Trial Design
This study was a multicenter, randomized, controlled, parallel-group clinical trial conducted at academic fertility centers of four public hospitals in the Chinese Mainland. The study enrolled a total of 732 women with a planned transfer of good-quality vitrified-warmed embryos between May 2020 and July 2021. The study was approved by the ethics committees of the participating hospitals. All couples provided voluntary written informed consent before participation. The study protocol was as previously published [23]. This trial was registered with chictr.org.cn (Identifier: ChiCTR2000033313). Final live birth outcomes were available in June 2021. The data were reviewed and approved by an external Data and Safety Monitoring Board.
Eligibility Criteria
Individuals were screened for eligibility. The inclusion criteria were: women aged 21–43 years at the time of IVF/ICSI treatment, participants undergoing IVF/ICSI treatment with a standard controlled ovarian stimulation (COS) protocol, participants had at least one frozen embryo remaining for transfer and had the initial FET cycle after a failed fresh ET. The exclusion criteria were: women with a body mass index (BMI) ≥ 28 Kg/m2, women with a natural cycle or mild stimulation for IVF/ICSI treatment, severe ovarian hyperstimulation syndrome (OHSS) during COS, history of unilateral oophorectomy or recurrent pregnancy loss, previously diagnosed with congenital or acquired uterine abnormalities, undergoing blastocyst biopsy for preimplantation genetic testing (PGT) or preimplantation genetic diagnosis (PGD), in vitro maturation (IVM) carried out, use of donor oocytes, or presence of hydrosalpinx, ovarian endometriosis cyst, or endometrial polyps that were not surgically treated. Eligible women signed a written consent form after counseling.
Randomization and Blinding
Women undergoing the initial FET cycle after a failed-fresh ET cycle were randomized into two groups (immediate FET and delayed FET) according to a computer-generated randomization list. Randomization was conducted 14 days after fresh ET (negative blood β-hCG test) for participants with a failed-fresh ET cycle. In addition, randomization was conducted by the project nurse, who was blinded from the entire recruitment and clinical management of the participants. The participants were allocated to four blocks and randomized using random numbers generated using SPSS software (Version 26.0, IBM Corp., Armonk) and were placed in opaque envelopes. Finally, participants were randomized into one of two groups: 1) The immediate FET group in which FET was performed after the first menstrual cycle following a failed-fresh ET cycle; 2) The delayed FET group in which FET was performed after two or more menstrual cycles following a failed-fresh ET cycle. Due to the nature of the intervention, we did not blind the participants, doctors, or trial outcome assessors on the assigned groups.
Trial Protocols
Women underwent IVF/ICSI treatment in the fertility centers as clinically indicated. Standard COS protocol with gonadotrophins was performed using a gonadotrophin-releasing hormone (GnRH) antagonist protocol. Furthermore, a fixed GnRH antagonist (GnRH-ant) (0.25mg, Cetrorelix; Merck Serono, Darmstadt, Germany) was used together with 112.5–225 IU/day of recombinant FSH (600IU, Puregon, Merck Sharp & Dohme B.V., Haarlem, Netherlands). The gonadotropin doses were determined based on individual patient characteristics. The oocyte retrieval was conducted under ultrasound transvaginal guidance, 34–36 hours after triggering with recombinant hCG (250µg, Ovitrelle®, Merck Serono S.p.A., Italy). After that, conventional IVF/ICSI was carried out depending on the partner’s semen quality per the standard protocols at the centers. Normal fertilization was assessed (a second polar body and two pronuclei) after 16–18 hours of conventional insemination or ICSI. Notably, an embryo with at least seven cells (grades Ⅰ and Ⅱ) on the third day after oocyte retrieval was defined as good quality. In addition, embryos with at least six cells with fragments < 50% were considered frozen. According to the standard protocol, all "good" embryos were vitrified using the cryopreservation method on the third day.
Hormone replacement treatment (HRT) was used for endometrial preparation. Treatment with estradiol valerate (E2, Progynova, Schering AG, Berlin, Germany) was commenced on the third day of the menstrual cycle at 4-6mg daily for 10–12 days. Moreover, vaginal progesterone (90mg, 8% Crinone, Merck-Serono, Switzerland) was administered at a dose of 90mg per day, upon the endometrial layer reaching a thickness of 8 mm as revealed by pelvic ultrasound scanning. FET using three-day-old embryos was also scheduled on the fourth day of commencing treatment with vaginal progesterone. Furthermore, one or two embryos with the best morphology were transferred using a soft embryo-transfer catheter under ultrasound guidance. Finally, the serum β-hCG levels were determined fourteen days following FET. The hormone therapy was stopped when the serum β-hCG was negative. Luteal phase support (using estradiol valerate 6 mg daily and vaginal progesterone gel 90 mg daily) was continued in a transvaginal-ultrasound confirmed pregnancy until ten weeks of gestation. The maternal and neonatal outcomes of the initial FET were obtained through a review of medical records.
Assessment for depression and stress
Standardized and validated questionnaires were administered to the study participants to assess depression and stress before the FET. The participants were asked to complete self-administered questionnaires determining major depression index[24, 25]. The symptoms were rated on a 6-point Likert scale. The scale determines how long the symptoms have been present during the past 14 days. The scale ranges from 0 (no depression) to 50 (extreme depression). It corresponds to the diagnostic criteria for depression in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (including an additional item of low self-esteem) and the International Classification of Diseases and Related Health Problems (ICD-10). Study participants were categorized into four in accordance to the ICD-10 depression categories: mild (two to three core symptoms and two to three additional symptoms); moderate (two core symptoms and four or more additional symptoms); and severe (three core symptoms and five or more additional symptoms). Participants who did not fulfill any of these criteria were categorized as ‘no depression’.
Emotional stress was assessed using the Cohen's Perceived Stress Scale. The participants completed a 10-item self-report questionnaire consisting of a 5-point Likert scale. The final score ranged from 0 (no stress) to 40 (extreme stress). The PSS scale was designed to assess the degree to which respondents found their lives unpredictable, uncontrollable, or overwhelming [26]. The PSS scale was designed for comparisons between -groups and is not a diagnostic tool. The PSS scores were dichotomized into < 19 or ≥ 19, where the latter represents ‘high stress’ based on a previous publication [27].
Trial Outcomes
The primary outcome was an ongoing pregnancy rate, which included natural conception. Ongoing pregnancy was defined as an intrauterine detectable fetal heartbeat after more than twelve weeks of gestation. Secondary outcomes were positive pregnancy rates, pregnancy loss rates, embryo implantation rates, ectopic pregnancy rates, multiple pregnancy rates, live birth rates, pregnancy-related complications, and obstetric complications. Maternal and neonatal outcomes, including preeclampsia, gestational diabetes, gestational hypertension, preterm delivery, low birth weight, infants born small or large for gestational age, and congenital anomalies, were recorded in pregnancies that continued beyond twenty weeks. Supplementary Table S5 provides definitions of all secondary outcomes.
Statistical analysis
The trial was designed as a non-inferiority study. The PASS software version 11.0 (NCSS, LLC. Kaysville, Utah, USA.) was used to determine the sample sizes for both groups. Notably, the ongoing pregnancy rate per embryo transfer was about 30% based on data from our reproductive center at the time of the trial design. The sample size calculation revealed that 329 women were to be included in each trial group to provide an 80% power to detect a minimal difference of 10% points between the immediate and delayed FET groups for the primary outcome of ongoing pregnancy (40% vs. 30%) at a two-sided α level of 0.05. Finally, the trial planned to include 732 women, with 366 participants in each arm to account for an expected 10% lost to follow-up.
Per protocol (PP) principle was used for the primary statistical analysis. Primary and secondary outcomes were assessed by comparing outcomes after the initial FET cycle. All women in an intention-to-treat (ITT) analysis were accounted for in the group to which they were randomised, irrespective of whether or not they received the treatment. The PP analysis included all women who adhered strictly to the study protocol. The as-treated analysis included women randomized to the immediate FET group who had immediate FET at the first menstrual cycle after a failed fresh ET and women randomized to the delayed FET group who received delayed FET at the second or subsequent menstrual cycle after a failed fresh ET. The ongoing pregnancy rate was determined, and relative risk was used to determine the difference. PP analyses were also performed for all reproductive outcomes. Continuous data were compared using Student’s t-test, and the results were presented as mean (standard deviation) or median (interquartile range). Categorical data were assessed using χ2 analysis and Fisher's exact test for expected frequencies less than five. A two-sided P value of less than 0.05 was considered statistically significant. Data analyses were conducted using SPSS version 26.0 and R statistical package version 4.0.0.
Multivariable logistic regression analysis was performed to determine variables independently associated with ongoing pregnancy, live birth, clinical pregnancy, or positive pregnancy and affecting outcomes. Female gender, age (< 35 yrs., ≥ 35 yrs.), anti-müllerian hormone (AMH) (< 1.2ng/ml, ≥ 1.2ng/ml), BMI (< 24 kg/m2, ≥ 24 kg/m2), follicle-stimulating hormone (FSH) (< 10 UI/L, ≥ 10 UI/L), antral follicle count (AFC) (< 10, ≥ 10), endometrial thickness before FET, No. of oocytes retrieved (≤ 9, > 9), No. of transferred frozen-thawed embryos (single, double), method of fertilization (IVF, ICSI), moderate/severe depression prior to FET (yes, no), and high-stress levels prior to FET (yes, no) were included in the analysis.