Endometriosis: Pathogenesis and Treatment 2014 Ed.

3. Visible and Invisible (Occult) Endometriosis

Khaleque Newaz Khan 

(1)

Department of Obstetrics and Gynecology, Graduate School of Biomedical Sciences, Nagasaki University, 1-7-1 Sakamoto, Nagasaki 852-8501, Japan

Khaleque Newaz Khan

Email: nemokhan@nagasaki-u.ac.jp

Abstract

Endometriosis is a chronic disease characterized by endometrial tissue located outside of the uterine cavity and is associated with chronic pelvic pain and infertility. However, an in-depth understanding of the pathophysiology of endometriosis is still elusive. Once generated within pelvis due to retrograde entry of menstrual debris, peritoneal endometriotic lesions time-dependently change their color appearance resulting from certain biochemical change within lesions. A variable pattern of endometriotic lesions within pelvis can be detected by laparoscopy as visible peritoneal endometriosis. It is generally believed that besides ovarian steroid hormones, the growth of endometriosis can be regulated by innate immune system in pelvic microenvironment by their interaction with endometrial cells and immune cells. We conducted a series of studies in perspectives of pelvic inflammation that is triggered primarily by bacterial endotoxin (lipopolysaccharide, LPS) and is mediated by Toll-like receptor 4 (TLR4) and showed their involvement in the growth regulation of visible peritoneal endometriosis. Even with the careful eye of an expert surgeon, we may sometimes miss to detect peritoneal lesion within peritoneal cavity or deep into peritoneum. In such a case, random collection of normal peritoneum may carry the possibility to identify some hidden endometriotic lesions by microscopy and these lesions can be named as invisible endometriosis. Here, we discuss the color appearance of peritoneal lesions, role of innate immune system in visible endometriosis, and finally our recent findings on invisible microscopic endometriosis and their biological and clinical significance.

Keywords

Visible endometriosisBacterial endotoxinInnate immunityTLR4Invisible endometriosis

3.1 Visible Endometriosis: Introduction

Endometriosis, the presence of functional endometrium outside of the uterine cavity, is a common disease, causing abdominal pain, dysmenorrhea, dyspareunia, and infertility in about 10 % of the female population [1]. Besides metaplastic transformation of endometrial and peritoneal mesothelial cells, the transplantation, implantation, and growth of exfoliated menstrual debris on the peritoneal and ovarian surfaces are the widely accepted mechanisms of endometriosis [23]. With the elapse of many decades and publication of abundant number of literatures, exact physiopathology or pathogenesis of endometriosis is still debatable. The potential role of ovarian steroid hormones in the regeneration of endometrium after menstruation and the growth of endometriosis has been demonstrated [4]. However, as a nonself lesion in pelvic environment, the growth or persistence of visible endometriosis can also be regulated by innate immune system. Therefore, we cannot rule out the involvement of an immuno-endocrine crosstalk in the pelvis of women with visible endometriosis.

3.1.1 Pathogenesis and Natural Course of Peritoneal Endometriosis

3.1.1.1 Common Concepts in the Pathogenesis of Endometriosis

In addition to transplantation and implantation theory of Sampson [2] and coelomic metaplasia theory of Meyer [5], immuno-surveillance of the refluxed endometrial cells is another attractive theory for the development of pelvic endometriosis. The immune tolerance or immune defect theory could be responsible for a deficiency in the rejection of the autologous cells derived from the eutopic endometrium in the peritoneal cavity after menstrual reflux. This rejection in the clearance of endometrial cells could be contributed by a dysfunctional immune response in the pelvic cavity [6].

According to the retrograde menstruation theory, endometrial fragments flow back through the patent fallopian tubes, reach the peritoneal cavity, attach on the pelvic mesothelium, invade the peritoneum, and develop into endometriotic lesions [2]. Limited information still exists regarding early endometrial-peritoneal attachment and invasion in the development of endometriosis.

3.1.1.2 Role of Cell Adhesion Molecules in Endometriosis

After overcoming a phase of immune tolerance, a key step in the development of early endometriosis is the ability of endometrial cells to adhere to mesothelium and invade the extracellular matrix. These effects are contributed by a number of intercellular adhesion molecules (ICAMs) and subcellular matrix degrading metalloproteinases (MMPs) [7]. The expressions of these ICAMs such as integrins and E-cadherin and MMPs are already detected in cells derived from menstrual effluent, endometrium, peritoneal fluid, peritoneum, and endometriosis [8].

3.1.1.3 Role of Heme Metabolism in Endometriosis

A potential implication of hemoglobin in the pathogenesis of peritoneal endometriosis has been recently reported [9]. A simple hypothesis is that hemoglobin, being released into the peritoneal cavity after red blood cell lysis, may activate cell adhesion molecules and induce cytokine production, cell proliferation, and the process of neovascularization. Degradation of hemoglobin yields biologically active molecules, heme, and its products of oxidative cleavage by heme oxygenases such as iron, carbon monoxide, biliverdin, and bilirubin. Accumulation of heme in the peritoneal cavity might have a number of deleterious effects including induction of oxidative stress, stimulation of cell adhesion, and cytokine production by macrophages (Mφ). All these biological events are finally involved in the generation of visible peritoneal lesions.

3.1.1.4 Color Changes of Visible Peritoneal Lesions

The lesions of early endometriosis are either transparent or translucent because they still lack formation of vasculatures around them. We named these early lesions as nonopaque lesions [10] because these lesions contain either of watery, serous, or mucinous secretion and there is no collection of blood in the stroma by histology. Once cellular attachment and invasion of endometrial cells are established, the subsequent growth or maintenance of endometriotic lesions is maintained by promotion of mitogenesis and angiogenesis with the continuation of menstrual cycle. The growth-promoting effect of endometriosis is contributed by an orchestrated action of estrogen and other inflammatory or proinflammatory mediators. Over proliferation of microvessels in the growing endometriotic lesions causes oozing of blood in the stroma and appears as blood-filled opaque red lesions by laparoscope [10]. With the progression of time, there is deoxygenation process from hemoglobin to methemoglobin or hemosiderin leading to color changes of these opaque red lesions to black lesion or related lesions. In this stage, collection of blood in the stroma disappears. Black lesion again changes to white lesion due to collection of bilirubin or biliverdin and accumulation of fibrous tissue. In this stage, gland gradually becomes smaller and stroma sometimes disappears due to deposition of fibrous tissue. Finally old lesions disappear and there is new focus of endometriosis due to continuation of menstrual reflux. These sequential events indicate that once exfoliated, the endometrium enters into the pelvic cavity and becomes attached to the mesothelial layer, and then a process of angiogenesis, heme metabolism, and fibrosis ensue to maintain the natural course of endometriosis (Fig. 3.1). A panel of nonopaque lesion, blood-filled opaque lesion, blue berry spots, and their corresponding histological pictures is shown in Fig. 3.2.

A313895_1_En_3_Fig1_HTML.gif

Fig. 3.1

Shows the diagrammatic representation of the natural course of visible peritoneal endometriosis in pelvic cavity. After initial attachment of refluxed endometrial cells with peritoneal cells producing early endometriotic lesions, the consequent events of mitosis, angiogenesis, metabolic degradation of heme, and appearances of fibrosis result in the generation of different morphological appearances of peritoneal endometriosis as shown in this figure

A313895_1_En_3_Fig2_HTML.gif

Fig. 3.2

Shows the laparoscopic and corresponding histologic appearance of nonopaque transparent/translucent lesion (A, B), blood-filled opaque red lesion (C, D), and blue berry spots (E, F) in pelvic cavity of women with peritoneal endometriosis. It can be noted here that nonopaque endometriotic lesions such as vesicular bleb (A) lack oozing of blood in stroma (B) and opaque red lesions such as blood bleb with ecchymosis (C) are accompanied by oozing of blood in the stroma of these lesions (D). In contrast, black lesions such as blue berry spots (E) are manifested by color change and disappearance of blood from the stroma of these lesions (F)

3.1.2 Role of Mφ in Pelvic Inflammation and Growth of Visible Endometriosis

3.1.2.1 Infiltration of Mφ in Eutopic and Ectopic Endometria

As a cell component of innate immune system, peritoneal fluid (PF) and eutopic/ectopic endometria derived from women with endometriosis have been shown to contain higher numbers of activated Mφthan in control women [1112]. This results in the secretion of higher concentrations of growth factors and cytokines in the PF as produced by the stimulated Mφ in these patients [11]. Red peritoneal lesions and their adjacent peritoneum had the greatest Mφ infiltration, compared with black/white lesions or chocolate cyst walls. These results indicate that early endometriosis with red peritoneal lesions induces a higher inflammatory response in the pelvic cavity than advanced endometriosis [12]. The inflammatory reactions in the eutopic and ectopic endometria suggest that the growth of endometriosis does not depend on the fibrotic extension of disease; rather, it depends on the tissue activity of endometriosis. We presume that extension of disease could be related to pelvic pain, but higher tissue activity of endometriosis associated with abundant recruitment and infiltration of Mφ could be related to infertility.

3.1.2.2 Role of Endotoxin or Lipopolysaccharide (LPS) in Mφ-Mediated Pelvic Inflammation

We examined the ability of peritoneal Mφ to synthesize different macromolecules in basal conditions and after treatment with lipopolysaccharide (LPS), a bacterial endotoxin derived from the cell wall extract of Gram-negative bacteria. We speculated that LPS could be a primary inflammatory mediator of Mφ stimulation in pelvic microenvironment. We found that activated Mφ synthesize and secrete variable amount of different secondary inflammatory mediators such as IL-1, IL-6, IL-10, TNF-α, and other growth factors in response to LPS [1314].

3.1.2.3 TLR4-Mediated Cytokine Production and Growth of Endometrial Cells

Bacterial endotoxin (LPS) is recognized by Toll-like receptor 4 (TLR4), a pattern recognition receptor, and transmits NF-κB-mediated cellular signals in association with other accessory molecules [15]. We confirmed gene and protein expression of TLR4 in Mφ, gland cells, and stromal cells derived from the eutopic and ectopic endometria of women with or without endometriosis [16]. In an attempt to examine that the stimulating effect of LPS in the production of cytokines and growth factors is mediated by TLR4, we pretreated peritoneal Mφ and glandular epithelial cells/stromal cells derived from eutopic/ectopic endometria with antibody against TLR4 and then again treated them with LPS. We found that blocking of TLR4 was able to significantly suppress Mφ-mediated production of HGF/VEGF/IL-6/TNF-α as well as growth of endometriotic cells [16]. These results indicate that LPS-mediated inflammatory reaction and growth of endometriotic cells are mediated by TLR4.

3.1.3 Source of Bacterial Endotoxin in Endometriosis

3.1.3.1 Endotoxin Levels in Body Fluids

We measured endotoxin levels in the menstrual fluid (MF) and PF of women with and without endometriosis. We found that the concentration of bacterial endotoxin is two- to fourfold higher in the MF when compared with that in PF. Endotoxin level in MF/PF was also significantly higher in women with endometriosis than in control women [16]. We found the highest endotoxin level during the menstrual phase and persistence of a small amount of endotoxin in the pelvis either in the proliferative phase or in the secretory phase of the menstrual cycle [16]. This indicates that MF of women with endometriosis is highly enriched with bacterial endotoxin followed by the presence of a modest amount in the PF.

3.1.3.2 Source of Endotoxin in Intrauterine and Pelvic Environment

There is a possibility that the lower genital tract of women with or without endometriosis is contaminated with a number of normal bacterial florae including Escherichia coli (E. coli). Therefore, we speculated that there might be an ascending migration of E. coli from the vaginal lumen up into the uterine cavity that causes contamination of menstrual blood and resulting in the subsequent release of endotoxin into menstrual blood and back to the peritoneal fluid. After bacteria culture analysis, we found a significantly higher colony formation (CFU/ml) of E. coli in the menstrual blood of women with endometriosis than that in control women [16]. The CFU of E. coli was also significantly higher in women containing red peritoneal lesions than in women having only chocolate cyst.

Based on these findings of our serial experiments, we proposed a new “bacterial contamination hypothesis” that may be involved in the growth regulation of visible peritoneal endometriosis via LPS/TLR4 cascade [16]. This E. coli contamination of menstrual blood is responsible for higher endotoxin levels in the MF and PF of women with endometriosis. In search of a mechanistic basis of bacterial contamination of menstrual blood, we found that higher concentrations of PGE2 in MF and PF of women with endometriosis were involved in E. coli growth by its direct bacterial proliferation effect or indirect immunosuppressive effect [17].

3.1.4 Inflammation, Stress Reaction, and TLR4 in Endometriosis

In addition to pelvic inflammation, a wide variety of stressful stimuli, such as heat shock, ultraviolet radiation, viral or bacterial infection, internal physical stress (cell growth, differentiation, invasion), chemical stress (ligand/receptor interaction), oxidative stress, neurogenic stress, pain sensation, and pelvic inflammation, may induce a variable degree tissue stress reaction in pelvis and release stress-induced proteins, such as heat shock proteins (HSPs) [1718]. As a danger signal, the effect of HSPs has been reported to be mediated by TLR4 either alone or in combination with LPS.

We demonstrated a variable amount of soluble HSP70 in MF, PF, and in different peritoneal lesions [1819]. These body fluid and tissue concentrations of HSP70 were significantly higher in women with endometriosis than in control women. In an in vitro cell culture system, we found that HSP70 was able to induce TLR4-mediated proinflammatory response and growth of endometriotic cells, and a combined effect was observed between HSP70 and LPS in further promoting pelvic inflammation and growth of endometriotic cells [19]. Although polymyxin B, a potent LPS antagonist, or anti-HSP70 antibody, was unable to suppress combined LPS+HSP70-mediated growth of endometriosis, blocking of TLR4 alone significantly suppressed LPS+HSP70-mediated inflammation and growth of endometriosis [19]. Recently it has been demonstrated that in addition to the effects of endogenous danger signals via TLRs, tissue oxidative stress itself may promote NF-κB-mediated or TLR4-mediated growth of endometriosis [20]. In fact, LPS itself has the capacity to produce ROS by Mφ [15]. These findings are consistent with the understanding that LPS, endogenous danger signals, and oxidative stress may promote the onset and progression of visible endometriosis after activation of TLR4 and/or NF-κB signaling.

3.1.5 Inflammation, Ovarian Steroid, and TLR4 in Endometriosis

Basically endometriosis is an estrogen-dependent disease and induces an inflammatory reaction in pelvic environment. Irrespective of the phases of the menstrual cycle, a small amount of estrogen is available in the PF of women with and without endometriosis [21]. Therefore, it is important to know the combined effect of estrogen and inflammation in the growth of endometriosis. Based on published reports, gland cells/stromal cells of eutopic/ectopic endometria and Mφ retain ER and PR [1821]. Recently, we reported that Mφ-mediated production of HGF/VEGF/IL-6/TNF-α in response to ovarian steroid was further enhanced after treatment with LPS [22]. An additive effect was observed between E2 and LPS on cytokine production and growth of eutopic/ectopic endometrial stromal cells when compared with their individual treatment. This combined effect of E2+LPS on pelvic inflammation and cell growth was markedly abrogated after pretreatment of cells with anti-TLR4 antibody and ICI, an ER antagonist [21]. Our findings suggest that a TLR4/ER-mediated immuno-endocrine crosstalk in pelvis may be involved in the growth or progression of endometriosis.

3.2 Invisible (Occult) Endometriosis: Introduction

The detection and visible diagnosis of peritoneal endometriosis is usually performed by laparoscopy, a gold standard modality, and is microscopically confirmed by histopathology. Even with the careful eyes of expert surgeons, there is obvious chance to miss or overlook hidden lesions in visually normal peritoneum. Therefore, immense interest could arise to randomly collect visually normal peritoneum from different anatomical location in pelvis and to investigate the nature of these visually undetectable lesions of endometriosis. The concept of microscopic endometriosis in visually normal peritoneum was first reported by Murphy in 1986 [23] and subsequently confirmed with an incidence rate of 6–13 % [24].

We histologically examined all biopsy specimens derived from visually normal peritoneum of women with and without endometriosis to detect the possible occurrence of invisible (occult) microscopic endometriosis (IME). The designation of visually normal peritoneum was based on modified criteria of Redwine [25]. The question still remains, if hidden endometriosis lesions could be detected in normal peritoneum, are these invisible lesions really inactive as proposed in a previous report [26], or do they truly retain some biological activity? If tissue activity of IME is there, this could be a clinically important issue. To address this question, we recently investigated the expression patterns of some tissue activity markers including ovarian steroid receptors and cell proliferation marker in histologically confirmed IME lesions.

3.2.1 Pattern of IME in Normal Peritoneum

After careful observation and analysis, we collected 227 visually normal peritoneal samples from 151 women with visible endometriosis and 78 samples from 62 women without any visible peritoneal lesions (control). We detected three patterns of IME: (I) presence of typical gland/stroma, (II) reactive hyperplastic change of endometrioid-like epithelium with surrounding stroma, and (III) single-layered mesothelium- or epithelium-lined cystic lesions with surrounding rim of stromal cells. All these IME lesions were confirmed by their immunoreactivity to Ber-EP4 (marker of gland epithelium), CD10 (marker of stroma), and nonreactivity to calretinin (marker of mesothelial cells) (Fig. 3.3).

A313895_1_En_3_Fig3_HTML.gif

Fig. 3.3

Microscopically detected three patterns of invisible (occult) endometriosis in visually normal peritoneum. Pattern I shows presence of typical gland/stroma; pattern II shows reactive hyperplastic change of endometrioid epithelial cells with surrounding stroma; and pattern III shows single-layered epithelium-lined cystic lesions with surrounding stromal cells (all in HE stain, upper column). The identification of glandular epithelial cells, stromal cells, and peritoneal mesothelial cells was confirmed by the immunoreaction to Ber-EP4, CD10, and calretinin, respectively, and are shown against each HE-stained slide. Flat mesothelial cells derived from normal peritoneum and mesothelioma cells as a positive control (inset) immunoreactive to calretinin are shown at the right panel. The immunoreactions to nonimmune mouse IgG as a negative control are shown on the extreme right panel. HE stain, hematoxylin and eosin stain. Magnification of slides (×200)

We could detect variable patterns of IME in the normal peritoneum derived from 23 women with endometriosis (biopsy samples, n = 27) and 4 control women (biopsy samples, n = 4) without visible endometriosis. The detection rate of IME was as follows: for endometriosis, 15.2 % (23/151) and 11.8 % (27/227) and for control women, 6.4 % (4/62) and 5.1 % (4/78) by the number of patients and number of collected samples, respectively. A higher tendency in the incidence of IME was found in women with visible endometriosis than in control women (p = 0.06 by patient number and p = 0.07 by sample number) [27].

A predominance of IME occurrence was observed in pouch of Douglas and uterovesical space than in other anatomical sites in pelvis. A dominant presence of r-ASRM stages I–II endometriosis and red/black lesions and complaints of dysmenorrhea were observed in women with visible endometriosis harboring IME in their peritoneum [27].

3.2.2 ER/PR/Ki-67 Expression in IME Lesions

A variable ER and PR immunoexpressions were observed in all IME lesions detected in women with visible endometriosis and control women. The immunoreaction of PR as measured by quantitative-histogram (Q-H) score appeared to be higher in all patterns of IME lesions comparing to ER expression.

A stronger immunoreaction of Ki-67 (cell proliferation marker) was found in pattern I/II IME lesions than in pattern III IME lesions detected in women with visible endometriosis. A weak Ki-67 expression was found in pattern I/III IME lesions diagnosed in control women. Ki-67 index (mean percentage of Ki-67-positive nuclei among total cells) was significantly higher in pattern I/II IME lesions than in pattern III IME lesions found in women with visible endometriosis [27].

We are against the argument by Donnez et al. [26] that IME lesions are quiescent and they are nonactive or inactive and that these lesions are clinically irrelevant. Our findings indicate that a proportion of occult lesions are indeed biologically active and a substantial amount of estrogen and different inflammatory mediators in pelvis might be involved in the growth promotion of IME lesions even if it is undetectable by naked eyes. With the influence of both systemic and local estrogen, these IME lesions, even if it is minute in size, may time-dependently increase in size to be recognized by histology. This could be responsible for the subsequent recurrence/occurrence of endometriotic lesions or persistence/recurrence of pain symptoms even after successful excision or ablation of visible peritoneal lesions by laparoscopy. Further studies in human or in primates are necessary to establish this possibility. From the logical point of view, it is difficult to trace these growing lesions on the peritoneal surface by repeated surgical procedures in human.

3.2.3 Origin of IME Lesions

The most alarming questions may arise now, “how can we decide the origin of IME lesions?” Or “is Sampson’s theory enough to explain IME lesions?” There is no definite answer at this moment. But we argue that we can link each and every theory supporting the origin of visible endometriosis [28] to the pathogenesis of IME lesions. If Sampson’s theory does not directly support the origin of IME lesions, it can be indirectly explained by lymphatic or hematogenous spread of menstrual debris and subsequent localization deep into peritoneum. We cannot exclude the possibility of genetic factor or metaplastic transformation of peritoneal mesothelial cells (metaplasia theory) in response to estrogen, inflammation, or environmental factors. Despite possible origin of IME as a result of epithelial-mesenchymal transition/mesenchymal-epithelial transition or from stem cells [2930], activation of cells derived from coelomic epithelium (mülleriosis, induction theory) within peritoneum may be another possible mechanism to explain the origin of IME [31]. Some hypothetical proposals that might be linked to the possible origin of IME lesions are shown in Fig. 3.4.

A313895_1_En_3_Fig4_HTML.gif

Fig. 3.4

Some hypothetical proposals that may be linked to the possible origin of invisible (occult) microscopic endometriosis (IME) lesions. After initial attachment of menstrual debris to the peritoneum, a sequence of epithelial-mesenchymal transition (EMT) or mesenchymal-epithelial transition (MET) may be involved in the generation of IME. Hematogenous or lymphogenous dissemination of shedding endometrial cells during menstruation could be another mechanism. In addition to possible origin from progenitor cells/stem cells, cellular change of flat mesothelium to cuboidal or columnar cells (metaplasia theory) or activated cells derived from residing coelomic epithelium within peritoneum (mülleriosis, induction theory) in response to pelvic inflammation could be linked to the origin of IME. Cells or tissues derived from müllerian rests in association with peritoneal pockets can be another explanation for the development of IME

3.3 Summary and Perspective

We proposed for the first time a new concept “bacterial contamination hypothesis” in endometriosis. Our results suggest that a substantial amount of endotoxin in peritoneal fluid due to reflux of menstrual blood is involved in pelvic inflammation and may promote TLR4-mediated growth of endometriosis. When there is persistence or recurrence of pain even after successful surgical ablation of visible endometriosis in pelvis, we should keep in mind about its possible link with invisible or occult endometriosis, although emerging evidence supporting this phenomenon is still unknown. In order to avoid confusion in the use of the term “invisible” or “non-visible” microscopic endometriosis among laparoscopists, we can also use the term “occult” microscopic endometriosis (OME) instead of IME to indicate any hidden lesion in visually normal peritoneum (27). Since the growth regulation of endometriosis is difficult to explain uniformly by a single factor, we believe that a mutual orchestration among inflammatory mediator, tissue stress reaction, and estrogen may be involved in the growth, maintenance, or progression of endometriosis once it is generated in the female pelvis. Our ongoing study targeting to find the evidence of a subclinical vaginal infection in women with endometriosis may hold new therapeutic potential in addition to conventional estrogen suppressing agent.

Acknowledgement

I gratefully thank Dr. Michio Kitajima of Nagasaki University Hospital and Dr. Akira Fujishita of Saiseikai Nagasaki Hospital for their kind assistance in sample collection; Prof. Masahiro Nakashima of Atomic Bomb Disease Institute, Nagasaki for his experimental advice; and Prof. Hideaki Masuzaki of Nagasaki University Hospital for reading/advice.

References

1.

Strathy JH, Molgaard CA, Coulman CB. Endometriosis and infertility: a laparoscopic study of endometriosis among fertile and infertile women. Fertil Steril. 1982;38:667–72.PubMed

2.

Sampson J. Peritoneal endometriosis due to menstrual dissemination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol. 1927;14:422–9.

3.

Thomas EJ, Prentice A. The etiology and pathogenesis of endometriosis. Reprod Med Rev. 1992;1:21–36.CrossRef

4.

Nisolle M, Casanas-Rouz F, Donnez J. Immunohistochemical analysis of proliferative activity and steroid receptor expression in peritoneal and ovarian endometriosis. Fertil Steril. 1997;68:912–9.PubMedCrossRef

5.

Meyer R. Ueber den Stand der Frage der Adenomyositis und Adenomyom im Allgemeinen und insbesonderes ueber Adenomyositis seroepithelialis und Adeno-myometritis sarcomatosa. Zbl Gynaekol. 1919;36:745–50.

6.

Oosterlynck DJ, Cornillie FJ, Waer M, Vandeputte M, Koninckx PR. Women with endometriosis show a defect in natural killer activity resulting in a decreased cytotoxicity to autologous endometrium. Fertil Steril. 1991;56:45–51.PubMed

7.

Lessey BA, Damjanovich L, Coutifaris C. Integrin adhesion molecules in the human endometrium. Correlation with the normal and abnormal menstrual cycle. J Clin Invest. 1992;90:188–95.PubMedCentralPubMedCrossRef

8.

van der Linden PJQ, van der Linden EPM, de Goeij AFPM, Ramaekers FC, Dunselman GAJ. Expression of integrins and E-cadherin in cells from menstrual effluent, endometrium, peritoneal fluid, peritoneum, and endometriosis. Fertil Steril. 1994;61:85–90.PubMed

9.

Langendonckt AV, Casanas-Roux F, Dolmans M-M, Donnez J. Potential involvement of hemoglobin and heme in the pathogenesis of peritoneal endometriosis. Fertil Steril. 2002;77:561–70.PubMedCrossRef

10.

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I, Ishimaru T. Higher activity by opaque endometriotic lesions than non-opaque lesions in women with endometriosis. Acta Obstet Gynecol Scand. 2004;83:375–82.PubMedCrossRef

11.

Halme J, Becker S, Haskill S. Altered maturation and function of peritoneal macrophages: possible role in pathogenesis endometriosis. Am J Obstet Gynecol. 1987;156:783–9.PubMedCrossRef

12.

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I, Ishimaru T. Differential macrophage infiltration in early and advanced endometriosis and adjacent peritoneum. Fertil Steril. 2004;81:652–61.PubMedCrossRef

13.

Halme J, White C, Kauma S, Estes J, Haskill S. Peritoneal macrophages from patients with endometriosis release growth factor activity in vitro. J Clin Endocrinol Metab. 1988;66:1044–9.PubMedCrossRef

14.

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Kohno T, Sekine I, Matsuyama T, Ishimaru T. Regulation of hepatocyte growth factor by basal and stimulated-macrophages in women with endometriosis. Hum Reprod. 2005;20:49–60.PubMedCrossRef

15.

Khan KN, Kitajima M, Hiraki H, Fujishita A, Sekine I, Ishimaru T, Masuzaki H. Toll-like receptors in innate immunity: role of bacterial endotoxin and toll-like receptor 4 (TLR4) in endometrium and endometriosis. Gynecol Obstet Invest. 2009;68:40–52.PubMedCrossRef

16.

Khan KN, Kitajima M, Hiraki K, Yamaguchi N, Katamine S, Matsuyama T, Fujishita A, Nakashima M, Ishimaru T, Masuzaki H. Escherichia coli contamination of menstrual blood and effect of bacterial endotoxin on endometriosis. Fertil Steril. 2010;94(7):2860–3.e3.PubMedCrossRef

17.

Asea A, Rehli M, Kabingu E, Boch JA, Bare O, Auron PE, Stevenson MA, Calderwood SK. Novel signal transduction pathway utilized by extracellular HSP70: role of toll-like receptor (TLR) 2 and TLR4. J Biol Chem. 2002;277:15028–34.PubMedCrossRef

18.

Khan KN, Kitajima M, Imamura T, Hiraki K, Fujishita A, Sekine I, Ishimaru T, Masuzaki H. Toll-like receptor 4 (TLR4)-mediated growth of endometriosis by human heat shock protein 70 (Hsp70). Hum Reprod. 2008;23:2210–9.PubMedCrossRef

19.

Khan KN, Kitajima M, Inoue T, Tateishi S, Fujishita A, Nakashima M, Masuzaki H. Additive effects of inflammation and stress reaction on Toll-like receptor 4-mediated growth of endometrioid stromal cells. Hum Reprod. 2013;28:2794–803.PubMedCrossRef

20.

Kajihara H, Yamada Y, Kanayama S, Furukawa N, Noguchi T, Haruta S, Yoshida S, Sado T, Oi H, Kobayashi H. New insights into the pathophysiology of endometriosis: from chronic inflammation to danger signals. Gynecol Endocrinol. 2011;27(2):73–9.PubMedCrossRef

21.

Khan KN, Kitajima M, Hiraki H, Fujishita A, Sekine I, Ishimaru T, Masuzaki H. Immunopathogenesis of pelvic endometriosis: role of hepatocyte growth factor, macrophages and ovarian steroids. Am J Reprod Immunol. 2008;60:383–404.PubMedCrossRef

22.

Khan KN, Masuzaki H, Fujishita A, Kitajima M, Sekine I, Matsuyama T, Ishimaru T. Estrogen and progesterone receptor expression in macrophages and regulation of hepatocyte growth factor by ovarian steroids in women with endometriosis. Hum Reprod. 2005;20:2004–13.PubMedCrossRef

23.

Murphy AA, Green WR, Bobbie D, dela Cruz ZC, Rock JA. Unsuspected endometriosis documented by scanning electron microscopy in visually normal peritoneum. Fertil Steril. 1986;46:522–4.PubMed

24.

Nisollle M, Paindaveine B, Bourdone A, Berliere M, Casanas-Rous F, Donnez J. Histologic study of peritoneal endometriosis in infertile women. Fertil Steril. 1990;53:984–8.

25.

Redwine DB. ‘Invisible’ microscopic endometriosis: a review. Gynecol Obstet Invest. 2003;55:63–7.PubMedCrossRef

26.

Donnez J, Langendonckt AV. Typical and subtle atypical presentations of endometriosis. Curr Opin Obstet Gynecol. 2004;16:431–7.PubMedCrossRef

27.

Khan KN, Fujishita A, Kitajima M, Hiraki K, Nakashima M, Masuzaki H. Occult microscopic endometriosis: an undetectable finding by laparoscopy in normal peritoneum. Hum Reprod. 2014;29(3):462–72.PubMedCrossRef

28.

Burney RO, Giudice LC. Pathogenesis and pathophysiology of endometriosis. Fertil Steril. 2012;98(3):511–9.PubMedCrossRef

29.

Gaetje R, Kotzian S, Herrmann G, Baumann R, Starzinski-Powitz A. Nonmalignant epithelial cells, potentially invasive in human endometriosis, lack the tumor suppressor molecule E-cadherin. Am J Pathol. 1997;150:461–7.PubMedCentralPubMed

30.

Sasson IE, Taylor HS. Stem cells and the pathogenesis of endometriosis. Ann N Y Acad Sci. 2008;1127:106–15.PubMedCentralPubMedCrossRef

31.

Redwine DB. Mülleriosis: the single best fit model of origin of endometriosis. J Reprod Med. 1988;33:915–20.