Three phases have been described in the human implantation process: apposition, adhesion (attachment), and invasion (penetration). The blastocyst communicates with the endometrium in the apposition phase. During the adhesion and invasion phases, the blastocyst attaches itself to the epithelium, orients itself 3-dimensionally and finally invades the underlying endometrial surface establishing a vascular relationship.
After 17β-estradiol priming for endometrial development and subsequent progesterone exposure, the endometrium becomes receptive for only a limited period of time. This window appears to be between days 19 and 21 of an idealized 28 day cycle (luteinizing hormone (LH) + 5 and 7). The dynamic changes in the endometrium described below enables embryo implantation .
Structural and molecular changes occur in the endometrium in the secretory phase during the WOI. Endocrine, paracrine, and autocrine factors are involved between the maternal tissue and the implanting blastocyst in a “cross-talk” that performs with limited efficiency in humans. Implantation requires the synchronous development of a competent blastocyst and an endometrium able to respond to the signals from the blastocyst [5]. The endometrium, with its orchestrated series of changes in preparation for implantation, will be destroyed leading to menstruation if the blastocyst fails to implant.
Endometrial receptivity is compounded of morphological features, molecular basis, and genetic evidence. Many approaches to assessing endometrial maturation and receptivity have been described and are outlined below along with their limitations [6, 7].
Morphology
Ultrasound
In contrast with invasive techniques, transvaginal ultrasound examination is a simple method to measure the structure of the endometrium. Three key aspects can be assessed with ultrasound: structure, contractility, and uterine perfusion.
Structure
The endometrium starts to thicken from a single thin line after menstruation under an estrogenic influence. It will achieve a hypoechoic, trilaminar appearance at the end of the follicular phase. After ovulation the change to a secretory state starts by increasing its echogenicity beginning at the periphery and progressing toward the midline [8]. More structural changes during the luteal phase will continue to transform the endometrium to a homogenous and hyperechoic pattern relative to the myometrium. There is only a slight increase in endometrial thickness during the transition from the mid-follicular to the secretory phase. Data from stimulated and IVF cycles regarding the thickness of the endometrium are conflicting. There seems to be a minimal thickness of about 5 mm below which implantation rarely occurs. Although, data from another study [9] show relatively high success rates with very poor endometrial development, the number of patients was relatively low. Although, studies on endometrial volume measurements with three-dimensional ultrasound show contradictory results, most of them conclude that endometrial volume does not predict endometrial receptivity [10].
The endometrial pattern at the time of human chorionic gonadotropin triggering has shown better correlation to implantation than endometrial volume [11]. The highest pregnancy rates were achieved in patients who had a trilaminar appearing endometrium when compared with the homogenous pattern. Unfortunately, most of this data on endometrial appearance derives from ovulation induction with either oral agents or injectable gonadotropins, and extrapolating this data with multifollicular development and supraphysiologic hormone levels may not be valid compared to the natural cycle. The change to a secretory appearance may be a reflection of a rise in progesterone (perhaps premature) rather than an intrinsic defect of the endometrium.
Contractility
The study of uterine myometrial contractions has shown that there is a reduction in uterine contraction frequency in the mid-luteal phase. Increased wavelike activity may act against implantation in natural cycles [12]. Higher frequency and amplitude contractions lowered implantation rates of transferred day three embryos in an ART study. Nevertheless, it is difficult to extrapolate the results of this study to patients with unexplained infertility.
Uterine Perfusion
An adequate uterine and endometrial blood flow has long been considered a marker of endometrial receptivity . Few studies have been done with conflicting results. One study showed a reduction in endometrial vascular perfusion (vascularization index (VI)) that represents the relative proportion of power Doppler data within the defined volume and flow index (FI) that is the mean signal intensity of this power Doppler information) in patients with unexplained infertility irrespective of serum hormone concentrations [13].
Histology
The first criteria for endometrial “dating” were established more than 60 years ago by Noyes et al. and are still in use today [14]. There are, however, significant limitations of this test described below. Other endometrial tests to evaluate the chance for successful implantation are now available [15].
The morphological description of specific histologic features in the endometrium throughout the menstrual cycle and that change following progesterone exposure has been considered the gold standard test for diagnosis of luteal phase defect. These criteria are still in use for endometrial dating, nevertheless their accuracy as a predictor of endometrial receptivity has been questioned. Balasch et al. evaluated more than 1000 endometrial biopsies showing that luteal phase assessment by histological dating was not related to outcome (pregnancy) in infertile patients [16].
The interobserver variation and difficulties in interpretation of the described histologic changes, much like the description in the original paper, limit its use in the clinical setting [17, 18]. Murray included in their study normal cycling and proven fertile women in contrast to the population evaluated in the study by Noyes et al. Relevant findings in this study included: (a) abnormal and delayed endometrium is common even in fertile women; (b) between-cycle variations in the histology of the secretory endometrium are as common in fertile women as in infertile women; and (c) interobserver variability among different pathologists is common, which may change the diagnosis and subsequently treatment. Although the methodology employed in this study was more rigorous than what is used in clinical practice, they concluded that traditional histologic dating is not a valid tool to diagnose luteal phase defect or to guide treatment for patients with reproductive failure.
Similar conclusions were achieved by Coutifaris et al. [19] in 2004 after the evaluation of 619 biopsies in volunteers in a multicenter, randomized prospective study. Histologic dating was not able to discriminate between a fertile or infertile status, and therefore they postulate that timed endometrial biopsy followed by dating of the endometrium does not provide clinically useful information as a screening test.
Unfortunately, in the endometrial evaluation of unexplained infertility, endometrial histology is not a reliable diagnostic test and is therefore not included in most clinics first line investigations in couples with unexplained infertility.
Pinopodes
Pinopodes (pinopods or uterodomes) identified on scanning electron microscopy (SEM) have been suggested to be ultrastructural markers of endometrial receptivity . They were first described in the rat in 1973 [20], and soon proposed as reliable biomarkers of the human WOI.
Pinopodes are smooth or balloon-like structures that arise from the apical surface of the luminal epithelium of the endometrium in response to progesterone around the time that implantation would be occurring. There is no consensus in the literature as to the timing, function, and clinical value of pinopodes . The quantification of pinopodes is also subjective, and the absence may be interpreted differently as they may have disappeared or not even yet appeared. Due to the inconsistency of pinopode expression during the WOI as well as the requirement of SEM to evaluate for the presence or absence of pinopodes, this does not appear to be useful in the human as a reliable and consistent marker of endometrial receptivity [21] .
Biomarkers
A number of factors produced by the endometrium during the so-called window of implantation (WOI) have been considered molecular biomarkers of endometrial receptivity.
Cytokines
Many cytokines participate in implantation [22]. Interleukin (IL)-1, IL-11, IL-6, IL-10, IL-15, IL-18, leukemia inhibitory factor (LIF) , colony-stimulating factor (CSF), tumour necrosis factor (TNF), and transforming growth factor-beta (TGF-beta) have all been shown to play essential roles in human implantation.
Interleukin-1
IL-1 is a proinflamatory cytokine located in multiple tissues. It is present in the human endometrium and in the maternal-trophoblast interface during implantation. The importance of this molecule was shown by the use of IL-1 receptor antagonist significantly reducing the number of implanted embryos in mice [23].
Interleukin-6
IL-6 is expressed during the proliferative phase and reaches its peak during the mid-secretory phase. The activity of IL-6 is mediated by a high affinity receptor complex with two membrane proteins (IL-6R and gp130). The IL-6-R is predominantly localized in glandular epithelium, and to a lesser extent in the stroma, throughout the menstrual cycle [24]. IL-6 secretion has been measured in endometrial biopsies comparing fertile and infertile women in day LH + 6 and LH + 13 finding no difference. However, the level of secretion varied enormously from patient to patient. Nevertheless significantly lower levels of sgp130 were secreted by endometrial biopsies taken between days LH + 6 and LH + 13 in the infertile patient group compared with fertile controls [25].
Endometrial biopsies in the mid-secretory phase from women with proven fertility and women with recurrent miscarriage were compared in another study. It showed a reduced IL-6 mRNA and IL-1 mRNA expression in women with recurrent miscarriage [26].
Cytokine profiling of endometrial secretions may offer a novel approach in the study of the endometrial factor in human implantation [27]. This study showed that a profile of mediators (17 soluble cytokines, chemokines, and growth factors) involved in implantation and endometrial maturation could be quantified in endometrial secretions aspirated with the embryo transfer catheter prior to an embryo transfer. It was also confirmed that this technique was safe and can be used in the clinical setting (sufficient material for analysis in 99.5 % of cases).
Leukemia Inhibitory Factor
LIF is a multifunctional cytokine of the IL-6 family. LIF expression is low in the endometrium during the follicular phase. It rises after ovulation and reaches its maximal expression during the mid-late secretory phase. LIF mRNA is expressed on days 18–28 (LH + 4 to LH + 14) of the menstrual cycle in the endometrium of fertile women [28]. The blastocyst also expresses LIF receptor, which highlights the essential crosstalk between both agents in the implantation process. LIF acts through specific receptors on the cell surface sharing gp130 subunit as a common accessory signal transduction molecule [29]. Measurement of LIF secretion in uterine flushing samples is a noninvasive technique that has been used to determine LIF concentration during the late luteal phase in different studies. Mutations in the LIF gene have also been described in nulligravid infertile women, and may be related with transcription abnormalities and decreased LIF expression [28] .
In women with unexplained infertility several studies have shown lower LIF concentrations compared to fertile women in uterine flushings, and also in endometrial explants, especially during the implantation window [30–32].
A recent study has been designed to investigate the expression of LIF and its receptor subunit gp130 in endometrium of infertile women in uterine flushing during the implantation window in patients with primary unexplained infertility [33]. LIF mRNA was expressed in the endometrium of all normal fertile women but was significantly decreased in infertile women. LIF was not detectable in 88 % of samples collected from infertile women. Gp130 mRNA was hardly detectable in both fertile and infertile women with no difference between them. Infertile women secreted significantly less LIF and gp130 molecules in the uterine flushing compared with normal fertile women. They concluded that the measurement of secreted LIF and gp130 molecules in uterine flushing could be another useful technique for predicting successful implantation.
Unfortunately, what appears to be a breakthrough in the laboratory does not always translate into clinical benefit. In a randomized controlled trial, the use of this molecular biomarker and supplementing women with rLIF in patients with unexplained implantation failure did not improve implantation or pregnancy rates compared with placebo [34]. Measurement of LIF concentrations in endometrial secretions and addition of LIF to culture media or into the endometrium at the time of embryo transfer are still in need of further investigation.
Cellular Adhesion Molecules Family
Integrins
Integrins are one of the families of the cell adhesion molecules (CAM). Integrin β-3 expression is up-regulated during the WOI , while α-4 integrin is down-regulated in the same period of time. Integrin αvβ3 and its ligand osteopontin (OPN) have been extensively studied. Apart from immunohistochemical methods, Horcajadas et al. showed that OPN is up-regulated during the WOI when compared with both the late proliferative phase and the early secretory phase with microarray technology [35]. Abnormal integrin expression was found in women with unexplained infertility [36, 37]. It has also been described an aberrant expression of αvβ3 integrin in patients with endometriosis and absence of expression in patients with IVF failure but adequate embryo quality and/or endometriosis. The hypothesis that this marker would return after a 3-month course of GnRH agonist, and that this would predict which patients with endometriosis would benefit from the treatment before an IVF cycle could not be proven [38].
Selectins
Selectins are a group of CAM’s that includes P-selectin, L-selectin, and E-selectin. Selectins seem to take part in the very early stages of blastocyst interaction with the maternal endometrial epithelium. L-selectine consists of a large, highly glycosylated extracellular domain, a single spanning trans-membrane domain, and a small cytoplasmic tail. It is expressed on vessel walls capturing leukocytes and binding them after activation at the site where they are needed [39]. A similar hypothesis can be made between leukocyte’s “rolling” phenomenon and the blastocyst apposition to the endometrial epithelium [40]. Selecting oligosaccharide-based ligand expression is also up-regulated during the WOI [41] . Immunohistochemical techniques of uterine epithelium for the expression of selectin ligands (MECA-79 and HECA-452) were compared in fertile women and in patients with unexplained infertility [42]. During the secretory phase in natural cycles, MECA-79 was more strongly expressed in fertile women compared to infertile patients. It was also shown that the expression of GlcNAc6ST-2 is decreased in infertile patients when compared with fertile women, which correlates with the decrease in MECA-79 expression.
The expression of L-Selectin was also compared in a pilot study in 20 patients with recurrent implantation failure and 20 fertile women [43]. In the RIF patients, those with negative result for the MECA-79 tests did not become pregnant, postulating that screening for the absence of this ligand may identify a poor prognosis group of patients.
Growth Factors
In the vast families of growth factors, the members of the epidermal growth factor (EGF) family, and the insulin-like growth factors (IGF) play decisive roles in the implantation process. Heparin-binding epidermal growth factor (HB-EGF) plays a role in implantation and embryonic development reaching its maximal expression during the period of endometrial receptivity [44, 45]. Following the hypothesis that women with unexplained infertility may show deficiencies in a series of biochemical markers reflecting impaired endometrial development, Aghajova studied HB-EGF endometrial expression. They compared the endometrium from women with unexplained infertility with endometrium from women with male factor infertility or healthy fertile controls. HB-EGF expression was lower in women with unexplained infertility and the authors postulated that abnormal expression of this member of the EGF family may contribute to infertility in some patients with unexplained infertility [46].
Proprotein Convertase 5/6 (PC6)
PC6 is a serine protease, which has an essential role as a regulator for implantation [47]. PC6 achieves its maximum expression during the WOI.
Uterine lavages were obtained in patients ( n = 103) diagnosed with unexplained infertility and compared to fertile controls in a recent study. In a subgroup of unexplained infertility patients, the PC6 activity was significantly lower than the activity in another group of unexplained infertility patients and highly significantly lower than fertile women. Unfortunately, there is not a description of the two subgroups of patients with unexplained infertility with such a different PC6 activity. The assessment of PC6 with in uterine fluid may lead to the establishment a less invasive marker for the evaluation of endometrial receptivity [48], although no treatment exists for those with reduced levels.
Molecular Approaches
New molecular technologies are being used to identify biomarkers of the WOI and thus to assess endometrial receptivity . During the last decade global gene expression analysis has tried to identify genes associated with human endometrial receptivity [49–59]. Many different genes have been evaluated and only a small fraction of them have been selected for their potential role in the endometrial receptivity . Selecting specific sets of genes important during the WOI is made more difficult due to discrepancies among studies in design, data analysis, and microarray platforms. Altmäe et al. published their experience in 2010 comparing endometrial biopsy samples from healthy women with proven fertility and patients with unexplained infertility demonstrating that the endometrial gene expression pattern at the time of implantation is different between these two populations of patients [60]. The endometrial samples were obtained on cycle day LH + 7. Microarray analysis was performed using the Whole Human Genome Oligo Microarray (Agilent Technologies) and gene expression profiles were compared. A total of 260 differentially expressed genes were identified, 145 were significantly up-regulated and 115 down-regulated in the endometria of infertile patients compared with fertile controls. Among other findings, there was a significant dysregulation in the leukocyte extravasation-signaling pathway in infertile women, in concordance with previous studies [40].