JBRA Assist. Reprod. 2009;13(4):20-24
ARTIGO ORIGINAL
doi: 10.5935/1518-0557.2010.13.4.05
Endometrial thickness assessment at embryo transfer in ICSI cycles using GnRH antagonists
Avaliação da espessura endometrial no momento da transferência embrionária em ciclos de ICSI utilizando antagonista de GnRH
G&O Barra - Ginecologia e Obstetrícia da Barra - Rio de Janeiro, RJ, Brasil
Local de realização do trabalho:G&O Barra - Ginecologia e Obstetrícia da Barra - Rio de Janeiro, RJ, Brasil.Não foram recebidos auxílios/financiamentos para execução deste trabalho.Trabalho apresentado no 9º Taller General - Red Latinoamericana de Reproducción Asistida - abril 2009, Cancún, México.
ABSTRACT
Objective: To evaluate the prognostic value of the endometrial thickness assessment on the day of embryo transfer (ET) on pregnancy outcome. The main outcome measures are pregnancy rates and live births.
Methods: Fifty-one patients, ≤37 years of age, undergoing their first cycle of ICSI, were prospectively studied from May 2004 to May 2005. They were prescribed to the same protocol for ovarian stimulation with recombinantfollicular stimulating hormone (r-FSH) and GnRH antagonist. During ET, the endometrial thickness was evaluated by abdominal ultrasonography (US).
Results: There were no statistically significant differences in age, duration of stimulation, total dose of follicular stimulating hormone (FSH), number of metaphase II (MII) oocytes retrieved, and number of good quality embryos transferred among pregnant (39%) and nonpregnant (61%) patients. Also, no statistically significant difference in endometrial thickness on the day of ET was observed. Similar results were observed for the comparison of patients who had a live birth (25%) with the rest (75%) showing no statistically significant difference in age, duration of stimulation, total dose of FSH, number of MII oocytes retrieved, and number of good quality embryos transferred. Furthermore, no statistically significant difference in endometrial thickness on the day of ET was observed.
Conclusions: The prognostic question of uterine receptivity has not yet been solved through the use of ultrasound technology. Our results did not show statistically significant relationship between the endometrial thickness on the day of ET and pregnancy rates or live births in patients using GnRH antagonists.
Keywords: ultrasonography, endometrial receptivity, endometrial thickness, embryo transfer, ICSI cycles, pregnancy rates.
RESUMO
Objetivos:Avaliar o valor prognóstico da aferição da espessura endometrial no dia da transferência embrionária (TE) no resultado gestacional. Os principais desfechos foram taxas de gestação e nascidos vivos.Métodos:Cinqüenta e uma pacientes, com idade ≤37 anos, em seu primeiro ciclo de ICSI, foram prospectivamente estudadas no período de Maio 2004 a Maio 2005. Todas foram submetidas ao mesmo protocolo para estimulação ovariana com FSH recombinante (r-FSH) e antagonista de GnRH. Durante a TE, a espessura endometrial foi avaliada por ultra-sonografia abdominal (US).Resultados:Não houve diferença estatisticamente significativa quanto à idade, duração da estimulação, dose total de FSH, número de oócitos em metáfase II (MII) recuperados e número de embriões de boa qualidade transferidos entre pacientes grávidas (39%) e não-grávidas (61%). Também não foi observada diferença estatisticamente significativa na espessura endometrial no dia da TE. Resultados semelhantes foram encontrados na comparação de pacientes que tiveram um nascido vivo (25%) com as demais (75%), não havendo diferença estatisticamente significativa quanto à idade, duração da estimulação, dose total de FSH, número de oócitos MII recuperados e número de embriões de boa qualidade transferidos. Além disso, não foi observada diferença estatisticamente significativa na espessura endometrial no dia da TE.Conclusões:A questão prognóstica da receptividade uterina ainda não foi totalmente esclarecida através da tecnologia da ultra-sonografia. Nossos resultados não demonstraram relação estatisticamente significativa entre a espessura endometrial no dia da TE e as taxas de gestação ou nascidos vivos em pacientes utilizando antagonistas de GnRH.
Palavras-chave: ultra-sonografia, receptividade endometrial, espessura endometrial, transferência embrionária, ciclos de ICSI, taxas de gestação.
INTRODUCTION
The endometrial ultrasonographic evaluation, both in terms of thickness and its pattern of echogenicity, has been studied for more than 20 years, aiming to correlate it to the endometrial receptivity and pregnancy rates, in ART cycles (Barbieri & Hornstein, 2004; Passos, 2004; Lass, 2003).
Oliveira et al. (1997) evaluated the endometrial thickness and its echographic pattern on the hCG day by transvaginal ultrasonography (TVUS), in 150 IVF patients. They concluded that the endometrial US, in terms of pattern and thickness, do not present prognostic value in IVF cycles on the hCG day. However, it became evident that a minimum thickness should be reached to achieve pregnancy (7.0 mm), and that the endometrial trilaminar pattern favors pregnancy. Serafini et al. (1994) also observed the preovulatory endometrium of trilaminar aspect as a pregnancy predictor.
Khalifa et al. (1992) studied the predictive value of endo- metrial thickness through the patterns on the hCG day. Weissman et al. (1999) tried to determine a maximum value for the endometrial thickness on the day of hCG, after which there would not be implantation.
Amir et al. (2007) evaluated the factors that contribute to endometrial thickness as well as the impact of this thickness on pregnancy rates in IVF/ICSI cycles, through TVUS on the day of hCG administration. Patients who became pregnant had a thicker endometrium, when compared with the rest (10.8±2.4 mm vs. 10.5±2.4 mm). However, this datum was only significant in patients above 35 years of age. The infertility etiology also influenced the endometrial thickness: women with infertility due to male factors had thicker endometrium, when compared with the other infertility causes.
Richter et al. (2007) retrospectively evaluated 1294 IVF cycles with blastocyst transfer. A gradual increase in pregnancy rates was observed as the endometrial thickness increased, suggesting that this may be a good uterine receptivity indicator. All cycles were evaluated under TVUS on the hCG day. The relationship observed was regardless of the patients’ age or quality of the embryo. First, the study aims to correlate the prognostic value of the endometrial thickness assessment on the day of ET with pregnancy outcome, in ICSI cycles that use GnRH antagonists. Second, to analyze the interference of other variables that could classically influence the pregnancy outcome following high complexity assisted reproduction, such as patients’ age, number of MII oocytes retrieved, number and quality of embryos transferred, and type of transfer.
MATERIALS AND METHODS
In this prospective cohort study, 55 patients who started ICSI cycles from May 2004 to May 2005 were included. The inclusion criteria were: age ≤ 37 and first ICSI cycle. The protocol was approved by The Ethical Committee of Centro Médico Barrashopping.
The ovulation induction protocol with recombinant-follicular stimulating hormone (r-FSH) (Gonal F®, Serono) began on the second day of the cycle, with a daily dose of 150-300 IU. On the sixth day of the cycle, there was a TVUS, evaluating the endometrial thickness and aspect, as well as the follicular response. The GnRH antagonist (Cetrorelix Acetate, Cetrotide®, Serono), 0.25 mg daily, subcutaneously, was administered in the presence of at least 1 follicle of ≥ 14 mm. After achieving at least 1 follicle ≥18 mm and 2 follicles ≥16 mm, ovulation was triggered with 250Mg of r-hCG (Ovidrel®, Serono). Oocyte was retrieved 34-36 h after, using a Wallace needle, guided by transvaginal US (GE Logiq 400 Pro Series equipment with an endocavitary transducer E721, in the frequency of 6 MHz).
The semen was collected by masturbation, by percutaneous epididymal aspiration or by testicular biopsy.
The embryos that presented 6-8 cells, of grades 1 or 2, were considered as good quality embryos (morphological criteria) by Redlara (Red Latinoamericana de Reproducción Asistida, 1998).
The transfers took place on day 3, guided by abdominal US, led by the same ultrasonographer. The measurement of endometrial thickness was taken in a longitudinal section of the uterus, positioning the calipers in the interfaces between the endometrium and the myometrium, in the portion of greater anteroposterior distance from the endometrial cavity.
For the embryo transfer, a Wallace catheter was located at a distance of 15-20 mm from the fundus of the endometrial cavity (Coroleu et al., 2002).
The transfers were classified into four categories: very easy (type 1) - when the catheter smoothly passed by the cervix, without any resistance; easy (type 2) - when the use of a stylet was necessary; difficult (type 3) - when the use of a stylet and a Pozzi holder was necessary; very difficult (type 4) - with cervical dilation with hysterometer or metallic cannula.
The supplementation of the luteal phase consisted of the daily use of vaginal progesterone.
The biochemical pregnancy was confirmed by detecting increasing serum β-hCG concentrations 14 days after ET. Statistical analysis: The Student’s t-test was used for comparison of numerical variables and the chi-square (X2)for qualitative data. The difference was considered statistically significant at p-value < 0.05.
RESULTS
The causes for infertility in couples were distributed as: 49% for male factors (25/51), 29% for female causes (15/51), 15.6% for unexplained infertility (8/51), and 6% for mixed causes (3/51).
Table 1 aims to outline a general profile of the 51 patients in the study.
Table 1. Descriptive analysis of the variables in study.
With respect to the type of transfer, the distribution of patients showed 56.9% of type 1 (29/51), 15.7% of type 2 (8/51), 19.6% of type 3 (10/51), and 7.8% of type 4 (4/51).
It was observed that 20 (39.2%) out of the 51 women were tested positive for pregnancy. After characterizing the groups as positive and negative, the next step was to check the significant difference among the patients with respect to mean age, number of stimulus days, total dose of gonadotrophin used, number of aspirated MII oocytes, the measure of the endometrial thickness on the day of ET, and the number of good quality embryos transferred. There was no significant difference in the mean age (p = 0.46), duration of stimulation (p = 0.15), total dose of FSH (p = 0.31), number of MII oocytes retrieved (p = 0.68), number of good quality embryos transferred (p = 0.96), and endometrial thickness on the day of ET (p = 0.55) among the pregnant and non-pregnant patients (Table 2). The types of transfer were grouped into 1 or 2 (n = 37), and 3 or 4 (n = 14), owing to the small number of cases observed in some categories. The subgroup of women under type 1 or 2 proportionally presented a significantly higher pregnancy rate (48.7%) than the subgroup type 3 or 4 (14.3%), with p = 0.025 (Fig. 1).
Table 2. Statistical analysis of variables according to pregnancy per transfer.
Figure 1. Pregnancy according to the type of transfer.
There is no significant relationship between the endometrial thickness on the day of ET and the pregnancy rate. The comparison of patients who had a live birth (13/51=25%) with the rest of the patients (38/51=75%) showed no statistically significant difference in age (31.4p3.9; 32.3p3.2), duration of stimulation (9.7p1.4; 10.7p2.5), total dose of FSH (2076.9p448.3; 2161.1p518.8), number of MII oocytes retrieved (7.6p5.0; 6.8p4.8), and number of good quality embryos transferred (1.7p1.3; 1.5p1.0). Also, no statistically significant difference in endometrial thickness on the day of ET (10.2p1.9 mm; 10.3p1.9 mm) was observed.
A section point was proposed for the endometrial thickness r10 mm, taking into consideration the median of this measure in women who became pregnant. Out of the 51 women studied, 29 (56.9%) presented thickness r10 mm, while 22 (23.1%) presented a thickness <10 mm. There was no significant difference in the proportion of pregnancy between the subgroup of women with endometrial thickness r10 mm (31%) and those with thickness <10 mm (50%) (Fig. 2).
Figure 2. Pregnancy according to endometrial thickness.
Out of the 20 pregnant women, there were 4 biochemical pregnancies and 3 losses in the first trimester (2 anembryonic eggs and 1 molar pregnancy). The take-home baby rate was 25.49% (13 patients had 14 babies, 13 born alive with only 1 gemelar case, besides the loss of a 32-week malformed stillborn).
DISCUSSION
To date, the importance to assess the endometrial thickness around the moment of ET is discriminatory, assuring the presence of a minimum thickness to allow implantation. The assessment of the endometrial pattern, trilaminar, or nontrilaminar, complements this evaluation.
An initial questioning involved the choice of methodology for the estimation of the endometrium before ET. When comparing the ways of evaluating the endometrium (Magalhães, 1993), the one through the vagina was considered superior, due to its greater proximity, absence of interposition of tissues, better quality and resolution of the image, besides discarding the discomfort of vesical repletion. Using TVUS, the assessed thickness was numerically superior than the abdominal assessment. However, the endometrial pattern evaluation proved to be concurrent in both the methods.
Although a TVUS may be better for the evaluation of the endometrium with respect to pathology detection, its use at the moment of ET would represent the introduction of a new variable. The abdominal US was, however, a part of the ET procedure in our protocol, confirmed by many earlier studies (Coroleu et al, 2002; Franco et al., 2004; Martins, 2007; Martins, 2004). Thus, with all the measurements being assessed through the abdomen, there was a uniform relationship among the conclusions Amir et al. (2007) studied the predictors for endome- trial thickness in 4518 fresco cycles with ET. The mean age was 33.5, and 35% of the patients were aged over 35. The endometrial thickness >14 mm on hCG day was correlated to better pregnancy rates in women aged over 35. Besides having three different nonrandomized protocols under agonist and antagonist analogs, it certainly included ovodonation cycles (ages from 19 to 56). In our study, 8 patients were over 35 years (15.6%), and patients with ovodonation were not included.
The several results regarding the importance of the endometrial thickness in assisted reproduction treatments considered different patients` selection criterion, which can cause confusion in the analysis of the findings (Friedler et al., 1996). Our study, the prospective design previously established assured the homogeneity of the parameters and data, reducing the possibilities of bias as long as all the patients included strictly used the same protocol.
We have not found a significant correlation between the endometrial thickness and pregnancy rate. Possible explanations for this diverging conclusion may be the difficulties in having a standard sagittal section of the uterus in the different ultrasonographic approaches (vaginal and abdominal), besides the different stimulation protocols used. There may also be differences in the frequencies of the probes used as well as the manufacturers of the available equipment.
The mean age was of 32.1 and the duration of stimulation was on average of 10.5, comparable with the literature data in cycles with antagonists (Kolibianakis et al., 2006, Tarlatzis et al., 2006, Kolibianakis et al., 2006). This choice of GnRH antagonist reflects the option that translates into fewer stimulus days, smaller quantity of gonadotrophin (costing less though), not to mention the diminishing of ovarian hyperstimulation syndrome (OHSS). There was no OHSS in this casuistic.
This population presented the number of MII oocytes on an average of 7 per puncture, transferring an average of 1.5 good quality embryos per patient.
The pregnancy rate per transfer was 39.2% (20 in 51 patients), which is adequate in relation to results in the Latin-American (33.1% in 2003 Red Lara), American (39.5%) (Gleicher et al., 2007) and European (28.7%) (Andersen et al., 2007) records. There was no significant difference in the correlation of variables studied, when comparing the pregnant with the non-pregnant women, especially with respect to endometrial thickness, mentioned earlier (Table 2). The take-home baby rate was of 25.49%.
Most studies conclude that the endometrial thickness on the hCG day only has prognostic value when extreme thicknesses are observed.
Thus, we tried to relate the type of transfer to the positive or negative pregnancy disclosure. The significant difference found (p=0.025) reveals a higher number of pregnancies (48.7%) in the transfer type 1 or 2 (Fig. 2).
The easy transfers correspond to 72.6% of the procedures and in these subgroups, the pregnancy proportion was of 48.7%, significantly greater than the transfers, involving greater manipulation (Fig. 2). It is important to emphasize the use of abdominal US as a guide in ET, as a facilitator factor (Coroleu et al, 2002; Franco et al., 2004; Martins, 2007; Martins, 2004).
As we did not find any association between the endometrial thickness on the day of ET and the outcome of pregnancy, we evaluated whether it would be possible to identify a better prognostic section point in the patients who became pregnant that was not conclusive.
Hartman et al. (2005) randomly evaluated the endometrium via transvaginal on the day of ET and established groups with thickness <8 mm, 8-12 mm, and >12 mm. They observed greater pregnancy rates in the endometrium >12 mm and considered canceling ET and cryopreserving the embryos, when the endometrium was <7 mm.
CONCLUSION
The prognostic question of uterine receptivity using ultrasound technology has not yet been solved. Our study did not detect any difference between the endometrial thickness values on the day of ET, with an interval from 7 to 15.8 mm, and the final result for pregnancy, neither for the pregnancy rates nor for live births.