Remember me
The coronavirus disease 2019 (COVID-19) pandemic, caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2), has significantly impacted various medical fields since its outbreak in 2019.[1] Following the development of an mRNA-based anti-COVID-19 vaccine in 2020, the vaccination rate has rapidly increased to over 87% in Korea.[2] The sub-effect of vaccination on other medical fields has not been fully elucidated because anti-COVID-19 vaccination was emergently instituted based on the catastrophic effects of the infectious pandemic.[3] Therefore, COVID-19 infection potentially detrimental effect has become a cause for concern across various aspects, including reproductive medicine. Moreover, concerning women health, it has significantly affected fertility-related decision-making.[4]
Since the first anti-COVID-19 vaccine was developed in 2020, several vaccine types have been introduced, including the Pfizer BioNTech mRNA (BNT162b2 mRNA), Moderna (mRNA-1273), and AZD1222 (ChAdOx1 nCov-19) vaccines, which have demonstrated reasonable effectiveness and efficacy.[5] Anti-COVID-19 vaccination is recommended for individuals older than 16 years of age, and the vaccinated population has grown rapidly over the past few years.[6] Furthermore, repeated vaccination has become increasingly imperative owing to the emergence of COVID-19 variants by mutation and immune escape ability.[7,8]
Considering the COVID-19 pandemic status and its requirement for additional vaccinations, many women are hesitant to get an additional vaccine dose due to uncertainties about the efficacy and safety of additional vaccine dose (boosting dose). Moreover, there are concerns about the potentially detrimental effect on fertility and women health such as increased menstrual flow, irregular menstrual cycles, and breakthrough bleeding which were not proven in clearly designed research.[9–12] Controlled ovarian hyperstimulation is process in vitro fertilization (IVF) technique for collecting additional mature oocytes from couples with infertility, and it potentially causes supraphysiologic levels of steroid hormones. According to majority of studies and review articles published so far, the impact of COVID-19 vaccination on IVF outcomes is mostly limited. However, studies on anti-COVID-19 vaccination, type of vaccination and its influence on IVF outcomes are few. Therefore, this study aimed to identify IVF outcome, the effect of follicular function, its steroidogenesis, and response to ovarian stimulation according to COVID-19 infection and vaccination.
2. Methods 2.1. Study designThis prospective study was performed at a single center where assisted reproductive technology (ART) is routinely utilized from December 2021 to October 2022. This study concept and procedure were approved by a public institutional review board (IRB), namely, the Korea National Institute for Bioethics Policy (IRB number: P01-202111-31-006). All consecutive patients provided informed consent and underwent oocyte retrieval in preparation for IVF procedures. The inclusion criteria were as follows: age between 19 and 44 years and planned ovarian stimulation for IVF, intracytoplasmic sperm injection, or oocyte cryopreservation. Individuals were excluded for the following reasons: age younger than 19 years or older than 44 years, failure to consent to study participation, poor ovarian response (according to the Bologna criteria or less than 3 mature follicles on oocyte-retrieval day), and having a male partner undergoing testicular sperm extraction or percutaneous testicular sperm fine-needle aspiration. To control the potential confounding factors, procedures such as ovarian stimulation and oocyte retrieval with follicular fluid sampling were consistently conducted by the same physician. Then, the collected samples were blinded and then analyzed by a different researcher. Finally, the research findings were de-identified, and a third researcher analyzed the data.
2.2. ParticipantsForty-five consecutive patients were finally included in this study and divided into the following 3 groups according to COVID-19 infection and vaccination: control (n = 5, patients without a history of COVID-19 infection and vaccination), COVID + (n = 10, patients with a history of COVID-19 infection), and vaccination (n = 30, patients with a history of vaccination [patients who completed second doses of vaccination of BNT162b2 and mRNA-1273 or at least one dose of Ad.26.COV2.S or ChAdOx1 nCov-19 or have a history of cross-vaccination with BNT162b2 or mRNA-1273] but without a history of COVID-19 infection). The vaccination group was further subdivided into the following 2 subgroups based on vaccine type: BNT162b2 and others. COVID-19 infection history and vaccination status were investigated after participants had provided informed consent. The baseline characteristics included age, body mass index (BMI), smoking status, antral follicle count (AFC), anti-Müllerian hormone, duration of infertility (total period (in years) of not occurring pregnancy after unprotected intercourse), previous parity, indication of IVF, age of male partner and sperm parameters.
2.3. Serum and follicular-fluid collectionTo investigate steroidogenesis and alterations in antibody titer during ovarian stimulation according to the vaccine type and infection history, blood samples were collected on cycle days 2 and 3 as well as on the ovulation trigger day (35–36 hours before oocyte retrieval) and centrifuged at 3000 g for 7 minutes. On oocyte- retrieval day, the dominant follicle/s was/were aspirated into an empty tube up to a total volume exceeding 5 mL following oocyte isolation. The FF was centrifuged at 1500 g for 7 minutes. Both serum and FF supernatants were frozen and stored at –80°C until analysis. Estradiol and progesterone concentration (basal serum, trigger-day serum, and follicular fluid [FF]) were measured in the 3 different frozen-stored samples, and Elecsys Estradiol III and progesterone III assays (Cobas e 801 analyzer; Roche Diagnostics GmbH, Mannheim, Germany) were used for dilutions.
2.4. SARS-CoV-2 antibody titer assayThe SARs-CoV-2 antibodies were measured in serum on the day of ovarian stimulation initiation and at the time of triggering, and in follicular fluid collected on the day of oocyte retrieval. The measurement was using Atellica SARS-CoV-2 Total (IgG + IgM) reagents from Siemens (NY). The Atellica SARS-CoV-2 Total assay detects an-ti-receptor-binding-domain antibodies. We conducted a quantitative analysis of the antibody titer by performing a 5-fold dilution and subsequent analysis. According to the assay precision design, we defined a difference as significant when there is a difference of 15% or more in index values below 10, and a difference of 12% or more in titer values when they are 10 or higher.
2.5. Statistical analysisStatistical analysis was conducted using SPSS statistics for Windows (version 27.0; IBM Corp., Armonk, NY). After confirming distribution normality using the Kolmogorov-Smirnov test, one-way analysis of variance and Kruskal-Wallis testing were appropriately applied to compare parametric data among the 3 groups. Fisher exact tests were used to compare non-parametric categorical data, as appropriate. Statistical significance was at P < .05.
3. Results 3.1. Baseline characteristics of recruited patientsThe detailed baseline characteristics of the recruited patients are summarized in Table 1. The mean ages were 35.4, 34.9, and 36.7 years in the control, COVID+, and vaccination groups, respectively. In the vaccination group, the mean ages of the BNT162b2 and other vaccine subgroups were 37.0 and 36. 0 years, respectively. The mean BMIs were 20.5, 21.5, and 23.2 kg/m2 in the control, COVID + and vaccination groups, respectively. No statistically significant differences in age (P = .442) and BMI (P = .090) were observed among the 3 groups. The mean AFCs were 9.8, 9.8, and 11.2 in the control, COVID+, and vaccination groups, respectively, with no statistical differences (P = .691). In the vaccination group, no statistical differences in AFC were noted between the BNT162b2 and other vaccine subgroups. Infertility duration lasted 2.6, 3.1, and 3.0 years in the control, COVID+, and vaccination groups, respectively, with no significance (P = .881). The mean ages of male partners were 39.8 in the control group, 38.2 in the COVID + group, and 39.1 years in the vaccination groups, respectively, with no significant difference observed (P = .811). Sperm parameters, including semen volume (P = .043), sperm concentration (P = .959), sperm strict morphology (P = .508), showed no statistical differences among the groups. However, sperm total motile rate exhibited a significant difference among the groups (P < .001) (Table 1).
Table 1 - Baseline characteristics in this study. Variables ControlAll values are presented as mean ± standard deviation. n is number of patients. p*are calculated for comparison of BNT162b2 and others within the vaccination group. Time interval is defined as the period (in mo) from the last vaccination date to ovarian stimulation. Semen parameters are represented 6th edition of World Health Organization (WHO) reference values.
AMH = anti-Müllerian hormone, BMI = body mass index, n = number.
The mean serum basal estradiol concentrations were 43.2, 36.1, and 34.2 pg/mL in the control, COVID+, and vaccination groups, respectively, with no statistical differences (P = .709). On trigger day, the control, COVID+, and vaccination groups exhibited serum mean estradiol levels of 2238, 2069, and 2340 pg/mL, respectively, with no statistical differences (P = .684). Regarding the mean FF estradiol concentration, no statistical differences were observed among the 3 groups (385.2, 603.4, and 580.6 ng/mL in the control, COVID+, and vaccination groups respectively) (P = .736). No significant differences were noted among the 3 groups in the mean basal (P = .814), trigger-day (P = .122), and FF (P = .052) progesterone levels.
The mean oocyte numbers were 18.4, 13.2, and 17.3 in the control, COVID+, and vaccination groups, respectively; however, no statistical differences were observed (P = .291). The mean numbers of mature oocytes were 9.0, 7.4, and 10.0 in the control, COVID+, and vaccination groups, respectively, with no significance (P = .416). The mean mature/total oocyte ratio demonstrated no statistical differences among the 3 groups (P = .494). The mean number of good-quality embryos was 2.8 in the control, 1.9 in COVID+, and 1.7 in the vaccination group with no significant differences (P = .694). Within the vaccination group, no statistical difference was observed in the mean number of good-quality embryos (P = .471). In terms of follicular function, the mean trigger-day estradiol level in MII oocytes was significantly lower in the vaccination group (152.3 pg/mL) than in the control (256.7 pg/mL) and COVID + (366.1 pg/mL) groups (P = .002). The mean trigger-day serum estradiol/MII oocyte level was significantly lower in the BNT162b2 subgroups (110.6 pg/mL) than in the other vaccine subgroups (289.5 pg/mL) (P = .006). Regarding the progesterone levels in MII oocytes, no statistical differences were noted among the 3 groups (P = .349) (Table 2).
Table 2 - The steroidogenic function of the follicle. Variables ControlPs*were calculated comparison of BNT162b2 and others within the vaccination group. GQE: good quality embryos: Blastocyst grade ≥ 4BB or cleavage embryo Grade 1 or 2, FF = Follicular fluid, hCG = Human Chorionic Gonadotropin, N = number. MII oocyte is defined as a metaphase II oocyte, a matured oocyte.
For IVF outcomes, there were no statistical differences in pregnancy rate (P = .621), implantation rate (P = .121), and clinical pregnancy rate (P = .270) among 3 groups (Fig. 1A). For the subgroup of vaccination, the pregnancy rate, implantation rate, and clinical pregnancy rate were also not significance between the 2 groups (Fig. 1B).
Figure 1.: The pregnancy, implantation, and clinical pregnancy rate in this study. (A) There were no statistical differences among control, COVID+, and vaccination groups. (B) For subgroup analysis, there were no statistical differences between BNT162b2 and other groups. The pregnancy rate included both chemical and clinical pregnancy. Clinical pregnancy: ultrasound confirmation of gestational sac with fetal cardiac activity.
In terms of SARS-CoV-2 serum antibody concentration, comparing the difference in serum antibody concentrations between the starting point of ovarian stimulation and the trigger day, the control group showed no change in 100% of cases. In the COVID + group, 10% of cases were observed to have a change, while 90% showed no change. In the vaccination group, 23.3% showed a change, while 76.7% showed no change. The proportion of individuals who exhibited changes in serum antibodies during ovarian stimulation did not show a statistically significant difference between the groups (P = .255).
On comparing antibody concentration of FF to serum, no difference in 80% and difference in 20% of the control group were observed. Ninety percent of the COVID + group displayed no difference, whereas 10% exhibited difference in the FF antibody level. In the vaccination group, the FF antibody level did not change in 46.7% and changed in 53.3%. Significant differences in the FF antibody titer compared to serum antibody were noted among the control, COVID+, and vaccination groups (P = .039) (Table 3). Subgroup comparisons revealed a significant difference in the rate of FF antibody change compared to basal serum antibody concentration between the BNT162b2 and other vaccine subgroups of the COVID + group but not between those of the control group (Fig. 2).
Table 3 - SARS-CoV-2 antibody in serum and follicular fluid presence change in this study. Serum Ab(IgG + IgM) ControlNumber of patients (%) was described. According to the assay precision design, Serum Ab change was defined as difference of antibody index before ovarian stimulation and after ovarian stimulation. For follicular fluid Ab, the difference was calculated between antibody index of follicular fluid and serum of before ovarian stimulation. Ps were calculated by Fisher exact test.
SARS-CoV-2 = severe acute respiratory syndrome coronavirus-2.
Figure 2.: Comparisons on SARS-COV-2 antibody change between BNT162b2 and others group. (A) In serum, there was no statistical difference in SARS-COV-2 antibody titer level between BNT162bw and others. (B) In follicular fluid, there was statistical significance for SARS-COV-2 antibody titer level between BNT162bw and other groups. SARS-CoV-2 = severe acute respiratory syndrome coronavirus-2.
4. DiscussionDue to the rapid spread of the COVID-19 pandemic, over 772,000,000 cases have been confirmed, including 6981,263 reported deaths.[13] In Korea, approximately 25,000,000 people have been infected with SARS- CoV-2, constituting half of the entire population. Following the development of an mRNA-based anti- COVID-19 vaccine in 2020, the vaccination rate has rapidly increased to over 87% in Korea.[14] The sub-effect of vaccination on other medical fields has not been fully elucidated because anti-COVID-19 vaccination was emergently instituted based on the catastrophic effects of the infectious pandemic.[3] Therefore, COVID-19 infection potentially detrimental effect has become a cause for concern across various aspects, including reproductive medicine.
Ovarian stimulation is an essential part of ART, and it promotes multiple egg production by the ovaries and elevates estradiol and progesterone to supraphysiologic levels with the administration of extrinsic gonadotropin.[15] Sex steroids are predominantly produced by the ovarian follicle and corpus luteum, which are potentially affected by COVID-19 infection or vaccination.[16] The spike protein of SARS-CoV-2 is known to bind to the host cell membrane by mimicking the human-like angiotensin-converting enzyme 2 (ACE2) receptor and type 2 transmembrane serine protease (TMPRSS2). In human ovaries, granulosa cells exhibit ACE2 expression, while cumulus cells present in the Graafian follicle display both ACE-2 and TMPRSS2 expression.[17] Granulosa cells are the critical center of estradiol production in women. Therefore, our study aimed to examine the effects of COVID-19 infection and vaccination on IVF outcome and follicular development after ovarian stimulation and estradiol production.
The analysis revealed no observable differences among the baseline characteristics across the control, COVID+, and vaccination groups. However, a significant difference was noted in sperm total motile rate among the groups. Post hoc analysis indicated an elevated sperm motility rate in the COVID + group compared to other groups (P = .020). Nonetheless, it is noteworthy that mean values across all groups were higher than the WHO reference value of 42%, which did not result in discernible differences in IVF outcomes among the groups. In previous study, temporary impairment of total motile count after COVID-19 vaccination, especially BNT162b2 was reported.[18] Hence, further investigation is warranted to comprehensively understand the potential influence of both vaccination and COVID-19 infection on sperm total motile count.
Bentov et al reported the impact of COVID-19 infection and BNT162b2 vaccination on ovarian follicular function. Their cohort study suggested no detrimental effect on IVF outcomes according to COVID-19 infection or BNT162b2 vaccination.[19] In addition, a recent larger retrospective study assessing ovarian function in ART regarding the type of vaccination revealed no detrimental effect on ART outcome.[20] In our study, after ovarian stimulation, the number of retrieved oocytes, MII oocytes, and good quality embryos were comparable between BNT162b2 and another vaccine group, as well as among 3 groups, which is consistent with the previous studies. It is noteworthy that the COVID-19 vaccine type did not affect IVF outcome. Therefore, IVF is a safe procedure regarding COVID-19 infection and types of vaccination.
Serum estradiol plays a crucial role in oocyte/follicular maturation and in preparing the uterine endometrium for implantation. Importantly, this study found the serum estradiol level per MII oocyte on trigger day to be significantly lower in the vaccination group. In particular, the BNT162b2 vaccine demonstrated lower levels with statistical significance, thus contradicting the findings of a previous study by Bentov et al, which showed no significant effect on steroidogenesis during ovarian stimulation according to COVID-19 infection and vaccination.[19] There might be an association between immune reactions and estradiol levels following COVID-19 vaccination. Given that the immune system can influence a woman hormone level, both the COVID-19 infection and vaccination could lead to changes in the menstrual cycle, which leads to irregular menstruation and increased premenstrual symptoms.[21] In our study, the ratio of mature oocytes to total retrieved oocytes yielded similar mean values (0.6), and we elicited trigger-day E2/MII oocyte production of estradiol from the cumulus-oocyte complex. Some previous studies aimed to predict IVF outcomes with E2/oocytes or E2/M2 oocytes and produced controversial results.[22–25] Future studies need to analyze implantation and clinical pregnancy outcomes among the subgroups to determine whether any differences affect ART outcomes.
Anti-COVID-19 antibody presence was detected in the COVID + and vaccination groups, including the BNT162b2 and other vaccine subgroups. All the patients in the vaccination group underwent secondary or tertiary vaccinations, as scheduled. Two patients who had initially received the JNJ-78436735 vaccine completed their vaccination with a BNT162b2 or mRNA-1273 secondary booster shot. In the control group, one patient was positive but had considerably low serum and FF anti-COVID-19 antibodies; this individual was potentially an asymptomatic COVID-19 patient. Furthermore, the patient exhibited decreased trigger-day serum and FF antibody levels. In the COVID + group, among the ten patients, 3 were not vaccinated and 7 patients acquired immunity from both infection and the vaccination. The hybrid immunized patients showed relatively higher titer of serum SARS-CoV-2 antibody level than the acquired immunity from actual viral infection. In a previous study regarding quantitative analysis of antibody titers in infected and non-infected patients after the third dose of COVID-19 vaccination, the hybrid immunized patients showed increased SARS-CoV-2 antibody titer to anti-S after the 4 months.[26] From these results, the acquired immunity solely from either infection or vaccination should be re-vaccinated at some time point and maybe re-infected.
As shown in the time interval in Table 1, no significant difference was noted between the COVID + group and vaccination group from the most recent vaccination or infection to the ovarian stimulation onset day. To exclude bias from the time interval, we compared the serum titer before and after ovarian stimulation and observed concentration changes in FF. For our study, the aspects of SARS-CoV-2 antibody titer changes in FF were noted among the 3 groups. Only FF antibody changes observed significantly different between the BNT162b2 and other vaccine subgroups. To the best of our knowledge, this is the first vaccination study to compare immunoglobulin concentration and hormonal level changes, including those of estradiol and progesterone, for steroidogenesis both in serum and follicular fluid according to vaccine type and COVID-19 infection.
Our study has certain limitations. This study had a relatively small population in each group due to prospective sample collection and ethical issues. The vaccination was actively performed during the COVID- 19 pandemic before obtaining the IRB approval, which made it very difficult to recruit the control group. To overcome this limitation, we tried to control the covariant factors in this prospective design. Among the demographic factors, as elements considering the confounding factors, they (such as age, BMI, antral follicle count, and anti-Müllerian hormone level) did not show any statistical differences between the groups. We recruited study participants from November 2021, the Omivron variant in South Korea was firstly reported in December 2021. We did not investigate the COVID-19 infection status and variants, which can possibly affect the sperm parameters, IVF outcomes and follicular steroidogenesis. We also excluded cases of male factor infertility and controlled physician-related factors by blinding and de-identification during data collection and analysis to minimize bias. However, as previously mentioned, some adverse events regarding gynecological problems have been reported after vaccine application, which can pose a potential threat to human reproductive health.[20] Diverse parameters may affect after vaccination, previous research also shares similar results about IVF outcomes that have no fertility problems. Our study also has similar results of no effect of IVF outcomes along with COVID-19 infection and vaccination despite the influence of follicular function in mature oocytes by anti-COVID-19 vaccination (Fig. 3). Therefore, multi-center trials should be needed to strengthen our conclusions in the future.
Figure 3.: The graphical illustration of the key findings in this study.
Although this prospective study has a small sample size, it has many potential values considering our results. Currently, the exact mechanism for how COVID-19 infection and vaccination affect serum estradiol levels has not been revealed. Since the serum estradiol level is important to pregnancy, future trials revealing this relationship will strengthen the results of this study and help to understand the immune response in the menstruation cycle. Moreover, this preliminary study shows that the COVID-19 vaccination can potentially be involved in follicle maturation by affecting the hormone level. Although age is the most important factor in IVF outcome, our study indicates that immune deregulation by vaccination also affects IVF outcomes so future research may needed to investigate the impact of IVF outcomes for the relatively elderly group.
5. ConclusionsCOVID-19 infection and vaccination do not influence IVF outcomes. Notably, anti-COVID-19 vaccination, especially BNT162b2, showed a significant impact on the follicular profile of mature oocytes. However, despite these differences in follicular function, the IVF outcomes remained consistent. Thus, it is believed that various vaccines, including BNT162b2 can be safely administered to women undergoing IVF.
AcknowledgmentsWe thank Hee Jun Ji, Ph.D., and his colleagues from the IVF laboratory of the Fertility Center, Bundang CHA Women Hospital, for assisting with sample collection and Youn-hee Choi for providing technical and data analysis assistance.
Author contributionsConceptualization: So Yeon Shin, Dong Hee Choi, Sun-Mi Cho.
Data curation: So Yeon Shin.
Investigation: So Yeon Shin.
Comments (0)