Abstract
Purpose
To evaluate the obstetric and perinatal outcomes of three routine endometrial preparation protocols in women with PCOS who underwent frozen embryo transfer (FET).
Methods
This was a retrospective study in women with PCOS who underwent FET in an academic reproductive medical center. A total of 2710 cycles were enrolled and classified into three groups according to different endometrial preparation protocols; human menopausal gonadotropin (HMG), letrozole + HMG, or hormone replacement therapy (HRT).
Results
The stimulation groups had reduced risks of hypertensive disorders of pregnancy (HDP), large for gestational age (LGA) infants, and cesarean delivery than the HRT group. After adjustment for different confounder combinations in the two models, the frequencies of LGA and HDP in the letrozole + HMG group and the HMG group were still significantly lower than those in the HRT group. The letrozole + HMG group exhibited a reduced risk of LGA than HMG group after adjustment of confounders. A trend toward risk reductions in HDP and LGA was observe in turns of HRT, HMG, and letrozole + HMG groups, and the trends were statistically significant (Ptrend = 0.031 and 0.001).
Conclusion
In patients with PCOS, ovarian stimulation protocols for endometrial preparation are associated with reduced risks of HDP and LGA compared to HRT cycles. The use of letrozole could further reduce risk of LGA compared to HMG only protocol. We propose that ovarian stimulation protocols can be used widely for endometrial preparation in FET cycles in women with PCOS, especially with the use of letrozole.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-023-02956-z.
Keywords: Frozen embryo transfer, Hormonal replacement, Hypertensive disorders of pregnancy, Large for gestational age, Letrozole, Polycystic ovary syndrome
Introduction
Polycystic ovary syndrome (PCOS) is the most common cause of anovulatory infertility [1], affecting 5%–20% women of childbearing age globally [2]. In vitro fertilization (IVF) embryo transfer (ET) is an effective treatment in PCOS women after recurrent failures of ovulation induction, or if there are coexisting indications such as tubal or male infertility factors. Notably however, women with PCOS undergoing IVF treatment are more likely to develop ovarian hyperstimulation syndrome (OHSS) [3, 4]. As a result, a freezing-all strategy is more commonly used in PCOS women to reduce the risk of OHSS. Accordingly, frozen embryo transfer (FET) is a routine practice in women with PCOS undergoing IVF.
The aim of FET protocols is to adequately prepare the endometrium for successful implantation of frozen-thawed embryos [5]. The endometrium must be in synchronization with embryo development before transfer [6, 7]. Various cycle regimens have been designed for endometrial preparation in women with PCOS undergoing FET cycles, including the natural cycle, ovarian stimulation (OS), and hormone replacement therapy (HRT) protocols. A natural cycle protocol is rarely used in PCOS women because of oligo-anovulation or irregular menstruation [8], whereas HRT and ovarian stimulation regimens are commonly used. In HRT cycles, a proper endometrial environment is created by programmed administration of exogenous estradiol and progesterone, and the protocol is favored due to its lower cancellation rate and greater flexibility with regard to scheduling. OS protocols are used to facilitate follicular development with ovulation induction medications and mimic the natural process of ovulation. Such protocols are usually an inferior choice due to the relatively frequent clinical visits required for ultrasound scanning and hormone testing. In a recent meta-analysis, however, stimulated protocols were associated with a significantly higher live birth rate and a significantly lower miscarriage rate than HRT protocols in PCOS women who underwent FET [9]. This suggests that OS cycles may achieve better pregnancy and live birth outcomes than HRT cycles in women with PCOS undergoing FET cycles. The associated better obstetrics and perinatal outcomes are arguably even more important. Given the greater risk of adverse pregnancy outcomes such as preeclampsia, very preterm birth, and gestational diabetes mellitus in PCOS women [10–12], it is important to consider the health of the woman throughout the gestational period and the safety of newborns before deciding the endometrial preparation strategy.
Accumulating evidence indicates that pregnancy complications and adverse perinatal outcomes such as hypertension disorder, cesarean section, and large for gestational age (LGA) infants are related to HRT cycles in the general population [13–15]. According to a comprehensive epidemiological study [16], pregnancies following HRT cycles had a higher prevalence of hypertensive disorders of pregnancy (HDP) as compared to pregnancies following a natural FET cycle. These findings provided evidence in favor of the notion that the absence of the corpus luteum (CL) in HRT cycles may contribute, at least partially, to the heightened risk of HDP associated with FET [17, 18].
However, data on how various endometrial preparation techniques affect maternal and neonatal outcomes in patients with PCOS are scant and controversial. Compared to HRT-FET, the use of letrozole was associated with a lower prevalence of HDP in PCOS patients in a large retrospective study [19]. In another study that evaluated PCOS-afflicted women undergoing FET, there were no significant differences in obstetric or neonatal outcomes between human menopausal gonadotropin (HMG) and HRT groups [20]. Stimulation regimens such as letrozole + HMG and HMG alone have been variously used in previous FET studies. To date, no study has investigated letrozole + HMG, HMG alone, and HRT at the same time in PCOS women undergoing FET cycles. Thus with respect to obstetric and perinatal outcomes, whether adding letrozole to HMG is better than HMG alone in women with PCOS when HRT is used as a reference remains unknown. Clarification of this is pertinent due to the increased prevalence of FET [21], particularly in PCOS patients. The current study compared the obstetric and perinatal outcomes of endometrial preparation protocols including letrozole + HMG, HMG alone, and HRT in women with PCOS who underwent FET.
Materials and methods
Study design and patients
A retrospective study was conducted at the Center for Reproductive Medicine at Shandong University. All FET cycles for women with PCOS from May 2008 to October 2019 were reviewed for potential inclusion. Only embryo transfer cycles in women with PCOS who had singleton pregnancies were enrolled. Modified Rotterdam criteria known as the China diagnosis criteria were used to diagnose PCOS. The criteria included menstrual abnormalities (irregular uterine bleeding, oligomenorrhea, or amenorrhea) combined with either hyperandrogenism or polycystic ovaries, as validated in the Chinese population [22]. Other inclusion criteria were giving birth to a singleton, blastocyst transfer, and being aged 20–35 years. Women with an anomalous intrauterine cavity, a history of recurrent miscarriage or repeated implantation failure, or diabetes and hypertension prior to pregnancy were excluded, as were women who underwent preimplantation genetic testing cycles. Patients enrolled in the study were divided into three groups based on the protocols used for endometrium preparation; an HMG alone group, a letrozole + HMG group, or an HRT alone group (Fig. 1).
Fig. 1.
Flowchart. PCOS, polycystic ovary syndrome; HMG, human menopausal gonadotropin; LE, letrozole; HRT, hormone replacement therapy; FET, frozen embryo transfer; PGT: preimplantation genetic testing
Endometrial preparation
In the HRT group, 4–8 mg of oral estrogen valerate (Progynova, Bayer, Germany) was given daily for at least 10 days, beginning from day 2 to day 5 of the menstrual cycle. Endometrial thickness was monitored via transvaginal ultrasound after 10–12 days of medication, along with serum estradiol and progesterone. When the endometrial thickness reached 8 mm, progesterone was administered both orally (40 mg/day; Duphastone, Solvay, Netherlands) and vaginally (200 mg/day; Utrogestan, Besins Manufacturing, Belgium) along with oral estradiol. Blastocyst transfer was conducted 5 days after progesterone administration.
In the HMG group, 75 IU/day HMG (Le Baode, Livzon) was administered intramuscularly starting on day 3–5 of the menstrual cycle. Serum luteinizing hormone (LH), estradiol (E2), and progesterone (P) levels, as well as ultrasound examination were used to monitor follicle and endometrium development. Ovulation occurred either spontaneously or was induced by human chorionic gonadotropin (hCG). After ovulation was confirmed, oral progesterone (30–40 mg/day; Duphastone, Solvay, Netherlands) was given for luteal support. The ovulation day was set as day 0, and the blastocyst was warmed and transferred on day 5.
In the letrozole + HMG group, patients received 2.5–5.0 mg/day letrozole (Liquzuo, Heng-Rui, Jiangshu, China) for 5 consecutive days beginning from day 2–5. After letrozole, 75.0 IU/day HMG was given to continually stimulate follicle and endometrium development, with dosage increases of 37.5–75 IU every 5–7 days as required. The same procedures as in the HMG group were used for follicular monitoring, triggering, and luteal phase support.
Luteal support continued until 11–12 weeks of gestation. On vaginal ultrasound, clinical pregnancy was detected via the presence of a gestational sac with a fetal heartbeat. Once clinical pregnancy was confirmed, trained nurses collected follow-up information such as perinatal complications, gestational weeks, birth date, delivery mode, and newborn gender and birth weight and entered it into the patient’s electronic medical records.
Outcome definitions
The primary outcomes were the rates of maternal complications and neonatal outcomes. Maternal complications included HDP (we included only patients with preeclampsia or gestational hypertension. Preeclampsia is diagnosed as new-onset hypertension that develops after 20 weeks of gestation and is accompanied by proteinuria or evidence of end-organ dysfunction. Gestational hypertension is similar to preeclampsia; however, the condition is defined as hypertension alone after 20 gestational weeks.), gestational diabetes mellitus (GDM; defined as any degree of abnormal glucose metabolism during pregnancy that was not clearly overt diabetes prior to gestation), postpartum hemorrhage (defined as cumulative blood loss ≥ 1000 mL or blood loss accompanied by symptoms of hypovolemia within 24 h after childbirth), abnormal placentation (including placenta previa and placental abruption), oligohydramnios (defined as the single deepest pocket ≤ 2 cm or amniotic fluid index(AFI) ≤ 5 cm), and polyhydramnios (defined as the single deepest pocket ≥8 cm or AFI≥ 25 cm). Neonatal outcomes were recorded in conjunction with neonatal sex and birthweight and included preterm birth (prior to 37 weeks), small for gestational age (SGA) and LGA (respectively defined as birth weight within the 10th and 90th percentiles of gender-specific reference values for Chinese newborns) [23], and stillbirth (intrauterine death at gestation > 20 weeks or birthweight > 500 g).
Statistical analysis
All analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 25.0 (IBM Corp.). Means and standard deviations were calculated for continuous variables with normal distributions, and between-group differences were compared via one-way analysis of variance. Least significant difference adjustment was used to correct p values. Medians and ranges were calculated for continuous variables with non-normal distributions, and comparisons were conducted via the Kruskal-Wallis test. Frequencies and percentages were calculated for categorical variables, and distributions were analyzed via the chi-square test or Fisher’s exact test. p < 0.05 was deemed to indicate statistical significance. Multiple logistic and linear regression analyses were performed to evaluate relationships between the type of endometrial preparation and maternal and perinatal outcomes after adjustment for confounding factors. Different regression models were adjusted for different confounder combinations, including maternal age, body mass index, type of infertility, LH/follicle-stimulating hormone (FSH) ratio, antral follicle count, insemination method, number of embryos transferred, basal serum estradiol, basal serum T0, endometrial thickness, parity, GDM, and HDP. In Model 1, we adjusted for a subset of baseline variables with a p-value < 0.1 among the three groups, as well as factors that could potentially impact the outcomes. In Model 2, we adjusted for a subset of baseline variables with a p-value < 0.1 among the three groups, as well as factors that could potentially impact the outcomes. In Model 2, we included the variables of GDM and HDP as additional factors for adjustment. These variables were reported to have an impact on perinatal outcomes.
Results
A total of 2710 patients with PCOS who fulfilled the inclusion criteria were included in the final analysis. Of these, 1451 underwent an HRT cycle and 1259 underwent a stimulated cycle (73% HMG alone and 27% letrozole + HMG). Baseline patient characteristics are presented in Table 1. Patients in the HRT group were significantly younger than those in the HMG group and the letrozole + HMG group. Women in the HRT group had a lower body mass index than those in the HMG group and the letrozole + HMG group. Basal serum estradiol, total testosterone levels, and antral follicle count (AFC) differed significantly in the three groups. There were no significant differences in type of infertility, basal serum FSH, basal serum LH, LH/FSH ratio, FET cycle rank, fertilization method, endometrial thickness, blastocyst development stage, number of embryos transferred, or embryo grade between the three groups.
Table 1.
Baseline characteristics of the three groups
| HMG (n = 918) | LE + HMG (n = 341) | HRT (n = 1451) | p value | |
|---|---|---|---|---|
| Age (years) | 28.65 ± 3.04 | 28.59 ± 3.02 | 28.28 ± 3.05 | 0.010* |
| Body mass index | 24.99 ± 4.09 | 24.95 ± 3.76 | 24.43 ± 3.94 | 0.002* |
| Infertility type (%) | 0.060 | |||
| Primary | 592 (64.5) | 211 (61.9) | 984 (67.8) | |
| Secondary | 326 (35.5) | 130 (38.1) | 467 (32.2) | |
| Basal serum FSH (IU/L) | 5.56 (4.84–6.54) | 5.59 (4.80–6.54) | 5.65 (4.87–6.52) | 0.555 |
| Basal serum LH (IU/L) | 8.05 (5.30–12.51) | 8.77 (5.92–12.91) | 8.21 (5.38–12.74) | 0.208 |
| Basal serum estradiol (pg/mL) | 37.30 (27.80–48.25) | 39.60 (31.15–51.30) | 37.00 (28.60–48.86) | 0.038* |
| Basal serum T0 (ng/dL) | 36.83 (27.11–49.55) | 41.00 (29.42–55.07) | 38.25 (27.99–50.45) | 0.016* |
| LH/FSH ratio | 1.68 ± 1.08 | 1.75 ± 0.96 | 1.71 ± 1.05 | 0.558 |
| AFC | 26.80 ± 12.49 | 28.63 ± 12.58 | 27.77 ± 11.10 | 0.028* |
| FET cycle rank (%) | 0.126 | |||
| First | 812 (88.5) | 300 (88.0) | 1317 (90.8) | |
| Second | 99 (10.8) | 38 (11.1) | 117 (8.1) | |
| High order | 7 (0.8) | 3 (0.9) | 17 (1.2) | |
| Fertilization method (%) | 0.097 | |||
| IVF | 666 (72.5) | 263 (77.1) | 1093 (75.3) | |
| ICSI | 200 (21.8) | 58 (17.0) | 300 (20.7) | |
| IVF + ICSI | 52 (5.7) | 20 (5.9) | 58 (4.0) | |
| Endometrial thickness (mm) | 0.94 ± 0.15 | 0.96 ± 0.15 | 0.94 ± 0.15 | 0.056 |
| Blastocyst development stage (%) | 0.355 | |||
| D5 | 788 (85.8) | 303 (88.9) | 1263 (87.0) | |
| D6 | 130 (14.2) | 38 (11.1) | 188 (13.0) | |
| Number of embryos transferred (%) | 0.068 | |||
| 1 | 823 (89.7) | 304 (89.1) | 1257 (86.6) | |
| 2 | 95 (10.3) | 37 (10.9) | 194 (13.4) | |
| Parity | 0.068 | |||
| 1 | 825(89.9) | 303(88.9) | 1259(86.8) | |
| 2 | 93(10.1) | 38(11.1) | 192(13.2) | |
| Embryo grade (%) | 0.109 | |||
| Excellent (AA) | 193 (19.1) | 73 (19.3) | 286 (17.4) | |
| Good (AB, BA) | 193 (19.1) | 63 (16.7) | 260 (15.8) | |
| Average (BB, BC) | 627 (61.9) | 242 (64.0) | 1099 (66.8) |
HMG human menopausal gonadotropin, LE letrozole, HRT hormone replacement therapy, FET frozen embryo transfer, AFC antral follicle count, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection, LH luteinizing hormone, FSH follicle-stimulating hormone
*p-value < 0.05 was considered as statistically significant
Comparisons of maternal and neonatal outcomes between the three groups are shown in Table 2. The rate of HDP in the HRT group (9.6%) was higher than that in the letrozole + HMG group (5.9%). The HMG group had the lower rate of cesarean delivery (71.4%) than the HRT group (76.3%). The risks of LGA were 28.2% in the letrozole + HMG group, 31.9% in the HMG alone group, and 36.8% in the HRT group.
Table 2.
Maternal complications and perinatal outcomes in the three groups
| HMG (n = 918) | LE + HMG (n = 341) | HRT (n = 1451) | p value | HMG vs. LE + HMG | HMG vs. HRT | HRT vs. LE + HMG | |
|---|---|---|---|---|---|---|---|
| Maternal complications | |||||||
| GDM (%) | 98 (10.7) | 40 (11.7) | 132 (9.1) | 0.232 | |||
| HDP (%) | 71 (7.7) | 20 (5.9) | 140 (9.6) | 0.046* | 0.255 | 0.111 | 0.027* |
| PPH (%) | 1 (0.1) | 1 (0.3) | 6 (0.4) | 0.399 | |||
| Abnormal placentation (%) | 8 (0.9) | 1 (0.3) | 16 (1.1) | 0.364 | |||
| Oligohydramnios (%) | 10 (1.1) | 2 (0.6) | 18 (1.2) | 0.582 | |||
| Polyhydramnios (%) | 2 (0.2) | 1 (0.3) | 4 (0.2) | > 0.999 | |||
| Mode of delivery (%) | 0.013* | 0.975 | 0.007* | 0.052 | |||
| Vaginal | 263 (28.6) | 98 (28.7) | 344 (23.7) | ||||
| Cesarean delivery | 655 (71.4) | 243 (71.3) | 1107 (76.3) | ||||
| Gestational age (weeks) | 38.29 ± 1.79 | 38.31 ± 2.35 | 38.43 ± 1.64 | 0.184 | |||
| Neonatal outcomes | |||||||
| Sex of neonate (%) | 0.290 | ||||||
| Male | 480 (52.3) | 174 (51.0) | 798 (55.0) | ||||
| Female | 435 (47.4) | 164 (48.1) | 642 (44.2) | ||||
| Unknown | 3 (0.3) | 3 (0.9) | 11 (0.8) | ||||
| Birth weight (g) | 3434.2 ± 557.0 | 3412.8 ± 591.2 | 3480.9 ± 576.5 | 0.051 | |||
| Weight category (%) | |||||||
| < 1500 g | 5 (0.5) | 5 (1.5) | 17 (1.2) | 0.211 | |||
| 1500–2500 g | 47 (5.1) | 15 (4.4) | 53 (3.7) | 0.223 | |||
| 4000–4500 g | 132 (14.4) | 46 (13.5) | 239 (16.5) | 0.227 | |||
| > 4500 g | 15 (1.6) | 6 (1.8) | 31 (2.1) | 0.668 | |||
| SGA (%) | 17 (1.9) | 9 (2.6) | 18 (1.2) | 0.147 | |||
| LGA (%) | 293 (31.9) | 96 (28.2) | 534 (36.8) | 0.002* | 0.199 | 0.015* | 0.003* |
| Stillbirth (%) | 3 (0.3) | 3 (0.9) | 11 (0.8) | 0.354 | |||
| Preterm birth (%) | 86 (9.4) | 36 (10.6) | 140 (9.6) | 0.817 |
HMG human menopausal gonadotropin, LE letrozole, HRT hormone replacement therapy, GDM gestational diabetes mellitus, HDP hypertensive disorders of pregnancy, PPH postpartum hemorrhage, SGA small for gestational age, LGA large for gestational age
Abnormal placentation includes placenta previa, placental abruption, and placenta implantation
*p-value < 0.05 was considered as statistically significant
The results of multiple regression analysis of obstetrics outcomes are shown in Table 3 and Supplement table. In Table 3, the HRT group was used as a reference. After adjustment for different confounder combinations in the two models, letrozole + HMG was associated with a lower risk of HDP (adjusted odds ratio [aOR] 0.59, 95% confidence interval [CI] 0.35–0.98) than HRT. The frequencies of LGA in the letrozole + HMG group and the HMG group were still lower than that the HRT group (Model 1 aOR 0.76, CI 0.63–0.91 and aOR 0.64, CI 0.48–0.84, respectively; Model 2 aOR 0.74, CI 0.62–0.89 and aOR 0.62, CI 0.47–0.82, respectively). The values of Ptrend in LGA and HDP were significant, indicating that a trend of risk reductions in HDP and LGA was observed in turns of HRT group, HMG group, and letrozole+ HMG group. The Supplement table showed regression analysis of two ovarian stimulation groups; the letrozole + HMG group exhibited a reduced risk of LGA than HMG group after adjustment of confounders (Model 1 aOR 0.84, CI 0.63–0.99; Model 2 aOR 0.84, CI 0.62–0.97).
Table 3.
Unadjusted and adjusted regression models of maternal complications and perinatal outcomes
| Crude OR/β (95% CI) | Model 1 OR/β (95% CI) | Model 2 OR/β (95% CI) | |
|---|---|---|---|
| GDM | |||
| HRT | Ref. | Ref. | |
| HMG | 1.19 (0.91–1.57) | 1.20 (0.89–1.61) | _ |
| HMG + LE | 1.33 (0.91–1.93) | 1.20 (0.79–1.81) | |
| Ptrend | – | 0.221 | |
| HDP | |||
| HRT | Ref. | Ref. | _ |
| HMG | 0.79 (0.58–1.06) | 0.72 (0.52–0.99) * | |
| HMG + LE | 0.58 (0.36–0.95) | 0.59 (0.35–0.98) * | |
| Ptrend | – | 0.031* | |
| Gestational age (weeks) | |||
| HRT | Ref. | Ref. | Ref. |
| HMG | (-0.14–0.21) | -0.85 (-0.14–0.20) | (-0.17–0.18) |
| HMG + LE | -0.03 (-0.40–0.09) | 0.27 (-0.38–0.14) | -0.23 (-0.39–0.11) |
| Ptrend | – | 0.541 | 0.921 |
| LGA | |||
| HRT | Ref. | Ref. | Ref. |
| HMG | 0.81 (0.68–0.96) * | 0.76 (0.63–0.91) * | 0.74 (0.62–0.89) * |
| HMG + LE | 0.67 (0.52–0.87) * | 0.64 (0.48–0.84) * | 0.62 (0.47–0.82) * |
| Ptrend | – | 0.001* | <0.001* |
| Preterm birth | |||
| HRT | Ref. | Ref. | Ref. |
| HMG + LE | 1.11 (0.75–1.63) | 1.04 (0.69–1.58) | 1.11 (0.73–1.69) |
| HMG | 0.97 (0.73–1.28) | 1.00(0.74–1.35) | 1.04 (0.77–1.42) |
| Ptrend | – | 0.998 | 0.753 |
OR odds ratio, CI confidence interval, HMG human menopausal gonadotropin, LE letrozole, HRT hormone replacement therapy, GDM gestational diabetes mellitus, HDP hypertensive disorders of pregnancy, LGA large for gestational age, Ref. reference
Model 1, adjusted for maternal age, body mass index, type of infertility, insemination method, number of embryos transferred, basal serum estradiol, basal serum T0, endometrial thickness, AFC, and LH/FSH ratio, parity Model 2, adjusted for Model 1 + GDM, and HDP
*p-value < 0.05 was considered as statistically significant
Discussion
In this retrospective study of 2710 PCOS patients who underwent FET, the two stimulation groups exhibited lower risks of HDP and LGA than the HRT group. When the HRT group was used as a reference, the risk of HDP and LGA was further reduced after letrozole supplementation compared to HMG alone, and there was a statistically significant difference between the three groups.
In recent years, concerns have been raised about whether assisted reproductive technology has adverse effects on the health of mothers and their offspring. PCOS is associated with increased risks of GDM, preeclampsia, preterm birth, and LGA due to metabolic and endocrine disorders [10–12]. HRT regimes are the most common endometrial preparation protocols used in PCOS patients, but they are also associated with adverse prenatal complications such as hypertension disorder, cesarean section, and LGA infants [24]. It is therefore necessary to develop a safer and more effective FET regimen for use in PCOS patients. In a previous meta-analysis [9], stimulated protocols achieved a better live birth rate than an HRT protocol in PCOS women undergoing FET. Notably however, few studies have focused on endometrial preparation regimes in PCOS patients in terms of maternal and perinatal complications, and their findings are inconsistent [19, 20]. Zhang, J., et al. reported that in a study that included both singleton and twin pregnancies in PCOS women undergoing FET cycles, letrozole-induced FET cycles were associated with a reduced risk of hypertensive disorders of pregnancy compared with HRT. In that study, most women were transferred cleavage-stage embryos, which was not the case in our current study. Also in this study, the OS protocol only enrolled letrozole protocol, with or without HMG. Another study [20] involving 1720 frozen embryo transfers in PCOS women compared pregnancy outcomes in natural, HMG-stimulated, and HRT cycles with single blastocyst transfer. There were no significant differences in maternal or infant complications between the three groups. This may have been due to the comparatively small size of the OS protocol group. In that study, the primary outcome was live birth, and obstetric and perinatal outcomes were secondary outcomes. Different embryo stages or an insufficient sample size may have resulted in those researcher drawing conclusions that differed from those indicated by the current study. Thus, additional large-sample studies incorporating letrozole + HMG, HMG, and HRT regimes and FET cycles in PCOS women, and only blastocyst stage embryo transfer, are required.
PCOS is evidently associated with poor pregnancy outcomes and a higher risk of pregnancy complications [25–27]. In the current study, the prevalence of pregnancy complications such as HDP and GDM, and some perinatal outcomes such as preterm birth and LGA were greater than those reported in a previous retrospective cohort analysis of 14,373 singletons delivered after FET to non-PCOS women at our reproductive center [28]. These adverse outcomes may be attributed to characteristic changes in PCOS women such as insulin resistance, hyperandrogenism, metabolic disorder, and obesity. PCOS increases the expression of androgen receptors such as Cyr61 and decreases the expression of the integrin avb3 [29]. Integrins are believed to have a role in initial attachment of the embryo to the endometrium, and reduced integrin expression during the luteal phase of the menstrual cycle is thought to have a negative effect on blastocyst implantation [30, 31]. Inadequate embryo implantation and shallow cytotrophoblast invasion may result in a degree of spiral artery remodeling failure, leading to the genesis of preeclampsia [32].
The etiology and pathophysiological mechanisms of apparent differences in perinatal outcomes between stimulated cycles and HRT cycles are focused on the presence of corpus luteum (CL). A concurrent clinical physiology investigation indicated that a lack of CL affects maternal circulation, particularly central and peripheral arterial compliance, as indicated by a reduced fall in carotid-femoral pulse wave velocity (cfPWV) and an increase in carotid-femoral pulse wave transit time (cfPWTT) [33]. In that same study, compared to women who conceived with one or several CLs, women who conceived without a CL had greater risks of preeclampsia and preeclampsia with severe characteristics. CL hormones are reportedly significant regulators of maternal adaptations in early pregnancy, as part of an endocrinological continuum that begins in the luteal phase [17]. Estradiol and progesterone are replaced during HRT-FET procedures, but other vasoactive CL products such as relaxin—which is significant with respect to maternal cardiovascular adaption to pregnancy—are absent [34–37] . This may be the principal reason for the high incidence of HDP in HRT cycles.
In the current study, adding letrozole to HMG cycles could further reduce the risk of LGA, beyond the reduction achieved via HMG alone. And in the trend analysis of the three groups, the risks of LGA and HDP were in the descending order HRT, HMG, and letrozole + HMG. This is an interesting finding, and the precise underlying mechanisms warrant further investigation. The mechanism of action of letrozole in OS involves increased ovarian follicle exposure to FSH via reduced negative feedback effects of ovarian estradiol production [38]. Letrozole is commonly initiated at approximately cycle day 3, when the follicles are 6–8 mm in diameter and are FSH-dependent, and administration will last for 5 days. Endogenous FSH inspiration coincides with the demand for 6–8-mm follicle development, and letrozole taken on day 3 increases intraovarian androgens in 6–8-mm follicles. Six-millimeter follicles are equipped with a high level of androgen receptors, and increased androgen levels at this time promote granulosa cell mitosis and the induction of FSH receptors, and help make follicles more resistant to atresia. On cycle day 9–10, there is a reduction in letrozole availability due to letrozole’s half-life, or because of enhanced aromatase expression primarily due to increased FSH and FSH receptor. This is followed by an increase in estrogen and a reduction in follicle androgen, which promote granulosa cell mitosis in larger follicles of 11–13 mm [39–41]. Compared to an HMG alone stimulation protocol, letrozole + HMG may better promote endogenous FSH and benefit granulosa cell mitosis, providing a basis for adequate CL function. Another reason for the better maternal and perinatal outcomes associated with letrozole + HMG may be related to a decreased estradiol levels compared to HMG alone. Supraphysiological steroid serum concentrations in stimulated cycles may adversely affect LH secretion and induce a CL function defect. It has also been proposed that supraphysiological estradiol levels affect endometrial receptivity [42, 43]. The high chance of single follicle development and relatively decreased estradiol levels associated with the letrozole protocol help to avoid the flaws of an HMG alone protocol.
Letrozole may enhance maternal outcomes by increasing the expression of uterine receptivity markers such as integrins, L-selectin, leukemia inhibitory factor, and pinopod formation [44, 45]. Integrins are evidently involved in initial attachment of the embryo to the endometrium and promote implantation [44]. Healthy pregnancy establishment requires adequate CL formation and correct implantation. The letrozole + HMG protocol with appropriate hormone levels and better endometrial receptivity may facilitate decidualization, benefit extravillous trophoblast cell invasion, and promote placentation. The above considerations may account for the decreased incidence of HDP and LGA in letrozole + HMG cycles.
HRT cycles have gained popularity among patients and physicians because to their facilitation of scheduling embryo transfers and reduction in the necessity for frequent monitoring. However, the potential benefits of HRT cycles may overshadow the heightened likelihood of negative maternal and newborn outcomes. Given the greater risk of adverse pregnancy outcomes in PCOS women, it is important to consider the health of the woman throughout the gestational period and the safety of newborns before deciding the endometrial preparation strategy. Ovarian stimulation protocols during FET exhibited a higher degree of safety in comparison to HRT interventions. It is not absolutely clear if the use of letrozole in addition to HMG could reduce the risks of HDP and LGA even further, for there is still less direct evidence in women with PCOS. Future prospective randomized studies are needed to verify this point. Therefore, we propose that ovarian stimulation protocols can be used widely for endometrial preparation in FET cycles in women with PCOS, especially with the use of letrozole.
The main strengths of the current study are the large size of the cohort and the strong evidence of a link between endometrial preparation regimens and maternal and perinatal problems associated with frozen transfer cycles in PCOS women. The present study provides evidence from a fresh perspective on current practice patterns and related clinical consequences, and it is the biggest study in this field to date by a wide margin. Stringent inclusion criteria were used in an effort to address confounding and bias, and logistic regression analysis was conducted to strengthen the reliability of the findings. The study also had some limitations. Not all potential confounders could be addressed, due to the inherent bias associated with the retrospective study design. For example, because of a lack of sufficiently detailed records, maternal behavioral factors such as smoking, alcohol use, and exercise habits could not be adjusted for. Endometrial preparation type was not randomly assigned, it was determined based on patients’ and physicians’ preferences. Lastly, FET cycle pretreatments and supplements such as metformin, aspirin, and vitamin D3 which may influence perinatal outcomes were not analyzed. A well-designed, large-scale, randomized controlled trial including different OS protocols is needed to further investigate a safer endometrial preparation method for women with PCOS undergoing FET.
Conclusions
In patients with PCOS, ovarian stimulation protocols for endometrial preparation are associated with reduced risks of HDP and LGA compared to HRT cycles. The use of letrozole could further reduce risk of LGA compared to HMG only protocol, and a trend of risk reductions in HDP and LGA was observed in turns of HRT group, HMG group, and letrozole+ HMG group. We propose that ovarian stimulation protocols can be used widely for endometrial preparation in FET cycles in women with PCOS, especially with the use of letrozole. Future prospective randomized studies are needed to verify the results of the current study.
Supplementary information
(DOCX 22 kb)
Acknowledgements
The authors thank the medical workers in the research group at the Reproductive Hospital of Shandong University and also thank the information engineer for assembling the data.
Abbreviations
- PCOS
Polycystic ovary syndrome
- FET
Frozen embryo transfer
- HMG
Human menopausal gonadotropin
- HRT
Hormone replacement therapy
- HDP
Hypertensive disorders of pregnancy
- LGA
Large for gestational age
- IVF
In vitro fertilization
- ET
Embryo transfer
- OHSS
Ovarian hyperstimulation syndrome
- OS
Ovarian stimulation
- LH
Luteinizing hormone
- GDP
Gestational diabetes mellitus
- SGA
small for gestational age
- PPH
Postpartum hemorrhage
- FSH
Follicle-stimulating hormone
- AFC
Antral follicle counta
- OR
Adjusted odds ratio
- CI
Confidence interval
- CL
Corpus luteum
- cfPWV
Carotid-femoral pulse wave velocity
- cfPWTT
Carotid-femoral pulse wave transit time
- ICSI
Intracytoplasmic sperm injection
- PGT
Preimplantation genetic testing
Funding
This study was supported by the National Key Research and Development Program of China (2021YFC2700404).
Declarations
Ethics approval
The ethics committee at the Reproductive Hospital of Shandong University approved the study protocol.
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Adams J, Polson DW, Franks S. Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J (Clin Res Ed). 1986;293(6543):355–359. doi: 10.1136/bmj.293.6543.355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Azziz R, Carmina E, Chen Z, Dunaif A, Laven JS, Legro RS, Lizneva D, Natterson-Horowtiz B, Teede HJ, Yildiz BO. Polycystic ovary syndrome. Nat Rev Dis Primers. 2016;2:16057. doi: 10.1038/nrdp.2016.57. [DOI] [PubMed] [Google Scholar]
- 3.Delvigne A, Rozenberg S. Epidemiology and prevention of ovarian hyperstimulation syndrome (OHSS): a review. Hum Reprod Update. 2002;8(6):559–577. doi: 10.1093/humupd/8.6.559. [DOI] [PubMed] [Google Scholar]
- 4.Tang T, Glanville J, Hayden CJ, White D, Barth JH, Balen AH. Combined lifestyle modification and metformin in obese patients with polycystic ovary syndrome. A randomized, placebo-controlled, double-blind multicentre study. Hum Reprod. 2006;21(1):80–89. doi: 10.1093/humrep/dei311. [DOI] [PubMed] [Google Scholar]
- 5.Groenewoud ER, Cantineau AE, Kollen BJ, Macklon NS, Cohlen BJ. What is the optimal means of preparing the endometrium in frozen-thawed embryo transfer cycles? A systematic review and meta-analysis. Hum Reprod Update. 2013;19(5):458–470. doi: 10.1093/humupd/dmt030. [DOI] [PubMed] [Google Scholar]
- 6.Casper RF, Yanushpolsky EH. Optimal endometrial preparation for frozen embryo transfer cycles: window of implantation and progesterone support. Fertil Steril. 2016;105(4):867–872. doi: 10.1016/j.fertnstert.2016.01.006. [DOI] [PubMed] [Google Scholar]
- 7.Groenewoud ER, Cohlen BJ, Al-Oraiby A, Brinkhuis EA, Broekmans FJ, de Bruin JP, van den Dool G, Fleisher K, Friederich J, Goddijn M, et al. A randomized controlled, non-inferiority trial of modified natural versus artificial cycle for cryo-thawed embryo transfer. Hum Reprod. 2016;31(7):1483–1492. doi: 10.1093/humrep/dew120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Escobar-Morreale HF. Polycystic ovary syndrome: definition, aetiology, diagnosis and treatment. Nat Rev Endocrinol. 2018;14(5):270–284. doi: 10.1038/nrendo.2018.24. [DOI] [PubMed] [Google Scholar]
- 9.Zhang Y, Wu L, Li TC, Wang CC, Zhang T, Chung JPW. Systematic review update and meta-analysis of randomized and non-randomized controlled trials of ovarian stimulation versus artificial cycle for endometrial preparation prior to frozen embryo transfer in women with polycystic ovary syndrome. Reprod Biol Endocrinol. 2022;20(1):62. doi: 10.1186/s12958-022-00931-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Roos N, Kieler H, Sahlin L, Ekman-Ordeberg G, Falconer H, Stephansson O. Risk of adverse pregnancy outcomes in women with polycystic ovary syndrome: population based cohort study. Bmj. 2011;343:d6309. doi: 10.1136/bmj.d6309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pan ML, Chen LR, Tsao HM, Chen KH. Relationship between polycystic ovarian syndrome and subsequent gestational diabetes mellitus: a nationwide population-based study. PLoS One. 2015;10(10):e0140544. doi: 10.1371/journal.pone.0140544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Palomba S, de Wilde MA, Falbo A, Koster MP, La Sala GB, Fauser BC. Pregnancy complications in women with polycystic ovary syndrome. Hum Reprod Update. 2015;21(5):575–592. doi: 10.1093/humupd/dmv029. [DOI] [PubMed] [Google Scholar]
- 13.Jing S, Li XF, Zhang S, Gong F, Lu G, Lin G. Increased pregnancy complications following frozen-thawed embryo transfer during an artificial cycle. J Assist Reprod Genet. 2019;36(5):925–933. doi: 10.1007/s10815-019-01420-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wang B, Zhang J, Zhu Q, Yang X, Wang Y. Effects of different cycle regimens for frozen embryo transfer on perinatal outcomes of singletons. Hum Reprod. 2020;35(7):1612–1622. doi: 10.1093/humrep/deaa093. [DOI] [PubMed] [Google Scholar]
- 15.Pan Y, Li B, Wang Z, Wang Y, Gong X, Zhou W, Shi Y. Hormone replacement versus natural cycle protocols of endometrial preparation for frozen embryo transfer. Front Endocrinol (Lausanne). 2020;11:546532. doi: 10.3389/fendo.2020.546532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Saito K, Kuwahara A, Ishikawa T, Morisaki N, Miyado M, Miyado K, Fukami M, Miyasaka N, Ishihara O, Irahara M, et al. Endometrial preparation methods for frozen-thawed embryo transfer are associated with altered risks of hypertensive disorders of pregnancy, placenta accreta, and gestational diabetes mellitus. Hum Reprod. 2019;34(8):1567–1575. doi: 10.1093/humrep/dez079. [DOI] [PubMed] [Google Scholar]
- 17.Conrad KP, Baker VL. Corpus luteal contribution to maternal pregnancy physiology and outcomes in assisted reproductive technologies. Am J Physiol Regul Integr Comp Physiol. 2013;304(2):R69–R72. doi: 10.1152/ajpregu.00239.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Singh B, Reschke L, Segars J, Baker VL. Frozen-thawed embryo transfer: the potential importance of the corpus luteum in preventing obstetrical complications. Fertil Steril. 2020;113(2):252–257. doi: 10.1016/j.fertnstert.2019.12.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhang J, Wei M, Bian X, Wu L, Zhang S, Mao X, Wang B. Letrozole-induced frozen embryo transfer cycles are associated with a lower risk of hypertensive disorders of pregnancy among women with polycystic ovary syndrome. Am J Obstet Gynecol. 2021;225(1):59.e51–59.e59. doi: 10.1016/j.ajog.2021.01.024. [DOI] [PubMed] [Google Scholar]
- 20.Man Y, Bian Y, Zhao S, Zhao R, Xu X, Wei D, Li L, Chen ZJ, Zhao H. The effect of different endometrial preparations on women with polycystic ovary syndrome undergoing initial frozen embryo transfer: a historical cohort analysis. Acta Obstet Gynecol Scand. 2021;100(6):1116–1123. doi: 10.1111/aogs.14058. [DOI] [PubMed] [Google Scholar]
- 21.De Geyter C, Calhaz-Jorge C, Kupka MS, Wyns C, Mocanu E, Motrenko T, Scaravelli G, Smeenk J, Vidakovic S, Goossens V. ART in Europe, 2014: Results generated from European registries by ESHRE: The European IVF-monitoring Consortium (EIM) for the European Society of Human Reproduction and Embryology (ESHRE) Hum Reprod. 2018;33(9):1586–1601. doi: 10.1093/humrep/dey242. [DOI] [PubMed] [Google Scholar]
- 22.Revised Consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS) Hum Reprod 2004. 2003;19(1):41–47. doi: 10.1093/humrep/deh098. [DOI] [PubMed] [Google Scholar]
- 23.Dai L, Deng C, Li Y, Zhu J, Mu Y, Deng Y, Mao M, Wang Y, Li Q, Ma S, et al. Birth weight reference percentiles for Chinese. PLoS One. 2014;9(8):e104779. doi: 10.1371/journal.pone.0104779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hu KL, Zhang D, Li R. Endometrium preparation and perinatal outcomes in women undergoing single-blastocyst transfer in frozen cycles. Fertil Steril. 2021;115(6):1487–1494. doi: 10.1016/j.fertnstert.2020.12.016. [DOI] [PubMed] [Google Scholar]
- 25.Katulski K, Czyzyk A, Podfigurna-Stopa A, Genazzani AR, Meczekalski B. Pregnancy complications in polycystic ovary syndrome patients. Gynecol Endocrinol. 2015;31(2):87–91. doi: 10.3109/09513590.2014.974535. [DOI] [PubMed] [Google Scholar]
- 26.Falbo A, Rocca M, Russo T, D'Ettore A, Tolino A, Zullo F, Orio F, Palomba S. Changes in androgens and insulin sensitivity indexes throughout pregnancy in women with polycystic ovary syndrome (PCOS): relationships with adverse outcomes. J Ovarian Res. 2010;3:23. doi: 10.1186/1757-2215-3-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kjerulff LE, Sanchez-Ramos L, Duffy D. Pregnancy outcomes in women with polycystic ovary syndrome: a metaanalysis. Am J Obstet Gynecol. 2011;204(6):558.e551–558.e556. doi: 10.1016/j.ajog.2011.03.021. [DOI] [PubMed] [Google Scholar]
- 28.Wang Z, Liu H, Song H, Li X, Jiang J, Sheng Y, Shi Y. Increased risk of pre-eclampsia after frozen-thawed embryo transfer in programming cycles. Front Med (Lausanne). 2020;7:104. doi: 10.3389/fmed.2020.00104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.MacLaughlan SD, Palomino WA, Mo B, Lewis TD, Lininger RA, Lessey BA. Endometrial expression of Cyr61: a marker of estrogenic activity in normal and abnormal endometrium. Obstet Gynecol. 2007;110(1):146–154. doi: 10.1097/01.AOG.0000269047.46078.28. [DOI] [PubMed] [Google Scholar]
- 30.Nardo LG, Bartoloni G, Di Mercurio S, Nardo F. Expression of alpha(v)beta3 and alpha4beta1 integrins throughout the putative window of implantation in a cohort of healthy fertile women. Acta Obstet Gynecol Scand. 2002;81(8):753–758. [PubMed] [Google Scholar]
- 31.Tei C, Maruyama T, Kuji N, Miyazaki T, Mikami M, Yoshimura Y. Reduced expression of alphavbeta3 integrin in the endometrium of unexplained infertility patients with recurrent IVF-ET failures: improvement by danazol treatment. J Assist Reprod Genet. 2003;20(1):13–20. doi: 10.1023/A:1021254620888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Conrad KP. Evidence for corpus luteal and endometrial origins of adverse pregnancy outcomes in women conceiving with or without assisted reproduction. Obstet Gynecol Clin North Am. 2020;47(1):163–181. doi: 10.1016/j.ogc.2019.10.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.von Versen-Höynck F, Schaub AM, Chi YY, Chiu KH, Liu J, Lingis M, Stan Williams R, Rhoton-Vlasak A, Nichols WW, Fleischmann RR, et al. Increased preeclampsia risk and reduced aortic compliance with in vitro fertilization cycles in the absence of a corpus luteum. Hypertension. 2019;73(3):640–649. doi: 10.1161/HYPERTENSIONAHA.118.12043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Conrad KP. Maternal vasodilation in pregnancy: the emerging role of relaxin. Am J Physiol Regul Integr Comp Physiol. 2011;301(2):R267–R275. doi: 10.1152/ajpregu.00156.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Novak J, Danielson LA, Kerchner LJ, Sherwood OD, Ramirez RJ, Moalli PA, Conrad KP. Relaxin is essential for renal vasodilation during pregnancy in conscious rats. J Clin Invest. 2001;107(11):1469–1475. doi: 10.1172/JCI11975. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Debrah DO, Novak J, Matthews JE, Ramirez RJ, Shroff SG, Conrad KP. Relaxin is essential for systemic vasodilation and increased global arterial compliance during early pregnancy in conscious rats. Endocrinology. 2006;147(11):5126–5131. doi: 10.1210/en.2006-0567. [DOI] [PubMed] [Google Scholar]
- 37.Smith MC, Murdoch AP, Danielson LA, Conrad KP, Davison JM. Relaxin has a role in establishing a renal response in pregnancy. Fertil Steril. 2006;86(1):253–255. doi: 10.1016/j.fertnstert.2005.11.070. [DOI] [PubMed] [Google Scholar]
- 38.Rose BI, Brown SE. A review of the physiology behind letrozole applications in infertility: are current protocols optimal? J Assist Reprod Genet. 2020;37(9):2093–2104. doi: 10.1007/s10815-020-01892-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Fauser BC, Van Heusden AM. Manipulation of human ovarian function: physiological concepts and clinical consequences. Endocr Rev. 1997;18(1):71–106. doi: 10.1210/edrv.18.1.0290. [DOI] [PubMed] [Google Scholar]
- 40.Kristensen SG, Mamsen LS, Jeppesen JV, Bøtkjær JA, Pors SE, Borgbo T, Ernst E, Macklon KT, Andersen CY. Hallmarks of human small antral follicle development: implications for regulation of ovarian steroidogenesis and selection of the dominant follicle. Front Endocrinol (Lausanne). 2017;8:376. doi: 10.3389/fendo.2017.00376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Franks S, Hardy K. Androgen action in the ovary. Front Endocrinol (Lausanne). 2018;9:452. doi: 10.3389/fendo.2018.00452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Tavaniotou A, Albano C, Smitz J, Devroey P. Impact of ovarian stimulation on corpus luteum function and embryonic implantation. J Reprod Immunol. 2002;55(1-2):123–130. doi: 10.1016/s0165-0378(01)00134-6. [DOI] [PubMed] [Google Scholar]
- 43.Bülow NS, Skouby SO, Warzecha AK, Udengaard H, Andersen CY, Holt MD, Grøndahl ML, Nyboe Andersen A, Sopa N, Mikkelsen ALE, et al. Impact of letrozole co-treatment during ovarian stimulation with gonadotrophins for IVF: a multicentre, randomized, double-blinded placebo-controlled trial. Hum Reprod. 2022;37(2):309–321. doi: 10.1093/humrep/deab249. [DOI] [PubMed] [Google Scholar]
- 44.Miller PB, Parnell BA, Bushnell G, Tallman N, Forstein DA, Higdon HL, 3rd, Kitawaki J, Lessey BA. Endometrial receptivity defects during IVF cycles with and without letrozole. Hum Reprod. 2012;27(3):881–888. doi: 10.1093/humrep/der452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Ganesh A, Chauhan N, Das S, Chakravarty B, Chaudhury K. Endometrial receptivity markers in infertile women stimulated with letrozole compared with clomiphene citrate and natural cycles. Syst Biol Reprod Med. 2014;60(2):105–111. doi: 10.3109/19396368.2013.862316. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DOCX 22 kb)