Issue: October 2019
October 17, 2019
4 min read

Should all pregnant women be screened for thyroid dysfunction?

Issue: October 2019
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact


Click here to read the Cover Story, "Conflicting findings, guidelines support cautious approach to mild thyroid dysfunction in pregnancy."


Universal screening would be beneficial for identifying overt thyroid disease.

We should not automatically accept that controversy in one area should serve as a rationale for doing nothing in another.

We recognize that thyroid function is important to pregnancy, and there are risks associated with abnormal thyroid states during pregnancy; however, there is controversy regarding the extent of those risks, what values should delineate the healthy reference range, and when thyroid treatment improves various pregnancy outcomes

Trevor E. Angell

With respect to thyroid function screening, researchers have identified key criteria that would be relevant for any screening test and applied them to evaluating the thyroid in pregnancy. The first criterion is that the condition is sufficiently common. Overt thyroid disease — either hypothyroid or hyperthyroid — where biochemically there is a clear abnormality, constitutes about 1% of pregnancies with screening. These are not rare conditions. A significant number of cases — reaching 30% in some estimates — would be missed in just a case-finding strategy.

The second criterion would be a reliable, available tool for screening that is acceptable for patients. We have this. Simple blood testing for thyroid-stimulating hormone and thyroxine are readily available, patients tolerate these tests, and they are reliable. Some have argued that measured free T4 levels are inaccurate in pregnancy; however, this happens predominantly in the second and third trimester. A number of large, prospective cohort studies looking at first trimester levels suggest that first trimester free T4 correlates better with TSH and with outcomes than do other measures.

There should be beneficial treatment if the condition is identified. Hypothyroidism increases risks for miscarriage, preeclampsia and gestational hypertension. For overt disease, there is a benefit to taking levothyroxine that is well recognized.

We do not have randomized controlled trial evidence to support a screening recommendation, because to not treat women with overt disease would be considered unethical. Finally, thyroid testing is inexpensive and in at least one study, researchers found it would be cost-effective to screen pregnant women for thyroid disease.

Yet, the clarify of the situation is clouded by the problem of mild (subclinical) thyroid disease, for which the risks to pregnancy and benefits of treatment are less clear. When clinicians screen women during pregnancy, they are going to find subclinical cases and may not know what to do about it. Those are inadvertent consequences. Even if there is uncertainty, much of the existing data would suggest that levothyroxine supplementation is not harmful, so many would favor a strategy where perhaps more people were getting treated. In the last year, some small randomized trials with levothyroxine, particularly in women with positive thyroid peroxidase antibodies, have been positive. This growing evidence in women with subclinical hypothyroidism would further support greater benefits if universal screening were to be adopted.

Trevor E. Angell, MD, is an endocrinologist at Brigham and Women’s Hospital. Disclosure: Angell reports he serves on an advisory board for IBSA Pharma, a maker of levothyroxine.



The evidence for universal screening is controversial, and we do not have an agreed-on screening policy.

Physiological changes occur during pregnancy that cause changes in binding proteins and in levels of thyroid hormones, making the interpretation of thyroid function difficult. Saying that this is a simple test that we can easily interpret is inaccurate.

Kristien Boelaert

I am not claiming that overt thyroid dysfunction does not require treatment in pregnancy. It does, and we need to identify women with overt disease early. However, when we screen a lot of people, we see many with mild thyroid dysfunction — or subclinical thyroid disease — isolated hypothyroxinemia and, in the case of hyperthyroidism, transient, gestational thyrotoxicosis. There is no consensus on what to do with these cases.

When looking at the criteria for universal screening, are these common conditions? Yes, indeed they are. Thyroid peroxidase (TPO) antibodies, as shown in the TABLET trial, are present in 5% to 14% of unselected pregnancies and are much higher in the subfertile or high-miscarriage population.

All guidelines tell us that, to make treatment decisions, we need to know women’s TPO antibody status. In women with TPO antibody positivity, interpretation of thyroid function becomes even more difficult because the expected physiological changes, such as a dip in serum TSH and a rise in free T4, do not occur.

The second criterion calls for an easily identifiable screening tool. However, interpretation is not straightforward. Most guidelines tell us to use gestation-specific references ranges. For most clinicians, that is not possible. We are continuously debating what the thyroid reference ranges during pregnancy should be. We all know that the interpretation of, for example, free T4 measurements, can be problematic, and guidelines tell us that, ideally, we should be measuring total T4, though this is only available in a few centers. Do we know that the intervention favorably influences disease? We have large trials powered to look at the neurocognitive effects of treatment with levothyroxine in offspring, and those trials have been negative. Levothyroxine intervention also did not favorably influence obstetric outcomes.

Do we have clear evidence that this intervention is safe? For many years, we all assumed that, while we may not have good evidence that levothyroxine is beneficial, we’re probably not doing much harm. We now have increasing evidence that overtreating women in pregnancy does indeed cause harm. In the Generation R study, Korevaar and colleagues found both high and low concentrations of T4 adversely affect neurocognitive outcomes in offspring. If you go to an overtreatment scenario, you may slide off the slope, indicating that overtreatment may cause harm.

Some argue screening is inexpensive. However, the cost of public education, personnel training and the burden of overdiagnosis and overtreatment, which can cause anxiety for many women, has not been calculated. We need to take these things into account.

It is strongly urged that not all pregnant women undergo screening for thyroid function and thyroid autoimmunity.

Kristien Boelaert, MD, PhD, MRCP, is head of the Institute of Metabolism and Systems Research and a Reader in Endocrinology at the University of Birmingham, U.K. Disclosure: Boelaert reports no relevant financial disclosures.