Questions Remain Over Thyroid Management in Pregnancy

Liam Davenport

November 17, 2019

BRIGHTON — The conundrum of how to identify women who have undiagnosed overt thyroid dysfunction either before or during pregnancy remains unsolved, as screening may not be an effective way of picking them up, indicate two studies presented at Society for Endocrinology BES 2019.

Moreover, the proportion of women with fertility issues who have subclinical hypothyroidism (SCH) varies widely depending on how it is defined, making it unclear how best to approach their management.


In the first study, Dr Rima Dhillon-Smith, clinical lecturer at the University of Birmingham, went back over data from the previously published TABLET study - Thyroid Antibodies and Levothyroxine.

As reported by Medscape Medical News, this trial of levothyroxine before conception in women with normal thyroid function but with anti-thyroid peroxidase antibodies (TPOAb) did not reduce the rate of miscarriages.

Now, their analysis revealed that, across the whole screened population of almost 20,000 women, 0.5% had overt thyroid dysfunction, and 9.5% were positive for TPOAb.

However, the proportion of women identified as having SCH ranged from 4.8% to 19.9%, depending on which thyroid stimulating hormone (TSH) threshold they used.

Dr Dhillon-Smith said that, for her, there are three take home messages from the study.

The first is that "we need to be aware that there are women out there who are unaware that they have thyroid disease".

Second, "if we lower the TSH to 2.5 mIU/l and define this as subclinical hypothyroidism, then we’ll be treating up to 20% of women".

Dr Dhillon-Smith said this raises the question of whether levothyroxine treatment is beneficial in women with "only mildly raised" TSH levels, "because if there isn’t a benefit, we need to stop treating this population".

She continued: "Finally, antibody positivity is not as high as we thought it was." Nor did antibody positivity differ between fertile and subfertile women, although it was associated with a body mass index (BMI) of  35.0 kg/m2or more and TSH concentrations ≥2.50 mIU/l.

The next question is around screening for TPOAb, as "although it’s linked with adverse outcomes we can’t reduce those, so should we really be screening for something that we can’t modify the outcome for?"

Finally, Dr Dhillon-Smith asked what the cost implications would be of screening "every single woman at risk of miscarriage or with subfertility to pick up the 0.5% of women" with overt thyroid dysfunction.

Second Study

In the second study, Rebecca Scott from the Division of Diabetes, Endocrinology and Metabolic Medicine, Imperial College London, and colleagues performed thyroid function tests in 416 women during the first trimester of pregnancy.

Of those, 259 (62.3%) had known thyroid conditions. The remaining 157 women were screened due to the presence of diabetes, including gestational diabetes (32.3%), hyperemesis (15.9%) and other endocrine disorders (3.2%), while no reason was recorded for the thyroid function test in 40.8%.

The researchers report that no new cases of overt or subclinical hypothyroidism were identified in the women without a history of thyroid conditions.

The authors therefore concluded that "this would argue against testing for hypothyroidism in women without a history of thyroid dysfunction, unless there is a strong suspicion of thyroid disease".

They nevertheless add that "clinicians must remain vigilant for new cases of hyperthyroidism, especially as the symptoms overlap with those of hyperemesis within the first trimester".

Common Condition

Dr Dhillon-Smith said that thyroid disease is "one of the most common medical conditions that we see in women of reproductive age".

She added that "the thing we all agree on is that women who have overt thyroid disease definitely need to be treated before they become pregnant and especially in the fertility setting before they undergo IVF treatment".

"The area that is a bit more grey, where there’s certainly disagreement between gynaecologists and endocrinologists, is how we manage subclinical hypothyroidism and thyroid autoimmunity."

Dr Dhillon-Smith said this is particularly the case in women "who are labelled as high risk, so have a known clinical history of miscarriage or women who have subfertility".

She explained that there has been "quite a long-standing debate" as to whether such women should be routinely screened before conception for thyroid disease and/or thyroid autoimmunity, "as we know that there are adverse effects in pregnancy", or be tested on a case-by-case basis.

"Depending on where you look," she added, "there’s lots of different guidelines, which will give you different advice depending on what you want to find.

"In general, I would say that endocrinologists are less likely...and gynaecologists are more likely to treat and screen."

However, Dr Dhillon-Smith pointed out that to know whether a screening test is going to be effective, the prevalence of the disease needs to be understood.

History of Miscarriage or Infertility

She and her team therefore turned to data from the TABLET study, which included women aged 16–40 years who wanted to conceive but had a history of at least one miscarriage or infertility.

Among others, they excluded women currently receiving treatment for thyroid disease or those with a history of cardiac disease, as well as those with a contraindication to levothyroxine.

They recruited women at 49 centres in England and Scotland, performing thyroid function tests in 19,213 women and testing 19,237 women for TPOAb, recording the age, body mass index (BMI), ethnicity and fertility history of all participants.

The results indicated that 95.2% of women were euthyroid, defined as a TSH level of 0.44–4.50 mIU/l and a free thyroxine (fT4) level of 10–21 pmol/l.

Overt hypothyroidism, defined as TSH >4.50 mIU/l and fT4 <10 pmol/l, was found in 0.2%, while 0.3% had overt hyperthyroidism, defined as TSH <0.44 mIU/l and fT4 >21 pmol/l.

When using an upper TSH concentration of 4.50 mIU/l, the prevalence of SCH was 2.4%. However, lowering the upper TSH limit to 2.50 mIU/l meant that 19.9% of women were classified as having SCH.

Multiple regression analyses demonstrated that SCH, defined as TSH >4.50 mIU/l, was significant associated with BMI ≥35.0 kg/m2, at an adjusted odds ratio of 1.71 (p=0.01), and Asian ethnicity, at an odds ratio of 1.76 (p<0.001).

SCH defined as TSH ≥2.50 mIU/l was significantly associated with subfertility, at an adjusted OR 1.16 (p=0.008), as well as increased BMI and Asian ethnicity.

The team found that 9.5% of women were TPOAb positive, with no difference between fertile and subfertile women.

Women with a BMI ≥35.0 kg/m2 were significantly more likely to be TPOAb positive than normal weight women, at an adjusted odds ratio of 1.54 (p<0.001).

Dr Dhillon-Smith was asked in the post-presentation discussion how she would counsel women who have a TSH level between 2.5 and 4.5 mIU/l.

She replied that the results from TABLET indicated that 8% of women with a TSH level of 3.5 mIU/l developed clinically relevant thyroid disease that required treatment.

However, she said that "a lot of it is down to antibody status".

"If they have a borderline TSH but they are antibody negative we don’t worry so much, but if they’re antibody positive I would be inclined to treat, especially if they are undergoing IVF treatment."

Dr Dhillon-Smith explained that this is because "IVF treatment has been shown to be a stress test for thyroid hormone [function] in antibody positive women".

The research by Dr Dhillon-Smith and colleagues was funded by the National Institute for Health Research. No other funding declared.

No conflicts of interest declared.

Society for Endocrinology BES 2019: OC4.5 and P398. Presented 12th November.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: