You are here

The effect of elevated progesterone levels before HCG triggering in modified natural cycle frozen-thawed embryo transfer cycles

Reproductive BioMedicine Online, Volume 34, Issue 5, May 2017, Pages 546 - 554

Editorial Comment

Increased progesterone levels during the late follicular phase adversely affect pregnancy rates after fresh embryo transfer. Logically, one would expect to observe a similar effect in frozen embryo transfer cycles. Yet, the incidence of premature progesterone rise and its effect on clinical outcome can vary between different endometrial preparation protocols. In this secondary analysis of randomized controlled trial data, Groenewoud and colleagues present interesting findings which can make you reconsider your practice.

Abstract

Recent studies suggest that elevated late follicular phase progesterone concentrations after ovarian stimulation for IVF may result in embryo–endometrial asynchrony, reducing the chance of successful implantation after fresh embryo transfer. It remains unclear to what extent elevated late follicular phase progesterone levels may occur in unstimulated cycles before frozen–thawed embryo transfer, or what affect they may have on outcomes. In this cohort study, 271 patients randomized to the modified natural cycle arm of a randomized controlled trial comparing two endometrial preparation regimens underwent late follicular phase progesterone and LH testing. A receiver operating characteristic curve was constructed to identify a progesterone cut-off level with the best predictive value for live birth (progesterone level ≥4.6 nmol/l). A total of 24.4% of patients revealed an isolated elevated serum progesterone of 4.6 nmol/l or greater, and 44.3% showed an elevated progesterone level in association with a rise in LH. Neither endocrine disruption affected outcomes, with live birth rates of 12.9% versus 10.6% (OR 0.6, 95% CI 0.19 to 1.9) and 11.9% versus 17.5% (OR 1.6, 95% CI 0.79 to 3.1), respectively. Whether monitoring of progesterone and LH in natural cycle frozen–thawed embryo transfer has added clinical value should studied further.

Keywords: Frozen-thawed embryo transfer, Live birth rate, Modified natural cycle, Progesterone.

Introduction

The incidence and effect of elevated late follicular phase progesterone levels on IVF treatment outcomes continue to be debated (Fatemi, Van Vaerenbergh, 2015 and Weinerman, Mainigi, 2014). Several studies of patients undergoing ovarian stimulation for IVF have shown that an elevated late follicular phase progesterone level is associated with diminished pregnancy rates and an increased risk of obstetric complications (Bosch et al, 2010 and Venetis et al, 2015). Elevated late follicular phase progesterone levels are usually defined as a serum progesterone of 4.77 nmol/l or more (≥1.5 ng/l) on completion of ovarian stimulation, before administration of HCG to trigger final oocyte maturation (Bosch et al, 2010 and Venetis et al, 2015).

The underlying mechanism for this phenomenon remains unclear, but supraphysiological oestrogen levels arising from ovarian stimulation may be indirectly implicated (Farhi et al, 2010 and Joo et al, 2010). Both abnormal oestrogen and progesterone levels lead to a spectrum of alterations in the endometrium, such as changes in histological features as well as gene and cytokine expression related to endometrial receptivity (Boomsma et al, 2010, Bosch et al, 2010, Labarta et al, 2011, Lass et al, 1998, and Ubaldi et al, 1997). In patients undergoing ovarian stimulation for IVF, these features are observed earlier when progesterone levels are elevated before HCG has been administered. This suggests an acceleration in the maturation of the endometrium that may lead to an asynchrony between optimal receptivity and timing of embryo transfer, resulting in lower pregnancy rates (Ubaldi et al, 1997 and Van Vaerenbergh et al, 2011). Whether the changes in the endometrium are the result of an elevated progesterone alone or perhaps a combination of supra-physiological hormone levels and other ovarian stimulation related factors remains unclear.

As natural cycle frozen–thawed embryo transfer requires no ovarian stimulation, it provides the opportunity to avoid the assumed negative effects of ovarian hyperstimulation on the endometrium. In natural cycle frozen–thawed embryo transfer, endocrine control of the menstruation cycle is not altered, and the endometrium should not be exposed to supra-physiological oestrogen and progesterone levels. This premise underlies the recommendation by some to freeze all embryos when an elevated late follicular phase progesterone is measured, and to delay transfer to an unstimulated cycle (Polotsky et al, 2009 and Shapiro et al, 2010).

In modified natural cycle frozen–thawed embryo transfer, an injection with HCG is given to trigger ovulation of the dominant follicle. The continuing presence of the dominant follicle on ultrasound is taken to indicate that LH surge and luteinization have yet to occur. Progesterone levels, however, start to rise about 12 h before the start of the LH surge, so it is possible that at the time of HCG administration, progesterone levels are elevated (Hoff et al., 1983). Given that maturation of the endometrium is induced by progesterone, significant progesterone elevation before HCG injection might lead to desynchronization between the endometrium and the embryo, and thereby diminishing pregnancy rates.

This concern has led some to monitor progesterone levels in patients undergoing modified natural cycle frozen–thawed embryo transfer, and cancel the cycle if progesterone levels are elevated (Weissman et al, 2009 and Weissman et al, 2011). The incidence and effect of elevated progesterone levels before ovulation triggering in modified natural cycle frozen–thawed embryo transfer, however, remain to be elucidated. As elevated progesterone levels have been shown to be related to multi-follicular development, we investigated the frequency of premature progesterone elevations in modified natural cycle frozen–thawed embryo transfer cycles and whether they are associated with detrimental effects on clinical outcomes in this context.

Materials and methods

Patients and procedures

Included patients were recruited for the purpose of a multi-centre randomized controlled trial (‘ANTARCTICA’ trial, NTR1586) (Groenewoud et al, 2012b and Groenewoud et al, 2016). Patients were randomized between an artificial frozen embryo transfer cycle and a modified natural cycle frozen–thawed embryo transfer. Participants meeting the inclusion criteria for this trial were aged 18–40 years, had a regular cycle and had cryopreserved embryos derived from of one of their first three fresh IVF or IVF and intracytoplasmic sperm injection (IVF–ICSI) cycles. Patients with a uterine anomaly or a contraindication for medication used in the artificial frozen embry transfer cycle were excluded, as were patients undergoing treatment with donor gametes. Patients randomized to modified natural cycle frozen–thawed embryo transfer were considered eligible for the analyses of the present study. Ethical approval for the randomized controlled trial was obtained from the institutional review board of the Isala Clinics in Zwolle on 4 March 2009 (reference number NL23273.075.09).

Patients randomized to the modified natural cycle frozen–thawed embryo transfer arm underwent ultrasound monitoring of the dominant follicle and the endometrium to plan thawing and transferring of the cryopreserved embryo. Commencement of ultrasound monitoring depended on the length of the menstrual cycle on day 10, 11 or 12, and was repeated every other day or daily depending on the size of the follicle. No minimal endometrial thickness was appointed given that the endometrial thickness during natural cycle frozen–thawed embryo transfer cannot be influenced given the characteristics of the treatment.

When the dominant follicle reached a diameter of 16–20 mm, a blood sample was drawn followed by a single injection with HCG (5000 IE Pregnyl® or 250 µg Ovitrelle®). Thawing and transferring was timed to coincide with the time of HCG injection and the developmental stage at the moment of cryopreservation. Cleavage stage embryos were thawed on the fourth or fifth day after HCG injection. Blastocyst embryos were thawed on the sixth day after HCG injection. Transfer was carried out on the day of, or the day after, thawing depending on local hospital logistics. All participating centres used a standardized scoring system based on the European Society of Human Reproduction and Embryology Istanbul consensus on embryo assessment to determine embryo quality (Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology, 2011). Morphological features were summarized and embryos were classified as good, reasonable, moderate or poor. A maximum of three embryos were transferred. All embryos were cryopreserved using a slow-freeze technique, and 92% of transfers occurred at the cleavage stage (Wong et al, 2014 and Zhu et al, 2015). No luteal support was given. All blood samples were collected and analysed centrally for oestrogen, progesterone and LH levels after finishing the trial. No additional endocrine monitoring was carried out during the frozen embryo transfer treatment.

Laboratory procedures

Samples were centrifuged at 1500 g after which the plasma was stored at −20°C. On completion of the ANTARTICA trial, samples were collected from the different study sites and analysed in a central laboratory. All samples were analysed according to electrochemiluminescence method on the Cobas8000, module602 (Roche diagnostics, Indianapolis, USA). Intra- and inter-assay coefficients for progesterone determinations varied between 0.9% and 2.9%. For LH determination, intra- and inter-assay coefficients varied between 0.6% and 1.2%.

Definitions

A commonly used definition of progesterone elevation is a serum progesterone level of 4.77 nmol/l or over in the absence of an LH surge (Bosch et al, 2010 and Venetis et al, 2015). This threshold, however, was derived from studies in patients undergoing ovarian stimulation with exogenous gonadotrophins to achieve multi-follicular development. Considering the major endocrine differences between patients undergoing ovarian stimulation and patients undergoing modified natural cycle frozen–thawed embryo transfer, we investigated whether another progesterone threshold might apply in this context.

A serum LH level of 10 IU/l or more was considered to indicate an LH surge (Griesinger et al, 2007 and Groenewoud et al, 2012a). The combination of an LH surge and an elevated progesterone level while the dominant follicle is still evident on ultrasound (so called premature luteinization) represents a different endocrine disruption than when progesterone levels alone are elevated. To be able to study the specific affect of an isolated elevated progestone, two subgroups were analysed: one consisting of patients with an isolated elevated progesterone and one with patients demonstrating premature luteinization.

Statistical analyses

Given the retrospective nature of this study, no formal sample size calculation before the start of this study could be carried out. By calculating Cohen's d based on the obtained odds ratios, the effect size was calculated. The observed effect size was extrapolated to a post-hoc sample calculation to determine whether or not the present study was adequately powered. All data were analysed using SPSS (version 22. IBM, Armonk, New York. USA). The primary end-point was live birth rate. Secondary end-points were clinical pregnancy rate, ongoing pregnancy rate and miscarriage rate. A receiver operating characteristic (ROC) curve was constructed to identify a progesterone cut-off level with the best predictive value for live birth. The optimal cut-off value was identified after bootstrapping (1000 times) and calculating the Youden J-index. Pregnancy, live birth and miscarriage rates were compared using a chi-square test, including calculation of odds ratios. Comparison of median levels of progesterone between patients with and without pregnancy, live birth, or both, was carried out using a Mann–Whitney U test. Analyses were repeated after identifying and excluding outliers to determine the robustness of our findings. Outliers were detected by using a generalized extreme studentized deviate test. Baseline characteristics were compared between patients included and excluded in this trial to rule out selection bias. Depending on the distribution of data, one-way analysis of variance or the Kruskal–Wallis test was used to test for baseline characteristics between the different subgroups. In order to test whether or not the discriminating capacity for pregnancy and live birth of progesterone was influenced by other parameters, possible confounding factors for live birth were identified with the use of recent literature. Those confounding factors with P < 0.2 were incorporated in a stepwise enter logistic regression analyses. Confounding factors with P < 0.05 were considered to be of significance.

Results

Of the 495 patients included in the modified natural cycle frozen-thawed embryo transfer of the ANTARTICA trial, 394 underwent a frozen thawed embryo transfer. Treatment was cancelled in the other 101 either because of inadequate survival of the cryopreserved embryos or premature ovulation shown by the disappearance of the dominant follicle on ultrasound. In 271 patients undergoing embryo transfer, a blood sample was available for analyses.

The mean age of participants entering the study was 33.2 years, and mean duration of infertility 3.0 years. The most common diagnoses were severe male infertility (120/291, 41.2%), infertility of unknown origin (65/291, 22.3%) and tubal pathology (44/291, 15.1%). Slightly more patients had undergone an IVF–ICSI as initial treatment than IVF 144 (53.1%) versus 127 (46.9%) and on average patients had 0.6 IVF or IVF–ICSI treatments before the IVF or IVF–ICSI treatment from which these frozen-thawed embryos originated. Further overall baseline characteristics are presented in Table 1. Significant differences in the number of IVF or IVF–ICSI treatments before the fresh IVF treatment from which the frozen embryo in the present study derived (P = 0.03) and progesterone concentration (P < 0.01) were observed between patients with and without signs of premature luteinization.

Table 1

Baseline characteristics.

 

Patient characteristic Overall P < 4.6 nmol/l P ≥ 4.6 nmol/l LH > 10 IU/l and P ≥ 4.6 nmol/l
Patients (number)a 271 85(31.4%) 66(24.4%) 120(44,3%)
Age (years)a 33.2(3.9) 33.8(3.6) 33.0(3.6) 32.9(4.2)
Age at ovum retrieval (years)a 32.8(4.0) 33.2(3.9) 32.6(3.7) 32.5(4.2)
Duration of subfertility (years)a 3.0(2.2) 2.5(1.9) 3.2(2.3) 3.2(2.3)
Fertility status n (%)a
 Primary subfertility 102(37.6) 31(36.5) 30(45.5) 41(34.2)
Initial treatment, n (%)a
 IVF 127(46.9) 37(43.5) 32(48.5) 58(48.3)
 IVF–ICSI 144(53.1) 48(56.5) 34(51.5) 62(51.7)
Outcome fresh cycle, n (%)a
 Live birth 63(23.2) 19(22.4) 13(19.7) 31(25.8)
Number of diagnoses, n (%)a
 Single diagnosis 246(90.8) 81(93.1) 58(87.9) 107(90.7)
Diagnoses, n (%)a
 Unknown 65(22.3) 22(23.9) 17(23.3) 26(20.6)
 Severe male infertility 120(41.2) 38(41.3) 29(39.7) 53(42.0)
 Moderate male infertility 21(7.2) 2(2.2) 6(8.2) 13(10.3)
 Tubal factor 44(15.1) 17(18.5) 8(11.0) 19(15.1)
 Endometriosis stage 1–2 12(4.1) 2(2.2) 4(5.5) 6(4.8)
 Endometriosis stage 3–4 8(2.8) 4(4.3) 1(1.4) 3(2.4)
 Hormonal factor 13(4.5) 4(4.3) 5(6.8) 4(3.2)
 Cervical factor 2(0.7) 0(0) 2(2.7) 0(0)
 Other 6(2.1) 3(3.3) 1(1.4) 2(1.6)
Number of IVF (ICSI) cycles before fresh cycle from which studied FET derived n (%)a and c 0.6(0.8) 0.64(0.78) 0.45(0.75) 0.69(0.83)
Number of FET before study ET, n (%)a 0.8(0.8) 0.72(0.82) 0.80(0.73) 0.79(0.83)
Interval between ovum retrieval and FET (days) 0.64(0.85) 0.71(0.98) 0.49(0.75) 0.66(0.81)
Duration of cryopreservation (years)a 0.6(0.8) 0.68(0.96) 0.49(0.76) 0.65(0.79)
Stadium of cryopreservation, n (%)a
 Cleavage stage 249(91.9) 79(92.9) 61(92.4) 109(90.8)
 Blastocyst stage 22(8.1) 6(6.9) 5(7.6) 11(9.3)
Survival n (%)a 90(20) 88(25) 87(26) 88(22)
Number of embryos transferred n (%)a 1.2(0.4) 1.2(0.45) 1.1(0.43) 1.2(0.39)
Embryo quality (score out of 4.0) n (%)a 3.2(0.9) 3.1(0.88) 3.0(1.0) 3.3(0.83)
Endometrial thickness (mm) n (%)a 9.0(2.0) 8.9(1.9) 8.7(2.2) 9.2(1.9)
Progesterone level (nmol/l)b, c, and d 5.9(0.89 to 156.8) 3.6(0.89 to 4.7) 5.9(4.6 to 156.8) 7.9(4.87 to 24.7)

a Presented mean with standard deviations or percentage.

b Median including ranges.

c Significant at P = 0.03.

d Significant at P ≤ 0.01.

ET, embryo transfer; FET, frozen embryo transfer; ICSI, intracytoplasmic sperm injection.

The overall live birth rate per embryo transfer procedure was 14.4%. With the exception of mean scored embryo quality (mean score of 3.5 out of 4 versus 3.0 out of 4; P = 0.02), no differences were observed in baseline characteristics between patients achieving clinical pregnancy and patients who did not. Included patients are presented in Figure 1. Around 31% of the original available cohort could not be included in the present analyses owing to absence of a blood sample. To ascertain whether this may introduce a selection bias into the present analysis, the baseline characteristics of the particiapants sampled were compared with those from whom the sample was missing. No significant differences were observed (Supplementary Table S1). We therefore conclude that the patients with a blood sample drawn represent a random sample of the whole study population.

Figure 1

Figure 1

Patients. FET, frozen embryo transfer; mNC-FT, modified natural cycle frozen-thawed embryo transfer; RCT, randomized controlled trial.

 

The ROC constructed for progesterone is shown in Figure 2. On the basis of the calculated Youden-J index a progesterone level of 4.6 nmol/l or over seemed to be associated with lower live birth rate. Sensitivity (51.3%) and specificity (63.2%), however, were both low. The area under the curve of 0.54 was not significantly different from the true area. To visualize the distribution of progesterone levels as well as identifying the outliers a Box-and-Whisker plot was constructed (Figure 3). Three outliers were identified by the use of a generalized electrostatic discharge test. The cut-off value for an outlier was defined as a progesterone level over 18.63 nmol/l. Repeating the construction of the ROC curve and the calculation of the Youden-J index excluding the identified outliers did not result in a different cut-off value. Analyses were conducted for both the cut-off progesterone level obtained in this study (4.6 nmol/l) as well as the cut-off value of 4.77 nmol/l or over, which is most commonly used in recent studies. An isolated elevated progessterone level 4.6 nmol/l or over was observed in 66 of the 271 patients included (24.4%). A progesterone level of 4.77 nmol/l or over was observed in 23.6% of patients (64/271). No significant difference in live birth rate was observed between cycles without or with an isolated elevated progesterone level of 4.6 nmol/l or over (12.9% versus 10.6%, OR 0.6, 95% confidence interval (CI) 0.19 to 1.94). Moreover, no significant difference was evident in median progesterone level between patients who did or did not achieve a live birth (6.97 nmol/l versus 7.10 nmol/l). Live birth rate, clinical pregnancy, ongoing pregnancy rate and miscarriage rates are presented in Table 2, as well as results for the analyses conducted with the threshold for an elevated progesterone level of 4.77 nmol/l or over. On the basis of the observed OR of 0.6 for live birth rate Cohen's d was −0.28, which can be interpreted as a small result (Cohen, 1988). Given the effect size 404 patients should have been included in order to obtain robust statistical power.

Figure 2

Figure 2

Receiver operator characterstic curve.

 

Figure 3

Figure 3

Box-and-Whisker plot comparing median progesterone levels between patients with and without an outcome. The cut-off value for an outlier was defined as a progesterone level greater than 18.63 nmol/l using a generalized extreme studentized deviate test. Red arrows indicate the highest values.

 

Table 2

Live birth, pregnancy and miscarriage rates in patients with normal or isolated elevated progesterone conccentrations. No significant difference was obtained for all comparisons.

 

Progesterone cut-off ≥4.6 nmol/l Progesterone cut-off ≥4.77 nmol/l
Normal progesterone level, n (%) Elevated progesterone level, n (%) OR with 95 % CI Normal progesterone level, n (%) Elevated progesterone level, n (%) OR with 95 % CI
Live birth 11/85(12.9) 7/66(10.6) 0.6, 0.19 to 1.9 11/89(12.4) 7/64(10.9) 0.9(0.32 to 2.4)
Clinical pregnancy 23/85(27.1) 12/66(18.2) 0.6, 0.27 to 1.3 23/89(25.8) 12/64(18.8) 0.7(0.30 to 1.5)
Ongoing pregnancy 13/85(15.3) 7/66(10.6) 0.7, 0.45 to 1.8 13/89(14.6) 7/64(10.9) 0.7(0.67 to 1.9)
Miscarriage 12/23(52.2) 5/12(41.7) 1.5, 0.37 to 6.3 12/23(52.2) 5/12(41.7) 1.5(0.37 to 6.3)

In 44.3% (120/271), a combination of both progesterone elevation (cut-off ≥4.6 nmol/l) and an LH surge was observed. The live birth rate was 11.9% in patients without premature luteinization compared with 17.5% in those with evidence of premature luteinization (OR 1.6, 95% CI 0.79 to 3.1). Similar results were observed for clinical pregnancy rate (23.2% versus 23.3%) and ongoing pregnancy rate (13.2% in patients without premature luteinization versus 17.5% with premature luteinization) (Table 3). For the cut-off level 4.77 nmol/l or over, similar results were obtained. Live birth rates were not influenced by the presence of an LH of 10 IU/l or over in case of a normal progesterone at cut-off of 4.6 nmol/l or over (11.1% [5/45]) versus 15.0% [6/40], OR 1.4, 95% CI 0.40 to 5.0). Similar effects were observed in clinical pregnancy rate, ongoing pregnancy rate and miscarriage rate (Table 4). These results were not altered after adjusting the cut-off progesterone level to 4.77 nmol/l or over.

Table 3

Live birth, pregnancy and miscarriage rates in patients with and without premature luteinization. No significant difference was obtained for all comparisons.

 

Progesterone cut-off ≥4.6 nmol/l Progesterone cut-off ≥4.77 nmol/l
No premature luteinization, n (%) Premature luteinization, n (%) OR with 95 % CI No premature luteinization, n (%) Premature luteinization, n (%) OR with 95 % CI
Live birth 18/151(11.9) 21/120(17.5) 1.6(0.79 to 3.1) 18/153(11.8) 21/118(17.8) 1.6(0.79 to 3.1)
Clinical pregnancy 35/151(23.2) 28/120(23.3) 1.0(0.57 to 1.8) 35/153(22.9) 28/118(23.7) 1.1(0.60 to 1.9)
Ongoing pregnancy 20/151(13.2) 21/120(17.5) 1.4(0.71 to 2.7) 20/153(13.1) 21/118(17.8) 1.4(0.74 to 2.8)
Miscarriage 17/35(48.6) 7/28(25.0) 2.8(0.96 to 8.4) 17/35(48.6) 7/28(25.0) 2.8(0.96 to 8.4)

Table 4

Live birth, pregnancy and miscarriage rates in patients with and without a LH ≥ 10 IU/l in the absence of an elevated progesterone level. No significant difference was obtained for all comparisons.

 

Progesterone cut-off ≥4.6 nmol/l Progesterone cut-off ≥4.77 nmol/l
No isolated raised LH Isolated raised LH OR with 95 % CI No isolated raised LH Isolated raised LH OR with 95 % CI
Live birth 5/45(11.1%) 6/40(15.0%) 1.4(0.40–5.0) 5/47(10.6%) 6/42(14.3%) 1.4(0.40–5.0)
Clinical pregnancy 9/45(20.0%) 14/40(35.0%) 2.1(0.81–5.7) 9/47(19.1%) 14/42(33.3%) 2.1(0.80–5.6)
Ongoing pregnancy 6/45(13.3%) 7/40(17.5%) 1.4(0.42–4.5) 6/47(12.8%) 7/42(16.7%) 1.4(0.42–4.5)
Miscarriage 4/9(44.4%) 8/14(57.1%) 0.6(0.12–3.2) 4/9(44.4%) 8/14(57.1%) 0.6(0.12–3.2)

Univariate regression analyses showed that average embryo score, number of embryos transferred, fertility status and age during the ovum retrieval all had a P-value of less than 0.20. These possible confounding factors were incorporated in addition to progesterone level in a multivariate logistic regression analyses. Using a stepwise enter method, the only significant predictor for live birth was average embryo score. The discriminating capacity of progesterone was not altered by any of the confounding factors.

Discussion

In this cohort study, an isolated elevated progesterone level of 4.6 nmol/l or over was observed in over one-fifth of patients undergoing modified natural cycle frozen-thawed embryo transfer. Signs of premature luteinization were observed in over 40%. Neither were associated with a detrimental effect on either live birth rate or pregnancy rates. Given the small sample size, confirmation studies should be carried out before refraining from routine monitoring of progesterone and LH levels in modified natural cycle frozen-thawed embryo transfer.

A previous, smaller retrospective study also reported a high incidence (28.4%) of progesterone elevation of 5 nmol/l or more before the LH surge in patients undergoing true natural cycle frozen-thawed embryo transfer (Lee et al., 2014). Overall, no differences in clinical pregnancy rate and ongoing pregnancy rate were observed between patients with and without elevated progesterone. A subgroup analyses within that study suggested it was not the level but the duration of progesterone exposure before the LH surge that is responsible for lower pregnancy rates. In the present study, an isolated elevated progesterone level at the moment of HCG injection was observed in 24.4% of all patients. The observed difference in incidence cannot be ascribed to the different approach in cycle monitoring. It may be ascribed to variations in progesterone assay performance (Boudou et al, 2001, Coucke et al, 2007, and Fleming, 2008). Park et al. (2015b) showed that progesterone levels at the point of ovum retrieval are not influenced by patient characteristics. Whether these results can be extrapolated to natural cycles is not known (Park et al., 2015b).

Most of the recent information about elevated progesterone levels is derived from studies conducted in patients undergoing ovarian stimulation. Elevated progesterone levels before the LH surge in spontaneous cycles, however, have been described in older studies (Laborde et al, 1976 and Pauerstein et al, 1978). Ovulation was confirmed by direct visualization using laparoscopy, histological examination of the corpus luteum, or both. Given the imprecise method of ovulation detection, the results of these studies cannot be translated to current practice. A recent study conducting ultrasound cycle analyses combined with endocrine monitoring in both serum and urine in healthy volunteers showed that in the presence of a dominant follicle on ultrasound a progesterone level greater than 5 nmol/l was observed in 10% of included patients (Roos et al., 2015). In contrast to the observed frequency of progesterone elevation in natural cycles, progesterone elevation before ovum retrieval is more frequently observed in patients undergoing ovarian stimulation, and is associated with lower implantation rates (Bosch et al, 2010 and Venetis et al, 2015). A large meta-analysis containing over 60,000 patients undergoing ovarian stimulation showed that pregnancy rates are diminished in patients with a progesterone level greater than 2.5 nmol/l (Venetis et al., 2013). The outcome of subsequent frozen embryo transer cycles were not negatively influenced by an elevated progesterone level at the moment of ovum retrieval during the index IVF treatment. Comparable effects have been observed in patients undergoing intrauterine insemination with mild ovarian stimulation (Requena et al., 2015). The incidence of premature progesterone elevation in our study is higher than that reported in intrauterine insemination with mild ovarian stimulation and IVF. This might be explained by the use of GnRH agonist and antagonist in IVF and the strict criterion of maximum dominant follicle size before administrating HCG in the study by Requena et al. (2015). Moreover, this may reflect a local variation in progesterone assay performance, as this has been shown to be subject to local conditions. Clinics are therefore advised to determine their own cut-off levels.

The exact mechanism by which high progesterone levels diminish pregnancy rates has not been fully elucidated, but dysregulation of genes determining the period of endometrial receptivity seem to play an important role (Haouzi et al, 2009, Labarta et al, 2011, and Li et al, 2011). Compensating for the dysregulation by personalizing the timing of FET based on endometrial receptivity array has been suggested to improve pregnancy rates in patients with repeated implantation failure (Ruiz-Alonso et al., 2013). Although the present study did not relate progesterone levels to endometrial gene expression, no relationship between elevated progesterone levels and clinical pregnancy rate was observed. It can therefore be proposed that elevated progesterone is simply a marker of a more complex endocrine and paracrine disruption, which is set in motion by ovarian stimulation and which ultimately results in the observed changes in endometrial receptivity and diminished pregnancy rates.

Surges in LH occurring at the time of HCG administration are known to occur in a substantial number of patients (Cantineau, Cohlen, 2007 and Groenewoud et al, 2012a). The incidence and effect of premature luteinization on pregnancy rates in this clinical context has been studied less. Both an isolated LH surge as well as premature luteinization seems to be without an evident effect on pregnancy rates. The presence of an LH surge at the time of HCG suggests that the ‘window of implantation’ might be opened earlier compared with cycles without an LH surge at the point of HCG injection. Premature luteinization would point to even further endometrial advancement. Ignoring such events could result in an error in estimating the start of the ‘window of implantation’ and hence the timing of embryo transfer. Whether or not determining endometrial receptivity by gene analysis or other means in patients experiencing both an isolated LH surge as well as premature luteinization in a natural cycle could improve outcomes is the subject of a current clinical trial (NCT02280798).

So far, only one study has been published on the influence of endocrine monitoring in natural cycle frozen-thawed embryo transfer (Park et al., 2015a). Extensive monitoring did not increase pregnancy rates after natural cycle frozen-thawed embryo transfer compared with ultrasound alone. Extensive monitoring however, is presented as the gold standard for planning of thawing and embryo transfer by some (Casper and Yanushpolsky, 2016). Given the increasing number of frozen embryo transfer procedures carried out worldwide, a more scientific foundation for monitoring in natural cycle frozen-thawed embryo transfer is mandatory. Optimization of monitoring protocols will increase patient convenience as well as improve cost-efficiency. The presented data indicate that endocrine monitoring may add little to simpler monitoring based on ultrasound assessment of follicle maturation. Confirmation studies, however, are necessary.

A limitation of this retrospective analysis is the number of patients included. In 123 out of 394 patients, a blood sample was, for various reasons not drawn. As there were no differences in baseline characteristics between patients included and excluded for analyses, and as oestrogen levels were not found to be markedly high in those tested, we deem the chance of selection bias to be small. As can be calculated based on the effect size, at least 404 patients should have been included for robust statistical power. The present study, therefore, does not have sufficient sample size to draw firm conclusions on both primary and secondary outcomes and, therefore, these results should be interpreted with caution. Further confirmation studies are warranted. The observed ongoing pregnancy rates are relatively low compared with other recent studies. This may reflect the low number of embryos transferred, the prevalent transfer of cleavage stage embryos rather than blastocysts, the use of relatively liberal cryopreservation criteria and slow freeze techniques, which have since been largely replaced by vitrification. Miscarriage rates are high compared with another reports on natural cycle frozen-thawed embryo transfer, given that baseline characteristics of patients included in the present study do not seem to differ from the previous study (Tomas et al., 2012). Moreover, treatment-related factors such as the use of slow-freeze cryopreservation, cleavage stage embryo transfer and embryo quality do not result in higher miscarriage rate. Therefore, we cannot explain the high miscarriage rate. In the present study, predominantly cleavage stage embryo cryopreservation using a slow-freeze technique was used. The results of the present study might therefore not be applicable to FET cycles using blastocyst stage embryos or vitrification. As clinical decisions on planning of thawing and transferring were made without any knowledge of progesterone and LH levels, the effect of rises in both hormones could be interrogated. Regretfully, no information on follicle size at the time of HCG injection was recorded, so follicle size could not be related to progesterone elevation.

In conclusion, elevated progesterone levels are observed frequently in patients undergoing modified natural cycle frozen-thawed embryo transfer but seem to have little effect on live birth and pregnancy rates. A combination of elevated progesterone with an LH surge occurs in more than 40% of all patients and does not seem to have any consequences on clinical pregnancy rate, ongoing pegnancy rate, live birth rate and miscarriage rate. As frozen embry transfer becomes more widely adopted as a routine alternative procedure to fresh embryo transfer, a great understanding of associated endocrine events such as premature luteinization on clinical outcomes is merited.

Acknowledgement

We would like to thank the women who participated in this study. We thank all participating hospitals and their staff for their contribution to this study, as well as the research nurses and office members of the Dutch Consortium for their hard work and dedication (www.studies-obsgyn.nl). We would also like to thank Jack van Dijk for his help with the processing of the samples.

Appendix. Supplementary material

The following is the supplementary data to this article:

Download file

Table S1

Comparison between sampled patients and those not sampled.

 

References

  • Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology, 2011 Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology. The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting. Hum. Reprod. 2011;26:1270-1283
  • Boomsma et al, 2010 C.M. Boomsma, A. Kavelaars, M.J. Eijkemans, B.C. Fauser, C.J. Heijnen, N.S. Macklon. Ovarian stimulation for in vitro fertilization alters the intrauterine cytokine, chemokine, and growth factor milieu encountered by the embryo. Fertil. Steril. 2010;94:1764-1768 Crossref
  • Bosch et al, 2010 E. Bosch, E. Labarta, J. Crespo, C. Simon, J. Remohi, J. Jenkins, A. Pellicer. Circulating progesterone levels and ongoing pregnancy rates in controlled ovarian stimulation cycles for in vitro fertilization: analysis of over 4000 cycles. Hum. Reprod. 2010;25:2092-2100 Crossref
  • Boudou et al, 2001 P. Boudou, J. Taieb, B. Mathian, Y. Badonnel, I. Lacroix, E. Mathieu, F. Millot, N. Queyrel, C. Somma-Delpero, M.C. Patricot. Comparison of progesterone concentration determination by 12 non-isotopic immunoassays and gas chromatography/mass spectrometry in 99 human serum samples. J. Steroid Biochem. Mol. Biol. 2001;78:97-104 Crossref
  • Cantineau, Cohlen, 2007 A.E. Cantineau, B.J. Cohlen, Dutch IUI Study Group. The prevalence and influence of luteinizing hormone surges in stimulated cycles combined with intrauterine insemination during a prospective cohort study. Fertil. Steril. 2007;88:107-112 Crossref
  • Casper, Yanushpolsky, 2016 R.F. Casper, E.H. Yanushpolsky. Optimal endometrial preparation for frozen embryo transfer cycles: window of implantation and progesterone support. Fertil. Steril. 2016;105:867-872 Crossref
  • Cohen, 1988 J. Cohen. Statistical Power Analysis for the Behavioral Sciences. second ed. (Erlbaum., Hillsdale, NJ, 1988) 1988
  • Coucke et al, 2007 W. Coucke, N. Devleeschouwer, J.C. Libeer, J. Schiettecatte, M. Martin, J. Smitz. Accuracy and reproducibility of automated estradiol-17beta and progesterone assays using native serum samples: results obtained in the Belgian external assessment scheme. Hum. Reprod. 2007;22:3204-3209 Crossref
  • , NCT02280798. Endometrial gene expression in different protocols of endometrial preparation for embryo transfer (ERA).
  • Farhi et al, 2010 J. Farhi, A. Ben-Haroush, N. Andrawus, H. Pinkas, O. Sapir, B. Fisch, J. Ashkenazi. High serum oestradiol concentrations in IVF cycles increase the risk of pregnancy complications related to abnormal placentation. Reprod. Biomed. Online. 2010;21:331-337 Crossref
  • Fatemi, Van Vaerenbergh, 2015 H.M. Fatemi, I. Van Vaerenbergh. Significance of premature progesterone rise in IVF. Curr. Opin. Obstet. Gynecol. 2015;27:242-248 Crossref
  • Fleming, 2008 R. Fleming. Progesterone elevation on the day of hCG: methodological issues. Hum. Reprod. Update. 2008;14:391-392 Crossref
  • Griesinger et al, 2007 G. Griesinger, M. Weig, A. Schroer, K. Diedrich, E.M. Kolibianakis. Mid-cycle serum levels of endogenous LH are not associated with the likelihood of pregnancy in artificial frozen-thawed embryo transfer cycles without pituitary suppression. Hum. Reprod. 2007;22:2589-2593 Crossref
  • Groenewoud et al, 2012a E.R. Groenewoud, B.J. Kollen, N.S. Macklon, B.J. Cohlen. Spontaneous LH surges prior to HCG administration in unstimulated-cycle frozen-thawed embryo transfer do not influence pregnancy rates. Reprod. Biomed. Online. 2012;24:191-196 Crossref
  • Groenewoud et al, 2012b E.R. Groenewoud, N.S. Macklon, B.J. Cohlen, ANTARCTICA Trial Study Group. Cryo-thawed embryo transfer: natural versus artificial cycle. A non-inferiority trial. (ANTARCTICA trial). BMC Womens Health. 2012;12:27 Crossref
  • Groenewoud et al, 2016 E.R. Groenewoud, B.J. Cohlen, A. Al-Oraiby, E.A. Brinkhuis, F.J.M. Broekmans, J.P. de Bruin, G. van den Dool, K. Fleisher, J. Friederich, M. Goddijn, A. Hoek, D.A. Hoozemans, E.M. Kaaijk, C.A. Koks, J.S. Laven, P.J. van der Linden, A.P. Manger, E. Slappendel, T. Spinder, B.J. Kollen, N.S. Macklon. A randomized controlled, non-inferiority trial of modified natural versus artificial cycle for cryo-thawed embryo transfer. Hum. Reprod. 2016;31:1483-1492 Crossref
  • Haouzi et al, 2009 D. Haouzi, S. Assou, K. Mahmoud, S. Tondeur, T. Reme, B. Hedon, J. De Vos, S. Hamamah. Gene expression profile of human endometrial receptivity: comparison between natural and stimulated cycles for the same patients. Hum. Reprod. 2009;24:1436-1445 Crossref
  • Hoff et al, 1983 J.D. Hoff, M.E. Quigley, S.S. Yen. Hormonal dynamics at midcycle: a reevaluation. J. Clin. Endocrinol. Metab. 1983;57:792-796 Crossref
  • Joo et al, 2010 B.S. Joo, S.H. Park, An B.M., K.S. Kim, S.E. Moon, H.S. Moon. Serum estradiol levels during controlled ovarian hyperstimulation influence the pregnancy outcome of in vitro fertilization in a concentration-dependent manner. Fertil. Steril. 2010;93:442-446 Crossref
  • Labarta et al, 2011 E. Labarta, J.A. Martinez-Conejero, P. Alama, J.A. Horcajadas, A. Pellicer, C. Simon, E. Bosch. Endometrial receptivity is affected in women with high circulating progesterone levels at the end of the follicular phase: a functional genomics analysis. Hum. Reprod. 2011;26:1813-1825 Crossref
  • Laborde et al, 1976 N. Laborde, M. Carril, S. Cheviakoff, H.D. Croxatto, E. Pedroza, J.M. Rosner. The secretion of progesterone during the periovulatory period in women with certified ovulation. J. Clin. Endocrinol. Metab. 1976;43:1157-1163 Crossref
  • Lass et al, 1998 A. Lass, D. Peat, S. Avery, P. Brinsden. Histological evaluation of endometrium on the day of oocyte retrieval after gonadotrophin-releasing hormone agonist-follicle stimulating hormone ovulation induction for in-vitro fertilization. Hum. Reprod. 1998;13:3203-3205 Crossref
  • Lee et al, 2014 V.C. Lee, Li R.H., Chai J., T.W. Yeung, W.S. Yeung, P.C. Ho, E.H. Ng. Effect of preovulatory progesterone elevation and duration of progesterone elevation on the pregnancy rate of frozen-thawed embryo transfer in natural cycles. Fertil. Steril. 2014;101:1288-1293 Crossref
  • Li et al, 2011 Li R., Qiao J., Wang L., Li L., Zhen X., Liu P., Zheng X. MicroRNA array and microarray evaluation of endometrial receptivity in patients with high serum progesterone levels on the day of hCG administration. Reprod. Biol. Endocrinol. 2011;9:29 Crossref
  • Park et al, 2015a I. Park, Sun H., Chi H., S. Kim, J. Park, S. Kwak, K. Lee, J. Jo. Frozen thawd embryo transfer with simple monitoring does not impair IVF outcomes in natural cycle. Fertil. Steril. 2015;104:197
  • Park et al, 2015b J.H. Park, B.C. Jee, S.H. Kim. Factors influencing serum progesterone level on triggering day in stimulated in vitro fertilization cycles. Clin. Exp. Reprod. Med. 2015;42:67-71
  • Pauerstein et al, 1978 C.J. Pauerstein, C.A. Eddy, H.D. Croxatto, R. Hess, T.M. Siler-Khodr, H.B. Croxatto. Temporal relationships of estrogen, progesterone, and luteinizing hormone levels to ovulation in women and infrahuman primates. Am. J. Obstet. Gynecol. 1978;130:876-886 Crossref
  • Polotsky et al, 2009 A.J. Polotsky, J.L. Daif, S. Jindal, H.J. Lieman, N. Santoro, L. Pal. Serum progesterone on the day of human chorionic gonadotropin administration predicts clinical pregnancy of sibling frozen embryos. Fertil. Steril. 2009;92:1880-1885 Crossref
  • Requena et al, 2015 A. Requena, M. Cruz, A. Pacheco, J.A. Garcia-Velasco. Ongoing pregnancy rates in intrauterine insemination are affected by late follicular-phase progesterone levels. Fertil. Steril. 2015;104:879-883 Crossref
  • Roos et al, 2015 J. Roos, S. Johnson, S. Weddell, E. Godehardt, J. Schiffner, G. Freundl, C. Gnoth. Monitoring the menstrual cycle: comparison of urinary and serum reproductive hormones referenced to true ovulation. Eur. J. Contracept. Reprod. Health Care. 2015;20:438-450 Crossref
  • Ruiz-Alonso et al, 2013 M. Ruiz-Alonso, D. Blesa, P. Diaz-Gimeno, E. Gomez, M. Fernandez-Sanchez, F. Carranza, J. Carrera, F. Vilella, A. Pellicer, C. Simon. The endometrial receptivity array for diagnosis and personalized embryo transfer as a treatment for patients with repeated implantation failure. Fertil. Steril. 2013;100:818-824 Crossref
  • Shapiro et al, 2010 B.S. Shapiro, S.T. Daneshmand, F.C. Garner, M. Aguirre, C. Hudson, S. Thomas. Embryo cryopreservation rescues cycles with premature luteinization. Fertil. Steril. 2010;93:636-641 Crossref
  • Tomas et al, 2012 C. Tomas, B. Alsbjerg, H. Martikainen, P. Humaidan. Pregnancy loss after frozen-embryo transfer – a comparison of three protocols. Fertil. Steril. 2012;98:1165-1169 Crossref
  • Ubaldi et al, 1997 F. Ubaldi, C. Bourgain, H. Tournaye, J. Smitz, A. Van Steirteghem, P. Devroey. Endometrial evaluation by aspiration biopsy on the day of oocyte retrieval in the embryo transfer cycles in patients with serum progesterone rise during the follicular phase. Fertil. Steril. 1997;67:521-526 Crossref
  • Van Vaerenbergh et al, 2011 I. Van Vaerenbergh, H.M. Fatemi, C. Bourgain. Premature progesterone rise and gene expression. Hum. Reprod. 2011;26:2913 author reply 2194 Crossref
  • Venetis et al, 2013 C.A. Venetis, E.M. Kolibianakis, J.K. Bosdou, B.C. Tarlatzis. Progesterone elevation and probability of pregnancy after IVF: a systematic review and meta-analysis of over 60 000 cycles. Hum. Reprod. Update. 2013;19:433-457 Crossref
  • Venetis et al, 2015 C.A. Venetis, E.M. Kolibianakis, J.K. Bosdou, G.T. Lainas, I.A. Sfontouris, B.C. Tarlatzis, T.G. Lainas. Estimating the net effect of progesterone elevation on the day of hCG on live birth rates after IVF: a cohort analysis of 3296 IVF cycles. Hum. Reprod. 2015;30:684-691 Crossref
  • Weinerman, Mainigi, 2014 R. Weinerman, M. Mainigi. Why we should transfer frozen instead of fresh embryos: the translational rationale. Fertil. Steril. 2014;102:10-18 Crossref
  • Weissman et al, 2009 A. Weissman, D. Levin, A. Ravhon, H. Eran, A. Golan, D. Levran. What is the preferred method for timing natural cycle frozen-thawed embryo transfer?. Reprod. Biomed. Online. 2009;19:66-71 Crossref
  • Weissman et al, 2011 A. Weissman, E. Horowitz, A. Ravhon, Z. Steinfeld, R. Mutzafi, A. Golan, D. Levran. Spontaneous ovulation versus HCG triggering for timing natural-cycle frozen-thawed embryo transfer: a randomized study. Reprod. Biomed. Online. 2011;23:484-489 Crossref
  • Wong et al, 2014 K.M. Wong, S. Mastenbroek, S. Repping. Cryopreservation of human embryos and its contribution to in vitro fertilization success rates. Fertil. Steril. 2014;102:19-26
  • Zhu et al, 2015 Zhu H.Y., Xue Y.M., Yang L.Y., Jiang L.Y., Ling C., Tong X.M., Zhang S.Y. Slow freezing should not be totally substituted by vitrification when applied to day 3 embryo cryopreservation: an analysis of 5613 frozen cycles. J. Assist. Reprod. Genet. 2015;32:1371-1377 Crossref
Unlabelled box

Eva Groenewoud, MD, obtained her medical degree in 2007 from the University of Groningen. She is currently Consultant in Obstetrics and Gynaecology and a PhD student. Her PhD study, which she hopes to finish april 2017, focuses on endometrium preparation methods for frozen thawed embryo transfer.

 

Key message

 

Isolated elevated late follicular phase progesterone concentrations were found in over one-fifth of patients undergoing modified natural cycle frozen-thawed embryo transfer. These elevated progesterone concentrations, however, seem to be without any consequence for live birth and pregnancy rates.

Unlabelled box

Footnotes

e Amphia Hospital, Department of Obstetrics and Gynaecology, Breda, The Netherlands

f Meander Medical Center, Department of Obstetrics and Gynecology, Amersfoort, The Netherlands

g University Medical Centre Utrecht, Department for Reproductive Medicine, Utrecht, The Netherlands

h Jeroen Bosch Hospital, Department of Obstetrics and Gynaecology, ‘s Hertogenbosch, The Netherlands

i Albert Schweitzer Hospital, Department of Obstetrics and Gynaecology, Zwijndrecht, The Netherlands

j University Medical Center Nijmegen, Department of Obstetrics and Gynecology, Nijmegen, The Netherlands

k Noordwest Ziekenhuis, Department of Obstetrics and Gynecology, Den Helder, The Netherlands

l Centre for Reproductive Medicine, Academic Medical Centre, University of Amsterdam, The Netherlands

m Department of Obstetrics and Gynecology, University Medical Center Groningen, The Netherlands

n Medisch Spectrum Twente, Department of Obstetrics and Gynaecology, Enschede, The Netherlands

o Onze Lieve Vrouwe Gasthuis, Department of Obstetrics and Gynaecology, Amsterdam, The Netherlands

p Máxima Medical Centre, Department of Obstetrics and Gynaecology, Veldhoven, Netherlands

q Department of Obstetrics and Gynecology, Erasmus Medical Center Rotterdam, The Netherlands

r Deventer Hospital, Department of Obstetrics and Gynecology, Deventer, The Netherlands

s Diakonessenhuis, Department of Obstetrics and Gynecology, Utrecht, The Netherlands

t Catharina Hospital, Department of Obstetrics and Gynaecology, Eindhoven, The Netherlands

u Medisch Centrum Leeuwarden, Department of Obstetrics and Gynecology, Leeuwarden, The Netherlands

a Department of Obstetrics and Gynecology, Noordwest Ziekenhuis, Location Den Helder, PO box 501, 1800 AM Alkmaar, The Netherlands

b Department of Obstetrics and Gynecology, Academic Unit of Human Development and Health, University of Southampton, University Road, Southampton SO17 1BJ, UK

c Department of Obstetrics and Gynaecology, University Hospital Zealand, Roskilde, Denmark

d Isala Fertility Center, Isala Clinics. PO box 10400, 8000 GK, Zwolle, The Netherlands

* Corresponding author.

1 ANTARTICA Study Group: Amani Al-Oraiby, Amphia Hospital, Department of Obstetrics and Gynaecology, Breda, The Netherlands; Egbert A Brinkhuis, Meander Medical Center, Department of Obstetrics and Gynecology; Amersfoort, The Netherlands; Frank JM Broekmans, University Medical Centre Utrecht, Department for Reproductive Medicine, Utrecht, The Netherlands; Jan-Peter de Bruin, Jeroen Bosch Hospital, Department of Obstetrics and Gynaecology, ‘s Hertogenbosch, Netherlands; Grada van der Dool, Albert Schweitzer Hospital, Department of Obstetrics and Gynaecology, Zwijndrecht, The Netherlands; Kathrin Fleisher, University Medical Center Nijmegen, Department of Obstetrics and Gynecology. Nijmegen, The Netherlands; Jaap Friederich, Noordwest Ziekenhuis, Department of Obstetrics and Gynecology, Den Helder, The Netherlands; Mariëtte Goddijn, Centre for Reproductive Medicine, Academic Medical Centre, University of Amsterdam, The Netherlands; Annemieke Hoek, Department of Obstetrics and Gynecology, University Medical Center Groningen, The Netherlands; Diederik A Hoozemans, Medisch Spectrum Twente, Department of Obstetrics and Gynaecology, Enschede, The Netherlands; Eugenie M Kaaijk, Onze Lieve Vrouwe Gasthuis, Department of Obstetrics and Gynaecology, Amsterdam, The Netherlands; Carolina AM Koks, Máxima Medical Centre, Department of Obstetrics and Gynaecology, Veldhoven, Netherlands; Joop SE Laven, Department of Obstetrics and Gynecology, Erasmus Medical Center Rotterdam, The Netherlands; Paul JQ van der Linden, Deventer Hospital, Department of Obstetrics and Gynecology, Deventer, The Netherlands; A Petra Manger, Diakonessenhuis, Department of Obstetrics and Gynecology, Utrecht, The Netherlands; Minouche van Rumpste, Catharina Hospital, Department of Obstetrics and Gynaecology, Eindhoven, Netherlands; Taeke Spinder, Medisch Centrum Leeuwarden, Department of Obstetrics and Gynecology, Leeuwarden, The Netherlands.