Thyroid disease affects approximately 5% of U.S. women of reproductive age and has been associated with complications in pregnant women and their children. This relatively high incidence in part is driving a heated debate over whether there should be universal screening in pregnancy consisting of a thyroid-stimulating hormone (TSH) test—followed by measurement of free thyroxine (FT4) if TSH is abnormal. While a strong body of research points to a clear relationship between overt thyroid disease and pregnancy complications—including miscarriage, pre-eclampsia, anemia, placental abruption, and postpartum hemorrhage—evidence about the effects of subclinical thyroid disease, especially hypothyroidism, is less clear. Given this nuanced view of thyroid disease in pregnancy, it perhaps is not surprising that medical groups have taken different stances on universal screening.
Overt hypothyroidism affects up to 1% of all pregnancies. But subclinical hypothyroidism, which has been estimated to affect between 3% and 15% of pregnancies, “is where the controversy lies,” said Alex Stagnaro-Green, MD, regional dean at the University of Illinois College of Medicine in Rockford, chair of the American Thyroid Association’s (ATA) Taskforce on Thyroid and Pregnancy, and lead author of 2011 ATA guidelines on thyroid disorders and pregnancy.
The debate began with a 1999 study that found children ages 7 to 9 born to women with undetected thyroid deficiency had IQ scores that averaged 7 points lower than IQ scores of children of treated control subjects. Advocates of screening note that certain risk factors for both clinical and subclinical thyroid disease—age older than 30, obesity, and diabetes—are common as women bear children at later ages, so it makes sense to offer all women relatively inexpensive TSH tests and treatment to avoid pregnancy complications and possible intellectual impairment in children. Others support targeted screening and treatment of overt disease in pregnancy, but oppose universal screening without more evidence of improved outcomes from universal screening and subsequent treatment of subclinical hypothyroidism.
Difference of Opinions
Obstetrical and endocrinology society guidelines on universal thyroid screening in pregnancy have differed and changed in recent years. The most recent and influential 2015 recommendations from the American Congress of Obstetricians and Gynecologists (ACOG) suggest testing only women at high risk of thyroid disease before they become pregnant or when they are early in pregnancy, before the fetal thyroid gland begins concentrating iodine and synthesizing thyroid hormone at about 12 weeks of gestation. ACOG opposes routine pregnancy TSH testing, as does the Society for Maternal-Fetal Medicine.
Joint guidelines issued in 2005 by the American Association of Clinical Endocrinologists (AACE), ATA, and the Endocrine Society had recommended universal screening based on the 1999 research and another observational study showing that universal screening and treatment improves outcomes. A limited number of randomized, controlled trials, including one by Stagnaro-Green, showed similar results.
But a 2012 paper detailing a large randomized study of screening and treatment of more than 21,800 pregnant women with subclinical hypothyroidism found no effect on cognitive function of their children at age 3 (N Engl J Med 2012; 366:493–501). That study did not begin thyroid hormone replacement therapy until almost the 14th week of pregnancy, which is after the critical period in fetal brain development.
This study led to different stances by the endocrinology groups in separate guidelines issued later that year. Endocrine Society guidelines recommend against universal screening for thyroid disease before pregnancy, and do not make a clear recommendation about such screening in pregnancy. Rather, they note that some authors favored screening all pregnant women during their first obstetrical visit or by 9 weeks of gestation, while other authors generally supported testing high-risk women only. Meanwhile, AACE guidelines take no stance on the issue because the authors could not agree on one.
Obstetricians and endocrinologists are now waiting for results of a trial that is evaluating whether levothyroxine treatment for pregnant women with mild hypothyroidism affects their children’s intellectual development at age 5. Led by Brian Casey, MD, at University of Texas Southwestern Medical Center in Dallas, the large randomized, blinded trial screened about 120,000 pregnant women and enrolled 1,203 women at up to 20 weeks’ gestation and their children, so some women fell outside of the window of important fetal brain development. The authors are now evaluating data and will report data on intellectual development, as well as on rates of attention deficit, behavioral and fetal growth problems, preterm delivery, pre-eclampsia, stillbirth, and postpartum maternal thyroid dysfunction.
For the Mother’s Health
Noting conflicting evidence about universal screening’s benefits for children, the 1999 study’s first author still supports widespread TSH testing in pregnancy to protect and improve mothers’ health. Sixty-four percent of that study’s untreated subjects had full-blown hypothyroidism 11 years after giving birth, noted James Haddow, MD, co-director of the Division of Medical Screening and Special Testing at Women and Infants Hospital of Rhode Island and research professor in the department of pathology and laboratory medicine at the Alpert Medical School of Brown University. “My view is that universal screening can be justified on that basis alone. Pregnancy offers an excellent situation for offering screening.”
Meanwhile, Sarah Kilpatrick, MD, PhD, staunchly opposes universal TSH screening at this time. “Screening to identify subclinical hypothyroidism, a case in which the mother isn’t sick and has slightly elevated TSH, is not indicated because the data associated with long-term poor childhood or maternal outcomes is quite poor, and there are no data to say treatment does anything,” explained Kilpatrick, who is chair of the Department of Obstetrics and Gynecology at Cedars-Sinai Medical Center in Los Angeles.
“For the question of universal thyroid screening in pregnancy, the jury is still out,” said Ann Gronowski, PhD, professor of pathology, immunology, obstetrics and gynecology at Washington University School of Medicine in St. Louis. Echoing Kilpatrick, she added, “Until there are more data, we don’t know if screening is worthwhile and treatment is beneficial and safe.”
Kilpatrick looks forward to seeing the results of Casey’s study. “No one really knows which direction it will go. The study will not be perfect, but I believe it will help us understand what the right thing to do is.”
Even if Casey’s data shows that treatment for subclinical disease is not effective, Stagnaro-Green believes benefits to pregnant women with overt disease make routine screening worthwhile. In a recent commentary, he pointed out that overt disease is as common and sometimes more frequent than many pregnancy-related conditions detected by routine monitoring.
Even though they disagree on universal thyroid screening in pregnancy, Haddow and Gronowski both urge laboratorians to help clinicians navigate conflicting advice and a dearth of solid reference ranges for pregnant women. “If clinicians choose to test thyroid markers, we can help them interpret findings,” Gronowski says.
While ATA’s 2011 guidelines suggest reference ranges for pregnancy, a recent review paper by researchers from Erasmus Medical Center in the Netherlands urged laboratories to develop their own because of large differences in TSH and FT4 reference intervals used in several studies of thyroid function in pregnancy. The paper noted that 90% of the studies used higher upper limits of TSH than the fixed TSH cutoff concentrations of 2.5 and 3.0 mU/L suggested by ATA.
“Communicate to doctors that lower TSH may be normal, especially in the first trimester,” Gronowski suggested, explaining that rising concentrations of human chorionic gonadotropin (hCG) during pregnancy cause a dip in TSH and sometimes increases in FT3 and FT4. “In addition, total T3 and total T4 are increased during pregnancy, up to 1.5 times the upper limit of the non-pregnant reference interval,” she added.
“A prime prerequisite for offering screening for thyroid deficiency during pregnancy is for the laboratory to develop gestational age-specific normative data, focusing especially on the first and second trimesters,” Haddow said. “At least one laboratorian in a given laboratory should either have or develop a knowledge base in this subject area, so that physician questions can be adequately addressed.”
Deborah Levenson is a freelance writer in College Park, Maryland. +Email: firstname.lastname@example.org