Overdiagnosis of Hypothyroidism in Pregnant Women

Hypothyroidism during pregnancy comes with substantial risk to mother and child, which may be the reason why health care professionals readily measure thyroid function in pregnant women. A recent study by Jennifer Yamamoto MD at the Alberta Children’s Hospital Research Institute in Canada and colleagues indicate that an eagerness to measure thyroid function might be leading to the overdiagnosis and treatment of women during and after pregnancy.

TSH reference limits

The paper by Yamamoto et al. looked at the current management of thyroid testing, using TSH levels, and treatment during pregnancy in Canadian women without any prior diagnosis of thyroid disease. TSH testing and reference guidelines for diagnosis are complicated at the best of times. During pregnancy, the TSH reference guidelines are less straightforward.

TSH, or thyroid stimulating hormone, is a chemical messenger which is part of a negative feedback loop regulating the release of thyroid hormones (THs) from the thyroid gland. When TSH levels are high, THs levels are low, indicating a hypothyroid state.

What then is a high TSH level? TSH levels can vary by age, sex, ethnicity, diet, and weight. The lower TSH limit, usually defined by the bottom 2.5^th percentile in a healthy adult population is relatively similar across people groups at ~0.4 mIU/L. According to the National Health and Nutrition Examination Survey III (NHANES-III), the upper limit of TSH levels identified by the 97.5 percentile reveal several population differences. For example, the upper percentile in TSH values for different ethnic women in the United States were reported as:

• African American women: 3.69 mIU/L

• Caucasian American women: 6.77 mIU/L

• Hispanic American women: 6.31 mIU/L

The survey also found that individuals less than 40 years old, regardless of ethnic background, typically have an upper TSH limit between 3.6-3.8 mIU/L. As age increases, TSH levels also increase, so the variation between ethnic women listed above could be influenced by the broad age range (12-80+ years) of the surveyed sample. The data was not parsed out further into age by gender by ethnicity.

Defining the euthyroidal TSH range is further complicated in pregnancy. During the first trimester it is common to see TSH levels decline as a result of rising placental chorionic gonadotropin and thyroid hormone production. Several guidelines have been offered for pregnant women, but again TSH levels are variable across ethnicities. Table 1 lists the TSH upper limits for pregnant women by trimester as recommended or determined by several groups.

Upper TSH limits (mIU/L) for pregnant women from different countries.

The American Thyroid Association (ATA) recognizes the limitations of defining a precise cutoff point for hypothyroidism in pregnant women. Without an available population and trimester defined reference range which is representative of the patient in age and ethnicity, the ATA suggests the following:

“If internal or transferable pregnancy-specific TSH reference ranges are not available, an upper reference limit of ~4.0 mIU/L may be used. For most assays, this limit represents a reduction in the nonpregnant TSH upper reference limit of ~0.5 mIU/L.”

TSH testing in a Canadian cohort

The retrospective paper by Yamamoto et al. looked at women who gave birth in the ethnically diverse Alberta province between October 2014 and September 2017. Data was collected on over 188,000 women without any previous thyroid disease. 59.2% of women had at least one TSH measure during their pregnancy, often during the 5^th or 6^th gestational week. The women most likely to be tested were:

• 35 years and older

• first time mothers (nulliparous)

• urbanites

• those who developed gestational hypertension or had other medical disorders

The team, following the ATA guidelines, divided the group into normal TSH levels (<= 4.0 mIU/L), subclinical hypothyroid (4.01-9.99 mIU/L), and overt hypothyroid (>= 10.00 mIU/L). More than 90% of patients fell into the normal, 4% were classified as subclinical, and approximately 1.5% had overt hypothyroid disease.

Hormone therapy in the lowest and highest TSH levels

Of those women who were tested, only 4.5% were placed on thyroid hormone therapy. Levothyroxine was prescribed to 99.6% of these patients. The median dose was 50 ug/day (IQR 25-50 ug/day) and many women began treatment during the 7^th gestational week.

Among the women who tested in the normal TSH range (<= 4.0 mIU/L), 2.2% were placed on hormone therapy. A third continued treatment post-partum and 25% of women continuing therapy for at least one year after delivery.

Substantially fewer women who were identified as having overt hypothyroidism (TSH >= 10.0 mIU/L) were placed on hormone therapy. In the paper, the authors commented on this unusual phenomenon,

“It is not clear from these data why some women with TSH of 10 mIU/L or higher did not receive treatment with thyroid hormone. This may have occurred in part because TSH fell below 4.01 mIU/L in more than 60% of those with repeat TSH measurement.”

Despite the small number of women medicated for overt hypothyroidism, 70% remained on levothyroxine post-partum and over half were still on treatment for hypothyroidism a year later.

Testing and treating subclinical hypothyroidism in pregnant women

Of the 4417 women (~ 4% of the tested cohort) with an initial TSH level in the 4.01-9.99 mIU/L, only 55.5% initiated hormone therapy. Of those on treatment, 47.9% maintained their treatment post-partum and a third continued for at least a year after delivery.

About a quarter of women in the subclinical hypothyroid range went untreated until a second TSH test was performed usually five weeks later. In the majority of these untreated women (68%), TSH levels dropped below 4.01 mIU/L, into the normal range. Still, 14.4% of them were placed on levothyroxine treatment after the second test.

In 30% of women the TSH levels remained in the subclinical hypothyroid range and most (~70%) were started on hormone treatments.

A small percentage of women (2.4%) had elevated TSH levels at or above 10.00 mIU/L by the second test. Again, only a small percentage (~30%) were placed on hormone therapy.

Overdiagnosis and overmedication

The researchers suggest the current pattern of measuring during early pregnancy when TSH are at their most volatile and basing decisions on a single TSH measure are the leading cause of overdiagnosis and medicating women for hypothyroidism during pregnancy. Corresponding author for the paper, Lois Donovan MD, Clinical Professor at Cumming School Medicine, University of Calgary and Medical Director of Diabetes in Pregnancy, Calgary, Canada, offers the following to health care professionals:

“The practice of TSH screening in pregnancy, among women without thyroid disease, may be contributing to unnecessary treatment of women during pregnancy, postpartum and for years to come. If a pregnant woman without thyroid disease is found to have a minor TSH elevation in early pregnancy, we recommend a repeat TSH since this frequently normalizes.”

We turned to Maria Brito MD, co-director of the Thyroid Center at Mount Sinai at Union Square and assistant professor of medicine at the Icahn School of Medicine at Mount Sinai in New York City, to see if she agrees with the conclusion that women are being over diagnosed or overmedicated during pregnancy.

Dr. Brito said, “Yes, especially in women with TSH levels in the high-norm range (i.e. >2.6-4.5mIU/L). Healthcare practitioners should discuss with their high-norm patients that there is no strong evidence that thyroid medication is particularly beneficial at this time. It is questionable that it may help optimize the chances of a successful pregnancy but there is also little risk of harm. So often it becomes a case of being safe rather than sorry.”

About the 31% of women who continued thyroid treatment after pregnancy, Dr. Brito said, “There is a difference between finding a disorder during pregnancy and developing a disorder as a result of pregnancy. Thyroid medications should not be perpetuated past delivery in patients without frank hypothyroidism.”

In agreement with Dr. Donovan’s comment regarding unnecessary treatment postpartum, Dr. Brito added, “TSH should be measured again about 6 weeks post-delivery to determine if hypothyroidism persists. If TSH levels continue to be elevated, then medications can be continued.”

Dr. Brito leaves us with three discussion points that healthcare professionals can use if testing reveals a high-normal TSH level in their pregnant patients:

1. It is arguable that thyroid medication may optimize successful pregnancy in view that there is a lack of strong evidence to support this, however there is little evidence of harm as well. Again, it may be better to be safe than sorry.

2. The medication will likely be discontinued after delivery since subclinical hypothyroidism can result from pregnancy alone.

3. Testing TSH levels several weeks post-partum will identify those women who need to restart or continue taking thyroid medication

The authors report no competing financial conflicts with regard to their involvement in discussing this study.