Impact of Thyroid Autoimmunity on Ovarian Reserve,Pregnancy Outcomes,and Offspring Health in Euthyroid Women Following In Vitro Fertilization/Intracytoplasmic Sperm Injection

Abstract

Background:

Thyroid autoimmunity (TAI) is the most frequent autoimmune disease among reproductive-aged women. It has been related to premature ovarian insufficiency, but the mechanisms remain elusive, and its association with ovarian reserve in euthyroid women is debatable. Moreover, the impact of TAI on assisted reproduction is controversial: especially for women with diminished ovarian reserve (DOR), few studies are available. Therefore, the present study was aimed to look for an association between TAI and DOR, and to evaluate the effect of TAI on pregnancy outcomes and offspring health following assisted reproductive technology stratified by ovarian reserve.

Methods:

A total of 6213 euthyroid women from the Reproductive Hospital Affiliated to Shandong University between 2012 and 2017 were retrospectively included. The prevalence of DOR in women with negative or positive TAI was calculated, and pregnancy and neonatal outcomes after in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) cycles were compared between the TAI-positive and TAI-negative groups both in women with DOR and in those with normal ovarian reserve (NOR). Longitudinal growth parameters and temperament type of the offspring were also observed in the TAI-positive and TAI-negative groups.

Results:

The prevalence of DOR in women with positive TAI and those with negative TAI was not significantly different (4.09% vs. 2.96%, p = 0.053), even after stratifying patients by age. In women with DOR, the live birth rate, pregnancy loss rate, neonatal complication rate, and offspring outcomes between the TAI-positive and TAI-negative groups were comparable (p > 0.05). In women with NOR, a higher rate of live births (44.94% vs. 40.34%, p = 0.027) and a higher prevalence of congenital anomalies (4.68% vs. 2.14%, p = 0.005) were observed in the TAI-positive group.

Conclusions:

TAI had no impact on ovarian reserve in euthyroid women and had no association with IVF/ICSI outcomes in women with DOR. Although an increased incidence of congenital anomalies in the TAI-positive group was observed in women with NOR, an association between neonatal anomalies and TAI cannot be demonstrated. Large cohort studies to evaluate the effects of TAI on offspring health are warranted, and further experimental studies are required to explore the underlying mechanisms.

Introduction

Thyroid autoimmunity (TAI), a leading cause of thyroid dysfunction, is the most frequent autoimmune disease among women of reproductive age and is characterized by the presence of anti-thyroid peroxidase antibodies (TPOAbs) and/or anti-thyroglobulin antibodies (TGAbs). TAI has been independently associated with reproductive function, even in euthyroid women, but the findings differ considerably among studies and conclusive evidence has remained elusive. Currently, no effective interventions are available for euthyroid women with positive TAI pursuing assisted reproductive technology treatment.

It has been observed that TAI occurs more frequently in women with premature ovarian insufficiency (POI) (1,2), with a prevalence of 14–27% compared with 9–15% in the general population (2,3). However, the association is yet to be definitively proven, and the exact mechanisms behind the association remain unknown. The ovary shares multiple antigens with the thyroid gland, which makes it a potential target for thyroid autoantibodies present in the circulation or follicular fluid (4). The organ-specific and systematic autoimmune disturbances induced by TAI might accelerate the apoptosis and atresia of ovarian follicles (5). A detrimental effect of TAI on ovarian reserve has been hypothesized, but studies evaluating the association between TAI and the ovarian reserve have yielded conflicting results. TAI has been correlated with both higher (6,7) and lower (8,9) ovarian reserve, while two recent large cross-sectional studies failed to find any significant association (10,11). More evidence is needed to establish the association between TAI and ovarian reserve.

TAI is also considered to be associated with oocyte and embryo quality (12) and with pregnancy outcomes and offspring health (13). A number of studies have observed adverse effects of TAI on pregnancy outcomes, including lower implantation and live birth rates and a higher risk of pregnancy loss or preterm delivery, following in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) cycles. Still, controversies have arisen due to variable study populations and research designs (14 –18), and the potential modifying effect of age and serum thyrotropin (TSH) levels along with TAI positivity cannot be ignored (17). Furthermore, neonatal complications such as preterm birth and perinatal mortality and child neurodevelopment scores, including scores on both cognitive and motor scales, have been reported to be associated with TAI (19), but confirmation in large populations is lacking. Of note, diminished ovarian reserve (DOR) per se has been shown to be related to poor oocyte and embryo quality (20); thus, it is unknown whether TAI aggravates the poor pregnancy and neonatal outcomes in the DOR population. Therefore, it is warranted to further evaluate the effect of positive TAI on IVF/ICSI outcomes and offspring health among women with different ovarian reserve status.

Here, we aimed to clarify the relationship between positive thyroid antibodies and ovarian reserve and to evaluate the impact of TAI on IVF/ICSI outcomes and offspring health stratified by ovarian reserve status.

Materials and Methods

Ethics approval

Written informed consent was obtained from each participant. The study was approved by the institutional review board of the Reproductive Hospital Affiliated to Shandong University.

Participants

Patients undergoing their first cycle of IVF/ICSI in the Reproductive Hospital Affiliated to Shandong University from January 2012 to December 2017 were screened for eligibility. Patients aged 20–40 years with normal thyroid function (TSH = 0.27–4.2 μIU/mL) were included. The exclusion criteria included the following: (a) history of thyroid disease, thyroid hormone/antithyroid medication, or thyroid surgery; (b) history of chemotherapy, radiotherapy, or ovarian surgery; (c) history of autoimmune disease; (d) X chromosome abnormalities; and (e) receiving preimplantation genetic diagnosis or preimplantation genetic screening. According to TAI status, participants were divided into two groups: the TAI-positive group, with the presence of TPOAb (>34 IU/mL) and/or TGAb (>115 IU/mL), and the TAI-negative group, with the absence of both TPOAb and TGAb. According to ovarian reserve, the patients were subdivided to the DOR group with basal follicle-stimulating hormone (FSH) levels ≥12 IU/L and the normal ovarian reserve (NOR) group with basal FSH levels <12 IU/L.

Study outcomes

Ovarian reserve and response

In evaluating the relationship between TAI and ovarian reserve, the prevalence of DOR (FSH level ≥12 IU/L) in the TAI-negative and TAI-positive groups was calculated. Age-stratified comparison of DOR occurrence was also performed. Ovarian response parameters included the duration of ovarian stimulation (from day of gonadotropin initiation to day of human chorionic gonadotropin [hCG] administration), total gonadotropin dosage, and estradiol (E2) and progesterone levels on the day of hCG trigger, and the number of oocytes retrieved was also recorded.

IVF/ICSI outcomes

Live birth, defined as the delivery of at least one live-born infant after 28 weeks of gestation, was the primary endpoint. Secondary outcomes included biochemical pregnancy (confirmed by serum hCG >25 IU/L at two weeks after embryo transfer), clinical pregnancy (diagnosed by observation of a gestational sac on transvaginal ultrasound scan at 5–7 weeks after embryo transfer), and pregnancy loss (fetal loss in a spontaneous abortion or therapeutic abortion that occurred at any point during gestation). Early pregnancy loss refers to miscarriage within the first 12 weeks of gestation, and late pregnancy loss refers to abortion between 13 and 28 weeks of gestation.

Neonatal outcomes

Neonatal adverse events were defined as follows (21): stillbirth, a baby born with no signs of life at or after 28 weeks of gestation; congenital anomalies, including structural and/or functional abnormalities present from birth and evaluated according to the International Classification of Diseases, 10th Revision (ICD-10) Chapter XVII; neonatal death, death of a live birth within 28 days after delivery; low birth weight, a birth weight of <2500 g regardless of gestational age; and very low birth weight, a birth weight of <1500 g. In the follow-up study, the length (height), weight, body mass index (BMI), head circumference, and temperament type (see Neonatal follow-up section) of the children were measured.

Biochemical measures and ultrasonography

Endocrine hormones, including FSH, luteinizing hormone (LH), prolactin, E2, and testosterone, were measured on days 2–4 of the menstrual cycle through chemiluminescence immunoassays (Roche Diagnostics). Anti-Müllerian hormone (AMH) was measured with an ultra-sensitive enzyme-linked immunosorbent assay (Kangrun Biotech). Free triiodothyronine (fT3), free thyroxine (fT4), TSH, TPOAb, and TGAb were measured by automatic chemiluminescence immunoassays (Roche Diagnostics). The intra- and inter-assay coefficients of variation were <10% and <15%, respectively. Antral follicle count (AFC) was recorded as the number of follicles (2–10 mm in diameter) through transvaginal ultrasonography in the early follicular phase.

IVF procedure

Patients enrolled in the present study received a standardized ovarian stimulation regimen, which included a gonadotropin-releasing hormone (GnRH) agonist short protocol, a GnRH agonist long protocol, a natural protocol, a microdose flare-up protocol, a GnRH antagonist protocol, and an ultra-long protocol (22). When at least two dominant follicles reached 18 mm in diameter, 4000–8000 IU urinary hCG (Livzon) was injected to induce ovulation, and oocyte retrieval was performed 34–36 hours after hCG administration. Based on sperm quality, oocytes were fertilized through conventional insemination or ICSI. On day 3 of the embryo culture, up to three embryos were selected and transferred. Luteal-phase support was initiated on the day of oocyte retrieval with oral dydrogesterone (Duphaston, Abbott) at a daily dose of 20 mg or vaginal progesterone (Utrogestan, Besins Healthcare) at a daily dose of 200 mg until 12 weeks after conception.

Neonatal follow-up

At 6 weeks after delivery, participants were followed up by telephone for recording birth weight and adverse events in the newborns. A total of 287 cases participated in our long-term follow-up program and were evaluated for child growth parameters (from 6 months to 3 years old) and/or temperament type (from 6 months to 2 years old). The temperament type was evaluated by assessing the following 9 parameters: activity level, rhythmicity, approach/withdrawal, adaptability, intensity of reaction, attention span (distractibility), persistence, quality of mood, and threshold of reaction. The Chinese Infant Temperament Scale (CITS) was modified from the Revised Infant Temperament Questionnaire (23) and validated in 620 infants throughout China (24). Currently, the CITS can be used in Chinese infants between 4 months and 1 year old (25). The Chinese Toddlers Temperament Scale was revised from the Toddler Temperament Scale (26) for 1–3 years old children and validated in 3486 toddlers from six great administrative districts in China (27).

Data analysis

The Statistical Package for the Social Sciences (version 23.0; IBM Corp., Armonk, NY) was used for data analysis. The Kolmogorov–Smirnov test and quantile–quantile plot were combined to test for normality. Continuous variables with a normal distribution were compared by Student's t-test and are presented as mean ± standard deviation. Continuous variables that were not normally distributed were compared with the Mann–Whitney U-test and are presented as median (interquartile range). Categorical variables were analyzed by chi-square or Fisher's exact test as appropriate and are presented as counts (percentage), and z-scores with regard to child growth parameters were calculated based on the World Health Organization 2006 Child Growth Standards for 0–60 months (28). Binary logistic regression was used to adjust for the effect of baseline characteristics and to calculate the odds ratios (ORs) and 95% confidence intervals (CIs). A fixed-effect binary logistic regression model was adopted. Post hoc subgroup analyses of TSH levels >2.5 IU/L on pregnancy outcomes were performed and presented as the relative risk (RR) and CI. A two-tailed value of p < 0.05 was considered statistically significant.

Results

Baseline characteristics

Among the recruited 6213 euthyroid patients, 1075 (17.3%) were positive for TAI and 5138 (82.7%) were negative for TAI. The TAI-positive group had a greater mean age than the TAI-negative group (31.04 ± 4.37 years old vs. 30.57 ± 4.34 years old, p = 0.001). No differences were observed in terms of BMI, duration of infertility, prevalence of nulliparity, or causes of infertility (p > 0.05). As expected, women with positive TAI had mildly higher serum TSH level (2.36 ± 0.94 μIU/mL vs. 2.30 ± 0.90 μIU/mL, p = 0.042), but comparable fT3 (5.00 ± 0.76 pmol/L vs. 5.02 ± 0.78 pmol/L, p = 0.465) and fT4 levels (17.56 ± 2.95 pmol/L vs. 17.52 ± 2.96 pmol/L, p = 0.677). The baseline characteristics of the participants are shown in Table 1.

Table 1.

Baseline Characteristics of the Participants

Characteristic	Total	Positive TAI	Negative TAI	p ^*
N	6213	1075	5138
Age (years)	30.65 ± 4.35	31.04 ± 4.37	30.57 ± 4.34	0.001
BMI (kg/m²)	23.88 ± 5.47	23.64 ± 3.71	23.93 ± 5.77	0.118
Duration of infertility (years)	3.5 (2–5)	3 (2–6)	3.5 (2–5)	0.382
Nulliparity	3321 (53.45)	570 (53.02)	2751 (53.54)	0.756
Cause of infertility
Male factor	1865 (30.02)	333 (30.98)	1532 (29.82)	0.624
Tubal factor	4744 (76.36)	831 (77.30)	3913 (76.16)
Endometriosis	130 (2.09)	25 (2.33)	105 (2.04)
PCOS/ovulation disorder	1234 (19.86)	198 (18.42)	1036 (20.16)
Unexplained	209 (3.36)	40 (3.72)	169 (3.29)
TSH (μIU/mL)	2.31 ± 0.91	2.36 ± 0.94	2.30 ± 0.90	0.042
fT3 (pmol/L)	5.01 ± 0.77	5.00 ± 0.76	5.02 ± 0.78	0.465
fT4 (pmol/L)	17.53 ± 2.96	17.56 ± 2.95	17.52 ± 2.96	0.677
TGAb^a (IU/mL)	16.37 (11.50–26.81)	257.10 (133.90–415.70)	14.65 (10.73–19.63)	<0.001
TPOAb^b (IU/mL)	8.88 (6.21–13.46)	40.64 (10.29–143.10)	8.23 (5.94–11.48)	<0.001

Data are expressed as the mean ± standard deviation, median (interquartile range), or number of observations (percentage).

Differences between the TAI-positive and TAI-negative groups.

Refers to 954 values below 5 IU/mL or above 600 IU/mL.

Refers to 1071 values below 10 IU/mL or above 4000 IU/mL.

BMI, body mass index; fT3, free triiodothyronine; fT4, free thyroxine; PCOS, polycystic ovarian syndrome; TAI, thyroid autoimmunity; TGAb, anti-thyroglobulin antibody; TPOAb, anti-thyroid peroxidase antibody; TSH, thyrotropin.

Ovarian reserve and response

The serum FSH level was higher in the TAI-positive group compared with the TAI-negative group (6.53 [5.59–7.80] IU/L vs. 6.41 [5.47–7.58] IU/L, p = 0.011), but the difference was no longer significant after adjusting for confounding factors including age, BMI, and serum TSH level (p _adj = 0.215). The AMH level and bilateral AFC were comparable between the two groups (p > 0.05). Women with positive TAI presented with a slightly higher prevalence of DOR, but the difference did not reach statistical significance (4.09% vs. 2.96%, p = 0.053) (Table 2). To attenuate the impact of age, patients were stratified into three age subgroups (20–30, 31–37, and 38–40 years) (29). However, no significant difference in DOR occurrence was observed between positive and negative TAI, that is, 2.45% vs. 1.75% in the 20–30 year age group (p = 0.272), 5.17% vs. 3.43% in the 31–37 year age group (p = 0.080), and 8.00% vs. 8.85% in the 38–40 year age group (p = 0.788) (Table 2). Subsequently, indicators for ovarian reserve were compared between patients as per TGAb and TPOAb status separately. Neither TGAb positivity nor TPOAb positivity was associated with high prevalence of DOR, after correcting for age, BMI, and serum TSH levels (Supplementary Tables S1 and S2).

Table 2.

The Impact of Thyroid Autoimmunity on Ovarian Reserve

Characteristic	Positive TAI	Negative TAI	p	p-adj^*	OR [CI]-adj^**
N	1075	5138
FSH (IU/L)	6.53 (5.59–7.80)	6.41 (5.47–7.58)	0.011	0.215	—
LH (IU/L)	4.82 (3.58–6.54)	4.82 (3.52–6.50)	0.755	0.763	—
E2 (pg/mL)	33.80 (26.30–44.93)	33.50 (25.10–45.00)	0.246	0.924	—
AMH^a (ng/mL)	3.22 (1.48–5.95)	3.21 (1.58–5.92)	0.628	0.316	—
Bilateral AFC	13 (10 –19)	13 (9 –19)	0.334	0.516	—
DOR^b	44/1075 (4.09)	152/5138 (2.96)	0.053	0.154	1.287 [0.910–1.820]
20 ≤ Age ≤30	13/530 (2.45)	48/2746 (1.75)	0.272	0.327	1.365 [0.733–2.543]
31 ≤ Age ≤37	23/445 (5.17)	68/1985 (3.43)	0.080	0.127	1.464 [0.898–2.386]
38 ≤ Age ≤40	8/100 (8.00)	36/407 (8.85)	0.788	0.784	0.893 [0.399–1.998]

Data are expressed as the median (interquartile range) or as the number of observations/total number (percentage).

Adjusted p-value after correcting for age, BMI, and serum TSH level through binary logistic regression.

Adjusted OR [CI] for DOR occurrence associated with positive TAI, estimated by binary logistic regression in which age, BMI, and serum TSH level were accounted for.

Refers to 5827 patients with AMH data, in which 136 values (2.33%) below or above the detection limit.

DOR was defined as basal FSH levels ≥12 IU/L.

AFC, antral follicle count; AMH, anti-Müllerian hormone; CI, 95% confidence interval; DOR, diminished ovarian reserve; FSH, follicle-stimulating hormone; E2, estradiol; LH, luteinizing hormone; OR, odds ratio; PRL, prolactin; T, testosterone.

As for parameters indicating ovarian response in IVF/ICSI cycles, the distribution of controlled ovarian hyperstimulation protocols was comparable between the TAI-negative and TAI-positive groups. No differences were observed in terms of the number of days of ovarian stimulation, total gonadotropin dosage, E2 level on hCG trigger day, or the number of oocytes retrieved between the two groups (p > 0.05) (Supplementary Table S3).

Pregnancy outcomes

Patients were further subdivided into DOR (FSH ≥12 IU/L) and NOR subgroups (FSH <12 IU/L) to evaluate the effect of TAI on pregnancy outcomes. In 123 patients with DOR receiving embryo transfer, no differences were observed between the TAI-positive and TAI-negative groups in terms of live birth rate (35.71% vs. 35.79%, p = 0.994), biochemical pregnancy rate (46.43% vs. 52.63%, p = 0.564), clinical pregnancy rate (42.86% vs. 44.21%, p = 0.899), or twin delivery rate (7.14% vs. 4.21%, p = 0.628). The pregnancy loss rates, including both biochemical and clinical pregnancy and early and late pregnancy loss, were also similar between the two groups (p > 0.05). Among women giving live birth, the preterm delivery rate was 10.00% in the TAI-positive group and 8.82% in the TAI-negative group (p = 1.000). The pregnancy outcomes in DOR patients with different TAI status are summarized in Table 3.

Table 3.

Pregnancy Outcomes Stratified by Thyroid Autoimmunity Status in Women with Diminished Ovarian Reserve (Follicle-Stimulating Hormone ≥12 IU/L)

	Positive TAI	Negative TAI	p	p-adj^**
Total transplantation cycles, N	28	95	—
Timing of embryo transferred
Day 2	8 (28.57)	16 (16.84)	0.253^*
Day 3	19 (67.86)	68 (71.58)		—
Day 5	1 (3.57)	11 (11.58)
No. of embryo transferred
1	10 (35.71)	39 (41.05)	0.739^*
2	18 (64.29)	55 (57.89)		—
3	0	1 (1.05)
Biochemical pregnancy rate^a	13/28 (46.43)	50/95 (52.63)	0.564	0.608
Clinical pregnancy rate^b	12/28 (42.86)	42/95 (44.21)	0.899	0.908
Live birth rate^c	10/28 (35.71)	34/95 (35.79)	0.994	0.753
Twin delivery	2/28 (7.14)	4/95 (4.21)	0.628^*	0.199
Any pregnancy loss rate^d
Among biochemical pregnancies	1/13 (7.69)	7/50 (14)	1.000^*	0.318
Among clinical pregnancies	2/12 (16.67)	8/42 (19.04)	1.000^*	0.497
Early^e	1/12 (8.33)	8/42 (19.04)	0.665^*	0.211
Late^f	1/12 (8.33)	0/42 (0)	0.222^*	0.993
Preterm delivery among live birth^g	1/10 (10)	3/34 (8.82)	1.000^*	0.925

Data are expressed as the number of observations (percentage) or as the number of observations/total number (percentage).

Fisher's exact test.

Adjusted p-value after correcting for age, BMI, serum TSH level, and timing and number of embryos transferred through binary logistic regression.

Biochemical pregnancy was defined as serum hCG >25 IU/L at 2 weeks after embryo transfer.

Clinical pregnancy was defined as the observation of a gestational sac on transvaginal ultrasound scan at 5–7 weeks after embryo transfer.

Live birth was defined as the delivery of at least one live-born infant after 28 weeks of gestation.

Pregnancy loss was defined as fetal loss in a spontaneous abortion or therapeutic abortion that occurred at any point during gestation.

Early pregnancy loss refers to abortion within the first 12 weeks of gestation.

Late pregnancy loss refers to abortion between 13 and 28 weeks of gestation.

Premature delivery was defined as delivery of live-born infant(s) after 28 weeks but before 37 weeks of gestation.

A significantly higher live birth rate was seen in the TAI-positive group compared with the TAI-negative group, both among all patients (44.57% vs. 40.21%, p = 0.032; Supplementary Table S4) and in the NOR subgroup (44.94% vs. 40.34%, p = 0.027; Supplementary Table S5). After correcting for age, BMI, serum TSH levels, and timing and number of embryos transferred, the difference was still statistically significant (p_adj < 0.05). Similarly, patients with positive TAI also achieved higher rates of biochemical pregnancies and clinical pregnancies (p_adj < 0.05). Other outcomes, including biochemical and clinical pregnancy loss, twin delivery, and preterm delivery, showed no differences between groups, even after ovarian reserve stratification (p > 0.05).

Neonatal outcomes

Among the 50 live newborns (12 with maternal positive TAI and 38 with maternal negative TAI) in the DOR subgroup, no neonatal adverse events were observed except for 4 live newborns with low birth weight in the TAI-negative group, but there were no statistically significant differences compared with the TAI-positive group (Table 4). Only three children underwent postnatal follow-up; thus, the children's growth parameters and temperament types could not be analyzed. As for women with NOR, the prevalence of neonatal congenital malformations was significantly higher in the TAI-positive group (4.68%, 18/385) compared with the TAI-negative group (2.14%, 37/1725) (p = 0.005). After adjusting for maternal age, BMI, TSH, number of embryos transferred, and gestational age at delivery, the difference remained significant (OR = 2.316 [CI 1.280–4.190]; p = 0.006). Detailed descriptions of the congenital anomalies are summarized in Supplementary Table S6. A total of 286 live newborns from 230 women with NOR were subsequently followed up for 3 years. The z-scores of BMI, length (height) for age, weight for age, and head circumference for age were comparable between the two TAI groups (Table 4). Among 128 children receiving temperament evaluation up to 2 years, no significant differences were observed (Table 4).

Table 4.

Neonatal Outcomes Stratified by Thyroid Autoimmunity Status in Women with Normal and Diminished Ovarian Reserve

	DOR (FSH ≥12 IU/L)			NOR (FSH <12 IU/L)
	Positive TAI	Negative TAI	p	Positive TAI	Negative TAI	p
Gestational age at delivery (weeks)	39.09 ± 1.39 (10)	38.89 ± 1.55 (34)	0.717	38.54 ± 2.11 (306)	38.42 ± 2.09 (1330)	0.411
Birth weight, g (n)
Singleton	3294 ± 603 (8)	3322 ± 390 (30)	0.873	3410 ± 552 (229)	3406 ± 541 (926)	0.736
Twin	3175 ± 260 (4)	2331 ± 504 (8)	0.011	2615 ± 431 (148)	2612 ± 473 (778)	0.261
Neonatal complications, n/N (%)
Stillbirth among all deliveries	0/10 (0)	0/34 (0)	—	1/307 (0.33)	2/1332 (0.15)	0.463^*
Congenital anomalies among live newborns^a	0/12 (0)	0/38 (0)	—	18/385 (4.68)	37/1725 (2.14)	0.005^**
Neonatal death among live newborns^a	0/12 (0)	0/38 (0)	—	2/385 (0.52)	10/1725 (0.58)	1.000^*
Low birth weight among live newborns^b	0/12 (0)	4/38 (10.53)	0.560^*	60/383 (15.67)	306/1713 (17.86)	0.306
Very low birth weight among live newborns^b	0/12 (0)	0/38 (0)	—	3/383 (0.78)	20/1713 (1.17)	0.785^*
Children growth (n)
BMI z-score	—	—	—	0.61 ± 1.29 (52)	0.78 ± 1.19 (234)	0.372
LFA z-score	—	—	—	0.42 ± 1.08 (52)	0.64 ± 1.15 (234)	0.204
WFA z-score	—	—	—	0.71 ± 1.21 (52)	0.97 ± 1.41 (234)	0.218
HCFA z-score	—	—	—	0.08 ± 0.71 (52)	0.28 ± 0.95 (234)	0.099
Temperament type, n/N (%)
Easy	—	—	—	14/20 (70)	57/108 (52.78)	0.605^*
Intermediate	—	—		6/20 (30)	43/108 (39.81)
Difficult	—	—		0/20 (0)	6/108 (5.56)
Show up to warm	—	—		0/20 (0)	2/108 (1.85)

Fisher's exact test.

A significant difference (P _adj = 0.006) was observed after correcting for maternal age, BMI, TSH level, timing and number of embryos transferred, and gestational age at delivery. The adjusted odds ratio was 2.316 [CI 1.280–4.190].

The denominator was the number of live newborns with neonatal following data.

The denominator was the number of live newborns with birthweight data.

HCFA, head circumference for age; LFA, length (height) for age; NOR, normal ovarian reserve; WFA, weight for age; z-score, standard score.

Discussion

In the present study, we demonstrate that in euthyroid women, TAI positivity was not associated with DOR as defined by elevated FSH level (≥12 IU/L), even when stratified into distinct age subgroups. Among women with DOR, no difference with respect to pregnancy outcomes was found between the TAI-positive and TAI-negative groups. However, in both the total and NOR population, higher pregnancy and live birth rates were observed in patients with positive TAI, indicating no compromising effect of positive TAI on the course of pregnancy. In addition, a trend toward a greater number of congenital anomalies in women with NOR with positive TAI was observed compared with those with negative TAI. To the best of our knowledge, this is the largest cohort study to clarify the relationship between TAI and ovarian reserve in euthyroid women and the first to focus on the impact of TAI on IVF/ICSI outcomes and offspring health in women with DOR.

TAI status and ovarian reserve

The relationship between ovarian reserve and TAI is controversial, and AMH has frequently been used as an indicator of ovarian reserve. Both higher (6,7) and lower (8,9) AMH levels have been reported in euthyroid women with autoimmune thyroid diseases or Hashimoto's thyroiditis. Two large cross-sectional studies stratified patients by age-specific AMH level but came to different conclusions. Polyzos et al. (10) found no difference in TPOAb positivity depending on ovarian reserve status, while Chen et al. (11) found that TPOAb positivity but not TGAb positivity was significantly associated with idiopathic low ovarian reserve, particularly in subgroup of women with unexplained infertility. More recently, a cohort study by Korevaar et al. (30) reported that infertility causes might affect the association between TAI and ovarian reserve. TPOAb positivity was negatively correlated with AFC in women with DOR or unexplained infertility but had no association with either AFC or day 3 FSH levels in the whole infertile population. Nevertheless, the negative association between TGAb positivity and AFC was only identified in the whole cohort.

Compared with previous studies, we did not observe significant differences in AMH or AFC between the two groups, and we failed to see a relationship between TAI and DOR. However, there were differences in study design that might have influenced the differences in outcomes. First, only euthyroid patients were recruited in our study, while previous studies also included subjects with hypothyroidism or subclinical hypothyroidism, and thyroid dysfunction was found to be a strong contributor to DOR (12,31). Second, in contrast to patients with mixed causes investigated by Polyzos et al. (10) and Korevaar et al. (30), recognized causes of compromised ovarian function were excluded in the present study. The distribution of different causes of infertility among recruited women might be another explanation. Both Chen et al. (11) and Korevaar et al. (30) identified a relationship of positive TPOAb with lower ovarian reserve or lower AFC among women with unexplained infertility. In our cohort, a higher frequency of DOR in positive TAI among women with unexplained infertility was also identified (10% [4/40] vs. 1.78% [3/169], p = 0.009). However, it only accounted for a minority of patients (3.36%, 209 of 6213) in the whole population. Third, different definitions and criteria for DOR have been used. In two previous studies, ovarian reserve was graded by age-specific AMH, and DOR was defined as being below the 10th percentile or 15th percentile for age-specific AMH. Nonetheless, DOR was defined by elevated FSH level (32) in the current study, and confounding factors such as age, BMI, and TSH were adjusted in a multivariable logistic regression model.

DOR and POI lie on the same spectrum of ovarian reserve diminishment (3), and they are considered to share a similar distribution of etiologies, including immune factors (33). Our findings regarding DOR are distinct from the relationship between TAI and POI reported previously (2,34,35). Regarding this disagreement, Busnelli et al. (17) proposed the “all or nothing effect” model in which TAI positivity only manifests itself in a small number of affected women and accelerates ovarian reserve exhaustion, otherwise the ovarian reserve remains intact. However, the exact mechanism through which TAI has a less detrimental impact on relatively mild DOR phenotype still needs further exploration in clinical and laboratory studies. Further studies in larger populations with different ovarian reserve status and large enough subgroups of different causes of DOR, such as women with unexplained infertility, might help unravel the true relationship between TAI and ovarian reserve. In addition, advanced age and thyroid dysfunction along with TAI might exert a dominant destructive effect on oocyte quality and might attenuate the compromising effects of other factors. Thus, different subgroups divided according to age, TSH level, or cause of DOR might help to elucidate the actual and independent effect of thyroid antibodies on ovarian reserve.

TAI status and pregnancy outcomes

Previous studies have been conducted to evaluate the impact of TAI on different stages of the reproductive process, but these results have also been controversial. Thangaratinam et al. (36) and Busnelli et al. (17) found an increased risk of pregnancy loss and a decreased rate of live births in patients with TAI positivity. Conversely, such trends were not detected by van den Boogaard et al. (19) in subjects with loose inclusion criteria or by He et al. (37) in women without subclinical hypothyroidism. Clinical evidence has also shown that thyroid autoantibodies might compromise embryo quality in IVF/ICSI (38,39).

Given the association of DOR per se with limited numbers of oocytes, deficient oocyte competence, and adverse pregnancy events, it was further explored whether TAI positivity can aggravate the poor outcomes in women with DOR. A recent study by Beydilli Nacak et al. (40) investigated the impact of TAI positivity on ICSI outcomes in women with DOR stratified by early ovarian aging (19–35 years old) and age-related poor ovarian reserve (36–43 years old). Women with early ovarian aging had higher TPOAb positivity and achieved a lower clinical pregnancy rate, suggesting a possible contribution of TAI to poor outcomes (RR = 2.8 [CI 1.2–6.3]). Different from their findings, no association between TAI and pregnancy outcomes in patients with DOR was observed in our data. Poor outcomes in women with DOR might be determined mainly by reduced oocyte quality and quantity, whereas the impact of TAI might only be modest. The limited number of subjects in the DOR subgroup might also explain the negative association seen in our study.

Both individual studies and meta-analyses have shown comparable or poor pregnancy outcomes with positive TAI. Also, basic research studies have confirmed the binding of TPOAb to preimplantation embryos in mouse models (41) and the presence of TGAb in human ovarian follicular fluid (42), which might generate antibody-mediated cytotoxicity and poor outcomes. However, we found that patients with TAI positivity had better outcomes, both in the overall population and in the NOR subgroup, with an approximate 5% increase in live birth rate and biochemical/clinical pregnancy rate. Other confounding factors, such as embryo quality, intrauterine environment, and pregnancy complications, might be unevenly distributed between the TAI-positive and TAI-negative groups and consequently contribute to the unexpected results. In addition, it is generally assumed that an absolute difference of 10% in live birth rate will be of clinical significance (43,44). Therefore, the small effect size concerning pregnancy outcomes in the current study suggested that TAI might not result in significant changes following IVF/ICSI treatment in our large cohort. Also, the identified association might be spurious, as we did not adjust the analysis for multiple comparisons.

TAI status and neonatal outcomes

Generally, TAI has been linked with neonatal complications and neurodevelopment impairment (13,19). Neonatal death, stillbirth, and congenital malformations in relation to TAI were addressed by Benhadi et al. (45), but no association was observed because no participants had adverse pregnancy outcomes. In the present study, there were no adverse neonatal events except for four live newborns with low birth weight among women with DOR. As for the NOR population, TAI positivity did not increase the risk of perinatal mortality, stillbirth, or low or very low birth weight but did exhibit a possible correlation with a higher incidence of congenital anomalies compared with those with absent TAI. Given that the biological plausibility of a relationship between TAI and congenital anomalies is unclear and that the anomalies recorded in our cohort were heterogeneous, the results should be interpreted carefully and no firm conclusions should be drawn. In addition, the level of thyroid autoantibodies during pregnancy was unavailable in the present study, and thus, the effect of TAI positivity during the gestational period on perinatal and postnatal outcomes requires further exploration.

Several studies have investigated the detrimental impact of TAI on neurodevelopment. The offspring from TPOAb-positive mothers exhibited lower scores on general cognitive and motor development at 25–30 months (46), increased risk for attention deficit/hyperactivity problems at age of 3 years of age (47), and a lower intelligence quotient that was prominent at 4 years of age but was attenuated at 7 years (48). According to temperament scale evaluation for children younger than 2 years old, those in the TAI-positive group did not present with a more difficult temperament type in our cohort. However, these infants or toddlers might not have been old enough to display any discrepancy, and the results might also have been influenced by the limited sample size.

Strengths and limitations

Our study does have some limitations. First, it is a retrospective and observational cohort study, and the inevitable increase in the age of women with TAI positivity might have generated bias and erroneous data interpretation. However, confounding factors were accounted for in a multivariable logistic model, and thus, their modifying effect could partly be reduced. Second, only FSH was used to define DOR in the present study. Despite limitations due to intercycle and intracycle variability, an abnormally elevated FSH is still frequently used in the clinic because of its high positive predictive value. Third, the limitation in sample size might hide the true relationship between TAI and DOR. In addition, the number of patients proceeding with embryo transplantation was limited in the DOR group due to inadequate numbers of oocytes and embryos, resulting in the comparisons being susceptible to type II error. Adverse events were only rarely observed and only in a limited number of patients or children; thus, further studies in a larger population will provide more solid evidence. Last, a TSH level of ≤2.5 mIU/L before IVF or ICSI as recommended by the American Society for Reproductive Medicine (49) and the American Thyroid Association (50) for women with TAI was not used as the cutoff value for euthyroidism in the present study because this allowed us to recruit a much larger sample size. Post hoc subgroup analysis using a TSH cutoff level of 2.5 mIU/L was subsequently performed. No modifying effect on pregnancy outcomes was found in patients with DOR (Supplementary Table S7), although the numbers of subjects were quite small, limiting the power of the subgroup analysis to detect a difference. Among patients with NOR, the difference of IVF/ICSI outcomes between TAI status was only observed in the TSH <2.5 mIU/L subgroup (Supplementary Table S8).

In conclusion, our data demonstrate that TAI might not impair ovarian reserve within the euthyroid range. In addition, TAI did not appear to affect IVF/ICSI outcomes in women with either DOR or NOR, except for a potential relationship observed with congenital anomalies in women with NOR. Further longitudinal investigations and larger studies are required to validate these results and to explore the mechanism behind these effects.

Footnotes

Acknowledgment

We are grateful to the subjects for taking part in our research.

Author Disclosure Statement

The authors declare that the study was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Information

This work was supported by the National Key Research and Developmental Program of China (Grant Nos. 2017YFC1001100, 2018YFC1003803, and 2016YFC1000604), the National Natural Science Foundation of China (Grant Nos. 81522018, 31601198, 81601245, and 81971352), and the Natural Science Foundation of Jiangsu Province (Grant No. BK20160372).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

Supplementary Table S5

Supplementary Table S6

Supplementary Table S7

Supplementary Table S8

References

Welt

. 2008. Primary ovarian insufficiency: a more accurate term for premature ovarian failure. Clin Endocrinol (Oxf), 68:499–509.

Goswami

, Marwaha

, Goswami

, Gupta

, Ray

, Tomar

, Singh

. 2006. Prevalence of thyroid autoimmunity in sporadic idiopathic hypoparathyroidism in comparison to type 1 diabetes and premature ovarian failure. J Clin Endocrinol Metab, 91:4256–4259.

European Society for Human

, Embryology Guideline Group on

POI

, Webber

, Davies

, Anderson

, Bartlett

, Braat

, Cartwright

, Cifkova

, de Muinck Keizer-Schrama

, Hogervorst

, Janse

, Liao

, Vlaisavljevic

, Zillikens

, Vermeulen

. 2016. ESHRE Guideline: management of women with premature ovarian insufficiency. Hum Reprod, 31:926–937.

Kosir Pogacnik

, Meden Vrtovec

, Vizjak

, Ursula Levicnik

, Slabe

, Ihan

. 2014. Possible role of autoimmunity in patients with premature ovarian insufficiency. Int J Fertil Steril, 7:281–290.

Forges

, Monnier-Barbarino

, Faure

, Bene

. 2004. Autoimmunity and antigenic targets in ovarian pathology. Hum Reprod Update, 10:163–175.

Magri

, Schena

, Capelli

, Gaiti

, Zerbini

, Brambilla

, Rotondi

, De Amici

, Spinillo

, Nappi

, Chiovato

. 2015. Anti-Mullerian hormone as a predictor of ovarian reserve in ART protocols: the hidden role of thyroid autoimmunity. Reprod Biol Endocrinol, 13:106.

Saglam

, Onal

, Ersoy

, Koca

, Ergin

, Erel

, Cakir

. 2015. Anti-Mullerian hormone as a marker of premature ovarian aging in autoimmune thyroid disease. Gynecol Endocrinol, 31:165–168.

Tuten

, Hatipoglu

, Oncul

, Imamoglu

, Acikgoz

, Yilmaz

, Ozcil

, Kaya

, Misirlioglu

, Sahmay

. 2014. Evaluation of ovarian reserve in Hashimoto's thyroiditis. Gynecol Endocrinol, 30:708–711.

Pirgon

, Sivrice

, Demirtas

, Dundar

. 2016. Assessment of ovarian reserve in euthyroid adolescents with Hashimoto thyroiditis. Gynecol Endocrinol, 32:306–310.

10.

Polyzos

, Sakkas

, Vaiarelli

, Poppe

, Camus

, Tournaye

. 2015. Thyroid autoimmunity, hypothyroidism and ovarian reserve: a cross-sectional study of 5000 women based on age-specific AMH values. Hum Reprod, 30:1690–1696.

11.

Chen

, Huang

, Tzeng

, Huang

, Chen

. 2017. Idiopathic low ovarian reserve is associated with more frequent positive thyroid peroxidase antibodies. Thyroid, 27:1194–1200.

12.

Weghofer

, Barad

, Darmon

, Kushnir

, Gleicher

. 2016. What affects functional ovarian reserve, thyroid function or thyroid autoimmunity?. Reprod Biol Endocrinol, 14:26.

13.

De Leo

, Pearce

. 2018. Autoimmune thyroid disease during pregnancy. Lancet Diabetes Endocrinol, 6:575–586.

14.

Chai

, Yeung

, Lee

, Li

, Ho

, Ng

. 2014. Live birth rates following in vitro fertilization in women with thyroid autoimmunity and/or subclinical hypothyroidism. Clin Endocrinol (Oxf), 80:122–127.

15.

Tan

, Dieterle

, Pechlavanis

, Janssen

, Fuhrer

. 2014. Thyroid autoantibodies per se do not impair intracytoplasmic sperm injection outcome in euthyroid healthy women. Eur J Endocrinol, 170:495–500.

16.

Karacan

, Alwaeely

, Cebi

, Berberoglugil

, Batukan

, Ulug

, Arvas

, Camlibel

. 2013. Effect of antithyroid antibodies on ICSI outcome in antiphospholipid antibody-negative euthyroid women. Reprod Biomed Online, 27:376–380.

17.

Busnelli

, Paffoni

, Fedele

, Somigliana

. 2016. The impact of thyroid autoimmunity on IVF/ICSI outcome: a systematic review and meta-analysis. Hum Reprod Update, 22:775–790.

18.

Lukaszuk

, Kunicki

, Kulwikowska

, Liss

, Pastuszek

, Jaszczolt

, Meczekalski

, Skowronski

. 2015. The impact of the presence of antithyroid antibodies on pregnancy outcome following intracytoplasmatic sperm injection-ICSI and embryo transfer in women with normal thyreotropine levels. J Endocrinol Invest, 38:1335–1343.

19.

van den Boogaard

, Vissenberg

, Land

, van Wely

, van der Post

, Goddijn

, Bisschop

. 2011. Significance of (sub)clinical thyroid dysfunction and thyroid autoimmunity before conception and in early pregnancy: a systematic review. Hum Reprod Update, 17:605–619.

20.

Devine

, Mumford

, Wu

, DeCherney

, Hill

, Propst A 2015 Diminished ovarian reserve in the United States assisted reproductive technology population: diagnostic trends among 181,536 cycles from the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System. Fertil Steril, 104:612.e613–619.e613.

21.

Chen

, Shi

, Sun

, Zhang

, Liang

, Cao

, Yang

, Liu

, Wei

, Weng

, Tian

, Hao

, Yang

, Zhou

, Shi

, Xu

, Li

, Yan

, Qin

, Zhao

, Zhang

, Legro

. 2016. Fresh versus Frozen embryos for infertility in the polycystic ovary syndrome. N Engl J Med, 375:523–533.

22.

Yang

, Sun

, Cui

, Sheng

, Tang

, Wei

, Qin

, Li

, Chen

. 2016. Younger poor ovarian response women achieved better pregnancy results in the first three IVF cycles. Reprod Biomed Online, 32:532–537.

23.

Carey

, McDevitt

. 1978. Revision of the infant temperament questionnaire. Pediatrics, 61:735–739.

24.

Zhang

, Yao

, Yang

, Hong

, Liu

. 1998. The compitation of the CITS National Norms. Chin J Child Health Case, 3:204–206.

25.

Liu

, Huang

, Zhong

. 2007. Survey and clinical application on temperament types of 4–12 months old babies in Daya Bay. Hainan Med J, 18:50–51.

26.

Fullard

, McDevitt

, Carey

. 1984. Assessing temperament in one-to three-year-old children. J Pediatr Psychol, 9:205–217.

27.

Hong

, Zhou

, Yao

, Liu

. 1998. Standardized and revised the toddler temperament scale. Chin J Child Health Case, 6:201–204.

28.

Group

WHOMGRS

. 2006. WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl, 450:76–85.

29.

La Marca

, Nelson

, Sighinolfi

, Manno

, Baraldi

, Roli

, Xella

, Marsella

, Tagliasacchi

, D'Amico

, Volpe

. 2011. Anti-Mullerian hormone-based prediction model for a live birth in assisted reproduction. Reprod Biomed Online, 22:341–349.

30.

Korevaar

TIM

, Minguez-Alarcon

, Messerlian

, de Poortere

, Williams

, Broeren

, Hauser

, Souter

. 2018. Association of Thyroid Function and Autoimmunity with ovarian reserve in women seeking infertility care. Thyroid, 28:1349–1358.

31.

Michalakis

, Mesen

, Brayboy

, Yu

, Richter

, Levy

, Widra

, Segars

. 2011. Subclinical elevations of thyroid-stimulating hormone and assisted reproductive technology outcomes. Fertil Steril, 95:2634–2637.

32.

Tal

, Seifer

. 2017. Ovarian reserve testing: a user's guide. Am J Obstet Gynecol, 217:129–140.

33.

Gleicher

, Weghofer

, Oktay

, Barad

. 2009. Do etiologies of premature ovarian aging (POA) mimic those of premature ovarian failure (POF)?. Hum Reprod, 24:2395–2400.

34.

Gleicher

, Weghofer

, Barad

. 2009. A pilot study of premature ovarian senescence: II. Different genotype and phenotype for genetic and autoimmune etiologies. Fertil Steril, 91:1707–1711.

35.

Ayesha, Jha

, Goswami

. 2016. Premature ovarian failure: an Association with Autoimmune Diseases. J Clin Diagn Res, 10:QC10–QC12.

36.

Thangaratinam

, Tan

, Knox

, Kilby

, Franklyn

, Coomarasamy

. 2011. Association between thyroid autoantibodies and miscarriage and preterm birth: meta-analysis of evidence. BMJ, 342:d2616.

37.

, Jing

, Gong

, Tan

, Lu

, Lin

. 2016. Effect of thyroid autoimmunity per se on assisted reproduction treatment outcomes: a meta-analysis. Taiwan J Obstet Gynecol, 55:159–165.

38.

Weghofer

, Himaya

, Kushnir

, Barad

, Gleicher

. 2015. The impact of thyroid function and thyroid autoimmunity on embryo quality in women with low functional ovarian reserve: a case-control study. Reprod Biol Endocrinol, 13:43.

39.

Andrisani

, Sabbadin

, Marin

, Ragazzi

, Dessole

, Armanini

, Dona

, Bordin

, Ambrosini

. 2018. The influence of thyroid autoimmunity on embryo quality in women undergoing assisted reproductive technology. Gynecol Endocrinol, 34:752–755.

40.

Beydilli Nacak

, Ozkaya

, Yayla Abide

, Bilgic

, Devranoglu

, Gokcen Iscan

. 2018. The impact of autoimmunity-related early ovarian aging on ICSI cycle outcome. Gynecol Endocrinol, 34:940–943.

41.

Lee

, Ng

, Lau

, Liu WM

O W

S, Yeung

, Kung

. 2009. Increased fetal abortion rate in autoimmune thyroid disease is related to circulating TPO autoantibodies in an autoimmune thyroiditis animal model. Fertil Steril, 91:2104–2109.

42.

Monteleone

, Parrini

, Faviana

, Carletti

, Casarosa

, Uccelli

, Cela

, Genazzani

, Artini

. 2011. Female infertility related to thyroid autoimmunity: the ovarian follicle hypothesis. Am J Reprod Immunol, 66:108–114.

43.

Wei

, Sun

, Liu

, Liang

, Zhu

, Shi

, Chen

. 2017. Live birth after fresh versus frozen single blastocyst transfer (Frefro-blastocyst): study protocol for a randomized controlled trial. Trials, 18:253.

44.

Shi

, Sun

, Hao

, Zhang

, Wei

, Zhang

, Zhu

, Deng

, Qi

, Li

, Ma

, Ren

, Wang

, Zhang

, Wang

, Liu

, Wu

, Wang

, Bai

, Li

, Zhou

, Sun

, Liu

, Li

, Zhang

, Chen

, Zhang

, Sun

, Legro

, Chen

. 2018. Transfer of Fresh versus Frozen Embryos in Ovulatory Women. N Engl J Med, 378:126–136.

45.

Benhadi

, Wiersinga

, Reitsma

, Vrijkotte

, Bonsel

. 2009. Higher maternal TSH levels in pregnancy are associated with increased risk for miscarriage, fetal or neonatal death. Eur J Endocrinol, 160:985–991.

46.

, Shan

, Teng

, Yu

, Li

, Fan

, Teng

, Guo

, Wang

, Li

, Chen

, Wang

, Chawinga

, Zhang

, Yang

, Zhao

, Hua

. 2010. Abnormalities of maternal thyroid function during pregnancy affect neuropsychological development of their children at 25–30 months. Clin Endocrinol (Oxf), 72:825–829.

47.

Ghassabian

, Bongers-Schokking

, de Rijke

, van Mil

, Jaddoe

, de Muinck Keizer-Schrama

, Hooijkaas

, Hofman

, Visser

, Roman

, Visser

, Verhulst

, Tiemeier

. 2012. Maternal thyroid autoimmunity during pregnancy and the risk of attention deficit/hyperactivity problems in children: the Generation R Study. Thyroid, 22:178–186.

48.

Wasserman

, Pillion

, Duggan

, Nelson

, Rohde

, Seaberg

, Talor

, Yolken

, Rose

. 2012. Childhood IQ, hearing loss, and maternal thyroid autoimmunity in the Baltimore Collaborative Perinatal Project. Pediatr Res, 72:525–530.

49.

Practice Committee of the American Society for Reproductive Medicine. 2015. Subclinical hypothyroidism in the infertile female population: a guideline. Fertil Steril, 104:545–553.

50.

Alexander

, Pearce

, Brent

, Brown

, Chen

, Dosiou

, Grobman

, Laurberg

, Lazarus

, Mandel

, Peeters

, Sullivan

. 2017. 2017 Guidelines of the American Thyroid Association for the Diagnosis and Management of thyroid disease during pregnancy and the postpartum. Thyroid, 27:315–389.