The Endocrine Society

The Endocrine Society

April 04, 2019
6 min read

Debate over universal screening for thyroid dysfunction in pregnancy centers on what to treat

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

NEW ORLEANS — The question about screening for thyroid dysfunction during pregnancy remains one with no clear answer, according to a debate at the Endocrine Society Annual Meeting.

Determining whether screening for any condition — including thyroid dysfunction — should be recommended is based on criteria established by the WHO, which calls for a common condition, an available and reliable screening tool, proof of favorable influences of intervention, relative cost-effectiveness and a uniform policy.

Alex Stagnaro-Green

Experts mostly agree that thyroid dysfunction in pregnancy is common. Alex Stagnaro-Green, MD, MHPE, MHA, of the University of Illinois College of Medicine in Rockford, who argued for universal screening during the debate, noted that in searching the current literature, the ranges for prevalence were up to 13.1% for overt hypothyroidism and up to 2.5% for overt hyperthyroidism. In addition, based on a study that appeared in Thyroid, Stagnaro-Green and colleagues found that the prevalence rates were 0.61% for overt hypothyroidism and 3.47% for subclinical hypothyroidism when using the 97.5th percentile of thyroid-stimulating hormone. The researchers noted a pooled prevalence range between 0.5% and 3.47% for each of the thyroid diseases, including overt and subclinical forms of hyperthyroidism and hypothyroidism.

“Based on prevalence and impact of overt thyroid disease in pregnancy, which is 1.15%, it is incumbent upon us to do universal screening whether or not the identification and treatment of subclinical hypothyroidism or thyroid autoimmunity has any beneficial effect,” Stagnaro-Green said during the debate.

Kristien Boelaert

In arguing against universal screening, Kristien Boelaert, MD, PhD, of the University of Birmingham in the U.K., agreed that the condition may be common, but with some important caveats. Based on her work in the TABLET trial, which examined the effect levothyroxine treatment had on pregnancy outcomes, Boelaert noted that only 85 of the 19,213 women screened for the study had overt thyroid dysfunction compared with higher prevalence for subclinical hypothyroidism (n = 451) and subclinical hyperthyroidism (n = 243).

Treatment effects

The criterion assessing whether the screening tool for thyroid dysfunction during pregnancy is easy to use, reliable and effective quickly exhibited diverging viewpoints.

“I looked at a lot of thyroid function testing during the TABLET trial, and interpretation is not straightforward,” Boelaert said. “Most guidelines tell us to use gestation-specific reference ranges. That is OK for people like me who work in large centers, but for the majority of us in reality and in real life, that is not possible.”


Boelaert also noted that unclear definitions of “normal reference ranges” and the fact that these are usually based on cross-sectional samples that ignore the effects of thyroid peroxidase (TPO) antibodies on thyroid function tests further damage the case. In addition, she said using measures of free thyroxine is “problematic” and that measuring total T4 concentrations, while preferable, is not necessarily possible since this test is not routinely available.

Stagnaro-Green countered by saying that the screening procedure for thyroid dysfunction in pregnancy, which comes in the form of a drawn blood sample, is a “slam dunk” and “imperceptible to the patient.” He also argued that the positive effects of treatment make such a screening procedure even more attractive. In treatment studies in the available literature, overt forms of thyroid dysfunction all had a strong correlation with adverse pregnancy outcomes, and therapy in these studies exhibited beneficial effects. Similar effects were observed for thyroid autoimmunity, Stagnaro-Green said, while stating that results were less clear for subclinical forms of thyroid dysfunction and isolated hypothyroxinemia. However, it is a focus on subclinical forms of the condition that may be halting progress toward consensus on screening.

“The discussion on whether or not to screen all women for thyroid disease in pregnancy has gotten lost and muddied in a discussion about subclinical hypothyroidism,” Stagnaro-Green said. “I’m not making the argument that we should screen all women to diagnose and treat subclinical hypothyroidism, and I am not making the argument that we should screen all women to diagnose and treat women with autoimmune antithyroid disease. My argument is very simple: We are missing the forest for the trees.”

Boelaert said she is less bullish on the positive effects of treatment. In the CATS trial, no significant difference was seen in the IQ of children of women who were treated for subclinical hypothyroidism with LT4 vs. those treated with placebo after 3 years. Similarly, in the NICHD trial, no significant difference was seen in the IQ of children of women who were treated for subclinical hypothyroidism or isolated hypothyroxinemia with LT4 vs. those treated with placebo after 3 to 5 years. In addition, in the TABLET trial, the rates of live births at 34 weeks were not significantly different for women with normal thyroid function but positive TPO antibodies who were treated with levothyroxine compared with those treated with placebo. This lack of significance held for other pregnancy outcomes, such as miscarriage.


Boelaert further argued that not only were the effects of treatment not statistically significant, but the risk for harm was higher than it might appear on the surface, noting that higher free T4 levels could be associated with autism and higher risk for ADHD based on evidence from the CATS trial and observational studies.

Cost-effectiveness of screening

On top of determining the ease of use and effects of treatment from screening, both Boelaert and Stagnaro-Green offered differing opinions on the cost-effectiveness of universal screening.

For Stagnaro-Green, the public health strategy easily meets this criterion. Based on cost-effectiveness analysis, which measured the costs of blood tests, extra visits and treatment against the benefits, such as decreased miscarriage and decreased preterm delivery, he argued that the incremental cost-effectiveness ratio for universal screening fell below the $50,000 threshold set in the United States for acceptable treatment. He also cited four studies that provided evidence for cost-effectiveness of universal screening in comparison with targeted screening, noting that one study included data that excluded any benefits of treatment for subclinical hypothyroidism.

Measuring cost-effectiveness in this way ignores additional expenses associated with universal screening, according to Boelaert. Specifically, she argued that the “burden of overdiagnosis and overtreatment” is not effectively addressed nor is the toll placed on patients and the health care infrastructure that universal screening would require.

“I don’t think we have sufficient proof that this is cost-effective, and certainly in the health care system where I work, and I presume even in the U.S., [for] those women who don’t have coverage this is not a cost-effective treatment,” Boelaert told Endocrine Today.

Target conditions

Disagreements about whom to treat are prevalent in this discussion, and both Stagnaro-Green and Boelaert acknowledged this roadblock. However, Stagnaro-Green said the European Thyroid Association, American Association of Clinical Endocrinologists and Italian Thyroid Association are among those that already recommend universal screening. Where he said he finds real uncertainty is in determining an agreed-upon stance on subclinical hypothyroidism and thyroid autoimmunity.

Boelaert said she views a medical community that is much further from consensus, noting that a debate over reference ranges, levothyroxine starting point, TPO antibodies and whether to treat isolated hypothyroxinemia complicate the matter.

Even after weighing the criteria, it appears that a consensus on universal screening remains an illusive target. For Stagnaro-Green, who is primarily concerned with screening as a means to diagnose and treat overt forms of thyroid dysfunction, it will take an easing of stringent practices.


“What we’re really saying is we are trying to avoid uncertainty at all costs. We’re putting our head in the sand, and we’re saying we only want to diagnosis that what we know how to handle. But that’s not how medicine works,” he said. “Our patients should know what risks they may or may not have and what we as professionals know and do not know.”

Conversely, Boelaert said she believes that more work must be done before an accurate, effective and agreed-upon strategy can be established.

“What we need are large randomized controlled trials where we start early, ideally preconception, and where we then by doing an intervention see if there are differences in outcomes,” Boelaert told Endocrine Today. “At the moment, what we have are trials that have been well-conducted but where treatment has been started too late, and it’s therefore very difficult to extrapolate what would happen if we started early in pregnancy.” – by Phil Neuffer


E22. Endocrine Debate: Thyroid Disease in Pregnancy: To Screen, or Not to Screen? Presented at: The Endocrine Society Annual Meeting; March 23-26, 2019; New Orleans.

Casey BM, et al. N Eng J Med. 2017;doi: 10.1056/NEJMoa1606205.

Dhillon-Smith RK, et al. N Eng J Med. 2019; doi:10.1056/NEJMoa1812537.

Dong AC, et al. Thyroid. 2019;doi:10.1089/thy.2018.0475.

Lazarus JH, et al. N Eng J Med. 2012;doi:10.1056/NEJMoa1106104.

Disclosures: Boelaert and Stagnaro-Green report no relevant financial disclosures.