International Journal of

Introduction
 
Thyroid hormones have profound variation during the life span and are associated with severe adverse health impacts (1, 2). Pregnancy, as an important reproductive event, has a profound but reversible effect on the thyroid gland and its functions. Pregnancy is actually a state of excessive thyroid stimulation leading to an increase in thyroid size by 10% in iodide sufficient areas and 20-40% in iodide deficient regions (3). Furthermore following the physiological and hormonal changes caused by pregnancy and human chorionic gonadotropin (HCG) the production of thyroxin (T4) and triiodothyronine (T3) increase up to 50% leading to 50% increase in a woman’s daily iodide need, while Thyroid-stimulating hormone (TSH) levels are decreased, especially in first trimester (4). In an iodide sufficient area, these thyroid adaptations during pregnancy are well tolerated, as stored inner thyroid iodide is enough; however in iodide deficient areas, these physiological adaptations lead to significant changes during pregnancy (5).
Furthermore in women who suffered from thyroid dysfunction prior to pregnancy, the hormonal changes mentioned are magnified, leading to possibly adverse pregnancy outcomes if not been treated appropriately. Furthermore the mode of delivery may additionally have adverse impact on fetal- pituitary- thyroid axis (6). The prevalence of thyroid dysfunction in pregnant women is relatively high so that overt thyroid dysfunction occurs in 2-3% of pregnancies, and subclinical dysfunction in 10% of pregnancies (7) and thyroid autoimmunity is even more prevalent (8).
Given the high prevalence of thyroid disturbances in pregnancy and lack of adequate review article summarizing the effect of thyroid dysfunction on pregnancy and neonatal outcomes, we aimed to summarize the adverse effects of thyroid dysfunction including hyperthyroidism, hypothyroidism and thyroid autoimmune positivity on pregnancy outcomes. Research question was: "are thyroid disorders in pregnant women associated with adverse effects on pregnancy outcomes"?
 
Evidence acquisition
This review study was conducted with a prospective protocol. We searched Medline (1985-2013), Embase (1985-2013) and the Cochrane Library (2012) for relevant English manuscripts. Using keywords including “thyroid”, “thyroid dysfunction”, “thyroid disorder”, “hyperthyroidism”, “hypothyroidism”, “euthyroidism”, “subclinical hypothyroidism”, “subclinical hyperthyroidism”, “thyroid autoantibodies”, “pregnancy outcome”, “miscarriage”, “abortion”, “pregnancy loss”, “preterm”, “premature”, “early labor”, “Thyroid peroxidase” and “cognitive” to generate a subset of citations relevant to our research question. Subclinical hypothyroidism (SCH) is defined as a serum TSH level above the upper limit of normal despite normal levels of serum free thyroxine and subclinical hyperthyroidism is defined as serum thyroid hormone levels within their respective reference ranges in the presence of low-undetectable serum TSH levels (9). Overt hypothyroidism is defined as a serum free T4 level lower than the upper limit of normal and overt hyperthyroidism is defined as a serum free T4 level above the upper limit of normal. Thyroid autoimmunity is defined as increase in thyroid auto antibodies above the upper limit of normal with or without thyroid disturbances.
The full manuscripts of all citations that met our study objective were selected and obtained. In cases of duplicate publications, we selected the most recent and complete versions. From the 4480 citations identified from electronic searches, at the beginning, we found 512 related articles; 130 studies on overt hypothyroidism, 203 on subclinical hypothyroidism, 69 on overt hyperthyroidism, 43 on subclinical hyperthyroidism, and 67 on thyroid immunity. Of these articles, 58 met our study objectives, including 11 on hyperthyroidism, 22 on hypothyroidism and 26 on thyroid immunity.
We included all qualified original articles on our study subject; including randomized clinical trials, cohort (prospective and retrospective), case-control and case reports. We excluded non-English manuscript, those conducted on non-pregnant women and those with poor quality methodology. The titles and abstracts of all of the studies were evaluated by two non-dependent persons and those met inclusion criteria were appraised.



Figure 1. The number of articles that were reviewed in the study.

Results
 
Hyperthyroidism and its adverse pregnancy and neonatal outcomes
The natural physiological changes during pregnancy can mimic some of the signs observed in hyperthyroidism, including increased in basal metabolism, heart rate, fatigue, anxiety, palpitations, heat intolerance, warm and wet skin, hand tremors and systolic murmur; as a result the diagnosis of hyperthyroidism during pregnancy could cause clinical difficulties (9-11). Pregnant women, who suffer from hyperthyroidism, have more severe tachycardia and thyromegaly, along with exophthalmia, and lack of weight gain despite receiving adequate food (10).
Overt hyperthyroidism during pregnancy was not prevalent and was reported in 2 out of 1000 pregnancies (0.2%), while subclinical hyperthyroidism was occurred in 1.7% of pregnancies (11, 12). The most prevalent reason for hyperthyroidism during pregnancy was the transient hyperthyroidism resulting from hyperemesis gravidarum (THHG) due to the thyroid stimulation of beta-HCG (13); it was more prevalent in Asian populations compared to Europeans (14).
Except for THHG, the etiologies of thyrotoxicosis during pregnancy are the same as for non-pregnant women; it is most prevalent in Grave’s Disease caused by thyrotropin receptor antibodies stimulating the thyroid (TRabs) (11, 15). It is well documented that overt hyperthyroidism has several adverse effects on pregnancy outcomes, e.g. miscarriage, stillbirth, preterm delivery, intrauterine growth retardation, preeclampsia (11, 16). Furthermore women with Graves' disease have antibodies that can stop or stimulate the fetal anti-TSH receptor of thyroid gland (15, 17, 18).
There is no consensus regarding the adverse effect of subclinical hypothyroidism on pregnancy or neonatal outcomes. Casey et al. reported no significant increase of placenta abruption, preterm labor and low birth weight in pregnancy complicated by subclinical hyperthyroidism in comparison with euthyroid ones (19). Table I summarizes the results of the most relevant studies on the impact of overt and subclinical hyperthyroidism on pregnancy and neonatal outcomes.
 
Hypothyroidism and its adverse pregnancy and fetal outcomes
Pregnancy can imitate some of the signs that are observed in hypothyroidism, including fatigue, anxiety, constipation, muscle cramps, and weight gain; as a result, the clinical diagnosis of hypothyroidism during pregnancy may be difficult (10, 27). Moreover, most signs of hypothyroidism can be hidden by a woman’s status following the increase in metabolism in pregnancy. Furthermore the thyroid hormonal profile in normal pregnancy can be mis-interoperated as hypothyroidism and as a result the interpretation of thyroid function tests needs trimester-specific reference intervals for a specific population (14, 15). Applying trimester-specific reference ranges of thyroid hormones prevents misclassification of thyroid dysfunction during pregnancy. Compared to hyperthyroidism, hypothyroidism is very common during pregnancy; 2-3% of pregnant women suffer from hypothyroidism (0.3-0.5% overt hypothyroidism and 2-2.5% subclinical hypothyroidism) (11, 28).
While the main etiology for hypothyroidism during pregnancy worldwide is iodide insufficiency, however in iodide sufficient areas its main cause is autoimmune thyroiditis (8). SCH is the most common thyroid dysfunction during pregnancy (11, 29). Its prevalence varies between 1.5-5% based on various definitions, different ethnicity, iodine consumption and nutrition life style as well as study designs (30). While the adverse effects of SCH accompanied with positive TPO antibodies or overt hypothyroidism on pregnancy outcome are well known, however there is controversy on negative impact of SCH without autoimmunity on pregnancy outcomes (27, 31-33). Pregnant women that possess the TPO antibodies during the initiation of their pregnancy are subjected to subclinical hypothyroidism during their pregnancy or thyroid dysfunction after childbirth (12).
Table II and III summarize the studies on adverse outcomes of overt and subclinical hypothyroidism, respectively. As it has been shown the overt hypothyroidism is associated with increase in prevalence of abortion, anemia, pregnancy-induced hypertension, preeclampsia, placental abruption, postpartum hemorrhage, premature birth, low birth weight, 
 
Table I. The adverse effects of hyperthyroidism (overt/subclinical) on pregnancy and neonatal outcomes



Table II. The adverse effects of overt hypothyroidism on pregnancy outcomes 



Table III: The adverse effects of subclinical hypothyroidism (with/without thyroid autoimmunity) on pregnancy outcomes



Table IV: The cognitive function of infants and children been affected by overt or subclinical hypothyroidism during pregnancy



Table V. The feto-maternal outcomes of thyroid autoimmune positivity in euthyroid pregnant women


Conclusion
 
Although it is well documented that overt hypothyroidism and overt hyperthyroidism have deleterious impacts on pregnancy and childhood outcomes, there is however no consensus on the potential impact of subclinical hypothyroidism and subclinical hyperthyroidism on maternal and fetal health. Furthermore there is debate on the association between miscarriage and preterm delivery in euthyroid women positive for TPO antibodies. As a result the universal screening of pregnant women has not been recommended yet, as the benefits of identification of those subclinical for of thyroid disturbances has not been proved. There is not adequate data on cost-benefit of treatment of pregnant women suffer from subclinical thyroid disorders. Studies are now focusing on these controversial issues to produce critically needed data on the impact of treating these subclinical forms of thyroid disease on the mother, fetus, and the future intellect of the unborn child. The present review article is limited by not including the non-English articles. Summarizing the studies which have been published, the following can be concluded: 1) Overt hyperthyroidism and hypothyroidism have several adverse effects on pregnancy outcomes, 2) The long term effect of overt hypothyroidism on cognitive function has been well documented, 3) There is debate on short and long term effect of subclinical hypothyroidism, 4) Thyroid antibody positivity is associated with adverse pregnancy outcomes, but there is no consensus on feto-maternal complication of pregnant women with TPO antibody positive and euthyroid status.
Future studies should include the following: 1) Studies of possible benefits of levo-T4 (L-T4) in euthyroid and subclinical hypothyroidism women with positive TPO antibody; 2) Larger randomized control trials of patients with maternal hypothyroidism are necessary to impact on neurocognitive function; 3) More comprehensive studies with controlled iodine intake checks (urinary tests, for example) are suggested.
 
Acknowledgment
 
There was no involvement of a pharmaceutical or any other company in funding of this manuscript. There is no one in charge of medical writing or editorial assistance with the preparation of this article.

Conflict of Interests
 
The authors report no declarations of interest.