Does subclinical hypothyroidism and/or thyroid autoimmunity influence the IVF/ICSI outcome? Review of the literature.

While overt hypothyroidism is a well-known risk factor for infertility, the association of subclinical hypothyroidism (SCH) or thyroid autoimmunity and reproductive failure has been still not elucidated. In this literature review, the current data on the effect of SCH and/or thyroid autoimmunity and human reproduction is presented. The main ART outcome measures are as follows: number of oocytes retrieved, fertilization rate, implantation rate, clinical pregnancy rate per embryo transfer, embryo quality, miscarriage rate, and live birth rate. Current guidelines on the management of women with SCH and/or thyroid autoimmunity undergoing ART cycles will be presented in this review.


Introduction
Global infertility prevalence rates among couples are difficult to determine but are generally believed to range between 10 and 15% [1] and has not changed significantly despite of the evolution of assisted reproduction technologies (ART). In recent years, the relationship between reproductive failure and autoimmune conditions, including thyroid disorders, becomes particularly relevant and attracts attention worldwide. Autoimmune thyroid disease (AITD) or autoimmune thyroiditis (AIT) is the most common organ-specific autoimmune disorder affecting 2-5% of the population in Western countries and among women of reproductive age. It is found 5-10 times more often in women than in men [2]. Clinical presentation varies from hyperthyroidism in Graves' disease to hypothyroidism in Hashimoto's thyroiditis.
Autoimmune thyroiditis (Hashimoto's thyroiditis, HT)chronic progressive disease characterized by lymphoid infiltration of the thyroid gland including T and B cells, resulting in inflammation and leading to the gradual extinction of thyroid function with a number of complications. Therefore, both cellular and humoral immunity have a role in the pathogenesis of thyroid autoimmunity. In Russia, the frequency of AIT reaches 45 cases per 1000 population, in the USA since 1997, AIT ranks third in terms of the prevalence of autoimmune diseases. Hashimoto thyroiditis is the most frequent autoimmune condition among reproductive-aged women. The condition is characterized as the presence of serum antibodies directed against a membrane-associated hemoglycoprotein expressed only in thyrocytes (thyroperoxidase, TPO-abs) or a glycoprotein homodimer produced predominantly by the thyroid gland (thyroglobulin, Tg-abs) [3]. It was demonstrated that in women presenting with thyroid autoimmunity, the prevalence of infertility was as high as up to 47% [4]. According to the recommendation of the American Society for Reproductive Medicine, the level of TSH before entering the IVF program should not exceed 2.5 mIU/L [5]. The prevalence of TPO-abs is 8-14% in women of reproductive age, while the incidence of subclinical hypothyroidism (SCH) in the same population is approximately 4-8% [5,6].
The contribution of AIT has been widely investigated in infertile women. Combined data suggest a significantly higher incidence of thyroid antibodies in infertile women compared to controls. In particular, Poppe et al. calculated an overall estimated relative risk of 2.1 (95% CI [1.7-2.6]; p < .0001), according to which, the prevalence of TAI among women attending infertility clinics is higher compared to the general population [7]. An association has been particularly observed in women with the polycystic ovarian syndrome (PCOS) and idiopathic infertility ( ̴ 25%) compared with that in fertile women (10%) [8]. Adverse obstetric outcomes in both spontaneous pregnancy and pregnancy achieved using assisted reproduction technologies, such as placental abruption, pregnancy loss, premature rupture of membranes, and neonatal death have also been established in women with subclinical hypothyroidism [9]. IVF/ICSI outcomes stimulating hormone FSH were used. Among the 26 patients of TAI positive group, 20 patients were on levothyroxine substitution (mean daily levothyroxine dose was 67.49 ± 29.40 mcg, treatment duration was 19.50 months). Similar results, in terms of the number of oocytes retrieved and fertilization rates, were obtained in both groups. However, the implantation rate was lower in the TAI positive group when compared to controls (p ¼ .092). Pregnancies rates per initiated cycle and per embryo-transfer cycle were significantly different between the TAI positive and the TAI negative group (30.8% versus 61.5%), p ¼ .026 and (34.8% versus 66.7%), p ¼ .029, respectively. Multivariate analysis showed that TAI positive women had significantly less chance to achieve pregnancy (OR 0.036, 95% CI [0.004-0.347], p ¼ .004) [10].
In the retrospective study of Zhong et al. [11], a total of 90 patients positive for antithyroid antibody and 676 infertile women negative for antithyroid antibody, undergoing IVF/ICSI, were included. Stimulation protocol was uniform in all patients, presented with long GnRH-a. There was no significant difference in the days of ovarian stimulation, total gonadotropin dose, and number of oocytes retrieved between the two groups. The fertilization rate (64.3% versus 74.6%, p < .001), implantation rate (17.8% versus 27.1%, p < .001) and pregnancy rate (33.3% versus 46.7%, p ¼ .002) following IVF-ET were significantly lower in women with TAI than in the control group. However, the authors did not report on thyroid function.
In a meta-analysis of nine studies that included 4396 women, Busnelli et al. [12] attempted to assess the association between TAI per se and the IVF/ICSI cycles outcomes. But no significant differences in the number of oocytes retrieved (standardized mean difference 0. 10 22) were found between patients with thyroid autoimmunity and controls. The authors themselves, relying on the theory that antithyroid antibodies may bind antigens expressed in the zona pellucida, disrupting its functional role [13], believe that a possible reason for the lack of difference in these values between the groups was due to the data analyzed only in ICSI protocols, but not in classic IVF protocols. Therefore, they made an assumption that ICSI could overcome the negative impact of thyroid autoantibodies on the surface of the oocytes, highlighting, at the same time, that definitive conclusions could not be drawn because of insufficient evidence to advocate the systematic use of ICSI in patients with TAI.
The most recent meta-analysis, conducted by Poppe et al. [8], included four studies on ICSI outcomes only in women with or without TAI. According to the studies included, serum TSH cutoff of 2.5 and 3.0 mIU/L was used to define SCH. Meta-analysis revealed similar fertilization rates (combined OR  [13,14].
The fact that when analyzing the four selected parameters in the IVF protocols, definitive conclusion about the effect of antithyroid antibodies and subclinical hypothyroidism on ART outcomes cannot be made, worth attention. For instance, Zhong et al. [11] reported lower fertilization rate, implantation rate and pregnancy rate following IVF-ET in women with TAI, while Medenica et al. [10] did not reveal any decrease in fertilization rate. Perhaps because in their group of subjects, some women also underwent ICSI treatment. Thus, the suggestion, made in 2016 [12], seems quite reasonable today, given the results from 2018 [8].

Embryo quality
The relationship between thyroid function and reproduction becomes even more pronounced in the setting of ovarian hyperstimulation, when elevation in estradiol (E2) occurs, leading to an increase in thyroid binding globulin (TBG). As the binding sites of thyroid hormones are increased, the levels of free hormones decrease, inducing TSH production [15].
Based on the fact, Werghofer et al. assumed that thyroid function may influence in vitro fertilization cycle outcomes and conducted a case-control study to investigate the effects of thyroid performance and TAI on embryo quality in IVF patients with low functional ovarian reserve. As a result, in women with low functional ovarian reserve, thyroid peroxidase antibodies (TPO-abs) significantly affected embryo quality in euthyroid women with low-normal TSH values (TSH 2.5 lIU/ml). In women with TPO-abs and high normal TSH levels, a trend towards impaired embryo quality, although not statistically significant (p ¼ .056), was observed [16]. Obtained results find reflection in Monteleone's theory [13] and support a detrimental effect of TAI that is independent of hypothyroidism. A trend towards improved pregnancy potential in the presence of lownormal TSH levels was observed [16].

Endometrium
Taking into account, that clinical pregnancy rate (CPR) in women with thyroid autoimmunity and TSH! 2.5 lIU/ml was only 15% [16], it could be assumed that circulating thyroid autoantibodies, likely, act a marker of a general immune imbalance with complex changes in the background of this organ specific disease affecting the immunological relationship between the mother and the embryo.
Both TSHRs and THRs are expressed in human endometrium, depending on the stage of the menstrual cycle [17], allowing to speculate about the role of thyroid hormones and TSH in human endometrial physiology (i.e. proliferation, maturation) and potentially in the implantation process and early blastocyst development, especially, in the setting of AITD. Moreover, TPO and Tg are also expressed in the endometrium, where they may be responsible for local thyroxine production, making it susceptible to the actions of anti-TPO and anti-Tg autoantibodies [18]. However, there are very few studies comprehensively assessing endometrium among women with thyroid autoimmunity. For instance, Kilic et al. evaluated endometrial volume along with other IVF outcomes, a surrogate to assess endometrial receptivity and the implantation chance, in 69 women with unexplained infertility. Patients were divided into three groups: TAI negative group (n ¼ 31), TAI positive group (n ¼ 23), and TAI positive and euthyroid with the medication group (n ¼ 15). However, they failed to document any difference in endometrial volume among the groups [19].
It is also possible that thyroid antibodies have a direct embryotoxic effect on the trophoblast, limiting its invasion and hindering the normal development of the fetus. In the murine model, there is evidence of cross-reactivity of thyroid autoantibodies and trophoblast antigens, resulting in increased fetal wastage, with lower fetal and placental weights [20]. Additionally, TAI could affect the post-implantation embryo as shown in a mice model study where TPO-abs influenced post-implantation embryo development, leading to fetal loss [21].

Miscarriage rate (MR)
Nowadays, it is now generally accepted that there is an association between SCH and/or thyroid autoimmunity and adverse pregnancy outcomes in both spontaneous pregnancies and those achieved using assisted reproduction technologies [12,22]. The incidence of thyroid antibodies in euthyroid women with habitual abortions appears to be significantly increased compared with the controls of reproductive age without previous abortions [23]. In a cohort study of 2497 Dutch women without overt thyroid dysfunction, the risk of miscarriage, fetal, and neonatal death increased with higher levels of maternal TSH, i.e. by 60% (OR 1.60, 95% CI [1.04-2.47]) for every doubling in TSH concentration [24].
The molecular mechanisms underlying the association between adverse pregnancy outcomes and the presence of thyroid antibodies are not fully understood, thus, theories have been proposed. According to the first, TAI serves as a marker of a general immune imbalance with the dysregulated activity of the immune system at the fetal-maternal interface via, both, direct and indirect ways [20,21,25]. The second theory states that women who otherwise have normal thyroid function but have tested positive for thyroid antibodies may progress to subclinical or overt hypothyroidism, resulting in a reduced ability of thyroid gland to adapt to the physiological changes of pregnancy [26,27] and leading to insufficient concentrations of circulating thyroid hormone for the given gestational age. Finally, increasing maternal age is a well-known "independent" risk factor for miscarriage [28].
Zhong et al. [11] reported that the abortion rate was significantly higher in patients with antithyroid antibodies (26.9% versus 11.8%, p ¼ .002); however, they did not report on thyroid function. Thangaratinam et al. [29] meta-analyzed 31 studies, in total on 12.126 women and observed a significant difference in the miscarriage rate between patients with and without TAI (OR 3.90, 95% CI [2.48-6.12], p < .001 for cohort studies and OR 1.80, 95% CI [1.25-2.60], p ¼ .002 for case-control studies). The authors also observed a 2.07-fold higher risk of preterm birth among women positive for thyroid autoantibodies (95% CI However, the authors also pointed out that the age and the TSH level were not comparable between the groups, suggesting that further evidence is warranted prior to drawing inferences on causality [12]. Similar results have been reported by other authors as well [13,30], highlighting the important effect of thyroid autoimmunity on female reproductive failures. Existence of the same trend after ICSI treatment is of a particular interest in this context. Recent meta-analysis on women with a clinical pregnancy, revealed no difference in miscarriage rates (respective combined OR 0.95, 95% CI [0.48-1.87], p ¼ 0.31) [8], indicating that women with TAI who become pregnant after an ICSI treatment had similar a risk for a first-trimester miscarriage that women without TAI.  [12].
Werhofer et al. [16] observed a trend toward improved pregnancy potential in the presence of low-normal TSH levels (13.9%) compared to 5% in the group with high-normal TSH levels; however, the authors indicate that statistical analysis on the impact of baseline TSH levels on pregnancy potential was not feasible in their study due to the small number of pregnancies.
Thangaratinam et al. [29]  In another retrospective cohort study, Unuane et al. [31] compared live birth delivery after 25 weeks' gestation after six IVF-ICSI cycles in patients with and without TAI. In total, 2406 women were included: 333 with TAI, 2019 without TAI. Only patients with TSH levels 0.1-5.0 mIU/L were enrolled. Among women with TAI 802 ICSI, cycles until the sixth cycle were performed with 156 deliveries recorded. Overall miscarriage rates were 18%. The crude cumulative delivery rate after six cycles was 47%, whereas the expected cumulative delivery rate was 65%. After 3,937 ICSI cycles in the control group, overall 948 deliveries were recorded until the sixth cycle. The miscarriage rate was 13.4% and the crude cumulative delivery rate after six cycles was 47%, whereas the expected cumulative delivery rate was 76%. Thus, the authors did not confirm an influence of TAI status in patients undergoing IVF-ICSI fertility treatment on cumulative delivery rates. Moreover, using Cox regression analysis, they revealed that thyroid function (TSH, FT4) did not influence the delivery rate. Importantly, they noted an age-related decline in delivery rates in the TAI-positive group and TAI-negative group (crude cumulative rates of 57%, 48%, and 33% in the 20-29, 30-37, and 38-45 years old patient groups, respectively). It was, thus, concluded that it is the age and not the presence of anti-thyroid antibodies, which is the most important predictor of ICSI cycles outcomes among patients with AITD.
Treatment strategies for women with TAI and/or SCH and clinical practice guidelines regarding thyroid and ovarian stimulation The recent American Thyroid Association guidelines on thyroid disorders and pregnancy stated that there is insufficient evidence to recommend for or against universal screening for abnormal TSH concentrations before conception, except in women planning ART or those known to be positive for TPO-abs [32]. The statement from the American Society for Reproductive Medicine mentions [5] that the available data support the routine measurement of TSH in infertile women attempting pregnancy, but not that of TPO-abs, unless TSH levels are !2.5 mIU/L.
According to a recent statement of the ARSM [5], there is good evidence that TAI and/or SCH (defined as a serum TSH level !4.0 mIU/L) is associated with miscarriage. LT4 treatment may improve pregnancy outcomes in women undergoing ART, especially if the serum TSH level is !2.5 mIU/L with TAI or >4.0 mIU/L in general. American Thyroid Association states [32], that administration of LT4 to TPO-abs positive euthyroid women undergoing ART may be considered given its potential benefits in comparison to its minimal risk. They also recommend thyroid function testing before or 7-14 d after ovarian stimulation (OS). In the case of TSH elevation after OS in women who do not achieve pregnancy, testing should be repeated after 2-4 weeks and clinical decisions should be made. In the case of pregnancy following OS, any TSH elevation should be treated according to guidelines that exist for pregnancy [32].

Conclusions
Following available literature, we can conclude that the present data concerning the role of subclinical hypothyroidism and thyroid autoimmunity on ART success rate is controversial. A novel suggestion of proposing ICSI as the preferred method of infertility treatment in women with TAI is of a particular interest in this context. More investigations are necessary to improve our knowledge on this issue.

Disclosure statement
The authors report no declaration of interest.