ART – Assisted Reproduction Technology; ICSI – Intracytoplasmic Sperm Injection; IVF – In-Vitro Fertilisation

Approximately 85% of couples will successfully conceive a pregnancy after 12 months of unprotected intercourse. Of the remaining 15% with a degree of subfertility, a proportion will embark on the journey to assisted reproductive technology (ART) technique whereby gametes are manipulated to improve the probability of pregnancy [1]. In-Vitro fertilisation (IVF) and/ or Intracytoplasmic Sperm Injection (ICSI) has led to successful live pregnancies in this group of patients. Although ART success rates have improved dramatically over the last two decades, only 36% of women aged less than 30 years and 9% of women aged between 40-44 years will fall pregnant through a single ART cycle [2]. Such low implantation and pregnancy rates combined with the high cost of in IVF have instigated an evaluation of factors that can predict success for this group of patients.

Embryo implantation is a major rate-limiting step in the success of ART. A variety of factors can impact embryo implantation including endometrial differentiation, embryo quality and the method of embryo transfer. It is well established that endometrial differentiation and receptivity is critical in determining the possibility of success in ART treatments. Defined as a physiological condition of the endometrium whereby it is ideal for embryo implantation, endometrial receptivity is a state mainly induced by ovarian steroid hormones [3]. In the normal menstrual cycle, the endometrium undergoes cyclic morphological changes beginning with mitotic growth of the functional layer in the follicular phase in response to oestrogen. The luteal phase under the control of progesterone, readies the endometrium for implantation [4,5]. However, the molecular mechanisms underpinning receptivity and readiness for implantation and the interplay between ovarian hormones, cytokines, growth factors and adhesion molecules are not well understood [6]. As endometrial morphology may reveal such “readiness”, endometrial blood flow, pattern and thickness have all been evaluated as markers of receptivity and consequently implantation and pregnancy in IVF [5,7].

Endometrial thickness as measured by ultrasonographic examination is a simple, non-invasive and reproducible means of assessing endometrial development and may act as a surrogate marker of receptivity [8]. It has been studied extensively as a predictor of successful pregnancy in IVF, albeit with conflicting evidence. It is well accepted that poor pregnancy rates occur with thin endometrial linings as shown on transvaginal ultrasound scan. A thin endometrium is associated with implantation failure and numerous studies have demonstrated that implantation rarely occurs with thicknesses of < 6-7 mm [9,10]. However, there is conflicting evidence surrounding the impact of extremely increased endometrial thickness and the optimal thickness for implantation to occur. Such conflicting evidence may be in part due to the multiple confounding factors that influence the likelihood of success in any IVF cycle. In light of this conflicting evidence, this review will evaluate the available evidence in regards to endometrial thickness on the day of human chorionic gonadatrophin administration and its influence on pregnancy outcome in IVF cycles.

The Cochrane library, MEDLINE and EMBASE databases were searched using different combinations of the terms: “assisted reproductive technologies” OR “in vitro fertlisation” OR “infertility” AND “endometrial thickness” OR “thin endometrium” OR “thick endometrium” AND “pregnancy”.

Inclusion criteria were as follows:

1. Articles that were published in English and between 1992 and 2012

2. Studies where endometrial thickness was measured on the day of hCG injection via transvaginal ultrasound in the midsagittal plane

3. Studies where gonadotrophins were used in the stimulation protocol

Exclusion criteria were as follows:

1. Studies that evaluated endometrial thickness in natural conception or intrauterine insemination

2. Studies that did not provide quantitative data on endometrial thickness or pregnancy rates

3. Studies in which frozen embryos were used

4. Studies in which clomiphene citrate was used for ovarian stimulation

5. Case reports

A total of 427 articles were identified from the search strategy of which 11 were included for the literature review.

Eligible studies (n=11) were published in the last 20 years (1992- 2012). A comparison of baseline data between the studies is shown in Table 1. A comparison of findings in regards to endometrial thickness is shown in Table 2. Seven studies did not find a significant difference in endometrial thickness between the pregnant and nonpregnant cohorts. Of those studies that found a significant difference in endometrial thickness between pregnant and non-pregnant groups (n=4), a thickened endometrium was associated with improved embryo implantation rates.

Table 1: Baseline data comparison between studies. Pregnancy is defined as a positive urinary pregnancy test 14 days after embryo transfer. Not pregnant is defined as a negative urinary pregnancy test 14 days after embryo transfer. Data presented as mean ± standard deviation. NS – not significant. NA – not applicable/ not reported in the study.

Table 2: Summary of endometrial thickness findings. Pregnancy is defined as a positive urinary pregnancy test 14 days after embryo transfer. Not pregnant is defined as a negative urinary pregnancy test 14 days after embryo transfer. Data presented as mean ± standard deviation. NS – not significant.

Multiple studies in the literature have demonstrated that endometrial thickness is significantly increased in pregnant patients compared to non-pregnant patients [7,8,11-14]. However, there are also multiple studies that do not support this conclusion [15-22].

A limited number of studies have reported a detrimental impact of greatly increased endometrial thickness on pregnancy outcomes. Weissman et al demonstrated that an endometrium > 14 mm resulted in significantly reduced implantation and pregnancy rates [23]. This was supported by the findings of Rashidi et al. [20] who found no pregnancy at an endometrial thickness > 12 mm. The mechanisms proposed for this include a heightened risk of endometrial trauma during embryo transfer and an abnormal endometrial histologic pattern that does not support a pregnancy [23]. Similarly, multiple studies report poor pregnancy rates with a thin endometrial lining [19,20]. However, anecdotal case reports of successful pregnancy outcomes have been reported at both extremes of endometrial thickness [24,25]. Until a consensus is reached on the optimal threshold for endometrial thickness, the current evidence suggests that extremely thin or thick endometrial thicknesses are not an absolute contraindication to embryo transfer in an IVF cycle.

The majority of studies focus on biochemical pregnancy as the main outcome measure as opposed to continuing pregnancy. Although this may reflect an optimal state for implantation, a continuing pregnancy is of more clinical significance. As a thickened endometrium may result in a higher early pregnancy loss rate [23,26], it would be useful if future studies investigated the relationship between continuing pregnancy and endometrial thickness. Furthermore, implantation is a complex and poorly understood process. Although ultrasound measurements of thickness may be reflective of receptivity, it is not the sole determinant. Ultrasound assessments of thickness may be too simple to determine pregnancy potential. Some studies have examined the use of threedimensional ultrasound to determine endometrial volume, which may be a better surrogate marker [27,28]. However, the limited number of studies using this technology makes it difficult to reach definitive conclusions at this time. In addition, in those studies that demonstrate a significant increase in pregnancy rates with increased endometrial thicknesses, it is possible that this reflects improved ovarian stimulation with gonadotrophins and the subsequent downstream endometrial effects, as opposed to an independently responsive endometrial lining [7].

The different measurement outcomes and stimulation protocols used in the various studies make direct comparisons difficult. In an attempt to make a homogenous comparison, the studies examined in this review looked at endometrial thickness on the day of hCG administration. This was chosen, as it was a common time point in many studies. However, differences in stimulation protocols, namely gonadotrophin agonist versus antagonist protocols adds ambiguity, as does discrepancies in the statistical methods used. Strength of some studies was the use of Receiver Operating Curves (ROC) to generate a critical threshold for endometrial thickness, as were adequately powered studies with large sample sizes. Several studies commented on the need to increase sample sizes in order to prove significance based on power calculations. Although difficult to avoid, the retrospective nature of the studies is a weakness as is the fact that only one study blinded thickness measurements to subjects and treating physicians. Furthermore, as 6/11 studies in this review documented significant age differences between the pregnant and non-pregnant cohorts, it is difficult to draw conclusions on an independent effect of endometrial thickness. Increasing age is known to negatively impact on IVF success rates. This suggests that the poor prognosis associated with advancing age cannot be overcome by favorable factors including ideal endometrial receptivity [3].

There is insufficient evidence to support the use of a critical endometrial thickness threshold that clearly improves implantation and pregnancy rates in fresh IVF cycles. There is evidence of lower pregnancy rates at extremes of endometrial thickness, which should be taken into account when counseling couples about the chances of conception in a particular cycle. If the endometrium is deemed unsuitable, it may be worthwhile freezing embryos until the next cycle. The small number of trials examined in this review and the different measurement outcomes between studies make it difficult to draw definitive conclusions at this time. Three-dimensional ultrasound, which allows determination of endometrial volume, may prove to be an improved surrogate marker of endometrial receptivity in the future.