Impact of Thyroid Autoimmunity on Assisted Reproductive Technology Outcomes and Ovarian Reserve Markers: An Updated Systematic Review and Meta-Analysis

Abstract

Background:

Thyroid autoimmunity (TAI) has a high prevalence among women of reproductive age. Investigating its possible impact on ovarian function and fertility is, thus, of utmost relevance. The aim of this systematic review and meta-analysis was to elucidate the effect of TAI on both assisted reproductive technology (ART) outcomes and ovarian reserve.

Methods:

This systematic review and meta-analysis was restricted to two groups of research articles investigating the association between TAI and: (1) autologous ART outcomes (i.e., fertilization rate [FR], implantation rate, clinical pregnancy rate [CPR], miscarriage rate, and live birth rate), (2) markers of ovarian reserve (i.e., anti-Müllerian hormone, basal follicle stimulating hormone, antral follicle count, and number of oocytes retrieved). Studies including women affected by overt hypo/hyperthyroidism were excluded. Relevant studies were identified by a systematic search in PubMed, MEDLINE, ClinicalTrials.gov, Embase, and Scopus, from database inception to May 1, 2022.

Results:

From a total of 432 identified publications, 22 studies were included in Group 1 and 26 studies in Group 2. The presence of TAI was associated with a higher risk of miscarriage (7606 participants, odds ratio [OR] 1.52, confidence interval [CI 1.14–2.01], p = 0.004, I ² = 53%), lower chance of embryo implantation (7118 participants, OR 0.72, [CI 0.59–0.88], p = 0.001, I ² = 36%), and live birth (11417 participants, OR 0.73, [CI 0.56–0.94], p = 0.02, I ² = 71%). These associations were no longer observed in a subgroup analysis of patients who exclusively underwent intracytoplasmic sperm injection (ICSI). The FR and CPR as well as the mean values of surrogate markers of oocyte quantity appeared not to be affected by TAI.

Conclusions:

This data synthesis suggest a higher risk of adverse ART outcomes in women with positive TAI. However, the reliability of these findings is hampered by the relatively low quality of the evidence and significant heterogeneity in many of the meta-analyses. The possible protective effect of ICSI is promising but should be confirmed in controlled prospective clinical trials.

PROSPERO Registration ID

: CRD42021236529.

Introduction

Ovarian reserve is the number of oocytes that a woman possesses at a particular time in her life and inversely correlates with age.¹ Both follicular atresia and ovulation over time cooperate in determining the slow depletion of the initial pool of non-growing follicles.¹ Ovarian reserve may be evaluated by measuring biochemical (i.e., Anti-Müllerian hormone [AMH], basal follicle stimulating hormone [FSH], estradiol [E2], or inhibin B serum concentration) and ultrasound (i.e., antral follicle count [AFC] and ovarian volume) parameters or by assessing the response to controlled ovarian hyperstimulation (COH) for in vitro fertilization (IVF) (i.e., number of oocytes retrieved [NOR]).^1,2 Serum AMH, which is produced by granulosa cells of gonadotropin-independent early follicles, and AFC are considered the most sensitive and reliable markers of oocyte quantity.¹ However, they have only a weak association with reproductive outcomes such as clinical pregnancy rate (CPR), and live birth rate (LBR).¹

Ovarian senescence is also characterized by a progressive impairment of oocyte quality, which relates to the potential of a fertilized oocyte to result in a live birth.² This process determines an ineluctable reduction of fecundity and reproductive potential.^2,3 The extent and speed of the decline in both ovarian reserve and oocyte quality with aging vary among women and mainly depend on genetic characteristics.² Importantly, there are other contributing factors. Among these, ovarian surgery certainly plays a role. Evidence from studies investigating the modification in AMH serum concentration after surgical excision of ovarian endometriomas showed a surgery-related damage to ovarian reserve.^4,5 Other authors claimed a role for an autoimmunity-mediated damage.⁶ In particular, an association between thyroid autoimmunity (TAI) and diminished ovarian reserve (DOR) has been hypothesized.⁷

The available data also suggest a detrimental effect of TAI on oocyte quality. Monteleone et al demonstrated the presence of thyroid antibodies in the follicular fluid of women with TAI. Further, they observed a decreased chance of oocyte fertilization and top-quality embryo development in women with positive TAI compared with negative controls.⁷ On this basis, they speculated that the presence of thyroid antibodies within the follicle may damage the maturing oocyte.⁸

The impact of thyroid antibodies on the ovarian reserve can be easily assessed by comparing the earlier mentioned surrogate markers between TAI-positive and -negative women. However, investigating their effect on oocyte quality is more challenging. To date, assisted reproductive technology (ART) is considered the most reliable study model since it is well known that oocyte quality considerably influences its outcomes (i.e., fertilization rate [FR], implantation rate [IR], CPR, miscarriage rate [MR], and LBR).^3,9

Although the premises support a possible negative impact of TAI on both the quantity and the quality of the remaining oocyte pool, the available evidence is inconclusive.^10,11 In a previous meta-analysis, we showed a detrimental impact of TAI on the course of pregnancy achieved through ART.⁷ However, our findings have been questioned by subsequent systematic reviews.^12,13 As for quantitative outcomes, information is often hidden among the baseline characteristics and does not attract the reader's attention. Considering the relatively high prevalence of TAI among women of reproductive age, such uncertainties and lack of knowledge are particularly frustrating for both clinicians and their patients.

The aim of this systematic review and meta-analysis was to provide an updated data synthesis about the effect of thyroid antibodies on both ovarian reserve and the reproductive potential of women undergoing ART.

Materials and Methods

Protocol registration and guidelines

This systematic review of the published literature was reported according to the PRISMA guidelines,^14,15 and the meta-analysis was reported according to the MOOSE guidelines.¹⁶ This study was, thus, exempt from institutional ethics review. The protocol was registered with PROSPERO (CRD42021236529).

Eligibility criteria

This systematic review and meta-analysis were restricted to two groups of published research articles:

Studies that investigated the association between TAI and ART (including IVF, intracytoplasmic sperm injection [ICSI], intrauterine insemination [IUI], or ovulation induction [OI] with timed intercourses) outcomes (i.e., FR [i.e., percentage of transformation of micro-injected oocytes into two pronuclei], IR [i.e., the number of gestational sacs observed divided by the number of embryos transferred], CPR (i.e., the number of clinical pregnancies [defined as a pregnancy diagnosed by ultrasonographic visualization of one or more gestational sacs or definitive clinical signs of pregnancy] divided by the number of women who started an ART cycle [only one cycle per patient included]), MR (i.e., the number of spontaneous loss of an intra-uterine pregnancy before 22 completed weeks of gestational age divided by the number of clinical pregnancies), and LBR (i.e., the number of deliveries that resulted in at least one live birth divided by the number of women who started an ART cycle [only one cycle per patient included])¹⁷;

Studies that investigated the association between TAI and surrogate markers of ovarian reserve (i.e., AMH serum concentration, basal FSH serum concentration, AFC and NOR after COH for IVF or ICSI).

Information sources and search strategy

We systematically searched PubMed, MEDLINE, ClinicalTrials.gov, Embase, and Scopus from database inception to May 1, 2022. Searches were limited to studies in humans and written in English. Search strings are shown in Supplementary Appendix SA1. All pertinent articles were retrieved, and the relative reference lists were systematically reviewed to identify further reports that could be included in the meta-analysis. Moreover, review articles and meta-analysis that focused on the impact of TAI on both ART outcomes and ovarian reserve were consulted and their reference list searched for potential additional studies.

No attempt was made to identify unpublished studies. Two authors (A.Bus. and C.B.) independently evaluated the title and abstract of all articles and excluded those deemed irrelevant by both observers. Reports were classified according to the study design into case-control studies, prospective and retrospective cohort studies. In the event of multiple publications from the same group or overlapping study populations among studies, data from the most recent study were included.

Selection of studies

Studies examining TAI and ART outcomes (Group 1) were eligible for inclusion only if: (1) the FR, IR, CPR MR, and LBR were separately reported for women with and without TAI; (2) the participants were involved in autologous cycles; and (3) the outcomes were reported per patient and not per cycle. Studies on TAI and surrogate markers of ovarian reserve (Group 2) were eligible for inclusion only if the authors reported: (1) the AMH or basal FSH serum concentration or the AFC or the NOR separately for women with and without TAI, or (2) the rate of women with positive TAI according to the ovarian reserve level (i.e., DOR, normal ovarian reserve, and high ovarian reserve). We considered the use of median and interquartile range (IQR) as an indicator that the data distribution was skewed.

We did not convert median and IQR into mean ± standard deviation (SD).¹⁸ Studies that reported continuous data (i.e., surrogate markers of oocyte quantity) as median and IQR were, thus, included only in the qualitative synthesis. Finally, studies could be included only if participants were in their reproductive age (i.e., between 15 and 49 years old), had a basal serum thyrotropin (TSH) within the normal range as defined by each single study, were not affected by overt hypothyroidism (i.e., increased TSH and/or decreased thyroxine [T4 level]) or hyperthyroidism, and were not taking medications to treat thyroid diseases.

Risk of bias and quality assessment

Two authors (A.Bus. and F.C.) independently critically appraised the included studies for risk of bias using the Newcastle-Ottawa Quality Assessment Scale.¹⁹

To account for potential confounders (i.e., basal TSH serum level, women's age, type of thyroid antibodies, method of conception), the following subgroup analyses were conducted: (1) studies reporting data about the difference in age and/or TSH between groups; (2) studies including only subjects with basal TSH ≤2.5 mIU/L; (3) studies including only women with proven thyroperoxidase antibodies (TPO-Abs) positivity (i.e., studies that included women who were positive for TPO-Abs but who were not tested for thyroglobulin antibodies [Tg-Abs] as well as for other thyroid antibodies); (4) studies including only women who underwent ICSI; and (5) pooling of only risk estimates adjusted for age.

To evaluate the effect of women's age and basal TSH values on both MR and LBR, a meta-regression analysis was also performed. To evaluate for potential publication bias, funnel plots were generated and both Egger's test and Begg's test were performed.¹⁸ A.Bus. and F.C. graded the quality of evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.²⁰ The quality of evidence was downgraded by one level in case of serious concerns and by two levels in case of very serious concerns for risk of bias, inconsistency, indirectness, imprecision, and publication bias.

Data extraction and statistical analysis

Two authors (A.L. and A. Bul.) independently analyzed all articles and extrapolated their data on respective standardized forms. A final official report was compiled from these reports after resolving all discrepancies in discussion with the remaining authors. The year of publication, location, study design, characteristics of the included subjects, thyroid antibodies detected (i.e., TPO-Ab and/or Tg-Ab), thyroid function tests assessed (i.e., TSH, free T4 [fT4], free triiodothyronine [fT3]), surrogate markers of oocyte quantity (AMH, AFC, basal FSH and NOR), and ART outcomes quality (FR, IR, CPR, MR, and LBR) were recorded.

The difference in surrogate markers of oocyte quantity as well as in covariates (i.e., serum TSH concentration and age) was expressed using a mean difference (MD) with confidence interval [CI]. The risk estimate was expressed using an odds ratio (OR) with CI. If spurious data were not reported, the MD and the OR were extracted from original studies and combined in meta-analysis using the generic inverse variance method with the DerSimonian and Laird random-effects model.^15,18

The inconsistency of the studies' results was evaluated using Cochrane Q and the I ² statistic.¹⁸ Risk estimates were combined in a meta-analysis using a fixed-effects model when the statistical heterogeneity among the studies was absent to moderate (0% ≤ I ² < 30%). When it was moderate, substantial, or considerable (I ² ≥ 30%), the DerSimonian and Laird method was used for a random-effects model.^15,18 All analyses were performed using Review Manager (RevMan) [Computer program], Version 5.4, The Cochrane Collaboration, 2020. The effect of both woman's age and basal serum TSH level on MR and LBR was assessed through restricted maximum likelihood random effects meta-regression. The meta-regression analysis was performed using the “metareg” macro (Stata Statistical Software: Release 15 [Computer program; Stata Corp., 2017]).

Results

Results of search and description of studies

Figure 1 summarizes the process of study identification and selection.¹² Our literature searches yielded 435 studies, from which 19 duplicates were removed. After a review of the titles and abstracts, 30 and 32 studies were identified as potentially eligible for inclusion in group 1 and 2, respectively. After a full review, we excluded 5 reviews,^21

–25 7 systematic reviews and/or meta-analyses,^{7,11
–13,26
–28} and 6 studies because they included hypothyroid and/or hyperthyroid women.^29

–34 The exact reasons for exclusion are reported in Supplementary Table S1.^33,34

FIG. 1.

Selection of studies. Group 1 includes studies selected to investigate the association between TAI and ART outcomes. Group 2 includes studies selected to investigate the association between TAI and surrogate markers of ovarian reserve. ART, assisted reproductive technology; TAI, thyroid autoimmunity.

Data on the association between TAI and ART outcomes were extracted from 22 studies^{35

–56} (Group 1). Data on the association between TAI and surrogate markers of ovarian reserve were extracted from 26 studies^{10,35

–48,52,54,57

–65} (Group 2). Of these, four were excluded from the quantitative synthesis since they reported surrogate markers of ovarian reserve as median and IQR.^35,36,52,54 None of the included studies have overlapping participants for the same outcome. Characteristics of included studies are reported in Table 1.

Table 1.

Characteristics of the Included Studies

Study	Year	Country	Study design	Study period	Study population	No. of included subjects	TSH range in cases	Thyroid function tests and antibodies	Surrogate markers of oocyte quantity	Surrogate markers of oocyte quality	ART protocol performed
Kutteh et al⁴⁹	1999	The United States	Retrospective cohort study	1996–1997	Infertile women undergoing their first ART procedure	143 TAI+ and 730 TAI− women	Euthyroid (0.45–4.5 μIU/mL)	TSH; TPO-Ab; Tg-Ab	N.R.	CPR; MR; LBR	IVF or ICSI
Muller et al⁵⁰	1999	The Netherlands	Nested case-control study	1994–1996	Infertile women undergoing ART	25 TAI+ and 148 TAI− women	2.3 ± 0.3 mIU/L	TSH; TPO-Ab	N.R.	CPR; MR	IVF
Poppe et al⁵¹	2003	Belgium	Prospective cohort study	1999–2000	Infertile women undergoing their first ART procedure	32 TAI+ and 202 TAI− women	1.6 (0.02–4.1) mIU/L	TSH; fT4; TPO-Ab	N.R.	CPR; MR; LBR	IVF or ICSI
Negro et al³⁷	2005	Italy	Prospective cohort study	1993–2003	Infertile women undergoing their first ART procedure	72 TAI+ and 412 TAI− women	1.9 ± 0.7 mIU/L	TSH; fT4; TPO-Ab	NOR	CPR; MR; LBR	IVF or ICSI
Negro et al³⁸	2007	Italy	Retrospective cohort study	2000–2006	Infertile women undergoing their first ART procedure	42 TAI+ and 374 TAI− women	2.8 mIU/L (group 1)/1.6 mIU/L (group 2)	TSH; fT4; TPO-Ab	NOR	CPR; MR; LBR	IVF or ICSI
Kilic et al³⁹	2008	Turkey	Cross-sectional study	2006–2007	Infertile women undergoing their first ART procedure	23 TAI+ and 31 TAI− women	1.99 ± 1.11 (group 1)/2.21 ± 1.25 mIU/L (group 2)	TSH; fT3; fT4; TPO-Ab; Tg-Ab	Basal FSH; NOR	CPR; MR	ICSI
Revelli et al⁵⁷	2009	Italy	Retrospective cohort study	2004–2008	Infertile women undergoing ART	38 TAI+ and 200 TAI− women	2.0 ± 1.2 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	AFC; basal FSH; NOR	Raw number not extractable	N.A.
Zhong et al⁴⁰	2012	China	Case-control study	2009–2010	Infertile women undergoing ART	90 TAI+ and 676 TAI− women	N.R.	TPO-Ab; Tg-Ab	Basal FSH; NOR	FR; IR; MR; CPR	IVF or ICSI
Karacan et al⁴¹	2013	Turkey	Prospective cohort study	2009–2012	Infertile women undergoing ART	34 TAI+ and 219 TAI− women	1.89 ± 1.18 mIU/L	TSH; fT4; TPO-Ab; Tg-Ab	NOR	FR; IR; CPR; MR	ICSI
Magri et al⁵⁸	2013	Italy	Retrospective cohort study	2011–2013	Infertile women undergoing their first ART procedure	55 TAI+ and 233 TAI− women	2.1 ± 1.2 mIU/L	TSH; TPO-Ab; Tg-Ab	AMH; basal FSH; NOR	N.R.	IVF
Mintziori et al⁵²	2014	Greece	Retrospective cohort study	2006–2010	Infertile women undergoing ART	15 TAI+ and 67 TAI− women	0.5–4.5 μIU/ml	TSH; fT3; fT4; TPO-Ab; Tg-Ab	NOR (not included in the quantitative analysis)	CPR; MR; LBR	IVF
Chai et al⁴²	2014	China	Retrospective cohort study	2007–2009	Infertile women undergoing their first ART procedure	89 TAI+ and 419 TAI− women	1.5 (0.35–2.47) mIU/L	TSH; TPO-Ab; Tg-Ab	AMH; basal FSH	CPR; MR; LBR	IVF or ICSI
Saglam et al⁵⁹	2014	Turkey	Case-control study	2012–2013	Euthyroid women recruited from endocrinological outpatient clinics	85 TAI+ and 82 TAI− women	2.7 ± 1.1 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	AMH; basal FSH; Inhibin B	N.R.	N.R.
Tan et al⁵³	2014	Germany	Retrospective cohort study	1997–2006	Infertile women undergoing ART	110 TAI+ and 725 TAI− women	1.7 ± 0.7 mIU/L	TSH; fT4; TPO-Ab; Tg-Ab	N.R.	CPR; MR; LBR	ICSI
Litwicka et al⁴³	2015	Italy	Prospective study	2011–2012	Infertile women undergoing their first ART procedure	30 TAI+ and 134 TAI− women	1.8 ± 0.6 mIU/L	TSH; TPO-Ab; Tg-Ab	AMH; basal FSH; AFC; NOR	CPR; MR; LBR	ICSI
Lukaszuk et al⁵⁴	2015	Poland	Retrospective cohort study	2011–2012	Infertile women undergoing their first ART procedure	114 TAI+ and 551 TAI− women	0.5–2.5 mIU/L	TSH; TPO-Ab	AMH; basal FSH; NOR (not included in the quantitative analysis)	FR; IR; CPR; MR; LBR	ICSI
Polyzos et al¹⁰	2015	Belgium	Cross-sectional study	2009–2011	All consecutive women attending the Centre for Reproductive Medicine of the University Hospital of Brussels	3929 women with normal, 487 with low, and 478 with high ovarian reserve	1.1–2.2 mIU/L	TSH; fT4; TPO-Ab	AMH	N.R.	N.R.
Sakar et al⁶⁰	2015	Turkey	Prospective case-control study	2013–2014	Infertile women undergoing ART	31 TAI+ and 121 TAI− women	2.35 ± 1.84 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	AMH; AFC; basal FSH	FR; CPR; MR; OPR	IVF
Chen et al⁴⁴	2017	China	Prospective cohort study	2015	Infertile women undergoing their first ART procedure	235 anti-Tg positive, 214 anti-TPO positive, and 844 TAI negative	N.R.	TPO-Ab; Tg-Ab	Basal FSH; NOR	IR; CPR; LBR; MR	IVF or ICSI
Chen et al⁶¹	2017	China	Cross-sectional study	2013–2016	Infertile women referring to the Center for Reproductive Medicine of Taipei Medical University Hospital	157 women with low, 725 with normal, and 158 with high ovarian reserve	1.02–2.38 mIU/L	TSH; TPO-Ab; Tg-Ab	AMH	N.R.	N.R.
Seungdamrong et al⁴⁵	2017	USA	Secondary analysis of data from two RCTs	2009–2014	Euthyroid infertile women selected for OI/IUI	123 TAI +1306 TAI− women	1.9 ± 1.0 mIU/L	TPO-Ab	Basal FSH, AMH, AFC	CPR; MR, LBR	IO/IUI
Unuane et al⁴⁶	2017	Belgium	Retrospective cohort study	2010–2014	Infertile women undergoing IUI	187 TAI+ and 2956 TAI− women	1.83 ± 1.44 mIU/L	TSH; fT4; TPO-Ab	AMH; basal FSH	CPR; MR; LBR	IUI
Osuka et al⁶²	2018	Japan	Case-control study	2008–2016	Infertile women referring to the fertility clinic of Nagoya University Hospital	30 anti-Tg-positive women, 28 anti-TPO-positive women, and 145 TAI-negative women	Euthyroid	TSH; TPO-Ab; Tg-Ab	AMH; basal FSH	N.R.	N.R.
Korevaar et al⁶³	2018	The United States	Study embedded within the EARTH Study	2004	Women who attended the Massachusetts General Hospital Fertility Center	46 TPO-Ab-positive women, 40 Tg-Ab-positive women	1.90 (0.68–4.70) mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	AFC; basal FSH	N.A.	N.A.
Devi et al⁴⁷	2019	India	Prospective cohort study	2014–2016	Infertile women undergoing ART	17 TAI+ and 64 TAI− women	0.5–2.5 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	Basal FSH; NOR	FR; IR; CPR	IVF
Gungor et al⁵⁵	2020	Turkey	Retrospective cohort study	2015–2016	Infertile women undergoing ART	39 TAI+ and 46 TAI− women	2.1 ± 0.3 mIU/L	TSH; TPO-Ab; Tg-Ab	N.R.	MR	ICSI
Liu et al³⁵	2020	China	Case-control study	2015–2017	Infertile women undergoing ART	49 TAI+ and 63 TAI− women	2.60 (1.90–3.06) mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	Basal FSH; NOR (not included in the quantitative analysis)	FR; IR; CPR; MR; LBR	IVF or ICSI
Rao et al⁶⁴	2020	China	Retrospective cohort study	2016–2018	Infertile women evaluated at the Reproductive Medical Center of the First Affiliated Hospital of Kunming Medical University	210 TAI+ and 1718 TAI− women	2.28 ± 1.28 mIU/L	TSH; fT3; fT4; TPO-Ab	AMH; AFC; basal FSH; NOR	N.R.	N.R.
Tokgoz et al⁴⁸	2020	Turkey	Retrospective cohort study	2018–2019	Infertile women undergoing ART	65 TAI+ and 250 TAI− women	1.8 ± 0.9 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	Basal FSH, AMH, NOR	FR; IR; MR	ICSI
Ke et al³⁶	2020	China	Case-control study	2012–2017	Infertile women undergoing ART	1075 TAI+ and 5138 TAI− women	2.36 ± 0.94 mIU/L	TSH; fT3; fT4; TPO-Ab; Tg-Ab	AMH; basal FSH; AFC; NOR (not included in the quantitative analysis)	CPR; MR; LBR	IVF or ICSI
Adamska et al⁶⁵	2020	Denmark	Case-control study	2016–2019	Euthyroid women in their reproductive age	21 TAI+ and 67 TAI− women	1.7 (1.2–2.4) mIU/L	TSH; fT3; fT4; TPO-Ab	Basal FSH	N.R.	N.R.
Huang et al⁵⁶	2021	China	Retrospective cohort study	2018–2019	Infertile women undergoing ART	778 TAI+ and 778 TAI− women	2.1 (1.5–2.9) mIU/L	TSH; fT4; TPO-Ab; Tg-Ab	Basal FSH; AMH; AFC; NOR	CPR; MR; LBR	IVF or ICSI

AFC, antral follicle count; AMH, anti-Mullerian hormone; ART, assisted reproductive techniques; CPR, clinical pregnancy rate; fT3, free triiodothyronine; fT4, serum free thyroxine; FR, fertilization rate; FSH, follicle-stimulating hormone; ICSI, intracytoplasmic sperm injection; IR, implantation rate; IUI, intra-uterine injection; IVF, in vitro fertilization; LBR, live birth rate; MR, miscarriage rate; NOR, number of oocytes retrieved; OI, ovulation induction; TAI+, positive thyroid autoimmunity; TAI−, negative thyroid autoimmunity; Tg-Ab, thyroglobulin antibodies; TPO-Ab, thyroid peroxidase antibodies; TSH, thyrotropin.

Association between TAI and ART pregnancy outcomes

Fertilization rate

Five studies reported the FR in women with TPO-Abs and/or Tg-Abs that underwent IVF/ICSI and in negative controls.^{40,41,43,54,55} The FR was not significantly different among the two groups of women (random-effects model, OR 0.87, [CI 0.62–1.24], p = 0.44, I ² = 92%) (Table 3) (Supplementary Fig. S1).

Table 3.

Association Between Thyroid Autoimmunity and Surrogate Markers of Ovarian Reserve

Surrogate marker	No. of studies	No. of included subjects	Surrogate marker MD	Possible confounding variables	No. of studies	Confounding variable MD	Surrogate marker MD (analysis limited to studies comparing the possible confounding variable)	Overall summary of evidence	Quality of evidence (GRADE)
Association between TAI and surrogate markers of ovarian reserve in women with TPO-Abs and/or Tg-Abs
Basal FSH	15	TAI-positive women: 1276; TAI-negative women: 9112	Fixed-effects model, MD 0.04 [−0.10 to 0.18], p = 0.60, I ² = 7%. Egger's test: intercept −0.0001 [CI −0.002 to 0.001], p = 0.84; Begg's test: Kendall's Tau 0.22, p = 0.26	Age	14	Random-effects model, MD 0.36 [−0.06 to 0.78], p = 0.09, I ² = 46%	Fixed-effects model, MD 0.01 [−0.17 to 0.19], p = 0.91, I ² = 11%	Absence of an association between the presence of TAI and basal FSH serum concentration	Low
Basal FSH	15	TAI-positive women: 1276; TAI-negative women: 9112		TSH	11	Random-effects model, MD 0.12 [−0.05 to 0.29], p = 0.16, I ² = 56%	Fixed-effects model, MD 0.16 [−0.15 to 0.46], p = 0.32, I ² = 0%		Low
AFC	6	TAI-positive women: 478; TAI-negative women: 3869	Random-effects model, MD −0.01 [−0.72 to 0.70], p = 0.98, I ² = 45%. Egger's test: intercept 0.0008 [CI −0.003 to 0.004], p = 0.60; Begg's test: Kendall's Tau 0.20, p = 0.58	Age	4	Random-effects model, MD 0.49 [−0.38 to 1.36], p = 0.27, I ² = 55%	Random-effects model, MD −0.21 [−1.36 to 0.94], p = 0.72, I ² = 51%	Absence of an association between the presence of TAI and AFC	Low
AFC	6	TAI-positive women: 478; TAI-negative women: 3869		TSH	3	Random-effects model, MD 0.23 [−0.27 to 0.73], p = 0.37, I ² = 77%	Random-effects model, MD −0.15 [−1.48 to 1.18], p = 0.83, I ² = 68%		Low
AMH	10	TAI-positive women: 818; TAI-negative women: 7282	Random-effects model, MD −0.20 [−0.49 to 0.09], p = 0.18, I ² = 92%. Egger's test: intercept −0.017 [CI −0.085 to 0.005], p = 0.59; Begg's test: Kendall's Tau −0.09, p = 0.72	Age	9	Random-effects model, MD 0.50 [−0.03 to 1.12], p = 0.06, I ² = 59%	Random-effects model, MD −0.21 [−0.56 to 0.14], p = 0.24, I ² = 93%	Absence of an association between the presence of TAI and AMH serum concentration	Low
AMH	10	TAI-positive women: 818; TAI-negative women: 7282		TSH	8	Random-effects model, MD 0.27 [0.08–0.46], p = 0.006, I ² = 63%	Random-effects model, MD −0.23 [−0.66 to 0.19], p = 0.28, I ² = 94%		Low
NOR	11	TAI-positive women: 1335; TAI-negative women: 3944	Random-effects model, MD −0.04 [−0.49 to 0.40], p = 0.85, I ² = 79%. Egger's test: intercept 2.87 [CI 1.17–4.58], p = 0.005; Begg's test: Kendall's Tau 0.24, p = 0.33	Age	9	Random-effects model, MD 0.10 [−0.16 to 0.35], p = 0.47, I ² = 85%	Fixed-effects model, MD 0.08 [−0.10 to 0.26], p = 0.41, I ² = 26%	Absence of an association between the presence of TAI and NOR	Low
NOR	11	TAI-positive women: 1335; TAI-negative women: 3944		TSH	6	Random-effects model, MD 0.12 [−0.10 to 0.34], p = 0.30, I ² = 73%	Random-effects model, MD 0.36 [−0.12 to 0.84], p = 0.14, I ² = 34%		Low
Association between TPO-Abs and surrogate markers of ovarian reserve
Basal FSH	6	TAI-positive women: 783; TAI-negative women: 7036	Random-effects model, MD 0.01 [−0.25 to 0.27], p = 0.93, I ² = 43%	Age	5	Random-effects model, MD 0.84 [0.19–1.48], p = 0.01, I ² = 46%	Random-effects model, MD 0.01 [−0.25 to 0.27], p = 0.93, I ² = 43%	Absence of an association between the presence of anti-TPO-Ab and basal FSH serum concentration	Low
Basal FSH	6	TAI-positive women: 783; TAI-negative women: 7036		TSH	3	Fixed-effects model, MD 0.30 [0.15–0.44], p < 0.0001, I ² = 0%	Fixed-effects model, MD 0.38 [−0.35 to 1.11], p = 0.30, I ² = 0%		Low
AFC	2	TAI-positive women: 333; TAI-negative women: 3024	Fixed-effects model, MD −0.01, [−0.70 to 0.68], p = 0.97, I ² = 0%	Age	1	MD −0.40 [−1.18 to 0.38], p = 0.32	MD −0.50 [−4.64 to 3.64], p = 0.81	Absence of an association between the presence of anti-TPO-Ab and AFC	Low
AFC	2	TAI-positive women: 333; TAI-negative women: 3024		TSH	//	//	//		Low
AMH	4	TAI-positive women: 548; TAI-negative women: 6125	Fixed-effects model, MD −0.06 [−0.15 to 0.04], p = 0.24, I ² = 0%	Age	3	Random-effects model, MD 1.03 [0.25–1.81], p = 0.009, I ² = 40%	Fixed-effects model, MD −0.04 [−0.15 to 0.07], p = 0.52, I ² = 0%	Absence of an association between the presence of anti-TPO-Ab and AMH serum concentration	Low
AMH	4	TAI-positive women: 548; TAI-negative women: 6125		TSH	2	Fixed-effects model, MD 0.30 [0.15–0.44], p < 0.0001, I ² = 0%	Fixed-effects model, MD −0.06 [−0.20 to 0.08], p = 0.41, I ² = 0%		Low
NOR	4	TAI-positive women: 509; TAI-negative women: 3512	Fixed-effects model, MD 0.01 [−0.18 to 0.20], p = 0.92, I ² = 0%	Age	3	Random-effects model, MD 0.39 [−0.28 to 1.05], p = 0.25, I ² = 95%	Fixed-effects model, MD 0.01 [−0.19 to 0.20], p = 0.95, I ² = 0%	Absence of an association between the presence of TAI and NOR	Low
NOR	4	TAI-positive women: 509; TAI-negative women: 3512		TSH	//	//	//		Low

GRADE's approach to rate the quality of evidence in observational studies: Initial quality of evidence: Low. The quality rate was downgraded if: risk of bias −1 serious, −2 very serious; inconsistency −1 serious, −2 very serious; indirectness −1 serious, −2 very serious; imprecision −1 serious, −2 very serious; publication bias −1 likely, and −2 very likely. The quality of evidence was upgraded if: large effect +1 large, +2 very large; dose response +1 evidence of a gradient; all plausible residual confounding +1 would reduce a demonstrated effect, +2 would suggest a spurious effect if no effect was observed.

Implantation rate

Six studies reported the IR in women with TPO-Abs and/or Tg-Abs who underwent IVF/ICSI and in negative controls.^{35,40,41,44,47,54} A significantly lower chance of embryo implantation was observed in TAI-positive women (random-effects model, OR 0.72, [CI 0.59–0.88], p = 0.001, I ² = 36%) (Table 3) (Supplementary Fig. S1).

Clinical pregnancy rate

Twenty studies reported the CPR in women with TPO-Abs and/or Tg-Abs who underwent ART and in negative controls.^{3

–47,49,51

–54,56,64} No differences in CPR emerged (random-effects model, OR 0.88, [CI 0.76–1.01], p = 0.07, I ² = 44%) (Table 3) (Supplementary Fig. S1).

Miscarriage rate

Twenty-one studies reported the MR in women with TPO-Abs and/or Tg-Abs who underwent IVF/ICSI and in negative controls.^{3

–46,48,49,51

–55,56,64} The MR was observed to be significantly higher in TAI women (random-effects model, OR 1.52, [CI 1.14–2.01], p = 0.004, I ² = 53%) (Table 3) (Supplementary Fig. S1).

Live birth rate

Fifteen studies reported the LBR in women with positive TAI and in negative controls.^{35

–38,42

–46,51

–54,56} Pooling of their results showed a significantly higher LBR in negative women (random-effects model, OR 0.73, [CI 0.56–0.94], p = 0.02, I ² = 71%) (Table 3) (Supplementary Fig. S1).

Confounding variables assessment

Results about the influence of possible confounding variables (i.e., female age and serum TSH concentration) on the reproductive outcomes as well as sensitivity analysis according to the method of conception and the restriction to studies including only women with proven TPO-Abs are reported in Table 2.

Table 2.

Association Between Thyroid Autoimmunity and Reproductive Outcomes

Reproductive outcome	No. of included subjects	Conception method	No. of studies	Association effect estimate	Possible confounding variables	No. of studies	Confounding variable MD	Association effect estimate (analysis limited to studies comparing the possible confounding variable)	Overall summary of evidence	Quality of evidence (GRADE)
Association between TAI and reproductive outcomes in women with TPO-Abs and/or Tg-Abs
FR	TAI-positive women: 2776; TAI-negative women: 18,834.	ART (cIVF and/or ICSI).	5	Random-effects model, OR 0.87, CI [0.62–1.24], p = 0.44, I ² = 92%. Egger's test: intercept 3.16 [CI −11.33 to 17.65], p = 0.54; Begg's test: Kendall's Tau 0.20, p = 0.62.	Age	4	Random-effects model, MD 0.23 [CI −0.73 to 1.19], p = 0.64, I ² = 50%	Random-effects model, OR 0.79 [CI 0.58–1.09], p = 0.15, I ² = 84%	Absence of an association between the presence of TAI and FR	Low
					TSH	2	Random-effects model, MD 0.41 [CI 0.14–0.68], p = 0.003, I ² = 32%	Fixed-effects model, OR 1.09 [CI 0.85–1.39], p = 0.49, I ² = 19%
					TSH <2.5 mIU/L	2	//	Fixed-effects model, OR 1.21 [CI 1.04–1.40], p = 0.01, I ² = 0%
		cIVF	//	//	//	//	//	//	//	//
		ICSI	3	Fixed-effects model, OR 1.16 [CI 1.01–1.33], p = 0.03, I ² = 0%	Age	2	Random-effects model, MD 0.11, [CI −2.24 to 2.46], p = 0.93, I ² = 83%	Fixed-effects model, OR 1.09 [CI 0.85–1.39], p = 0.49, I ² = 19%	Absence of an association between the presence of TAI and FR	Very low
		ICSI	3		TSH	2	Random-effects model, MD 0.41 [CI 0.14–0.68], p = 0.003, I ² = 32%	Fixed-effects model, OR 1.09 [CI 0.85–1.39], p = 0.49, I ² = 19%		Very low
IR	TAI-positive women: 1285; TAI-negative women: 5833	ART (cIVF and/or ICSI)	6	Random-effects model, OR 0.72 [CI 0.59–0.88], p = 0.001, I ² = 36%. Egger's test: intercept 0.01 [CI −3.72 to 3.74], p = 0.99; Begg's test: Kendall's Tau 0.07, p = 0.85.	Age	5	Fixed-effects model, MD −0.29 [CI −0.97 to 0.39], p = 0.41, I ² = 0%	Fixed-effects model, OR 0.67 [CI 0.57–0.78], p < 0.00001, I ² = 0%	Significantly lower embryo IR in women with positive TAI	Very low
					TSH	2	Random-effects model, MD −0.01 [CI −0.28 to 0.26], p = 0.95, I ² = 31%	Fixed-effects model, OR 0.83 [CI 0.49–1.41], p = 0.49, I ² = 0%
					TSH <2.5 mIU/L	2	//	Fixed-effects model, OR 0.94 [CI 0.68–1.29], p = 0.69, I ² = 0%
		cIVF	1	OR 0.61 [CI 0.23–1.62], p = 0.32	Age	1	MD −0.90 [CI −2.75 to 0.95], p = 0.34	OR 0.61 [CI 0.23–1.62], p = 0.32	Absence of an association between the presence of TAI and IR	Very low
					TSH	1	MD −0.10 [CI −0.33 to 0.13], p = 0.39	OR 0.61 [CI 0.23–1.62], p = 0.32
					TSH <2.5 mIU/L	1	//	OR 0.61 [CI 0.23–1.62], p = 0.32
		ICSI	2	Fixed-effects model, OR 0.98 [CI 0.73–1.32], p = 0.91, I ² = 0%	Age	1	MD −1.10 [CI −2.50 to 0.30], p = 0.12	OR 0.95 [CI 0.51–1.77], p = 0.88	Absence of an association between the presence of TAI and IR	Very low
					TSH	1	MD 0.20 [CI −0.23 to 0.63], p = 0.37	OR 0.95 [CI 0.51–1.77], p = 0.88
					TSH <2.5 mIU/L	1	//	OR 0.99 [CI 0.71–1.39], p = 0.96
CPR	TAI-positive women: 2317; TAI-negative women: 14,510	ART (cIVF and/or ICSI) or IUI or OI	20	Random-effects model, OR 0.87 [CI 0.75–1.01], p = 0.07, I ² = 44%. Egger's test: intercept −1.02 [CI −2.21 to 0.16], p = 0.08; Begg's test: Kendall's Tau −0.06, p = 0.70.	Age	16	Random-effects model, MD 0.41 [CI −0.08 to 0.90], p = 0.10, I ² = 80%	Random-effects model, OR 0.85 [CI 0.72–1.02], p = 0.08, I ² = 53%	Absence of an association between the presence of TAI and CPR	Low
					Age adjustment	3	//	Random-effects model, OR 1.00 [CI 0.63–1.59], p = 0.99, I ² = 59%
					TSH	10	Random-effects model, MD 0.19 [CI 0.07–0.32], p = 0.002, I ² = 67%	Fixed-effects model, OR 1.08 [CI 0.95–1.23], p = 0.23, I ² = 19%
					TSH <2.5 mIU/L	5	//	Random-effects model, OR 0.87 [CI 0.69–1.10], p = 0.25, I ² = 0%
		IUI or OI	2	Fixed-effects model, OR 0.86 [CI 0.63–1.18], p = 0.34, I ² = 0%	Age	1	MD 1.50 [CI 0.82–2.18], p < 0.0001	OR 0.98 [CI 0.62–1.56], p = 0.94	Absence of an association between thr presence of TAI and CPR	Very low
					Age adjustment	2	//	Fixed-effects model, OR 0.86 [CI 0.63–1.18], p = 0.34, I ² = 0%
					TSH	1	MD 0.30 [CI 0.15–0.45], p < 0.0001	OR 0.98 [CI 0.62–1.56], p = 0.94
		cIVF	3	Random-effects model, OR 1.22 [CI 0.55–2.69], p = 0.63, I ² = 40%	Age	2	Fixed-effects model, MD −0.35 [CI −1.50 to 0.80], p = 0.55, I ² = 0%	Random-effects model, OR 1.37 [CI 0.42–4.42], p = 0.60, I ² = 62%	Absence of an association between the presence of TAI and CPR	Very low
					TSH	2	Random-effects model, MD 0.72 [CI −1.13 to 2.56], p = 0.45, I ² = 85%	Random-effects model, OR 1.37 [CI 0.42–4.42], p = 0.60, I ² = 62%
					TSH <2.5 mIU/L	1	//	OR 0.70 [CI 0.22–2.27], p = 0.55
		ICSI	5	Fixed-effects model, OR 0.83 [CI 0.64–1.06], p = 0.14, I ² = 0%	Age	4	Random-effects model, MD −0.06 [CI −1.15 to 1.02], p = 0.91, I ² = 60%	Fixed-effects model, OR 0.81 [CI 0.59–1.11], p = 0.19, I ² = 0%	Absence of an association between the presence of TAI and MR	Low
					TSH	4	Random-effects model, MD 0.30 [CI 0.11–0.48], p = 0.002, I ² = 43%	Fixed-effects model, OR 0.81 [CI 0.59–1.11], p = 0.19, I ² = 0%
					TSH <2.5 mIU/L	3	//	Fixed-effects model OR 0.85 [CI [0.65–1.11], p = 0.24, I ² = 25%
MR	TAI-positive women: 1398; TAI-negative women: 6208	cIVF and/or ICSI or IUI or OI	21	Random-effects model, OR 1.52 [CI 1.14–2.01], p = 0.004, I ² = 53%. Egger's test: intercept 1.34 [CI 0.16–2.52], p = 0.03; Begg's test: Kendall's Tau 0.07, p = 0.67.	Age	17	Random-effects model, MD 0.42 [CI −0.05 to 0.89], p = 0.08, I ² = 79%	Random-effects model, OR 1.76 [CI 1.26–2.47], p = 0.0009, I ² = 55%	Significantly higher risk of miscarriage in women with positive TAI	Low
					Age adjustment	3	//	Random-effects model, OR 0.94 [CI 0.30–2.91], p = 0.91, I ² = 73%
					TSH	11	Random-effects model, MD 0.19 [CI 0.09–0.29], p = 0.0001, I ² = 60%	Random-effects model, OR 1.43 [CI 1.00–2.05], p = 0.05, I ² = 33%
					TSH <2.5 mIU/L	5	//	Random-effects model, OR 1.43 [CI 0.78–2.61], p = 0.25, I ² = 36%
		IUI or OI	2	Fixed-effects model, OR [CI 0.49–4.46], p = 0.49, I ² = 62%	Age	2	MD 1.07 [CI 0.19–1.95], p < 0.0001	Random-effects model, OR 1.47 [CI 0.49–4.46], p = 0.49, I ² = 62%	Absence of an association between TAI and MR	Very low
					Age adjustment	2	//	Random-effects model, OR 1.47 [CI 0.49–4.46], p = 0.49, I ² = 62%
					TSH	1	MD 0.30 [CI 0.15–0.45], p < 0.0001	OR 0.73 [CI 0.21–2.54], p = 0.62
		cIVF	2	Fixed-effects model, OR 2.26 [CI 0.64–7.94], p = 0.20, I ² = 0%	Age	1	MD 0.00 [CI −1.48 to 1.48], p = 1.00	OR 2.13 [CI 0.51–8.85], p = 0.30	Absence of an association between TAI and MR	Very low
		cIVF	2		TSH	1	MD 1.80 [CI 0.38–3.22], p = 0.01	OR 2.13 [CI 0.51–8.85], p = 0.30	Absence of an association between TAI and MR	Very low
		ICSI	7	Fixed-effects model, OR 1.42 [CI 0.92–2.19], p = 0.11, I ² = 2%	Age	6	Random-effects model, MD −0.04 [CI −0.78 to 0.70], p = 0.91, I ² = 41%	Fixed-effects model, OR 1.83 [CI 1.13–2.95], p = 0.01, I ² = 0%	Absence of an association between TAI and MR	Low
					TSH	6	Random-effects model, MD 0.20 [CI 0.05–0.35], p = 0.01, I ² = 59%	Fixed-effects model, OR 1.83 [CI 1.13–2.95], p = 0.01, I ² = 0%
					TSH <2.5 mIU/L	4	//	Random-effects model, OR 1.59 [CI 0.71–3.53], p = 0.26, I ² = 46%
LBR	TAI-positive women: 1801; TAI-negative women: 9616	cIVF and/or ICSI or IUI or OI	15	Random-effects model, OR 0.73 [CI 0.56–0.94], p = 0.02, I ² = 71%. Egger's test: intercept −1.77 [CI −2.98 to −0.55], p = 0.008; Begg's test: Kendall's Tau −0.2762, p = 0.15.	Age	11	Random-effects model, MD 0.78 [CI 0.21–1.35], p = 0.008, I ² = 81%	Random-effects model, OR 0.67 [CI 0.48–0.94], p = 0.02, I ² = 71%	Significantly lower LBR in women with positive TAI	Low
					Age adjustment	3	//	Random-effects model, OR 0.67 [CI 0.48–0.93], p = 0.02, I ² = 0%
					TSH	6	Random-effects model, MD 0.23 [CI 0.08–0.39], p = 0.004, I ² = 82%	Random-effects model, OR 0.85 [CI 0.49–1.46], p = 0.56, I ² = 58%
					TSH <2.5 mIU/L	4	//	Random-effects model, OR 0.87 [CI 0.55–1.37], p = 0.54, I ² = 0%
		IUI or OI	2	Random-effects model, OR 0.74 [CI 0.37–1.47], p = 0.39, I ² = 31%	Age	2	Random-effects model, MD 1.12 [CI 0.14–2.10], p = 0.03, I ² = 72%	Random-effects model, OR 0.74 [CI 0.37–1.47], p = 0.39, I ² = 31%	Absence of an association between the presence of TAI and LBR	Very low
					Age adjustment	2	//	Random-effects model, OR 0.74 [CI 0.37–1.47], p = 0.39, I ² = 31%
					TSH	1	MD 0.37 [CI 0.23–0.51], p < 0.00001	OR 1.37 [CI 0.39–4.77], p = 0.62
		ICSI	3	Fixed-effects model, OR 1.02 [CI 0.58–1.80], p = 0.93, I ² = 17%	Age	2	Random-effects model, MD 0.62 [CI −0.43 to 1.66], p = 0.25, I ² = 47%	Random-effects model, OR 1.01 [CI 0.17–5.96], p = 0.99, I ² = 58%	Absence of an association between the presence of TAI and LBR	Low
					TSH	2	Random-effects model, MD 0.34 [CI 0.04–0.63], p = 0.02, I ² = 80%	Random-effects model, OR 1.01 [CI 0.17–5.96], p = 0.99, I ² = 58%
					TSH <2.5 mIU/L	3	//	Fixed-effects model, OR 1.02 [CI 0.58–1.80], p = 0.93, I ² = 17%
Association between TPO-Abs and reproductive outcomes
FR	TPO-Abs positive women: 768; TPO-Abs negative women: 3613	ART (cIVF and/or ICSI)	1	OR 1.20 [CI 1.02–1.40]	Age	//	//	//	//	//
		ART (cIVF and/or ICSI)	1	OR 1.20 [CI 1.02–1.40]	TSH	//	//	//	//	//
		cIVF	//	//	//	//	//	//	//	//
		ICSI	1	OR 1.20 [CI 1.02–1.40]	Age	//	//	//	Slightly increased FR in women with positive TPO-Abs	Very low
		ICSI	1	OR 1.20 [CI 1.02–1.40]	TSH	//	//	//	Slightly increased FR in women with positive TPO-Abs	Very low
IR	TPO-Abs positive women: 676; TPO-Abs negative women: 2772	cIVF and/or ICSI	2	Random-effects model, OR 0.83 [CI 0.62–1.11], p = 0.20, I ² = 55%	Age	1	MD 0.30 [CI −5.74 to 6.34], p = 0.92	OR 0.73 [CI 0.59–0.90], p = 0.004	Absence of an association between the presence of TPO-Abs and IR	Very low
		cIVF and/or ICSI	2		TSH	//	//	//		Very low
		ICSI	1	OR 0.99 [CI 0.71–1.39], p = 0.96	Age	//	//	//	Absence of an association between the presence of TPO-Abs and IR	Very low
		ICSI	1	OR 0.99 [CI 0.71–1.39], p = 0.96	TSH	//	//	//		Very low
CPR	TPO-Abs positive women: 1378; TPO-Abs negative women: 8041	cIVF and/or ICSI or IUI or OI	11	Fixed-effects model, OR 0.88 [CI 0.77–1.00], p = 0.05, I ² = 22%	Age	6	Random-effects model, MD 0.97 [CI 0.09–1.86], p = 0.03, I ² = 83%	Random-effects model, OR 0.98 [CI 0.71–1.34], p = 0.90, I ² = 48%	Absence of an association between the presence of TPO-Abs and CPR	Low
		cIVF and/or ICSI or IUI or OI	11		TSH	3	Random-effects model, MD 0.27 [CI −0.03 to 0.58], p = 0.08, I ² = 71%	Random-effects model, OR 1.19 [CI 0.76–1.87], p = 0.44, I ² = 37%		Low
		ICSI	//	//	Age	//	//	//	//	//
		ICSI	//	//	TSH	//	//	//	//	//
MR	Pregnancies of TPO-Abs positive women: 552; Pregnancies of TPO-Abs negative women: 2760	cIVF and/or ICSI or IUI or OI	11	Random-effects model, OR 1.63 [CI 1.09–2.43], p = 0.02, I ² = 44%	Age	7	Random-effects model, MD 0.97 [CI 0.09–1.86], p = 0.03, I ² = 83%	Fixed-effects model, OR 1.83 [CI 1.13–2.95], p = 0.01, I ² = 0%	Absence of an association between the presence of TPO-Abs and MR	Very low
		cIVF and/or ICSI or IUI or OI	11		TSH	3	Random-effects model, MD 0.27 [CI −0.03 to 0.58], p = 0.08, I ² = 71%	Fixed-effects model, OR 1.82 [CI 0.74–4.47], p = 0.19, I ² = 44%		Very low
		ICSI	1	OR 0.45 [CI 0.13–1.55]	Age	//	//	//	Absence of an association between the presence of TPO-Abs and MR	Very low
		ICSI	1	OR 0.45 [CI 0.13–1.55]	TSH	//	//	//		Very low
LBR	TPO-Abs positive women: 1253; TPO-Abs negative women: 5192	cIVF and/or ICSI or IUI or OI	9	Random-effects model, OR 0.72 [CI 0.56–0.93], p = 0.01, I ² = 45%	Age	6	Random-effects model, MD 1.04 [CI 0.26–1.83], p = 0.009, I ² = 82%	Random-effects model, OR 0.59 [CI 0.38–0.90], p = 0.02, I ² = 58%	Significantly lower LBR in women with positive TPO-Abs	Low
		cIVF and/or ICSI or IUI or OI	9		TSH	2	Random-effects model, MD 0.25 [CI −0.02 to 0.51], p = 0.07, I ² = 76%	Random-effects model, OR 0.62 [CI 0.15–2.60], p = 0.51, I ² = 72%	Significantly lower LBR in women with positive TPO-Abs	Low
		ICSI	//	//	Age	//	//	//	//	//
		ICSI	//	//	TSH	//	//	//	//	//

GRADE's approach to rate the quality of evidence in observational studies: initial quality of evidence: low. The quality rate was downgraded if: risk of bias −1 serious, −2 very serious; inconsistency −1 serious, −2 very serious; indirectness −1 serious, −2 very serious; imprecision −1 serious, −2 very serious; publication bias −1 likely, and −2 very likely. The quality of evidence was upgraded on the basis of: i) magnitude of the effect (+1 if the magnitude of the effected was deemed ‘large’, +2 if the magnitude of the effect was deemed ‘very large’; ii) dose response gradient (+1 if there was evidence of a dose response gradient); iii) all plausible residual confounding (+1 if they would reduce a demonstrated effect, +2 if they would suggest a spurious effect if no effect was observed). Since the present meta-analysis included only observational studies, particularly relevant in rating up quality were the magnitude of the effect estimate and the presence of a dose-response gradient.

CI, confidence interval; cIVF: conventional in vitro fertilization; GRADE, Grading of Recommendations Assessment, Development and Evaluation; OR, odds ratio; MD, mean difference.

Meta-regression analysis

Seventeen^{29

–33,35
–37,39,41,43,44,50,53,54,56,58} and 11 studies^{29

–32,37,39,41,43,44,54,56} could be included in the meta-regression analysis exploring the possible influence of women's age on MR and LBR, respectively. Eleven^{30,31,33,36,37,39,44,50,53,56,58} and 6 ^{30,31,37,39,44,56} studies could be included in the meta-regression analysis focusing on the impact of basal TSH values on MR and LBR, respectively. No significant relationship between women's age and log OR for both MR (p = 0.12) and LBR (p = 0.09) was observed. Similarly, basal serum TSH seems to not have an effect on both MR (p = 0.24) and LBR (p = 0.11).

Association between TAI and surrogate markers of ovarian reserve (i.e., basal FSH and AMH serum concentration, AFC and NOR)

Three studies reported FSH serum concentration as median and IQR. None reported a significant difference between TAI-positive and -negative women.^35,36,54 Fifteen studies reported enough data to calculate the mean basal FSH difference between women with TPO-Abs and/or Tg-Abs and negative controls.^{39,40,42,45

–48,57

–62,64,65} Pooling of results showed the absence of a significant difference (Table 3) (Supplementary Fig. S1).

Two studies reported AMH serum concentration as median and IQR and did not show any difference between TAI-positive and -negative women.^36,54 Pooling of results of studies reporting the mean ± SD AMH failed to show a difference between women with TPO-Abs and/or Tg-Abs and negative controls.^{42,43,45,46,48,58
–60,62,64} (Table 3) (Supplementary Fig. S1). Two studies compared the prevalence of TPO-Abs according to the AMH serum concentration category.^10,61 No differences emerged (low ovarian reserve vs. normal ovarian reserve) (random-effects model, OR 1.38 [CI 0.89–2.15], p = 0.15, I ² = 59%).

One study reported the bilateral AFC as median and IQR without showing any difference between TAI-positive and -negative women.³⁶ Pooling of results of studies reporting the mean ± SD AFC confirmed the absence of a difference between women with TPO-Abs and/or Tg-Abs and in negative controls (Table 3) (Supplementary Fig. S1).^{43,45,57,60,63,64}

Two studies compared the median NOR between women with positive and negative TAI without observing any difference.⁵² The meta-analysis of studies reporting the mean ± SD NOR did not show any difference between women with TPO-Abs and/or Tg-Abs and negative controls (Table 3) (Supplementary Fig. S1).^{37
–39,40,41,43,44,47,48,56,64}

Confounding variables assessment

Results about the influence of possible confounding variables (i.e., female age and serum TSH concentration) on the difference in ovarian reserve surrogate markers between groups as well as the findings of studies that included only women with proven TPO-Abs are reported in Table 3.

Risk-of-bias and quality assessment results

Risk-of-bias assessment for observational studies is summarized in Supplementary Table S2. Overall, included studies showed a moderate or low risk of bias. Among the nine applicable stars assessing the three main categories of selection, comparability, and outcomes, the eligible studies received between six and eight stars. Visual evaluation of Funnel Plots and the Egger's test results suggest the reliability of the association between TAI status and NOR; MR and LBR might be hampered by publication bias (Supplementary Fig. S2).

The quality of evidence evaluated according to the GRADE system is summarized in Tables 2 and 3. Owing to the retrospective design of the included studies, the lower boundaries of CIs of risk estimate measures being close to 1, and the inability to properly adjust for the effect of possible confounding factors, the quality of evidence was between low and very low.

Discussion

Main findings

The present systematic review and meta-analysis showed that TAI-positive women undergoing ART cycles had a higher risk of miscarriage and a lower chance of live birth. Both sub-analyses and meta-regression suggest that the association between thyroid antibodies and MR was not significantly influenced by age or serum TSH concentration. However, a confounding role of age in the association between TAI and LBR cannot be excluded. Although both meta-regression and meta-analysis of adjusted risk estimates did not show a significant effect, the restriction of meta-analysis studies providing the mean (±SD) women's age showed that TAI-positive patients were significantly older. We also observed a reduced rate of embryo implantation in women with thyroid antibodies. However, the reliability of this finding can be questioned as the association disappeared after limiting the analysis to studies including only women with proven TPO-Abs. Moreover, the CPR was not significantly associated with the presence of TAI.

The increased risk of adverse pregnancy outcomes in TAI-positive women had already been observed in a previous meta-analysis and prompted many authors to formulate theories to explain the possible effect of thyroid antibodies on fertility.^{7,12,66
–68} The most accepted one considers TAI as a marker of generalized immune imbalance, possibly leading to the failure of fertilization, implantation, and sustained pregnancy.^66
–68 Alternatively, TAI might have a direct pathogenic effect on ovarian tissue. Monteleone et al hypothesized that thyroid antibodies may pass the “blood-follicle” barrier and cause mediated cytotoxicity in the growing ovarian follicle.^8,66
–68

According to other authors, TAI causes a degree of relative hypothyroidism that adversely impacts the reproductive function. Importantly, ART can facilitate the development of this condition. In a systematic review and meta-analysis, we demonstrated a significant increase in the serum TSH concentration in subjects undergoing COH for IVF/ICSI. In particular, approximately one out of four euthyroid women are expected to exceed the serum TSH level threshold of 2.5 mIU/L during or immediately after COH.^66

–70 On this basis, it has been hypothesized that the supplementation of levothyroxine (LT4) before the initiation of ART cycle could reverse the negative impact mediated by TAI.^71,72 Unfortunately, data published so far are not reassuring. A recent Cochrane review demonstrated that treatment with LT4 neither increased LBR nor decreased the MR in euthyroid TAI-positive women undergoing ART.⁷³ These findings indirectly refute the third mentioned theory.⁷

Our subgroup analyses also provided useful insights. In women who exclusively underwent ICSI, no association between the presence of thyroid antibodies and any of the considered reproductive outcomes was observed. The interpretation of these findings is suggested by the previous evidence that showed that zona pellucida may be a target for thyroid antibodies.^71

–74 Accordingly, in women with TAI, the interaction between the oocyte and the sperm cell and, as a consequence, the normal embryogenesis process could be hampered.^74

–77 The ICSI, which involves the direct injection of a single spermatozoon into the ooplasm, might thus overcome the detrimental effect mediated by TAI.¹²

As for the hypothesized association between TAI and oocyte quantity, we observed that the mean values of surrogate markers of ovarian reserve (i.e., basal FSH serum concentration, AMH serum concentration, AFC and NOR after COH) did not significantly differ between women with and without thyroid antibodies. Restriction to studies including only women with proven TPO-Abs positivity confirmed these findings.

We also did not observe a difference in the proportion of women with TAI between the various ovarian reserve categories defined on the basis of the AMH serum level. Unfortunately, data from only two studies could be pooled and our findings do not allow reliable conclusions to be drawn.^10,61

Our results partially conflict with the previous literature. Some studies showed a high prevalence of thyroid antibodies positivity in patients with premature ovarian insufficiency (POI).^78
–80 Specifically, TAI was reported in 37% of POI women with Turner Syndrome and in 15% of POI patients with 46, XX karyotype.^71,81 Accordingly, the European Society of Human Reproduction and Embryology (ESHRE) suggests TAI screening in all women diagnosed with spontaneous POI.⁸² Since DOR in young women and POI have been hypothesized to share at least a part of their pathophysiological mechanism, the recent guidelines published by the European Thyroid Association recommend to screen women with DOR not only for thyroid dysfunction but also for TAI.⁸³

In a recent meta-analysis, Hasegawa et al reported a lower AMH level in adult TAI-positive women. Although statistically significant, the calculated MD entity (0.12 ng/mL) is likely of modest clinical importance. However, adolescent girls with thyroid antibodies were shown to have a higher AMH serum concentration when compared with negative controls.²⁸ Our approach differs from the previous ones, as it is aimed at investigating the impact of TAI on different ovarian reserve markers. Interestingly, our findings suggest that the presence of TAI in women seeking fertility care should not be considered a risk factor for DOR and for poor ovarian response to COH.

Limitations and reason for cautions

Some limitations should be considered in the interpretation of our results. First, the reliability of the evidence could be undermined by possible confounders (e.g., woman's age and BMI, basal TSH value, indication to ART, etc.) and we rated most of the evidence as of a relatively low quality. Many of our meta-analyses were subject to statistically significant heterogeneity, and there was some concern about publication bias.

Another limitation concerns the type of thyroid antibody tested in the included studies. Some investigators included TPO-Abs- and/or Tg-Abs-positive subjects without providing separate data for women with one or both positive antibody titers. To minimize the interference of this factor, we conducted a sub-analysis that included only patients who were positive for TPO-Abs that has been defined as the most sensitive marker of TAI and linked to the risk of (sub)clinical hypothyroidism.⁶⁷ In the great majority of cases, restriction to such studies confirmed the results of the principal analysis.

Further, some studies have included subjects with TSH values >2.5 mIU/L. Although the impact of minimally increased serum TSH levels (i.e., 2.5 mIU/L < TSH ≤4.0 mIU/L) on both natural fertility and ART outcomes is still debated, the European Thyroid Association guidelines suggest treatment with LT4 in infertile women with TAI and TSH >2.5 mIU/L.⁸³ Even in the absence of a difference in mean TSH values between the two study groups, a negative effect determined by the concomitant presence of TSH levels >2.5 mIU/L and thyroid antibodies cannot be ruled out. We, therefore, conducted a sub-analysis including only women with a TSH value below 2.5 mIU/L, which confirmed the results of the main one.

Finally, by pooling the results of non-spontaneous pregnancy outcomes regardless of the method of conception, we might have introduced a source of heterogeneity. To overcome this issue, we performed sub-analyses taking this variable into account. Unfortunately, studies including only women selected for conventional IVF or IUI/OI are still very few and prevent clinical inferences from being made. The reliability of the evidence regarding ICSI, although supported by a larger sample size, is weakened by other factors. In particular, the use of ICSI, despite the available evidence.,⁸⁴ has increased substantially in couples with non-male factor infertility.

The absence of a shared vision regarding ICSI indications makes its adoption rate extremely variable from one Fertility Unit to another, introducing a relevant confounding factor above all in retrospective studies.

Conclusions

In conclusion, in the present systematic review and meta-analysis, we observed an association between TAI and adverse ART outcomes. However, considering the extent of the calculated risk estimates and the reported limitations, a causal relationship cannot be determined.⁸⁵ Noteworthy, the possible protective effect of ICSI is promising and reassuring. This technique could potentially be a simple solution to limit the widely spread trend toward pharmacological overtreatment of TAI-positive women undergoing ART. In this regard, randomized controlled trials comparing the reproductive outcomes in TAI-positive women treated with either classical IVF or ICSI should be a research priority. Importantly, we also demonstrated the absence of a difference in surrogate markers of ovarian reserve between women with positive and negative thyroid antibodies.

These findings tend to refute the theory of a damaging process on the residual oocyte patrimony mediated by TAI. At the current state of knowledge, considering our results and the absence of treatments of proven efficacy, testing euthyroid infertile women selected for ART for thyroid antibodies is questionable.

Footnotes

Authors' Contributions

A.Bus. and C.B. conceived and designed the study. A.Bus. and F.C. independently assessed the included studies for risks of bias and graded the quality of evidence. A.Bul. and A.L. evaluated all articles and extrapolated the data on a standardized form. A.Bus. and C.B. performed the statistical analysis. A.Bus. and P.E.L.-S. drafted the first version of the article. All authors revised it critically and approved the final version to be published.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

Supplementary Appendix SA1

Supplementary Figure S1

Supplementary Figure S2

Supplementary Table S1

Supplementary Table S2

References

Hansen

, Knowlton

, Thyer

, et al. A new model of reproductive aging: The decline in ovarian non-growing follicle number from birth to menopause. Hum Reprod, 2008;23(3):699–708; doi: 10.1093/humrep/dem408.

Practice Committee of the American Society for Reproductive Medicine 2020 Testing and interpreting measures of ovarian reserve: A committee opinion. Fertil Steril 114(6):1151–1157; doi: 10.1016/j.fertnstert.2020.09.134.

Busnelli

, Navarra

, Levi-Setti

. Qualitative and quantitative ovarian and peripheral blood mitochondrial DNA (mtDNA) alterations: Mechanisms and implications for female fertility. Antioxidants, 2021; 10(1):55; doi: 10.3390/antiox10010055.

Somigliana

, Berlanda

, Benaglia

, et al. Surgical excision of endometriomas and ovarian reserve: A systematic review on serum antiMüllerian hormone level modifications. Fertil Steril, 2012;98(6):1531–1538; doi: 10.1016/j.fertnstert.2012.08.009.

Somigliana

, Benaglia

, Paffoni

, et al. Risks of conservative management in women with ovarian endometriomas undergoing IVF. Hum Reprod Update, 2015;21(4):486–499; doi: 10.1093/humupd/dmv012.

Silva

, Yamakami

, Aikawa

, et al. Autoimmune primary ovarian insufficiency. Autoimmun Rev, 2014;13(3–4):427–430; doi: 10.1016/j.autrev.2014.01.003.

Busnelli

, Paffoni

, Fedele

, et al. The impact of thyroid autoimmunity on IVF/ICSI outcome: A systematic review and meta-analysis. Hum Reprod Update, 2016;22(6):775–790; doi: 10.1093/humupd/dmw019.

Monteleone

, Parrini

, Faviana

, et al. Female infertility related to thyroid autoimmunity: The ovarian follicle hypothesis. Am J Reprod Immunol, 2011;66(2):108–114; doi: 10.1111/j.1600-0897.2010.00961.x.

Busnelli

, Lattuada

, Rossetti

, et al. Mitochondrial DNA copy number in peripheral blood: A potential non-invasive biomarker for female subfertility. J Assist Reprod Genet, 2018;35(11):1987–1994; doi: 10.1007/s10815-018-1291-5.

10.

Polyzos

, Sakkas

, Vaiarelli

, et al. Thyroid autoimmunity, hypothyroidism and ovarian reserve: A cross-sectional study of 5000 women based on age-specific AMH values. Hum Reprod, 2015;30(7):1690–1696; doi: 10.1093/humrep/dev089.

11.

Simopoulou

, Sfakianoudis

, Maziotis

, et al. The impact of antibodies on IVF treatment and outcome: A systematic review. Int J Mol Sci, 2019;20(4):892; doi: 10.3390/ijms20040892.

12.

Poppe

, Autin

, Veltri

, et al. Thyroid autoimmunity and intracytoplasmic sperm injection outcome: A systematic review and meta-analysis. J Clin Endocrinol Metab, 2018;103(5):1755–1766; doi: 10.1210/jc.2017-02633.

13.

Venables

, Wong

, Way

, et al. Thyroid autoimmunity and IVF/ICSI outcomes in euthyroid women: A systematic review and meta-analysis. Reprod Biol Endocrinol, 2020;18(1):120; doi: 10.1186/s12958-020-00671-3.

14.

Page

, Moher

, Bossuyt

, et al. PRISMA 2020 explanation and elaboration: Updated guidance and exemplars for reporting systematic reviews. BMJ, 2021;372:n160; doi: 10.1136/bmj.n160.

15.

Deeks

, Higgins

JPT

, Altman

. Analysing data and undertaking meta-analyses. In: Cochrane Handbook for Systematic Reviews of Interventions, version 6.0. (Higgins JPT, Churchill R, Chandler J, Cumpston MS. eds.) Cochrane Collaboration; 2018. Available from: www.training.cochrane.org/handbook [Last accessed: May 1, 2022].

16.

Stroup

, Berlin

, Morton

, et al. Meta-analysis of observational studies in epidemiology: A proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA, 2000;283(15):2008–2012; doi: 10.1001/jama.283.15.2008.

17.

Zegers-Hochschild

, Adamson

, Dyer

, et al. The International Glossary on Infertility and Fertility Care, 2017. Fertil Steril, 2017;108(3):393–406; doi: 10.1016/j.fertnstert.2017.06.005.

18.

Higgins

JPT

, Thomas

, Chandler

, et al. Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane. Available from: www.training.cochrane.org/handbook [Last accessed: May 1, 2022].

19.

Wells

, Shea

, O'Connell

The Newcastle–Ottawa Scale (NOS) for Assessing the Quality of Nonrandomized Studies in Meta-Analyses. Ottawa, Canada: The Ottawa Health Research Institute. Available from: https://www. ohri.ca/programs/clinical_epidemiology/oxford.Htm [Last accessed: May 1, 2022].

20.

Balshem

, Helfand

, Schünemann

, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol, 2011;64(4):401–406; doi: 10.1016/j.jclinepi.2010.07.015.

21.

Safarian

, Gzgzyan

, Dzhemlikhanova Lyailya

, et al. Does subclinical hypothyroidism and/or thyroid autoimmunity influence the IVF/ICSI outcome? Review of the literature. Gynecol Endocrinol, 2019;35(Supp 1):56–59; doi: 10.1080/09513590.2019.1653564.

22.

Unuane

, Velkeniers

Impact of thyroid disease on fertility and assisted conception. Best Pract Res Clin Endocrinol Metab, 2020;34(4):101378; doi: 10.1016/j.beem.2020.101378.

23.

Mintziori

, Goulis

. In vitro fertilization/intracytoplasmic insemination and thyroid function: Reviewing the evidence. Metabolism, 2018;86:44–48; doi: 10.1016/j.metabol.2018.03.021.

24.

Wang

, Liao

, Li

Thyroid autoimmunity in adverse fertility and pregnancy outcomes: Timing of assisted reproductive technology in AITD women. J Transl Int Med, 2021;9(2):76–83; doi: 10.2478/jtim-2021-0001.

25.

Vissenberg

, Manders

, Mastenbroek

, et al. Pathophysiological aspects of thyroid hormone disorders/thyroid peroxidase autoantibodies and reproduction. Hum Reprod Update, 2015;21(3):378–387; doi: 10.1093/humupd/dmv004.

26.

Leiva

, Schwarze

, Vasquez

There is no association between the presence of anti-thyroid antibodies and increased reproductive loss in pregnant women after ART: A systematic review and meta-analysis. JBRA Assist Reprod, 2017;21(4):361–365; doi: 10.5935/1518-0557.20170057.

27.

Rao

, Zeng

, Zhou

, et al. Effect of levothyroxine supplementation on pregnancy loss and preterm birth in women with subclinical hypothyroidism and thyroid autoimmunity: A systematic review and meta-analysis. Hum Reprod Update, 2019;25(3):344–361; doi: 10.1093/humupd/dmz003.

28.

Hasegawa

, Kitahara

, Osuka

, et al. Effect of hypothyroidism and thyroid autoimmunity on the ovarian reserve: A systematic review and meta-analysis. Reprod Med Biol, 2021;21(1):e12427; doi: 10.1002/rmb2.12427.

29.

Morales-Martínez

, Sordia-Hernández

, Ruiz

, et al. Association between thyroid autoimmunity and ovarian reserve in women with hypothyroidism. Thyroid Res, 2021;14(1):6; doi: 10.1186/s13044-021-00095-0.

30.

Pirgon

, Sivrice

, Demirtas

, et al. Assessment of ovarian reserve in euthyroid adolescents with Hashimoto thyroiditis. Gynecol Endocrinol, 2016;32(4):306–310; doi: 10.3109/09513590.2015.1116510.

31.

Özalp Akın

, Aycan

Evaluation of the ovarian reserve in adolescents with Hashimoto's thyroiditis using serum anti-Müllerian hormone levels. J Clin Res Pediatr Endocrinol, 2018;10(4):331–335; doi: 10.4274/jcrpe.0047.

32.

Bahri

, Tehrani

, Amouzgar

, et al. Overtime trend of thyroid hormones and thyroid autoimmunity and ovarian reserve: A longitudinal population study with a 12-year follow up. BMC Endocr Disord, 2019;19(1):47; doi: 10.1186/s12902-019-0370-7.

33.

Samsami

, Ghasmpour

, Moradi Alamdarloo

, et al. Women with autoimmune thyroiditis have lower reproductive life span or not? A cross- sectional study. Int J Community Based Nurs Midwifery, 2020;8(4):305–310; doi: 10.30476/ijcbnm.2020.84255.1207.

34.

, Zhao

, Wang

, et al. Correlation analysis between ovarian reserve and thyroid hormone levels in infertile women of reproductive age. Front Endocrinol, 2021;12:745199; doi: 10.3389/fendo.2021.745199.

35.

Liu

, Wu

, Tian

, et al. Protein expression profile in IVF follicular fluid and pregnancy outcome analysis in euthyroid women with thyroid autoimmunity. ACS Omega, 2020;5(20):11439–11447; doi: 10.1021/acsomega.0c00463.

36.

, Hu

, Zhao

, et al. Impact of thyroid autoimmunity on ovarian reserve, pregnancy outcomes, and offspring health in euthyroid women following in vitro fertilization/intracytoplasmic sperm injection. Thyroid, 2020;30(4):588–597; doi: 10.1089/thy.2018.0657.

37.

Negro

, Mangieri

, Coppola

, et al. Levothyroxine treatment in thyroid peroxidase antibody-positive women undergoing assisted reproduction technologies: A prospective study. Hum Reprod, 2005;20(6):1529–1533; doi: 10.1093/humrep/deh843.

38.

Negro

, Formoso

, Coppola

, et al. Euthyroid women with autoimmune disease undergoing assisted reproduction technologies: The role of autoimmunity and thyroid function. J Endocrinol Invest, 2007;30(1):3–8; doi: 10.1007/BF03347388.

39.

Kilic

, Tasdemir

, Yilmaz

, et al. The effect of anti-thyroid antibodies on endometrial volume, embryo grade and IVF outcome. Gynecol Endocrinol, 2008;24(11):649–655; doi: 10.1080/09513590802531112.

40.

Zhong

, Ying

, Wu

, et al. Relationship between antithyroid antibody and pregnancy outcome following in vitro fertilization and embryo transfer. Int J Med Sci, 2012;9(2):121–125; doi: 10.7150/ijms.3467.

41.

Karacan

, Alwaeely

, Cebi

, et al. Effect of antithyroid antibodies on ICSI outcome in antiphospholipid antibody-negative euthyroid women. Reprod Biomed Online, 2013;27(4):376–380; doi: 10.1016/j.rbmo.2013.07.002.

42.

Chai

, Yeung

, Lee

, et al. Live birth rates following in vitro fertilization in women with thyroid autoimmunity and/or subclinical hypothyroidism. Clin Endocrinol, 2014;80(1):122–127; doi: 10.1111/cen.12220.

43.

Litwicka

, Arrivi

, Varricchio

, et al. In women with thyroid autoimmunity, does low-dose prednisolone administration, compared with no adjuvant therapy, improve in vitro fertilization clinical results? J Obstet Gynaecol Res 2015;41(5):722–728; doi: 10.1111/jog.12615.

44.

Chen

, Mo

, Huang

, et al. Association of serum autoantibodies with pregnancy outcome of patients undergoing first IVF/ICSI treatment: A prospective cohort study. J Reprod Immunol, 2017;122:14–20; doi: 10.1016/j.jri.2017.08.002.

45.

Seungdamrong

, Steiner

, Gracia

, et al. Eunice Kennedy Shriver National Institute of Child Health and Human Development Reproductive Medicine Network 2017 Preconceptional antithyroid peroxidase antibodies, but not thyroid-stimulating hormone, are associated with decreased live birth rates in infertile women. Fertil Steril, 2017;25(5):S0015; doi: 10.1016/j.fertnstert.2017.08.026.

46.

Unuane

, Velkeniers

, Bravenboer

, et al. Impact of thyroid autoimmunity in euthyroid women on live birth rate after IUI. Hum Reprod, 2017;32(4):915–922; doi: 10.1093/humrep/dex033.

47.

Devi

, Prasad

, Koner

. Antithyroid antibodies and fertility outcome in euthyroid women undergoing in vitro fertilisation. J Clin Diagn Res 2019;13(5):QC05-QC08; doi: 10.7860/JCDR/2019/40760.12872.

48.

Tokgoz

, Isim

, Tekin

. The impact of thyroid autoantibodies on the cycle outcome and embryo quality in women undergoing intracytoplasmic sperm injection. Middle East Fertil Soc J 2020;25(12); doi: 10.1186/s43043-020-00023-6.

49.

Kutteh

, Schoolcraft

, Scott

Jr . Antithyroid antibodies do not affect pregnancy outcome in women undergoing assisted reproduction. Hum Reprod 1999;14(11):2886–2890; doi: 10.1093/HUMREP/14.11.2886.

50.

Muller

, Verhoeff

, Mantel

, et al. Thyroid autoimmunity and abortion: A prospective study in women undergoing in vitro fertilization. Fertil Steril, 1999;71(1):30–34; doi: 10.1016/s0015-0282(98)00394-x.

51.

Poppe

, Glinoer

, Tournaye

, et al. Assisted reproduction and thyroid autoimmunity: An unfortunate combination? J Clin Endocrinol Metab 2003;88(9):4149–4152; doi: 10.1210/jc.2003-030268.

52.

Mintziori

, Goulis

, Gialamas

, et al. Association of TSH concentrations and thyroid autoimmunity with IVF outcome in women with TSH concentrations within normal adult range. Gynecol Obstet Invest, 2014;77(2):84–88; doi: 10.1159/000357193.

53.

Tan

, Dieterle

, Pechlavanis

, et al. Thyroid autoantibodies per se do not impair intracytoplasmic sperm injection outcome in euthyroid healthy women. Eur J Endocrinol, 2014;170(4):495–500; doi: 10.1530/EJE-13-0790.

54.

Łukaszuk

, Kunicki

, Kulwikowska

, et al. 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, 2015;38(12):1335–1343; doi: 10.1007/s40618-015-0377-5.

55.

Gungor

, Dokuzeylul Gungor

Antithyroid antibodies may predict serum beta HCG levels and biochemical pregnancy losses in euthyroid women with IVF single embryo transfer. Gynecol Endocrinol, 2021;37(8):702–705; doi: 10.1080/09513590.2020.1830968.

56.

Huang

, Chen

, Lian

, et al. Impact of thyroid autoimmunity on in vitro fertilization/intracytoplasmic sperm injection outcomes and fetal weight. Front Endocrinol, 2021;12:698579; doi: 10.3389/fendo.2021.698579.

57.

Revelli

, Casano

, Piane

, et al. A retrospective study on IVF outcome in euthyroid patients with anti-thyroid antibodies: Effects of levothyroxine, acetyl-salicylic acid and prednisolone adjuvant treatments. Reprod Biol Endocrinol, 2009;7(137); doi: 10.1186/1477-7827-7-137.

58.

Magri

, Capelli

, Gaiti

, et al. Impaired outcome of controlled ovarian hyperstimulation in women with thyroid autoimmune disease. Thyroid, 2013;23(10):1312–1318; doi: 10.1089/thy.2013.0022.

59.

Saglam

, Onal

, Ersoy

, et al. Anti-Müllerian hormone as a marker of premature ovarian aging in autoimmune thyroid disease. Gynecol Endocrinol, 2015;31(2):165–168; doi: 10.3109/09513590.2014.973391.

60.

Sakar

, Unal

, Atay

, et al. Is there an effect of thyroid autoimmunity on the outcomes of assisted reproduction? J Obstet Gynaecol 2016;36(2):213–217; doi: 10.3109/01443615.2015.1049253.

61.

Chen

, Huang

, Tzeng

, et al. Idiopathic low ovarian reserve is associated with more frequent positive thyroid peroxidase antibodies. Thyroid, 2017;27(9):1194–1200; doi: 10.1089/thy.2017.0139.

62.

Osuka

, Iwase

, Goto

, et al. Thyroid autoantibodies do not impair the ovarian reserve in euthyroid infertile women: A cross-sectional study. Horm Metab Res, 2018;50(7):537–542; doi: 10.1055/a-0637-9430.

63.

Korevaar

TIM

, Mínguez-Alarcón

, Messerlian

, et al. Association of thyroid function and autoimmunity with ovarian reserve in women seeking infertility care. Thyroid, 2018;28(10):1349–1358; doi: 10.1089/thy.2017.0582.

64.

Rao

, Wang

, Zhao

, et al. Subclinical hypothyroidism is associated with lower ovarian reserve in women aged 35 years or older. Thyroid, 2020;30(1):95–105; doi: 10.1089/thy.2019.0031.

65.

Adamska

, Łebkowska

, Krentowska

, et al. Ovarian reserve and serum concentration of thyroid peroxidase antibodies in euthyroid women with different polycystic ovary syndrome phenotypes. Front Endocrinol, 2020;11:440; doi: 10.3389/fendo.2020.00440.

66.

Dosiou

Thyroid and fertility: Recent advances. Thyroid, 2020;30(4):479–486; doi: 10.1089/thy.2019.0382.

67.

Unuane

, Velkeniers

. Impact of thyroid disease on fertility and assisted conception. Best Pract Res Clin Endocrinol

Metab

. 2020;34(4):101378; doi: 10.1016/j.beem.2020.101378.

68.

Poppe

MANAGEMENT OF ENDOCRINE DISEASE: Thyroid and female infertility: More questions than answers?! Eur J Endocrinol 2021;184(4):R123–R135; doi: 10.1530/EJE-20-1284.

69.

Busnelli

, Cirillo

, Levi-Setti

. Thyroid function modifications in women undergoing controlled ovarian hyperstimulation for in vitro fertilization: A systematic review and meta-analysis. Fertil Steril 2021;116(1):218–231; doi: 10.1016/j.fertnstert.2021.01.029.

70.

, Hu

, Meng

, et al. Changes in thyroid function during controlled ovarian hyperstimulation (COH) and its impact on assisted reproduction technology (ART) outcomes: A systematic review and meta-analysis. J Assist Reprod Genet, 2021;38(9):2227–2235; doi: 10.1007/s10815-021-02206-0.

71.

Wang

, Gao

, Chi

, et al. Effect of levothyroxine on miscarriage among women with normal thyroid function and thyroid autoimmunity undergoing in vitro fertilization and embryo transfer: A randomized clinical trial. JAMA, 2017;318(22):2190–2198; doi: 10.1001/jama.2017.18249.

72.

Pelliccione

, Lania

, Pizzocaro

, et al. Levothyroxine supplementation on assisted reproduction technology (ART) outcomes in women with subtle hypothyroidism: A retrospective study. Gynecol Endocrinol, 2018;34(12):1053–1058; doi: 10.1080/09513590.2018.1499087.

73.

Akhtar

, Agrawal

, Brown

, et al. Thyroxine replacement for subfertile women with euthyroid autoimmune thyroid disease or subclinical hypothyroidism. Cochrane Database Syst Rev, 2019;6(6):CD011009; doi: 10.1002/14651858.CD011009.pub2.

74.

Kelkar

, Meherji

, Kadam

, Gupta

, et al. Circulating auto-antibodies against the zona pellucida and thyroid microsomal antigen in women with premature ovarian failure. J Reprod Immunol, 2005;66(1):53–67; doi: 10.1016/j.jri.2005.02.003.

75.

Lee

, Ng

, Lau

, Liu

, et al. Increased fetal abortion rate in autoimmune thyroid disease is related to circulating TPO autoantibodies in an autoimmune thyroiditis animal model. Fertil Steril, 2009;91(5Suppl):2104–2109; doi: 10.1016/j.fertnstert.2008.07.1704.

76.

Twig

, Shina

, Amital

, et al. Pathogenesis of infertility and recurrent pregnancy loss in thyroid autoimmunity. J Autoimmun, 2012;38(2–3):J275–J281; doi: 10.1016/j.jaut.2011.11.014.

77.

Palermo

, Joris

, Devroey

, et al. Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet, 1992;340(8810):17–18; doi: 10.1016/0140-6736(92)92425-f.

78.

Novosad

, Kalantaridou

, Tong

, et al. Ovarian antibodies as detected by indirect immunofluorescence are unreliable in the diagnosis of autoimmune premature ovarian failure: A controlled evaluation. BMC Womens Health, 2003;3(1):2; doi: 10.1186/1472-6874-3-2.

79.

Goswami

, Marwaha

, Goswami

, et al. Prevalence of thyroid autoimmunity in sporadic idiopathic hypoparathyroidism in comparison to type 1 diabetes and premature ovarian failure. J Clin Endocrinol Metab, 2006;91(11):4256–4259; doi: 10.1210/jc.2006-1005.

80.

Gleicher

, Weghofer

, Oktay

, et al. Do etiologies of premature ovarian aging (POA) mimic those of premature ovarian failure (POF)? Hum Reprod 2009;24(10):2395–2400; doi: 10.1093/humrep/dep256.

81.

Bakalov

, Gutin

, Cheng

, et al. Autoimmune disorders in women with turner syndrome and women with karyotypically normal primary ovarian insufficiency. J Autoimmun, 2012;38(4):315–321; doi: 10.1016/j.jaut.2012.01.015.

82.

European Society for Human Reproduction and Embryology (ESHRE) Guideline Group on

POI

, Webber

, Davies

, Anderson

, et al. ESHRE Guideline: Management of women with premature ovarian insufficiency. Hum Reprod, 2016;31(5):926–937; doi: 10.1093/humrep/dew027.

83.

Poppe

, Bisschop

, Fugazzola

, et al. European Thyroid Association guideline on thyroid disorders prior to and during assisted reproduction. Eur Thyroid J, 2021;9(6):281–295; doi: 10.1159/000512790.

84.

Dang

, Vuong

, Luu

, et al. Intracytoplasmic sperm injection versus conventional in-vitro fertilisation in couples with infertility in whom the male partner has normal total sperm count and motility: An open-label, randomised controlled trial. Lancet, 2021;397(10284):1554–1563; doi: 10.1016/S0140-6736(21)00535-3.

85.

Grimes

, Schulz

. False alarms and pseudo-epidemics: The limitations of observational epidemiology. Obstet Gynecol 2012;120(4):920–927; doi: 10.1097/AOG.0b013e31826af61a.