Objective
The purpose of this study was to assess the receptivity of the homogeneous endometrium in the late follicular phase in infertile women with natural cycles.
Study Design
Twenty-eight infertile women with ultrasonographically homogeneous (group 1) or trilaminar (group 2) endometria in the late follicular phase underwent endometrial biopsies. Some molecular markers and development of pinopodes were evaluated.
Results
In the late follicular phase, the mean level of vascular endothelial growth factor was significantly lower in group 1 than in group 2 (0.96 ± 0.37 marks vs 1.39 ± 0.46 marks; P = .010). In the mid luteal phase, a decreased leukemia inhibitory factor and integrin alpha v beta 3 levels were found in group 1 (1.58 ± 0.99 marks vs 2.59 ± 0.61 marks; 1.85 ± 0.72 marks vs 2.60 ± 0.73 marks; 1.92 ± 0.91 marks vs 2.83 ± 0.57 marks; P = .003; P = .011; P = .004). The rate of fully developed pinopodes in the mid luteal phase was significantly decreased in group 1 ( P = .018).
Conclusion
An ultrasonographically homogeneous endometrium in the late follicular phase was associated with poor receptivity in infertile women with natural cycles.
Studying serial sonography in in vitro fertilization-embryo transfer cycles, 1 study reported that, in the late proliferative phase, there was a trilaminar pattern where the endometrium appears multilayered with hyperechogenic outer walls and a well-defined central echogenic line. Smith et al also described the next most common pattern (the isoechogenic) with a poorly defined central echogenic line and the least common pattern (the homogeneous hyperechogenic [HH]) with no visualization of a central echogenic line.
The first study to demonstrate poor pregnancy rates with the HH pattern in the late proliferative phase after in vitro fertilization-embryo transfer was by Gonen and Casper in 1990. Because this study involved the use of clomiphene citrate in the controlled ovarian hyperstimulation protocol, it was not clear whether this association was with only clomiphene citrate cycles. However, subsequent studies that evaluated pregnancy rates after in vitro fertilization–embryo transfer with the HH pattern in the late proliferative phase with the use of gonadotropins during controlled ovarian hyperstimulation reported a similar problem. Thereafter, several researchers of in vitro fertilization–embryo transfer also showed a significant higher pregnancy rate in women with trilaminar endometrium than women with HH endometrium in the late proliferative phase.
The regulatory mechanism that is involved in the development of HH endometrium in the late proliferative phase is still unclear. The changes of gene expression in the endometrium from the early to mid luteal transition, differences in molecular phenotyping, and biologic processes were the possible factors thought to be responsible for the formation of a homogeneous endometrium. Although endometrial growth and the transition from phase to phase were ultimately under the control of the sex steroids, these hormones that included estrogen and progesterone were thought to exert their effects indirectly through some biochemical factors. Vascular endothelial growth factor (VEGF), leukemia inhibitory factor (LIF), matrix metalloproteinase–9, ‡ and integrin alpha v beta 3, were considered to be several important markers that are associated with endometrial receptivity and pregnancy rate. The pinopode formation on the endometrial surface was considered to be an important morphologic occurrence that was correlated to the implantation of the embryo. The objective of the present study was therefore to assess various molecular markers in the late follicular phase and mid secretory phase of infertile women with natural cycles to determine whether either an increase or decrease of these factors may be associated with HH endometrial pattern in the late follicular phase that has been associated with low pregnancy rates.
Materials and Methods
Subjects
Twenty-eight apparently healthy women, 14 of whom had an ultrasonographic HH echo (group 1; HH group) and 14 of whom had “ triple-line” (group 2; trilaminar group) endometrium in late follicular phase, were recruited from 2638 infertile women with >8900 ovulation-monitoring cycles in our Department between September 2010 and November 2011. The natural cycle was defined as a spontaneous ovulation or menstrual cycle without any follicular-stimulating drugs. The inclusion criteria for both groups were ≤40 years old, regular menstrual cycles (25-35 days), and no use of intrauterine devices or oral contraceptives for at least 3 months before the study. Exclusion criteria were women with polycystic ovarian syndrome, endometriosis, premature ovarian failure, and abortion within 1 year or a history of pelvic inflammatory disease. All subjects gave verbal informed consent. The study was approved by our Institutional Ethics Committee.
Ovulation monitoring
The women were followed for 1 cycle; blood and urine samples were obtained as detailed later. Ovulation was monitored with vaginal ultrasonography (Aloka-1000, UST-985 transvaginal probe; Aloka Co Ltd, Tokyo, Japan). HH or trilaminar endometrial pattern was assessed according to the classification proposed by Smith et al or Gonen and Caspar as previously mentioned. An isoechogenic pattern with a poorly defined central echogenic line was the incomplete or mixed pattern that was excluded from this study. The day of the luteinizing hormone (LH) surge was determined in urine samples when the LH surge reached its highest value. The peripheral blood samples were obtained by vein puncture on the day of the LH surge and 7 days after ovulation. The serum was stored at –20°C until the concentrations of LH, follicle-stimulating hormone, estradiol, and progesterone were assayed.
Endometrial biopsy
Two endometrial biopsies were performed during a single menstrual cycle for each subject. The first biopsy was performed on the preovulatory day; the second biopsy was performed on ovulation day + 7. The endometrial sample was collected from the uterine cavity with an aspiration biopsy device (#4164 Probet; Gynetics, Lommel, Belgium). Each sample was divided into 3 pieces and immediately fixed. Samples for scanning electron microscopy were fixed in a solution that contained 2.5% (weight/volume) glutaraldehyde, 0.5% paraformaldehyde, 0.1 mol/L sucrose, 0.1 mol/L sodium cacodylate, and 3 mmol/L calcium chloride (pH, 7.4). Samples for immunohistochemistry and histologic dating were fixed in 4% formalin for a maximum of 24 hours and dehydrated in 75%, 85%, 95%, and 100% ethanol and finally embedded. The histologic dating was evaluated according to the criteria of Noyes et al. An out-of-phase biopsy was defined as a ≥3-day lag between the chronologic and the histologic day; an in-phase biopsy was defined as a ≤2 day lag.
Scanning electron microscopy
Samples were washed twice in a buffer that contained 0.15 mol/L sodium cacodylate and 3 mmol/L calcium chloride (pH, 7.4) and once in distilled water. The specimens were dehydrated first in increasing concentrations of ethanol (70%, 95%, and 99.5%) and then in acetone; they were dried in a critical-point dryer with the use of carbon dioxide. The specimens were mounted, coated, and examined with a scanning electron microscope (S-3000N; Hitachi, Tokyo, Japan). The morphologic evaluation of pinopodes was defined as no pinopode, developing pinopode, fully developed pinopode, and regressing pinopode ( Figure 1 ). Five fields were sampled and analyzed for the development and distribution of pinopodes. The pinopode density was evaluated in the following manner: no pinopode was found; pinopode area was <20% of field; pinopode area was 20-50% of field; and pinopode area was >50% of field.
Immunohistochemistry
Fifty-six biopsy samples were analyzed. For LIF, integrin alpha v, VEGF, and matrix metalloproteinase-9 studies, 4-μm paraffin-embedded sections were deparaffinized for 20 minutes in acetone at 70°C. Endogenous peroxidase was blocked with 3% H 2 O 2 for 10 minutes. Antigen retrieval was done by boiling for 5 minutes in 0.01 mol/L citrate buffer at pH 6.0. Then the sections were incubated overnight at 4°C with the specific antibodies that were diluted at 1:50. The antibodies that were used in this study were polyclonal antibodies LIF, integrin alpha v, VEGF, and matrix metalloproteinase-9 (BA 1239, BA0957, BA0407, and BA0573; Boster Co Ltd, Wuhan, China). The sections were rinsed in phosphate-buffered saline solution, blocked with 10% normal goat serum for 30 minutes and then incubated with goat anti-rabbit biotinylated immunoglobulin (SV0002; Booster Co Ltd) at 37°C for 30 minutes. For integrin beta 3 study, the sections were incubated with polyclonal antibodies integrin beta 3 (sc-6626; Santa Cruz, CA) diluted at 1:100 overnight at 4°C. Polymer Helper (PV-9003-1; Zhong-Shang Co Ltd, Beijing, China) was added to the section at room temperature; at 20-minute intervals, polyperoxidase anti-goat immunoglobulin (PV-9003-2) was added. The slides were then washed in phosphate-buffered saline solution and colored by 3,3-diaminobenzidine in H 2 O 2 (DAB; Invitrogen, Carlsbad, CA). Staining was evaluated semiquantitatively by the following formula: H-score = ΣPi(i+1), in which i is the intensity of staining with a value of 1 (weak), 2 (moderate), or 3 (strong), and Pi is the percentage of stained cells for each intensity, which varied from 0-100%.
Statistical analysis
The Statistics Package for Social Science (version 13.0; SPSS Inc, Chicago, IL) was used for statistical analyses. The mean (± SD) was evaluated with the Student t test, and qualitative data was analyzed with the χ 2 test. Statistical significance was set at .05 (2-tailed).
Results
Characteristics of participants and hormone levels
The mean age of the women were 31.9 ± 3.7 years, and the mean body mass index was 21.0 ± 2.5 kg/m 2 . Infertility duration ranged from 2-11 years. There were no significant differences in the age, duration of infertility, body mass index, dimensions of ovary, and hormone levels between groups 1 and 2. The mean menstrual cycle was longer in group 1 (31.3 ± 3.4 days) than in group 2 (27.9 ± 2.6 days; P = .006). Participant characteristics are given in Table 1 .
| Characteristic | Group 1 (n = 14) | Group 2 (n = 14) |
|---|---|---|
| Age, y a | 31.8 ± 3.3 | 31.9 ± 4.3 |
| Body mass index, kg/m 2 a | 21.4 ± 2.7 | 20.5 ± 2.1 |
| Duration of infertility, y a | 4.5 ± 2.6 | 4.1 ± 2.5 |
| Menstrual cycle, d a | 31.3 ± 3.4 | 27.9 ± 2.6 b |
| Duration of luteal phase, d a | 13.1 ± 1.1 | 13.2 ± 1.1 |
| Ovarian diameter, mm a | 24.9 ± 2.7 | 23.6 ± 2.4 |
| Unilateral follicle, n a | 6.5 ± 1.8 | 6.5 ± 1.7 |
| Infertility, n (%) | ||
| Primary | 5 (35.71) | 6 (42.86) |
| Secondary | 9 (64.29) | 8 (57.14) |
| Hormone levels a | ||
| Day 3 | ||
| Follicle-stimulating hormone, IU/L | 7.43 ± 1.76 | 7.77 ± 1.52 |
| Luteinizing hormone, IU/L | 4.93 ± 1.79 | 4.68 ± 1.96 |
| Estradiol, pmol/L | 153.41 ± 68.29 | 161.38 ± 64.25 |
| Progesterone, nmol/L | 2.52 ± 1.43 | 2.04 ± 1.26 |
| Prolactin, μg/L | 11.16 ± 3.71 | 14.60 ± 7.43 |
| Testosterone, nmol/L | 1.25 ± 0.46 | 1.22 ± 0.49 |
| Preovulatory day | ||
| Luteinizing hormone, IU/L | 37.69 ± 25.33 | 44.04 ± 25.74 |
| Estradiol, pmol/L | 1234.69 ± 609.63 | 946.86 ± 428.91 |
| Progesterone, nmol/L | 3.92 ± 1.43 | 5.05 ± 2.85 |
| Ovulation + 7 days | ||
| Estradiol, pmol/L | 840.98 ± 202.37 | 762.84 ± 275.39 |
| Progesterone, nmol/L | 58.62 ± 12.79 | 70.42 ± 22.11 |
a Values are given as mean ± SD;
Histologic dating and expressions of molecular markers
In the late follicular phase
In group 1, 6 of 14 preovulatory biopsy specimens (42.9%) showed in-phase endometria, and 8 specimens showed out-of-phase endometria (57.14%). Of the latter, 6 specimens were mid proliferative phase, and 2 specimens were early proliferative phase endometria. In group 2, 9 in-phase endometria (64.3%) and 5 out-of-phase endometria were demonstrated. The mean endometrial thickness was 10.14 ± 1.75 mm in group 1 and 11.29 ± 1.86 mm in group 2 ( P = .106) and 10.60 ± 2.06 mm in those that were in-phase and 10.85 ± 1.68 mm in those that were out-of-phase ( P = .734). The mean expression of VEGF was 0.96 ± 0.37 marks in group 1, which was significantly lower than in group 2 (1.39 ± 0.46 marks; P = .010). In early dated preovulatory biopsy specimens with and without HH, the mean VEGF levels was 0.95 ± 0.44 marks and 1.51 ± 0.41 marks, respectively ( P = .008).
In the mid luteal phase
In group 1, 10 of 14 mid luteal biopsy specimens (71.4%) were out-of-phase, and only 28.6% of women showed in-phase endometria. All out-of-phase endometria showed asynchronous development with some fields in the proliferative phase. In group 2, however, 11 of 14 mid luteal biopsy specimens (78.6%) showed an in-phase situation, and 3 specimens were out-of-phase ( P = .021). The mean endometrial thickness in those with in-phase and out-of-phase endometrial biopsy specimens were 11.13 ± 1.46 mm vs 10.80 ± 2.77 mm ( P = .117). The mean LIF expressions in the endometrial glands were 2.59 ± 0.61 marks in group 2 and 1.58 ± 0.99 marks in group 1 ( P = .003). The mean integrin alpha v beta 3 expressions in the gland were significantly lower (1.85 ± 0.72 marks and 1.92 ± 0.91 marks) in group 1 compared with group 2 (2.60 ± 0.73 marks and 2.83 ± 0.57 marks; P = .011, P = .004; Table 2 and Figure 2 ).
| Variable | Late follicle phase a | Mid luteal phase a | P value 1 b | P value 2 c | P value 3 d | P value 4 e | P value 5 g | P value 6 h | ||
|---|---|---|---|---|---|---|---|---|---|---|
| Group 1 (1) | Group 2 (2) | Group 1 (3) | Group 2 (4) | |||||||
| Gland | ||||||||||
| Leukemia inhibitory factor | 0.92 ± 0.66 | 1.01 ± 0.88 | 1.58 ± 0.99 | 2.59 ± 0.61 | .784 | .118 | .000 | .074 | .000 | .003 |
| Integrin alpha v | 1.62 ± 0.66 | 1.23 ± 0.71 | 1.85 ± 0.72 | 2.60 ± 0.73 | .189 | .043 | .000 | .321 | .001 | .011 |
| Integrin beta 3 | 1.12 ± 0.81 | 0.83 ± 0.73 | 1.92 ± 0.91 | 2.83 ± 0.57 | .329 | .002 | .000 | .015 | .000 | .004 |
| Vascular epithelial growth factor | 0.96 ± 0.37 | 1.39 ± 0.46 | 1.45 ± 0.64 | 1.99 ± 0.92 | .010 | .778 | .051 | .021 | .001 | .087 |
| Matrix metalloproteinase-9 | 2.29 ± 0.49 | 2.64 ± 0.48 | 2.77 ± 0.52 | 3.07 ± 0.60 | .076 | .513 | .026 | .032 | .001 | .170 |
| Stroma | ||||||||||
| Leukemia inhibitory factor | 0.55 ± 0.52 | 0.46 ± 0.63 | 1.06 ± 0.70 | 1.27 ± 0.89 | .691 | .025 | .018 | .060 | .014 | .487 |
| Integrin alpha v | 1.19 ± 0.73 | 1.16 ± 0.70 | 1.36 ± 0.68 | 1.82 ± 0.63 | .934 | .477 | .013 | .527 | .023 | .073 |
| Integrin beta 3 | 0.77 ± 0.62 | 0.67 ± 0.54 | 1.47 ± 0.62 | 1.85 ± 0.59 | .637 | .001 | .000 | .009 | .000 | .113 |
| Vascular epithelial growth factor | 0.90 ± 0.44 | 1.05 ± 0.36 | 1.20 ± 0.45 | 1.12 ± 0.73 | .334 | .362 | .696 | .095 | .343 | .749 |
| Matrix metalloproteinase-9 | 2.17 ± 0.58 | 2.47 ± 0.69 | 1.87 ± 0.63 | 2.33 ± 0.62 | .229 | .023 | .454 | .234 | .506 | .063 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
