Classically, anatomical distortion and structural changes are thought to be a factor causing subfertility in women with revised American Society of Reproductive Medicine (r-ASRM) stage III and IV endometriosis with adhesions involving the fallopian tubes, the ovaries or both, which may be dense or filmy. Infertile women with stage I and II disease have been shown to have an increased likelihood of fimbrial pathology compared to normally fertile women with the same stage disease, which may contribute to difficulties in conceiving in this group with less severe disease [10]. Descriptions of the types of structural changes involving the fallopian tubes include fimbrial clubbing, blunting, agglutination or phimosis, tortuosity of the tubes, complete encasement by dense adhesions, and endometriotic nodules of the mesosalpinx which disrupt the normal tubular anatomy. Hydro- or haematosalpinges may develop due to tubal blockage which may contribute to infertility in a similar manner to those following pelvic inflammatory disease, with distortion of cilia and retrograde flow of inflammatory fluid into the uterine cavity.

In addition to external anatomical factors contributing to infertility, in vitro studies of fallopian tubes exposed to peritoneal fluid from women with endometriosis have demonstrated reduced ciliary beat frequency probably resulting in impairment of tubal transport [11].

Fertilisation of the oocyte occurs within the ampulla or distal end of the fallopian tube. This area is in close contact with the peritoneal microenvironment, and therefore changes within the peritoneal fluid can have follow on effects for conception.

There is evidence that the peritoneal microenvironment is altered in women with endometriosis. In particular a localised inflammatory state is present with increased macrophage activity implicated in the secretion of cytokines such as tumour necrosis factor α (TNF-α), interleukin-1β, interleukin-8 and interleukin-6 (IL-6) [12, 13]. Secretion of oestradiol and progesterone by endometriotic implants has been shown to activate vascular endothelial growth factor (VEGF) promoting neovascularisation and perpetuation of the disease [14].

Most evidence (but not all [15]) points to increased macrophage activity within the peritoneal fluid of women with endometriosis which may be responsible for increased sperm immobilisation and phagocytosis [13]. Oxidative stress and apoptosis are thought to be the mechanisms by which TNF-α contributes to the damage of sperm DNA. In addition, sperm binding is reduced in general by many of the abovementioned cytokines, but also within the isthmus of the fallopian tubes resulting in fewer sperm pooling in this area, further impairing fertility [16].

As well as the changes seen in the peritoneal microenvironment, the follicular environment also appears to be abnormal in women with endometriosis. Inflammatory markers such as TNF-α and IL-6 are present in higher quantities within the granulosa cells and there is a higher rate of apoptosis.

Endometriomas also have a direct effect on ovarian tissue and function, with reduced capacity for oocyte production and a reduction in ovulation in affected ovaries. The natural decline in available follicles which is seen with increasing age appears to take place earlier for women with endometriosis. Anti-mullerian hormone (AMH) is a marker of ovarian reserve which remains stable throughout the menstrual cycle. Women with endometriosis have been shown to have lower AMH concentration than age-matched controls, and AMH seems to be further depleted by the presence of ovarian endometriomas and ovarian cystectomy [17]. Although follicle numbers are reduced, the fall in follicle numbers seen in women with endometriosis do not have the corresponding reduction in oocyte quality that is seen in women with advancing age as the cause of the follicle depletion. As a result, assisted reproductive technology (ART) data for women with endometriosis show a poor response to controlled ovarian stimulation but there is debate as to whether pregnancy rates are similarly reduced.

Emerging data show that women with endometriosis may have a reduction in eutopic endometrial receptivity, which may impair implantation. Systemically, estrogen and progesterone hormones are largely normal in women with endometriosis. Local hormonal factors within the endometrial microenvironment of women with endometriosis include increased production of oestradiol related to inflammation and reduced sensitivity to progesterone both of which may contribute to a reduction in endometrial receptivity within the uterus.

Integrins, specifically α_Vβ_3, on the surface of the cells of the endometrium mediate intracellular signals to promote receptivity. These integrins are in turn stimulated by Homeobox A 10 (HOXA10), a gene expressed in normal endometrium. Women with endometriosis have reduced expression and methylation of this gene and may contribute to deficiencies in α_Vβ₃ and a resultant reduction in endometrial receptivity.

It is unclear why some women have a successful pregnancy with moderate to severe endometriosis and others do not. Potentially, the presence of adenomyosis may have a role, with evidence showing reduced fertility in baboons with adenomyosis [38]. Human studies point to a similar association between adenomyosis and subfertility [39] with an effect on uterine peristalsis and sperm transport [40] as well as a higher rate of implantation failure with ART [41]. Adenomyosis and endometriosis have a clinical correlation with the more advanced stages of endometriosis [42]. These women may have higher success rates with ART [29, 43]; however, the data are limited due to the difficulty of diagnosing adenomyosis.