Review question
What is the most effective method for endometrial preparation in women undergoing embryo transfers with frozen embryos or embryos derived from donor oocytes?
Background
Couples undergo infertility treatments due to male factor, female factors or unexplained infertility. After an unsuccessful fresh embryo transfer cycle, a frozen-thawed embryo transfer can be performed when frozen embryos are available. Adequate hormonal preparation of the endometrium is of utmost importance for both egg donor and frozen embryo replacement cycles to provide the optimal chances of pregnancy. Many drugs and various modes of administration have been tried by several investigators in order to optimise implantation rates and consequently improve the success rates of the embryo transfer procedures: stimulated cycles (to generate endogenous oestradiol), programmed cycles (administering exogenous oestradiol) or natural cycles (allowing the ovaries to produce oestradiol without stimulation) are some of the options; avoiding spontaneous ovulation with gonadotropin-releasing hormone (GnRH) agonists and antagonists could have some impact; or using some other drugs such as aspirin or steroids that could potentially enhance the endometrial receptivity were also evaluated.
Study characteristics
We found 31 randomised controlled trials comparing different interventions such as the dose and route of administration of oestrogens and progestogen, the use of drugs that stop the patient from ovulating prematurely (GnRH agonists), and the use of other medications to improve the endometrium in a total of 5426 women. The evidence is current to June 2020.
Key results
We are uncertain whether a stimulated cycle (with letrozole) compared to a programmed cycle, for endometrial preparation, improves live birth. The evidence suggests that if the chance of live birth following a programmed cycle is assumed to be 24%, the chance following a stimulated cycle would be between 13% and 51%. We are also uncertain of the impact on miscarriage rate and endometrial thickness. A stimulated cycle may improve clinical pregnancy rate. Data were lacking on multiple pregnancy, cycle cancellation and other adverse effects.
We are uncertain whether a natural cycle improves the live birth rate, pregnancy rate, miscarriage rate and endometrial thickness in comparison with a programmed cycle. Data were lacking for all other outcomes.
We are uncertain if transdermal (delivered via the skin) oestrogens compared with oral (by mouth) oestrogens improve clinical pregnancy rate and miscarriage rate. Data were lacking for all other outcomes in this comparison.
Starting progestogen on the day of the donor oocyte retrieval or the day after probably increases the clinical pregnancy rate and probably reduces the cycle cancellation rate. We are uncertain if it reduces the miscarriage rate. Data were lacking for all other outcomes.
A cycle with GnRH agonist compared to without may improve live birth rate. We are uncertain of the effect of a GnRH cycle compared to no GnRH for the outcomes of clinical pregnancy rate, miscarriage rate, and endometrial thickness. No study reported on the other outcomes for this comparison.
We are uncertain if any GnRH agonist is better than other: a cycle with daily leuprolide or with deposit tryptorelin improves clinical pregnancy rate, or if daily acetate leuprolide or daily nafarelin reduces the miscarriage rate. Other outcomes were not reported.
GnRH antagonists compared to agonists probably improve clinical pregnancy rate. We are uncertain of the effect on miscarriage rate and multiple pregnancy rate. No study reported the other outcomes.
We are uncertain whether a cycle with aspirin compared to a cycle without improves live birth, clinical pregnancy rate or endometrial thickness. Data were lacking for all other outcomes.
We are also uncertain whether a cycle with steroids compared to a cycle without steroids improves live birth rate, clinical pregnancy rate or miscarriage rate. No study reported on the other outcomes.
Quality of the evidence
The evidence was of moderate to very low-quality. The main limitations in the evidence were poor reporting of study methods, and lack of precision in the findings for live birth.
There is insufficient evidence on the use of any particular intervention for endometrial preparation in women undergoing fresh donor cycles and frozen embryo transfers. In frozen embryo transfers, low-quality evidence showed that clinical pregnancy rates may be improved in a stimulated cycle compared to a programmed one, and we are uncertain of the effect when comparing a programmed cycle to a natural cycle. Cycle cancellation rates are probably reduced in a natural cycle. Although administering a GnRH agonist, compared to without, may improve live birth rates, clinical pregnancy rates will probably be improved in a GnRH antagonist cycle over an agonist cycle.
In fresh synchronised oocyte donor cycles, the clinical pregnancy rate is probably improved and cycle cancellation rates are probably reduced when starting progestogen the day of or day after donor oocyte retrieval.
Adequately powered studies are needed to evaluate each treatment more accurately.
A frozen embryo transfer (FET) cycle is when one or more embryos (frozen during a previous treatment cycle) are thawed and transferred to the uterus. Some women undergo fresh embryo transfer (ET) cycles with embryos derived from donated oocytes. In both situations, the endometrium is primed with oestrogen and progestogen in different doses and routes of administration.
To evaluate the most effective endometrial preparation for women undergoing transfer with frozen embryos or embryos from donor oocytes with regard to the subsequent live birth rate (LBR).
The Cochrane Gynaecology and Fertility Group trials register, CENTRAL, MEDLINE, Embase, PsycINFO, LILACS, trials registers and abstracts of reproductive societies' meetings were searched in June 2020 together with reference checking and contact with study authors and experts in the field to identify additional studies.
Randomised controlled trials (RCTs) evaluating endometrial preparation in women undergoing fresh donor cycles and frozen embryo transfers.
We used standard methodological procedures recommended by Cochrane. We analysed all available interventions versus placebo, no treatment, or between each other. The primary review outcome was live birth rate. Secondary outcomes were clinical and multiple pregnancy, miscarriage, cycle cancellation, endometrial thickness and adverse effects.
Thirty-one RCTs (5426 women) were included. Evidence was moderate to very low-quality: the main limitations were serious risk of bias due to poor reporting of methods, and serious imprecision.
Stimulated versus programmed cycle
We are uncertain whether a letrozole-stimulated cycle compared to a programmed cycle, for endometrial preparation, improves LBR (odds ratio (OR) 1.26, 95% confidence interval (CI) 0.49 to 3.26; 100 participants; one study; very low-quality evidence).
Stimulating with follicle stimulating hormone (FSH), letrozole or clomiphene citrate may improve clinical pregnancy rate (CPR) (OR 1.63, 95% CI 1.12 to 2.38; 656 participants; five studies; I2 = 11%; low-quality evidence). We are uncertain if they reduce miscarriage rate (MR) (OR 0.79, 95% CI 0.36 to 1.71; 355 participants; three studies; I2 = 0%; very low-quality evidence). Endometrial thickness (ET) may be reduced with clomiphene citrate (mean difference(MD) -1.04, 95% CI -1.59 to -0.49; 92 participants; one study; low-quality evidence). Other outcomes were not reported.
Natural versus programmed cycle
We are uncertain of the effect from a natural versus programmed cycle for LBR (OR 0.97, 95% CI 0.74 to 1.28; 1285 participants; four studies; I2 = 0%; very low-quality evidence) and CPR (OR 0.79, 95% CI 0.62 to 1.01; 1249 participants; five studies; I2 = 60%; very low-quality evidence), while a natural cycle probably reduces the cycle cancellation rate (CCR) (OR 0.60, 95% CI 0.44 to 0.82; 734 participants; one study; moderate-quality evidence). We are uncertain of the effect on MR and ET. No study reported other outcomes.
Transdermal versus oral oestrogens
From low-quality evidence we are uncertain of the effect transdermal compared to oral oestrogens has on CPR (OR 0.86, 95% CI 0.59 to 1.25; 504 participants; three studies; I2 = 58%) or MR (OR 0.55, 95% CI 0.27 to 1.09; 414 participants; two studies; I2 = 0%). Other outcomes were not reported.
Day of starting administration of progestogen
When doing a fresh ET using donated oocytes in a synchronised cycle starting progestogen on the day of oocyte pick-up (OPU) or the day after OPU, in comparison with recipients that start progestogen the day prior to OPU, probably increases the CPR (OR 1.87, 95% CI 1.13 to 3.08; 282 participants; one study, moderate-quality evidence). We are uncertain of the effect on multiple pregnancy rate (MPR) or MR. It probably reduces the CCR (OR 0.28, 95% CI 0.11 to 0.74; 282 participants; one study; moderate-quality evidence). No study reported other outcomes.
Gonadotropin-releasing hormone (GnRH) agonist versus control
A cycle with GnRH agonist compared to without may improve LBR (OR 2.62, 95% CI 1.19 to 5.78; 234 participants; one study; low-quality evidence). From low-quality evidence we are uncertain of the effect on CPR (OR 1.08, 95% CI 0.82 to 1.43; 1289 participants; eight studies; I2 = 20%), MR (OR 0.85, 95% CI 0.36 to 2.00; 828 participants; four studies; I2 = 0%), CCR (OR 0.49, 95% CI 0.21 to 1.17; 530 participants; two studies; I2 = 0%) and ET (MD -0.08, 95% CI -0.33 to 0.16; 697 participants; four studies; I2 = 4%). No study reported other outcomes.
Among different GnRH agonists
From very low-quality evidence we are uncertain if cycles among different GnRH agonists improves CPR or MR. No study reported other outcomes.
GnRH agonists versus GnRH antagonists
GnRH antagonists compared to agonists probably improves CPR (OR 0.62, 95% CI 0.42 to 0.90; 473 participants; one study; moderate-quality evidence). We are uncertain of the effect on MR and MPR. No study reported other outcomes.
Aspirin versus control
From very low-quality evidence we are uncertain whether a cycle with aspirin versus without improves LBR, CPR, or ET.
Steroids versus control
From very low-quality evidence we are uncertain whether a cycle with steroids compared to without improves LBR, CPR or MR. No study reported other outcomes.