglas pouch, an observation that would not be expected if the lesions have an extraperitoneal origin (Vercellini et al., 2000
). Finally, it also cannot be ruled out that the same disease may originate from several different pathogenetic mechanisms. In this context, it is interesting to note that Fedele et al. (1998
) reported four cases of bladder endometriosis resulting from the extension of adenomyosis lesions of the anterior uterine wall to the bladder. However, albeit that this may be possible, this extremely particular pathogenetic mechanism cannot explain the large majority of cases of bladder endometriosis (Chapron et al., 2002
).
In conclusion, results from the present investigation and from current available studies do not support the idea that deep endometriosis should be considered as external adenomyosis arising from metaplasia of Müllerian remnants.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Anaf V, Simon P, Fayt I and Noel J (2000b) Smooth muscles are frequent components of endometriotic lesions. Hum Reprod 15,767771.
Anaf V, Simon, P, El Nakadi I, Fayt I, Simonart T, Buxant F and Noel JC (2002) Hyperalgesia, nerve infiltration and nerve growth factor expression in deep adenomyotic nodules, peritoneal and ovarian endometriosis. Hum Reprod 17,18951900.
ASRM (1997) Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil Steril 67,817821.[CrossRef][ISI][Medline]
Chapron C, Boucher E, Fauconnier A, Vieira M, Dubuisson JB and Vacher-Lavenu MC (2002) Anatomopathological lesions of bladder endometriosis are heterogeneous. Fertil Steril 78,740742.[CrossRef][ISI][Medline]
Cornillie FJ, Oosterlynck D, Lauweryns JM and Koninckx PR (1990) Deeply infiltrating pelvic endometriosis: histology and clinical significance. Fertil Steril 53,978983.[ISI][Medline]
Eskenazi B and Warner ML (1997) Epidemiology of endometriosis. Obstet Gynecol Clin North Am 24,235258.[ISI][Medline]
Fauconnier A, Chapron C, Dubuisson JB, Vieira M, Dousset B and Breart G (2002) Relation between pain symptoms and the anatomic location of deep infiltrating endometriosis. Fertil Steril 78,719726.[CrossRef][ISI][Medline]
Fedele L, Piazzola E, Raffaelli R and Bianchi S (1998) Bladder endometriosis: deep infiltrating endometriosis or adenomyosis? Fertil Steril 69,972975.[CrossRef][ISI][Medline]
Koninckx PR, Meuleman C, Demeyere S, Lesaffre E and Cornillie FJ (1991) Suggestive evidence that pelvic endometriosis is a progressive disease, whereas deeply infiltrating endometriosis is associated with pelvic pain. Fertil Steril 55,759765.[ISI][Medline]
Nisolle M and Donnez J (1997) Peritoneal endometriosis, ovarian endometriosis, and adenomyotic nodules of the rectovaginal septum are three different entities. Fertil Steril 68,585596.[CrossRef][ISI][Medline]
Pearson EG and Hartley HO (1970) Biometrika Tables for Statisticians, 3rd edn. Vol. 1. Cambridge University Press, Cambridge.
Vercellini P, Aimi G, Panazza S, Vicentini S, Pisacreta A and Crosignani PG (2000) Deep endometriosis conundrum: evidence in favor of a peritoneal origin. Fertil Steril 73,10431046.[CrossRef][ISI][Medline]
Zondervan KT, Cardon LR and Kennedy SH (2002) What makes a good casecontrol study? Design issues for complex traits such as endometriosis. Hum Reprod 17,14151423.
Submitted on April 17, 2003; resubmitted on June 23, 2003; accepted on September 16, 2003.