TSH Level and Pregnancy Loss

Peter Kovacs, MD, PhD


November 18, 2010

Increased Pregnancy Loss Rate in Thyroid Antibody Negative Women With TSH Levels Between 2.5 and 5.0 in the First Trimester of Pregnancy

Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T, Stagnaro-Green A.
J Clin Endocrinol Metab. 2010;95:E44-48


The thyroid gland is primarily involved in the control of metabolism. Abnormal thyroid function directly and indirectly affects reproduction as well. Infertility and adverse pregnancy outcomes are more common when the thyroid gland is hypo- or hyperactive. Higher miscarriage rate, more frequent preterm deliveries, increased hypertension, diabetic complications, higher risk for placental abruption, and adverse fetal effects have all been reported with thyroid dysfunction in pregnancy.

At least half of implanted embryos will not survive to delivery, and on average 20% of clinical pregnancies are lost. Most pregnancy losses occur in the first trimester.[1]Various factors (genetic, endocrine, anatomic, infectious, immune, hematologic, etc) are responsible for these losses. By correcting the treatable causes, we could make a positive impact on pregnancy outcome.

Hypo- and hyperthyroidism are both associated with increased pregnancy loss. In addition, it has been shown that women with normal thyroid-stimulating hormone (TSH) levels, but with evidence for thyroid autoimmunity, are at approximately a 2-fold increased risk for pregnancy loss.[2] Women are more likely than men to have antibodies against the thyroid, and with age the frequency increases. Those with antithyroid antibodies are more likely to develop abnormal thyroid function later on. Patients who are diagnosed with subclinical hypothyroidism (normal free thyroxine levels and slightly elevated TSH levels) are at an increased risk of developing hypothyroidism.[3]

During pregnancy, a 30%-40% increased need for thyroid hormones is the result of increased placental uptake, higher thyroid-binding globulin levels, and greater blood volume. Those with subclinical hypothyroidism and/or high-normal TSH levels at the beginning of pregnancy may not be able to meet these needs and may show signs of thyroid insufficiency during pregnancy.

This study evaluated whether women who have higher TSH levels during early pregnancy have any evidence for adverse pregnancy outcome.

Study Summary

The study reports pregnancy outcomes in 4123 women who had their TSH levels measured at the first medical visit during pregnancy. Only women who had no thyroid antibodies were included. All pregnancies were singleton and spontaneously conceived. Two groups were created on the basis of TSH levels: < 2.5 mIU/l (n = 3481) and 2.5-5.0 mIU/l (n = 642). Age and parity were comparable. The mean gestational age for the first prenatal visit was also similar (between 8-9 weeks). The pregnancy loss rate was 3.6% in women with a TSH level of < 2.5 mIU/l, and 6.1% in women with TSH levels between 2.5 and 5.0 mIU/l. This difference was statistically significant. Within the 2 TSH groups, age, obstetric history, and thyroid function test did not differ between those who miscarried and those who did not. The risk for miscarriage increased by 15% for each 1 mIU/l elevation of the TSH level on the basis of logistic regression analysis. The risk for preterm delivery and very preterm delivery did not differ on the basis of TSH levels.


Normal thyroid function is needed for a successful pregnancy. In the general population, a TSH level between 0.45 and 4.5 mIU/l is considered normal and indicates euthyroidism.[4] Studies, however, have suggested that the normal TSH level during pregnancy should be lower than this and have suggested using 2.5 mIU/l as the upper range cutoff.[4] The risk for miscarriage and preterm delivery were increased when the level was higher. The presence of thyroid antibodies may further complicate this situation. Women with positive antibodies are at a 2-fold increased risk for miscarriage.[2]

The highest risk for pregnancy loss is in the first trimester. Most losses have a genetic etiology and occur during the first few weeks. Most studies that have looked at TSH levels made the measurements at the time of or after the first prenatal visit, which typically took place between weeks 8-10 of the pregnancy. By this time most losses have already occurred. Studies such as the one discussed here report very low loss rates (3%-6%), but it needs to be pointed out that these are losses beyond week 10.

Data are limited on the management of thyroid antibodies during pregnancy. A randomized study found reduced pregnancy loss rate with levothyroxine replacement during pregnancy when thyroid antibodies were present. Thyroid function tests were done at around 10 weeks, and treatment was initiated only subsequently. Women with positive antibodies and no replacement therapy were found to have higher miscarriage rates than women with positive antibodies and levothyroxine replacement (13.8% vs 3.5%).[5] Another randomized study in patients undergoing in vitro fertilization (IVF) treatment did not find an association between the presence of antibodies and pregnancy rates. The miscarriage rate, however, was increased 2-fold in the antibody-positive group. Women who had antibodies but were given levothyroxine had pregnancy loss rates similar to the antibody-negative control population.[6]

Without question, women with hypothyroidism need to be treated prior to and during pregnancy. It appears that a lower upper-range cutoff should be used for normal TSH levels during pregnancy. However, it has not been settled whether all women or only those who are at risk for thyroid abnormality should be screened. The presence of antibodies should lower the threshold for treatment, although further studies are needed to support treatment in these cases. Women with otherwise unexplained previous pregnancy losses probably should be treated if their TSH level is above 2.5 mIU/l and/or they have a positive test for antibodies. It is, however, not clear at this point whether all women planning to undergo fertility treatment need to be screened for thyroid antibodies and whether they should be placed routinely on replacement if antibodies are found. It is also not clear what upper normal level of TSH needs to be used for women who only plan a pregnancy or fertility treatment. Should it be 2.5 mIU/l or 4.5 mIU/l? If above 2.5 mIU/l, should they be screened for antibodies and be treated if positive? If the TSH level is between 2.5 and 4.5 mIU/l and antibodies are not found, should the TSH be repeated very early on in pregnancy and is treatment supposed to be offered at that point? Must all women with positive antibodies regardless of their previous obstetric history receive treatment? Future studies will hopefully provide evidence to guide the management of these cases.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.