The impact of coronavirus disease 2019 on frozen-thawed embryo transfer outcomes

Study design and participants

This was a single-centre retrospective cohort study. Data in this study were retrieved from the electronic medical records system of the Center for Reproductive Medicine at Tianjin Central Hospital of Obstetrics and Gynaecology. Prior to IVF treatment, all infertile couples underwent a systematic infertility-related assessment, including collection of the couple’s medical history, assessment of the female serum endocrine level, ultrasound examination, fallopian tube patency examination, hysteroscopy if necessary, and routine semen analysis. All of the data were recorded in the electronic medical records system. The study group consisted of patients who underwent FET cycles between 1 January 2023 and 31 March 2023. Oocyte retrieval for all patients was performed prior to SARS-CoV-2 infection. On the day of the FET, each patient volunteered to complete a questionnaire about COVID-19 while waiting for embryo transfer in the waiting area. The questionnaire included questions concerning whether the patient had been infected with SARS-CoV-2, how the infection was confirmed, the time of infection, the symptoms at the time of infection, the duration of symptoms, and whether the menstrual pattern changed in terms of the cycle length and bleeding volume after infection (details in Supplemental Information). According to the questionnaire information provided by the participants, some women were confirmed to be infected with COVID-19 by a real-time PCR assay conducted via nucleic acid testing of nasopharyngeal swabs from medical institutions, and some were confirmed via a positive self-administered rapid SARS-CoV-2 antigen test kit. However, in the context of the nationwide outbreak of the epidemic and the serious shortage of medical resources, some women did not undergo PCR or antigen testing to confirm COVID-19 infection. To avoid the loss of this population, patients with fever and/or respiratory symptoms and an epidemiological history were classified into the symptom-based group in this study. The control group consisted of patients with no history of past infection who underwent IVF treatments with FET during the same period and those who were treated between 1 January 2021 and 31 March 2021 (SARS-CoV-2 nondiagnosed, nonvaccinated).

The inclusion criteria for participants were (1) aged 20–42 years and (2) receiving only the first FET cycle following recovery from COVID-19. Women who fulfilled the following conditions were excluded: (1) women with a history of three or more failed FETs; (2) women with repeated spontaneous miscarriages (defined as two or more previous spontaneous pregnancy losses); (3) infertile couples with chromosomal abnormalities; (4) women with uterine factors, including uterine malformations, intrauterine adhesions, uterine submucosal myoma, uterine surgical history and other factors affecting endometrium receptivity, such as hydrosalpinx; (5) male use of a sperm donor; and (6) sequential FET cycles.

This study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the ethics committee of Tianjin Central Hospital of Obstetrics and Gynaecology (approval number: ZY2021004). Written informed consent was obtained from each participant prior to data collection.

Endometrial preparation and embryo transfer

Endometrial preparation protocols were individually tailored by the treating physician, largely dependent on the patient’s menstruation and clinical condition, and included both natural cycles and hormone replacement cycles, per usual institutional protocols. The embryos were frozen and thawed according to the instructions provided with the vitrified freezing/resuscitation solution (Japan Kato). After thawing, good-quality embryos on day 3 were defined as those with 7–10 cells, homogeneous or slightly uneven blastomere sizes, and no fragments or fragments ≤10% (Desai et al., 2000). The blastocysts were graded according to the Gardner and Schoolcraft quality system, and the good-quality blastocysts were those with an expansion score ≥4, an inner cell mass ≥B, and a trophectoderm score ≥B (Gardner et al., 2000). Embryo transfers were performed by highly experienced senior physicians. Up to two embryos were transferred per cycle. Luteal support was carried out from the day of transfer via dydrogesterone administered orally (20 mg/d; Duphaston, Abbott Biologicals, San Diego, CA, USA) or vaginally (90 mg/d; Crinone, Merck Serono, Switzerland) and maintained until 10 weeks of gestation, when pregnancy was established.

Pregnancy follow-up

Serum hCG levels were initially measured 14 days after FET. Transvaginal ultrasound was performed 6∼7 weeks after FET to confirm clinical pregnancy. Subsequent pregnancy outcomes were obtained by doctors or nurses who contacted the patients by phone and recorded them in the electronic health system.

Observation indicators

The primary outcome was live birth, which was defined as the delivery of any neonate with signs of life at 24 weeks of gestation or later (Wang et al., 2023). The secondary outcomes were as follows. Biochemical pregnancy was defined as an increase in the serum hCG level to greater than 25 mIU/mL 14 days after embryo transfer. The implantation rate was defined as the ratio of the total number of amniotic sacs in all patients to the total number of embryos transferred in all patients. Clinical pregnancy was defined as the presence of a gestational sac with or without a foetal heartbeat observed via transvaginal ultrasound. Early abortion was defined as the spontaneous termination of an intrauterine pregnancy before 12 weeks. Ongoing pregnancy was defined as an intrauterine sac with a foetal heartbeat that continued for more than 12 weeks. Preterm birth was defined as <37 weeks gestation. Low birth weight (LBW) referred to neonates born weighing less than 2,500 g. Preterm premature rupture of membranes (PPROM) was the rupture of membranes during pregnancy before 37 weeks of gestation. Gestational diabetes mellitus (GDM) was defined as a fasting plasma glucose level ≥5.1 mmol/L or a glucose tolerance test of ≥10.0 mmol/L (1 h after sugar intake) or ≥8.5 mmol/L (2 h after sugar intake). Preeclampsia was defined as having a systolic pressure ≥ 140 mmHg and/or diastolic pressure ≥ 90 mmHg after 20 weeks of gestation and proteinuria ≥ 0.3 grams in a 24-hour urine sample or random urine protein (+).

Statistical analyses

Continuous variables are expressed as the means ± SDs and were assessed for normality. The Shapiro–Wilk test was used to evaluate the normality of the data. A normal distribution diagram (frequency histogram) was used to visually check whether the data were approximately normally distributed. Normally or approximately normally distributed data were compared using Student’s t-test, whereas skewed data were compared via the Mann–Whitney U test. Categorical variables were summarized via counts and percentages, and Pearson’s chi-square test or Fisher’s exact test, when appropriate, was used to compare differences between groups. Binary logistic regression analysis was performed to determine the effects of COVID-19 on reproductive outcomes, and the data are presented as adjusted odds ratios (ORs) and 95% CIs. Variables that differed across the groups and those of clinical relevance were included as confounders in the logistic model. Possible confounders, including female age, body mass index (BMI), infertility duration, endometrial thickness, number of embryos transferred, transfer of ≥1 good-quality embryo and the luteal support protocol, were included in the logistic regression model. COVID-19 was stratified by the number of COVID-19 symptoms, duration of symptoms, days from COVID-19 infection to transfer and whether menstruation changed, and logistic regression analysis was used to further evaluate the impact of COVID-19 infection on the live birth rate. Statistical analysis was performed with SPSS (version 26.0. SPSS, Inc., Chicago, IL, USA). All tests were two-sided, and P < 0.05 was considered to indicate statistical significance.

Patient information

From 1 January 2023 to 31 March 2023 and 1 January 2021 to 31 March 2021, we collected 306 cycles and 270 cycles, respectively, at our IVF centre. After applying the inclusion and exclusion criteria, 160 and 167 cycles were included in the two groups. The outcomes for all participants were tracked to determine whether the patients had a live birth (Fig. 1).

The clinical characteristics of the women in the two groups are shown in Table 1. The general characteristics, including female age at oocyte pick-up (OPU) or FET, BMI, infertility duration, infertility cause, infertility type, and number of previous pregnancies and deliveries, did not significantly differ between the two groups.

For the COVID-19 group, we used three methods to determine whether women were infected with SARS-CoV-2, including nucleic acid testing (25.6%), antigen testing (41.2%), and symptom-based diagnosis (33.1%). The distribution of symptoms per woman varied from 1 to 10 when infected with SARS-CoV-2. Only one patient had 17 symptoms. We used isometric grouping to divide the number of symptoms into three grades as follows: 1∼3 (45.6%), 4∼6 (40.0%), and ≥ 7 (14.4%). The symptom duration was classified according to the natural week as follows:1∼7 days (36.2%), 7∼14 days (36.9%), 14∼21 days (10.0%), 21∼28 days (8.1%), and ≥28 days (8.8%). We also stratified the COVID-19 group by time from SARS-CoV-2 infection to embryo transfer into ≤2 months (52.5%) and >2 months (47.5%). In our study, an indicator of changes in menstruation after COVID-19 was also collected, and 31.8% of female patients reported self-perceived menstrual changes. Details of menstrual changes included longer or shorter menstrual cycles, increased or decreased menstrual volume, and changes in menstrual periods. Among the patients, 12.5% reported a prolonged menstrual cycle. Unfortunately, we did not collect information on how long the menstrual cycle was shortened or prolonged or whether such changes reached a level of clinical concern (Table 1).

FET characteristics and clinical outcomes

The clinical characteristics related to FET in both groups are presented in Table 2. There were no differences in the endometrial preparation protocol, endometrial thickness, stage, or quality of the transferred embryos between the two groups. However, the number of embryos transferred, according to the proportions of single embryo transfer and double embryo transfer, significantly differed between the two groups (P < 0.05).

The pregnancy and perinatal outcomes of FET in both groups are presented in Table 3. There was no significant difference in the live birth rates between the two groups (43.8% vs. 43.1%, P > 0.05). Other pregnancy-related indicators, including the implantation rates (41.2% vs. 39.3%), biochemical pregnancy rates (56.3% vs. 56.3%), clinical pregnancy rates (52.5% vs. 52.1%), ongoing pregnancy rates (46.9% vs. 44.3%) and early abortion rates (8.3% vs. 12.6%) were also similar (P > 0.05). The incidences of maternal and neonatal complications such as PPROM, GDM, preeclampsia, preterm birth, and LBW were also similar between the two groups (P > 0.05). Binary logistic regression analysis was performed to determine the effect of a history of COVID-19 on pregnancy outcomes. Table 4 presents the results of both the crude analyses and the analyses adjusted for age, BMI, infertility duration, endometrial thickness, number of embryos transferred, transfer of ≥1 good-quality embryo and the luteal support protocol. The results revealed that the live birth rate did not decrease after SARS-CoV-2 infection after adjusting for confounding factors, and the adjusted OR and 95% CI were 0.953 (0.597∼1.523). A history of SARS-CoV-2 infection did not increase the risk of early miscarriage, and the adjusted OR and 95% CI were 0.694 (0.250∼1.931).

A logistic regression model for pregnancy rates was applied to the subgroups (Fig. 2). The COVID-19 group was divided into subgroups by FET age (<35 years and ≥35 years) and the number of embryos transferred (single embryo transfer and double embryo transfer) for the binary logistic regression analysis while adjusting for confounding factors such as age, BMI, infertility duration, endometrial thickness, number of embryos transferred, transfer of ≥1 good-quality embryo and the luteal support protocols were adjusted.

The results revealed that SARS-CoV-2 infection had no adverse effect on the live birth rate, regardless of whether the patients were stratified by age or the number of embryos transferred. The detailed results are shown in Fig. 2.

Subgroup analysis of SARS-CoV-2-infected women

A stratifying analysis of the impact of COVID-19 on the live birth rate was performed. As demonstrated in Fig. 3, neither the number nor duration of symptoms at the time of COVID-19 nor the interval from infection to transplant had a significant influence on the live birth rate. A total of 14 patients (8.8%) had symptoms lasting ≥28 days. The live birth rate in this group was only 28.5% (4/14); although there was a downwards trend, which was probably affected by the sample size, the difference was not statistically significant. Changes in menstruation after infection were also not associated with the live birth rate in FET cycles (see Fig. 3 for details).

Conclusion

In conclusion, our work provides evidence that SARS-CoV-2 infection after oocyte retrieval has no detrimental effects on subsequent FET outcomes. Women can consider receiving FET after recovering from COVID-19 without having to wait 2∼3 months or more. Future larger studies are needed to validate our observations.