Thyroid Screening in Pregnancy: Too Much of a Good Thing?

Regularly screening pregnant women for thyroid dysfunction may lead to overdiagnosis, Canadian researchers reported.

In an analysis of over 188,000 pregnant women, about 60% underwent at least one measurement of thyroid-stimulating hormone (TSH), according to Jennifer Yamamoto, MD, of the University of Calgary, and colleagues.

Less than 6% of these pregnancies had free thyroxine measured at least once, and only 2.5% had thyroid peroxidase antibodies measured at least once during the pregnancy, they reported in CMAJ, the Canadian Medical Association journal.

TSH measurement was most typically conducted in early pregnancy, between gestational weeks 5 and 6, while thyroid hormone therapy was usually initiated by about week 7.

However, among women whose first TSH measurement reflected subclinical hypothyroidism -- a TSH between 4.01 to 9.99 mU/L -- but who did not immediately start hormone treatment, about 68% had a "normal" (below 4 mU/L) reading on a repeat TSH measurement.

Among the women who were started on thyroid hormone therapy during pregnancy, nearly half remained on this treatment after giving birth.

Timing is particularly important here as both the initial TSH measurement and initiation of thyroid hormone therapy "occurred in a period when it is well established what TSH level falls rapidly because of normal physiologic changes of pregnancy," specifically between weeks 5 to 7 of gestation, the researchers pointed out.

"This raises concerns about overmedicalization during pregnancy, given that minor, untreated TSH elevation usually normalized, as indicated by repeat measurement," Yamamoto's group stated.

Upper limit thresholds for diagnosing hypothyroidism in pregnancy have also seen some change over the years.

Earlier guidelines suggested a TSH cutoff of recommendations for a TSH upper reference limit of 2.5 mU/L during the first trimester and 3.0 mU/L for both the second and third trimesters. However, updated guidelines in 2017 from the American Thyroid Association (ATA) suggest the upper reference limit of 4.0 mU/L should be applied beginning with the late first trimester (weeks 7 to 12), followed by a "gradual return" towards the nonpregnant range during the second and third trimesters.

For this retrospective cohort study, Yamamoto's group drew upon records in the Alberta Perinatal Health Program database. This included women between the ages of 15 and 49 who delivered a child in the province. Women were excluded if they had evidence of thyroid disease within the prior 2 years before delivery, which was defined as filling a prescription for any thyroid medication, had a diagnosis, or TSH measurement less than 0.20 mU/L or 5.00 mU/L.

During pregnancy, subclinical hypothyroidism was defined according to the 2017 ATA guidelines, with a TSH range between 4.01 to 9.99 mU/L. Overt hypothyroidism was defined as a TSH of 10.00 mU/L or higher. Initiation of thyroid hormone therapy during pregnancy was defined as the filling of a prescription for levothyroxine, desiccated thyroid, or liothyronine after conception but prior to delivery.

Nearly all pregnant women (99.6%) who initiated thyroid hormone therapy were started on levothyroxine, at an average dose of 44.1 μg/day. The few other remaining pregnant women (0.4%) were instead started on desiccated thyroid therapy.

Among the over 1,700 pregnancies where the mother had an initial TSH measurement of 10.00 mU/L or higher (overt hypothyroidism), thyroid hormone therapy was only started in about 18% of these cases.

"It is not clear from these data why some women with TSH of 10 mU/L or higher did not receive treatment with thyroid hormone," the researchers stated. "This may have occurred in part because TSH fell below 4.01 mU/L in more than 60% of those with repeat TSH measurement." They also posed the idea that these numbers may be attributed to women not filling their prescriptions.

On the other hand, hormone therapy was initiated in about 8.5% of the pregnancies whose highest TSH measurement fell between 2.51 to 4.00 mU/L -- below the upper limit threshold. For the remaining women who had an initial TSH measurement in the subclinical hypothyroidism range (4.01 to 9.99 mU/L), about 56% of women were started on hormone therapy.

"Clinical practice guidelines are needed to give clinicians a stepwise approach, based on the best existing evidence, for deciding whether and when TSH testing should occur," Yamamoto's group suggested. "Guidance is also needed as to when it is appropriate to initiate treatment in pregnancy and continue treatment in the postpartum period."

Comment