Abstract
Purpose
This study aimed to investigate the association between mild elevation of thyroid-stimulating hormone (TSH) levels and pregnancy outcomes of in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) treatments in women with the first fresh embryo transfer.
Methods
Large single-center retrospective cohort study of 15,728 patients from January 2018 to December 2022 were enrolled in the analyses. Clinical pregnancy rates, live birth rates, miscarriage rates, and ectopic pregnancy rates were compared between the TSH levels < 2.5 mIU/L group (N = 10,932) and TSH levels ≥ 2.5 mIU/L group (N = 4796). Subgroup analysis was performed for patients with TSH levels ≥ 2.5 mIU/L, dividing them into the thyroid peroxidase antibody (TPO)–negative group (N = 4524) and the TPO-positive group (N = 272).
Results
There were no significant differences in the aforementioned pregnancy outcomes between the TSH levels < 2.5 mIU/L group and TSH levels ≥ 2.5 mIU/L group. Similarly, no significant differences were observed in the pregnancy outcomes between the TPO-negative group and the TPO-positive group.
Conclusion
Mildly elevated pre-conception TSH levels in thyroid-normal infertile patients did not have an impact on pregnancy outcomes of IVF/ICSI treatments.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-023-03014-4.
Keywords: Thyroid-stimulating hormone, Thyroid peroxidase antibody, In vitro fertilization, Intracytoplasmic sperm injection, Fresh embryo transfer
Introduction
Thyroid hormones play a vital role in maintaining reproductive and metabolic health in the human body, exerting direct or indirect effects on the hypothalamic-pituitary reproductive (HPR) axis through a myriad of pathway [1]. Thyroid dysfunction can affect the maintenance of normal reproductive function in women and contribute to adverse pregnancy outcomes [2]. Additionally, epidemiological data suggests a relatively high prevalence of thyroid disorders among women at childbearing age [3, 4]. Therefore, screening for and vigilantly monitoring thyroid diseases in women who are planning for pregnancy prior to conception is imperative to safeguard the well-being of pregnant women and their fetal development.
Considered as the most sensitive indicator of thyroid function, serum thyroid-stimulating hormone (TSH) is commonly used to assess thyroid dysfunction during pregnancy [5]. The reference range for TSH levels in early pregnancy may vary due to differences in population and laboratory methodologies. The American Thyroid Association (ATA) and the American College of Obstetricians and Gynecologists recommend a reference range of 0.1–4.0 mIU/L [6, 7]. However, several clinical guidelines suggest an upper limit of 2.5 mIU/L for TSH in early pregnancy [8–11].
Despite several studies indicating that pregnant women undergoing in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) treatment with TSH levels within the normal range (dichotomized at 2.5 mIU/L) do not show significant difference in clinical pregnancy rates, live birth rates, or miscarriage rates [12–14], some studies supported the use of TSH ≥ 2.5 mIU/L (within the reference range) as a basis for therapeutic intervention [15]. However, it is worth noting that these studies had relatively small sample sizes. A meta-analysis by Viviane et al. including data from 17 studies concluded that elevated TSH levels are not associated with adverse pregnancy outcomes [16]. Unfortunately, their analysis was limited by the presence of potential unadjusted confounding variables such as infertility duration, treatment regimens, gonadotropin (Gn) dosage and duration, or endometrial thickness, which may affect the results. Therefore, there is an ongoing debate regarding the use of TSH ≥ 2.5 mIU/L as the indication for levothyroxine (LT-4) administration in IVF patients.
In light of the ongoing debate, our research aimed to address the issue by conducting a comprehensive retrospective cohort study to investigate the associations between TSH levels and various pregnancy outcomes, including clinical pregnancy rates, miscarriage rates, live birth rates, and ectopic pregnancy rates among infertile patients with normal thyroid function undergoing IVF/ICSI treatment and the first fresh embryo transfer. Additionally, we performed a subgroup analysis for patients with TSH levels higher than 2.5 mIU/L. This subgroup was stratified based on thyroid peroxidase antibody (TPOAb) test results. The ultimate objective of our study is to provide clinical guidance for the management of patients who have normal thyroid function with elevated TSH levels exceeding 2.5 mIU/L.
Materials and methods
This retrospective cohort study was approved by the Institutional Review Committee of the Ethics Committee of Peking University Third Hospital (reference: IRB-2022–383-01). Considering that patients with frozen-thawed embryo transfer (FET) were more likely to receive LT-4 treatment before embryo transfer, our study only enrolled patients with fresh embryo transfer to lessen the effect of LT-4 treatment. A total of 27,489 infertile patients who underwent IVF/ICSI treatment and the first fresh embryo transfer at the Reproductive Medicine Center of Peking University Third Hospital from January 2018 to December 2022 were included in this study. After patients with no oocyte retrieved, fertilization failure, and no embryo obtained were excluded, remaining patients meeting the following criteria were excluded from the study: (1) chromosomal abnormalities in either partner; (2) intrauterine adhesions or uterine abnormalities such as unicornuate uterus, rudimentary horn uterus, septate uterus, and bicornuate uterus; (3) coexisting metabolic disorders affecting reproductive hormone secretion, such as hyperprolactinemia, diabetes, and thyroid dysfunction; (4) undergoing in vitro maturation or fertility preservation; (5) receiving LT-4 treatment; (6) male azoospermia; (7) with missing data. The total thyroid hormone (T4), free thyroid hormone (FT4), and TSH levels of the patients included in the study were all within the reference ranges of our institution.
A standardized controlled ovarian stimulation (COS) protocol [17] was applied to all patients for the processes of oocyte retrieval, fertilization, and fresh embryo transfer. (1) Antagonist protocol: on menstrual cycle days 2 to 3, transvaginal ultrasound was performed to assess ovarian follicles with a diameter exceeding 10 mm and serum estradiol (E2) levels below 200 pmol/L. If these criteria were met, Gn stimulation was initiated. Daily Gn injections were administered at doses ranging from 150 to 225 U, with adjustments based on ultrasound monitoring. GnRH antagonist (GnRH-ant, Cetrorelix, Merck) was introduced at a dose of 0.25 mg/day after 5 to 6 days of Gn stimulation or when follicles reached a diameter of ≥ 12 mm, continuing until the day of human chorionic gonadotropin (HCG) trigger. (2) Long protocol: commence with the use of a short-acting GnRH agonist (e.g., Triptorelin, Ipsen) at a daily dose of 0.1 mg or a long-acting GnRH agonist (Triptorelin) at 1.8 mg once follicle-stimulating hormone (FSH), luteinizing hormone (LH), and E2 levels had reached down-regulation criteria. Gn stimulation was initiated when appropriate. (3) Ultralong protocol: on menstrual cycle days 2 to 3, administer a long-acting GnRH agonist at a dose of 3.75 mg or 1.8 mg. After 28 days, administer the second injection. Conduct vaginal ultrasound, FSH, LH, and E2 assessments 2 to 3 weeks later, and initiate Gn stimulation based on ultrasound and endocrine evaluations. (4) Short protocol: on menstrual cycle days 2 to 3, commence subcutaneous injections of a short-acting GnRH agonist at a dose of 0.05 to 0.1 mg/day until the day of HCG trigger. Concurrently, initiate Gn stimulation and adjust the dose as necessary, based on ultrasound-monitored follicular development. (5) Micro-stimulation protocol: from days 2 to 6 of menstruation, administer 2.5 mg letrozole (Femara, Novartis Pharma AG, Basel, Switzerland) or 50 mg clomiphene citrate (Fertilan, Medochemie Ltd, Republic of Cyprus) daily, 150 IU of recombinant FSH was initiated from day 5, and administer GnRH antagonist consistent with the antagonist protocol. (6) Luteal-phase ovulation induction protocol: beginning approximately 3 days after ovulation was confirmed by ultrasound or after oocyte retrieval, when the follicle diameter was less than or equal to 10 mm. The dosage of gonadotropin was determined based on the patient’s age, body mass index (BMI), and anti-Müllerian hormone (AMH) levels.
When transvaginal ultrasound examination revealed two or more follicles with an average diameter of ≥ 18 mm, Gn administration is discontinued. On the same evening, a recombinant HCG injection (Ovidrel, Merck) of 250 μg was administered. Following 36 to 38 h, egg retrieval was conducted under transvaginal ultrasound guidance, and standard luteal phase support treatment was initiated post-procedure. IVF/ICSI was performed within 4 to 6 h after egg retrieval, and fertilization status was assessed after 17 to 19 h. When two pronuclei (2PN) are observed within the oocyte cytoplasm, it was considered normal fertilization. On the third day post-fertilization, embryos with a cleavage-stage blastomere count of 4 to 8 and graded as 2 or higher were considered suitable for transfer. Cycles in which fresh embryo transfer was not performed for various reasons after Gn usage are classified as canceled cycles. For embryo transfer cycles, blood levels of beta-human chorionic gonadotropin are measured 14 days after the transfer.
Clinical pregnancy was defined as the presence of at least one gestational sac observed by ultrasound after embryo transfer, at least 35 days post-transfer. Live birth was defined as the delivery of at least one surviving newborn. Miscarriage was defined as the loss of a clinical pregnancy before 28 weeks of gestation. Ectopic pregnancy was defined as the implantation of a pregnancy outside the uterine endometrium, typically occurring in locations such as the fallopian tubes, uterine cornua, cervix, ovaries, abdominal cavity, or pelvic cavity.
All patients underwent thyroid function testing within 3 months before the start of COS. This testing was performed using an automated chemiluminescence immunoassay analyzer (ADVIA Centaur XP, Siemens Healthcare Diagnostics) to measure T4, FT4, TSH, and thyroglobulin antibodies (TGAb) and thyroperoxidase antibodies (TPOAb). The reference values were as follows: T4 reference range of 4.50–10.9 ng/dL, TSH reference range of 0.55–4.78 uIU/mL, FT4 reference range of 0.89–1.80 ug/dL, TGAb reference range of 0–60 IU/mL, and TPOAb reference range of 0–60 IU/mL. TPOAb levels below 60 IU/mL were defined as negative.
Statistical analysis
Continuous variables were presented as means (standard deviations) for normally distributed data and as medians (25th–75th percentiles) for non-normally distributed data. Categorical variables were presented as frequencies (percentages). Continuous variables were compared using Student’s t-test (for normally distributed data) or the Mann–Whitney U test (for non-normally distributed data), while categorical variables were compared using the chi-squared test.
Logistic regression models were used for analysis after adjusting for relevant factors, and odds ratios (OR) with 95% confidence intervals (CI) were calculated. Statistical significance in all analyses was defined as a two-tailed p-value < 0.05.
Restrictive cubic spline (RCS) regression models were established using R 4.3.1 to analyze the relationship between TSH levels and clinical pregnancy rates, live birth rates, miscarriage rates, and ectopic pregnancy rates.
Results
A total of 27,489 infertile patients who underwent IVF/ICSI treatment and the first fresh embryo transfer at the Reproductive Medicine Center of Peking University Third Hospital from January 2018 to December 2022 were included in this study. After excluding patients as Fig. 1 showed, a total of 15,728 patients were selected based on the 2.5th and 97.5th percentiles of pre-IVF/ICSI serum TSH levels. Then, the patients were divided into two groups based on TSH levels: TSH < 2.5 mIU/L (N = 10,932) and TSH ≥ 2.5 mIU/L (N = 4796) for analysis.
Fig. 1.
Flow chart of the patient cohort selection. IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; TSH, thyroid-stimulating hormone; TPOAb, thyroperoxidase antibodies
Baseline data for both groups were presented in Table 1. There were statistically significant differences in BMI (22.78 ± 3.53 vs. 23.18 ± 3.72, p < 0.001), infertile duration (3 (1–4) vs. 3 (1–4), p = 0.037) and parity (10.3% vs. 8.8%, p = 0.002), parity (10.3% vs. 8.8%, p = 0.002), and TSH levels (1.67 ± 0.46 vs. 3.20 ± 0.53, p < 0.001) between the two groups. No significant differences were found in age, infertility type, gravidity, FSH, LH, E2, AMH, number of total antral follicle, TGAb, and TPOAb.
Table 1.
Baseline characteristics of patients
| Characteristics | TSH < 2.5 mIU/L (N = 10,932) | TSH ≥ 2.5 mIU/L (N = 4796) | p |
|---|---|---|---|
| Age, median (IQR), year | 32.99 (4.46) | 32.99 (4.47) | 0.961 |
| BMI, mean (SD), kg/m2 | 22.78 (3.53) | 23.18 (3.72) | < 0.001 |
| Infertility type, no. (%) | |||
| Primary | 6251 (57.2%) | 2752 (57.4%) | 0.815 |
| Secondary | 4681 (42.8%) | 2044 (42.6%) | |
| Infertile duration, median (IQR), y | 3 (1–4) | 3 (1–4) | 0.037 |
| Gravidity, no. (%) | |||
| 0 | 6478 (59.3%) | 2848 (59.4%) | 0.883 |
| ≥ 1 | 4454 (40.7%) | 1948 (40.6%) | |
| Parity, no. (%) | |||
| 0 | 9804 (89.7%) | 4376 (91.2%) | 0.002 |
| ≥ 1 | 1128 (10.3%) | 420 (8.8%) | |
| FSH, median (IQR), mIU/ml/ | 6.54 (5.65–8.13) | 6.57 (5.60–8.13) | 0.633 |
| LH, median (IQR), mIU/ml/ | 3.74 (2.51–5.52) | 3.72 (2.48–5.48) | 0.579 |
| E2, median (IQR), pmol/L/ | 143 (115–186) | 143 (113–185) | 0.271 |
| AMH, median (IQR), ng/ml/ | 2.07 (1.15–3.54) | 2.07 (1.15–3.56) | 0.937 |
| No. of total antral follicle, median (IQR) | 10 (7–14) | 10 (7–14) | 0.779 |
| TSH, mean (SD), mIU/ml | 1.67 (0.46) | 3.20 (0.53) | < 0.001 |
| TGAb, median (IQR), IU/ml | 15 (0–19) | 15 (0–19.7) | 0.283 |
| TPOAb, median (IQR), IU/ml | 31.2 (0–31.2) | 30.8 (0–31.2) | 0.209 |
BMI body mass index, FSH follicle-stimulating hormone, LH luteinizing hormone, E2 estradiol, AMH anti-Müllerian hormone, TSH thyroid-stimulating hormone, TGAb thyroglobulin antibodies, TPOAb thyroperoxidase antibodies
The protocols of COS and relevant IVF/ICSI and embryo transfer data of two patient group were shown in Table 2. No statistically significant differences were found. There were no statistically significant differences observed in terms of COS protocol selection, duration of Gn administration, dosage of Gn administration, endometrial thickness, number of total oocyte retrieval, number of 2PN, number of 2PN embryos, number of metaphase II oocytes, number of high-quality embryos, embryo transfer stage, number of embryos transferred, or the method of fertilization.
Table 2.
The protocols of COS and relevant IVF/ICSI and embryo transfer data of patients
| Characteristics | TSH < 2.5 mIU/L (N = 10,932) | TSH ≥ 2.5 mIU/L (N = 4796) | p |
|---|---|---|---|
| Protocol of COS, No. (%) | |||
| Ultralong GnRH agonist | 955 (8.7%) | 436 (9.1%) | 0.484 |
| Long GnRH agonist | 2085 (19.1%) | 893 (18.6%) | |
| Short GnRH agonist | 75 (0.7%) | 28 (0.6%) | |
| GnRH agonist | 7757 (71.0%) | 3421 (71.3%) | |
| Others | 60 (0.5%) | 18 (0.4%) | |
| Gn duration, mean (SD), day | 11 (2.23) | 11.03 (2.22) | 0.402 |
| Gn dosage, median (IQR), U | 2787 (1125) | 2819 (1125) | 0.100 |
| Endometrial thickness, mean (SD), mm | 9.99 (1.48) | 9.97 (1.50) | 0.459 |
| No. of total oocyte retrieval, median (IQR) | 9 (6–12) | 9 (6–12) | 0.525 |
| No. of MII oocytes, median (IQR) | 8 (6–12) | 8 (6–11) | 0.675 |
| No. of 2PN, median (IQR) | 5 (3–8) | 5 (3–8) | 0.850 |
| No. of 2PN embryos, median (IQR) | 5 (3–8) | 5 (3–8) | 0.903 |
| No. of high-quality embryos, median (IQR) | 3 (2–4) | 3 (2–4) | 0.730 |
| Embryo transfer stage, no. (%) | |||
| Cleavage stage embryo | 10,397 (95.1%) | 4567 (95.2%) | 0.793 |
| Blastocysts | 535 (4.9%) | 229 (4.8%) | |
| No. of embryos transferred, no. (%) | |||
| 1 | 1814 (16.6%) | 789 (16.5%) | 0.825 |
| 2 | 9118 (83.4%) | 4007 (83.5%) | |
| Method of fertilization, no. (%) | |||
| IVF | 7568 (69.2%) | 3283 (68.5%) | 0.128 |
| ICSI | 3268 (29.9%) | 1483 (30.9%) | |
| Half ICSI | 96 (0.9%) | 30 (0.6%) | |
(1) Others include the micro-stimulation protocol and the luteal-phase ovulation induction protocol
GnRH gonadotropin-releasing hormone, Gn gonadotropin, MII metaphase II, 2PN 2-pronuclei, IVF in vitro fertilization, ICSI intracytoplasmic sperm injection
Single-factor analysis was employed to compare the clinical pregnancy rates, live birth rates, miscarriage rates, and ectopic pregnancy rates between the two groups of patients, and no statistically significant differences were observed. Furthermore, after adjusting for age and potential confounding factors(p ≤ 0.10) identified in Tables 1 and 2 through binary logistic regression analysis, the results indicate that there remained no significant differences in the aforementioned pregnancy outcomes between the two patient groups. The results of both analyses were presented in Table 3.
Table 3.
Single-factor and binary logistic regression analysis of pregnancy outcomes and TSH levels
| Outcomes | Unadjusted | Adjusted | ||||
|---|---|---|---|---|---|---|
| OR | 95%CI | p | OR | 95%CI | p | |
| Clinical pregnancy | 0.970 | 0.935–1.006 | 0.103 | 0.974 | 0.938–1.011 | 0.172 |
| Live birth | 0.965 | 0.898–1.036 | 0.322 | 0.981 | 0.912–1.055 | 0.602 |
| Miscarriage | 1.027 | 0.951–1.110 | 0.500 | 1.008 | 0.931–1.090 | 0.852 |
| Ectopic pregnancy | 1.063 | 0.914–1.237 | 0.429 | 1.059 | 0.910–1.233 | 0.460 |
Model was adjusted for age, BMI, infertile duration, parity and Gn dosage
OR odds ratio, CI confidence intervals
We also conducted RCS analysis to explore the continuous relationship between pre-conception TSH levels as a continuous variable and pregnancy outcome indicators. The results were shown in Fig. 2. After adjusting the same factors, the non-linear correlation analysis revealed that among IVF/ICSI patients, there were no statistically significant differences observed in clinical pregnancy rates, live birth rates, miscarriage rates, or ectopic pregnancy rates in relation to pre-conception TSH levels. These findings suggested that pre-conception TSH levels did not exhibit a statistically significant association with the aforementioned pregnancy outcomes.
Fig. 2.
The fitting curve of odds ratios by different levels of preconception thyroid stimulating hormone. Model was adjusted for age, BMI, infertile duration, parity, and Gn dosage. Four knots were used. A Clinical pregnancy odds ratio, model likelihood ratio = 403.94, p < 0.001, R2 = 0.034, TSH = 2.00 when OR = 1 and p = 0.4498; B live birth odds ratio, model likelihood ratio = 170.36, p < 0.001, R2 = 0.040; TSH = 1.98 when OR = 1 and p = 0.2540; C miscarriage odds ratio, model likelihood ratio = 193.15, p < 0.001, R2 = 0.049; TSH = 1.98 when OR = 1, p = 0.3814; D ectopic pregnancy odds ratio, model likelihood ratio = 2.31, p = 0.9699, R2 < 0.001; TSH = 1.98 when OR = 1, p = 0.4688. OR, odds ratio; CI, confidence intervals; TSH, thyroid-stimulating hormone
Patients with pre-conception TSH levels exceeding 2.5 mIU/L were further stratified based on TPOAb results, resulting in two subgroups: the TPO-negative group (N = 4524) and the TPO-positive group (N = 272). Single-factor analysis and multi-factor analysis were conducted for both subgroups. Results were shown in Table 4. Remarkably, across all analyses, clinical pregnancy rates, live birth rates, miscarriage rates, and ectopic pregnancy rates exhibited no statistically significant differences between the two subgroups. The multi-factor analysis incorporated balanced variables such as age, infertility type, gravity, FSH, TSH, TGAb, protocol of COS, duration of Gn administration, number of metaphase II oocytes, number of 2PN, number of 2PN embryos, and number of high-quality embryos.
Table 4.
Single-factor and binary logistic regression analysis of pregnancy outcomes and TPOAb results
| Outcomes | Unadjusted | Adjusted | ||||
|---|---|---|---|---|---|---|
| OR | 95%CI | p | OR | 95%CI | p | |
| Clinical pregnancy | 0.928 | 0.724–1.189 | 0.555 | 0.849 | 0.648–1.110 | 0.231 |
| Live birth | 0.831 | 0.522–1.322 | 0.434 | 0.803 | 0.490–1.316 | 0.385 |
| Miscarriage | 1.071 | 0.637–1.799 | 0.796 | 1.148 | 0.659–2.002 | 0.625 |
| Ectopic pregnancy | 1.663 | 0.704–3.927 | 0.246 | 1.614 | 0.659–3.950 | 0.295 |
Model was adjusted for age, infertility type, gravity, FSH, TSH, TGAb, protocol of COS, Gn duration, number of metaphase II oocytes, number of 2PN, number of 2PN embryos and number of high-quality embryos
OR odds ratio, CI confidence intervals
Discussion
The aim of our study was to investigate whether mild elevation of TSH levels in infertile patients with normal thyroid function undergoing IVF/ICSI treatment was associated with pregnancy outcomes, including clinical pregnancy rates, miscarriage rates, preterm birth rates, and ectopic pregnancy rates. To achieve this, we conducted an extensive retrospective cohort analysis. TSH occupies a pivotal position within the hypothalamic-pituitary-thyroid axis and serves as the most sensitive indicator of thyroid function, making it clinically relevant. Additionally, TSH plays a crucial role in maintaining female reproductive health. Although the exact mechanisms by which TSH influences the HPR axis are not fully understood, current research suggests that TSH can modulate the HPR axis both directly through its interaction with FSH receptors and indirectly by affecting leptin secretion, which regulates kisspeptin-neurokinin B-dynorphin neurons [18].
Our study determined the reference range (0.70–4.50 mIU/L) for TSH levels in pre-pregnancy patients undergoing IVF/ICSI treatment. The specific reference range was based on the 2.5th and 97.5th percentiles of TSH levels of patients, smaller than the reference range established by our institution. It is worth mentioning that a meta-analysis conducted in China, which included 11 studies from nine cities, reported variations in TSH reference ranges between pregnant and non-pregnant women [19]. However, it is important to note that this analysis did not consider infertile patients undergoing IVF/ICSI treatment and had a limited sample size. In contrast, our study benefited from a large enough sample size to establish specific TSH reference ranges for infertile patients undergoing IVF/ICSI treatment in our institution.
Our study’s findings showed no significant differences between the two groups in terms of clinical pregnancy rates, miscarriage rates, live birth rates, and ectopic pregnancy rates. These results support the modification of the 2011 ATA guidelines in 2017, which revised the TSH cutoff value to 2.5 mIU/L. According to the 2017 ATA guidelines, pregnant women with early pregnancy TSH levels exceeding 2.5 mIU/L but remaining below the upper limit of the reference range are stratified based on TPOAb results. TPOAb-positive patients are considered for LT-4 therapy [6]. Studies by Zhu et al. and Li et al. evaluated the diagnostic efficacy of the 2011 and 2017 ATA guidelines for subclinical hypothyroidism in the Chinese population [20, 21]. Both studies concluded that the 2017 guidelines were superior to the 2011 guidelines. They found that TPOAb-negative pregnant women with pre-pregnancy TSH levels exceeding 2.5 mIU/L but within the reference range did not experience adverse pregnancy outcomes. Viviane et al.‘s retrospective cohort study, supported by a meta-analysis of 17 studies, confirmed these findings [16]. However, Li et al. reported that pre-pregnancy TSH levels ≥ 2.1 mIU/L in patients undergoing IVF treatment were associated with adverse outcomes such as miscarriage, small-for-gestational-age infants, large-for-gestational-age infants, and preterm birth [22]. It is important to note that their study did not adhere to strict inclusion and exclusion criteria for patients enrolling patients and did not collect additional thyroid function data. This limits the applicability of their findings. Zhou et al. found that IVF/ICSI patients with TSH levels ≤ 3 mIU/L had a higher live birth rate compared to patients with cutoff values of 2.5 or 4 mIU/L [13]. However, their study had a relatively small sample size and did not consider the presence of other thyroid-related conditions, such as anemia, hypertension, diabetes, or hepatic and renal dysfunction. Additionally, they did not differentiate patients based on TSH levels. Clinical trials by Cai et al. and prospective cohort studies by Zhang et al. did not support the strict recommendation to lower TSH levels to below 2.5 mIU/L in the 2017 ATA guidelines for patients receiving LT-4 treatment [14, 23]. Therefore, there is a need to reconsider whether a TSH cutoff value of 2.5 mIU/L should continue to be used as a diagnostic criterion for LT-4 treatment in infertile patients undergoing IVF/ICSI treatment.
Furthermore, our study conducted subgroup analyses based on TPOAb results for patients with TSH levels exceeding 2.5 mIU/L, in line with the 2017 ATA guidelines. However, both subgroups showed no significant differences in the pregnancy outcomes of interest. He et al. conducted a study where they grouped patients based on TSH levels and TPOAb results and reached the similar conclusion with us [24]. However, their study had a relatively smaller sample size, especially in the TSH ≥ 2.5 mIU/L and TPO-positive group. Raffaella et al. conducted a meta-analysis which found no associations between thyroid function-normal but TPO-positive IVF/ICSI patients and pregnancy-related complications such as gestational hypertension, preeclampsia, placental abruption, miscarriage, or neonatal intensive care unit admission [25]. However, they did find differences in observational studies regarding preterm birth. Another study by Liu et al. suggested lower clinical pregnancy rates and live birth rates among TPO-positive patients with normal thyroid function, along with higher miscarriage rates [26]. Their study further conducted proteomics analysis and found the possible association of angiotensinogen and fetuin-B with reproduction. However, it is important to note that their study did not consider TSH levels and the variations in study populations may explain some discrepancies in the results. Therefore, whether LT-4 treatment should be considered for patients with TSH ≥ 2.5 mIU/L and TPOAb positivity remains a topic of debate.
The strengths of our study include a large sample size for robust analysis, which was the first single-center study that covered more than 10,000 patients selected by the exclusion criteria and was the largest sample size as we knew. The large sample size was only met by a meta-analysis including 17 studies from 2006 to 2022 [16], which did not consider the difference between the fresh embryo transfer and FET. In our study, undergoing the first fresh cycle was ensured among each patient included to rule out the impact of multiple cycles on patients and compared with the FET patients, the shorter time interval between IVF/ICSI and thyroid function assessment could help us find more direct association between TSH and outcomes of IVF/ICSI. The use of both logistic regression models and RCS models is also worth noting, which consistently yielded similar results to support our conclusion. However, there are limitations in our study. The retrospective design may have potential issues with data completeness and homogeneity. Additionally, the study only included patients undergoing fresh-cycle embryo transfer, excluding FET patients. However, it is worth noting that FET patients in our institution also undergo thyroid function assessment preceding embryo transfer but in the six months, making the two groups comparable.
In conclusion, our study suggests that a mild elevation of TSH levels in infertile patients with normal thyroid function undergoing IVF/ICSI treatment does not lead to adverse pregnancy outcomes. Therefore, we argue that patients with TSH levels exceeding 2.5 mIU/L but within the reference range do not require LT-4 treatment. However, it is important to emphasize that these conclusions should be validated through large-sample prospective cohort studies.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank all the patients participating the study.
Funding
This study was supported by the National Key Research and Development Program of China (no. 2022YFC2702500).
Data availability
Due to the nature of this research and ethical restrictions, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.
Declarations
Ethics approval
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Institutional Review Committee of the Ethics Committee of Peking University Third Hospital (reference: IRB-2022–383-01).
Consent for publication
Patients signed informed consent regarding publishing their data.
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.
Ming-Mei Lin, Di Mao and Kai-Lun Hu contributed equally to this work.
References
- 1.Brown EDL, Obeng-Gyasi B, Hall JE, Shekhar S. The thyroid hormone axis and female reproduction. Int J Mol Sci. 2023;24.10.3390/ijms24129815. [DOI] [PMC free article] [PubMed]
- 2.Mazzilli R, Medenica S, Di Tommaso AM, Fabozzi G, Zamponi V, Cimadomo D, et al. The role of thyroid function in female and male infertility: a narrative review. J Endocrinol Invest. 2023;46:15–26. doi: 10.1007/s40618-022-01883-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Strikić Đula I, Pleić N, Babić Leko M, Gunjača I, Torlak V, Brdar D et al. Epidemiology of hypothyroidism, hyperthyroidism and positive thyroid antibodies in the Croatian population. Biology (Basel). 2022;11(3):394. 10.3390/biology11030394. [DOI] [PMC free article] [PubMed]
- 4.Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA, et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III) J Clin Endocrinol Metab. 2002;87:489–499. doi: 10.1210/jcem.87.2.8182. [DOI] [PubMed] [Google Scholar]
- 5.Lee SY, Pearce EN. Assessment and treatment of thyroid disorders in pregnancy and the postpartum period. Nat Rev Endocrinol. 2022;18:158–171. doi: 10.1038/s41574-021-00604-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Alexander EK, Pearce EN, Brent GA, Brown RS, Chen H, Dosiou C, et al. 2017 Guidelines of the American Thyroid Association for the Diagnosis and Management of Thyroid Disease During Pregnancy and the Postpartum. Thyroid. 2017;27:315–389. doi: 10.1089/thy.2016.0457. [DOI] [PubMed] [Google Scholar]
- 7.Schroeder BM. ACOG practice bulletin on thyroid disease in pregnancy. Am Fam Physician. 2002;65(2158):2161–2162. [PubMed] [Google Scholar]
- 8.Garber JR, Cobin RH, Gharib H, Hennessey JV, Klein I, Mechanick JI, et al. Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocr Pract. 2012;18:988–1028. doi: 10.4158/ep12280.Gl. [DOI] [PubMed] [Google Scholar]
- 9.Pfeifer S, Butts S, Dumesic D, Fossum G, Goldberg J, Gracia C et al. Subclinical hypothyroidism in the infertile female population: a guideline. Fertil Steril. 2015;104(3):545-53. 10.1016/j.fertnstert.2015.05.028. [DOI] [PubMed]
- 10.Bahn Chair RS, Burch HB, Cooper DS, Garber JR, Greenlee MC, Klein I, et al. Hyperthyroidism and other causes of thyrotoxicosis: management guidelines of the American Thyroid Association and American Association of Clinical Endocrinologists. Thyroid. 2011;21:593–646. doi: 10.1089/thy.2010.0417. [DOI] [PubMed] [Google Scholar]
- 11.Poppe K, Bisschop P, Fugazzola L, Minziori G, Unuane D, Weghofer A. 2021 European Thyroid Association Guideline on Thyroid Disorders prior to and during assisted reproduction. Eur Thyroid J. 2021;9:281–295. doi: 10.1159/000512790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jin L, Wang M, Yue J, Zhu GJ, Zhang B. Association between TSH level and pregnancy outcomes in euthyroid women undergoing IVF/ICSI: a retrospective study and meta-analysis. Curr Med Sci. 2019;39:631–637. doi: 10.1007/s11596-019-2084-5. [DOI] [PubMed] [Google Scholar]
- 13.Zhou D, Deng H, Xia M, Li R, Ye H. The relationship between TSH levels and clinical pregnancy outcomes for patients who undergo in vitro fertilization/intracytoplasmic sperm injection: a retrospective study. Transl Pediatr. 2022;11:1301–1310. doi: 10.21037/tp-22-79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cai Y, Zhong L, Guan J, Guo R, Niu B, Ma Y, et al. Outcome of in vitro fertilization in women with subclinical hypothyroidism. Reprod Biol Endocrinol. 2017;15:39. doi: 10.1186/s12958-017-0257-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xia J, Yang Z, Ou X, Wu G, Yang J. Evaluation of the relationship between serum TSH levels and pregnancy outcomes of IVF/ICSI patients in follicular phase long-acting long protocol. Gynecol Endocrinol. 2022;38:651–655. doi: 10.1080/09513590.2022.2091540. [DOI] [PubMed] [Google Scholar]
- 16.d'Assunção VRN, Montagna E, d'Assunção LEN, Caldas MMP, Christofolini DM, Barbosa CP, et al. Effect of thyroid function on assisted reproduction outcomes in euthyroid infertile women: a single center retrospective data analysis and a systematic review and meta-analysis. Front Endocrinol (Lausanne) 2022;13:1023635. doi: 10.3389/fendo.2022.1023635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Xu H, Zeng L, Yang R, Feng Y, Li R, Qiao J. Retrospective cohort study: AMH is the best ovarian reserve markers in predicting ovarian response but has unfavorable value in predicting clinical pregnancy in GnRH antagonist protocol. Arch Gynecol Obstet. 2017;295:763–770. doi: 10.1007/s00404-016-4274-8. [DOI] [PubMed] [Google Scholar]
- 18.Petrine JCP, Franci CR, Del Bianco-Borges B. Leptin actions through the nitrergic system to modulate the hypothalamic expression of the kiss1 mRNA in the female rat. Brain Res. 2020;1728:146574. doi: 10.1016/j.brainres.2019.146574. [DOI] [PubMed] [Google Scholar]
- 19.Gao X, Li Y, Li J, Liu A, Sun W, Teng W, et al. Gestational TSH and FT4 reference intervals in Chinese women: a systematic review and meta-analysis. Front Endocrinol (Lausanne) 2018;9:432. doi: 10.3389/fendo.2018.00432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhu P, Chu R, Pan S, Lai X, Ran J, Li X. Impact of TPOAb-negative maternal subclinical hypothyroidism in early pregnancy on adverse pregnancy outcomes. Ther Adv Endocrinol Metab. 2021;12:20420188211054690. doi: 10.1177/20420188211054690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Li MF, Ma L, Feng QM, Zhu Y, Yu TP, Ke JF, et al. Effects of maternal subclinical hypothyroidism in early pregnancy diagnosed by different criteria on adverse perinatal outcomes in Chinese women with negative TPOAb. Front Endocrinol (Lausanne) 2020;11:580380. doi: 10.3389/fendo.2020.580380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li M, He Y, Mao Y, Yang L, Chen L, Du J, et al. Preconception thyroid-stimulating hormone levels and adverse pregnancy outcomes. Clin Endocrinol (Oxf) 2022;97:339–346. doi: 10.1111/cen.14668. [DOI] [PubMed] [Google Scholar]
- 23.Gao X, Liu A, Wang X, Han Y, Wang H, Li J, et al. Postpartum levothyroxine adjustment and its impact factors in women with hypothyroidism in pregnancy. Endocr Pract. 2022;28:578–585. doi: 10.1016/j.eprac.2022.03.002. [DOI] [PubMed] [Google Scholar]
- 24.He Q, Zhang Y, Qiu W, Fan J, Zhang C, Kwak-Kim J. Does thyroid autoimmunity affect the reproductive outcome in women with thyroid autoimmunity undergoing assisted reproductive technology? Am J Reprod Immunol. 2020;84:e13321. doi: 10.1111/aji.13321. [DOI] [PubMed] [Google Scholar]
- 25.Di Girolamo R, Liberati M, Silvi C, D'Antonio F. Levothyroxine supplementation in euthyroid pregnant women with positive autoantibodies: a systematic review and meta-analysis. Front Endocrinol (Lausanne) 2022;13:759064. doi: 10.3389/fendo.2022.759064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Liu Y, Wu Y, Tian M, Luo W, Zhang C, Liu Y, et al. Protein expression profile in IVF follicular fluid and pregnancy outcome analysis in euthyroid women with thyroid autoimmunity. ACS Omega. 2020;5:11439–11447. doi: 10.1021/acsomega.0c00463. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Due to the nature of this research and ethical restrictions, participants of this study did not agree for their data to be shared publicly, so supporting data is not available.