In a new study, researchers screened more than 10,000 pregnant women and identified a group with previously undetected hypothyroidism. One in three of these women would not have been identified using the screening approach currently recommended, says a consultant.
During the first half of pregnancy, the foetus relies on thyroid hormones from the mother, partly for normal brain development. Nevertheless, some pregnant women at this stage have severely reduced thyroid function without knowing it because they are never tested.
Today, both in Denmark and internationally, thyroid function is primarily measured in pregnant women who are already considered at increased risk – for example, because they have previously had thyroid disease themselves or because it runs in the family. But what about women who actually have hypothyroidism without appearing to be at risk?
“We know that we are not identifying everyone with the current approach. As a result, some pregnant women have undiagnosed hypothyroidism because they do not meet the criteria for being tested. That is why we wanted to see whether we could develop a model that identifies precisely those women whose thyroid function is so low that they need treatment,” explains Stine Linding Andersen, Clinical Professor and Lead Consultant at the Department of Clinical Biochemistry at Aalborg University Hospital and the Department of Clinical Medicine at Aalborg University.
Those were precisely the women the new model was designed to identify. The research has been published in the European Journal of Endocrinology.
A blood sample already taken can reveal the condition
Severely reduced thyroid function during pregnancy is also associated with an increased risk of preterm birth and other pregnancy-related complications. The question was therefore whether these overlooked cases could be identified without introducing another test during pregnancy.
The researchers took advantage of the fact that a blood sample is already taken around weeks nine to ten as part of screening for foetal chromosomal abnormalities in Denmark. In a two-year study in the North Denmark Region, they used residual serum from these samples to investigate whether all pregnant women could be tested for hypothyroidism without another blood draw.
In the laboratory, they first measured thyroid-stimulating hormone (TSH), which rises when the thyroid gland does not produce enough hormone. The threshold was set at a TSH level above 6 mIU/L. In women above this threshold, they also measured the thyroid hormone fT4 to assess how pronounced the hypothyroidism was. Only women with no known thyroid disease and clearly abnormal results were referred back to their general practitioner for prompt assessment and treatment.
The point was to test everyone but only act on findings with clear clinical significance. This is similar to newborn screening with the heel-prick test, in which only serious findings lead to further investigation.
“There is a grey zone of minor changes in thyroid function in pregnant women that do not necessarily reflect actual thyroid disease. That is why, with this model, we specifically identify the cases in which thyroid disease is present and treatment is needed,” explains Stine Linding Andersen.
The risk criteria could not identify every woman
A total of 11,841 blood samples from 10,403 pregnant women were analysed. Among them, the model identified precisely the group that the risk-based approach struggles to pick up.
In 41 women – 0.39%, or about four in every 1,000 – the tests revealed previously undetected hypothyroidism. In some, the reduced thyroid function was pronounced, whereas others had a milder form. The number was in line with expectations based on previous studies.
All 41 women also had antibodies directed against the thyroid gland. This supports the conclusion that the results were not merely random fluctuations but were linked to autoimmune disease, in which the immune system attacks the thyroid gland and can gradually impair its production of thyroid hormones.
“This means that we only need to measure a single marker, namely TSH, to identify these women. This makes the procedure relatively simple and less costly, which is important if the model is ever to be used on a large scale,” says Stine Linding Andersen.
Even perfect risk-based screening would have missed one in three
One obvious question was whether the women had simply been overlooked because the existing recommendations are not always followed. An earlier study had already suggested this: around one in five of the women whose thyroid function had not been measured had at least one risk factor that should have prompted testing. International studies have also shown that even correctly implemented risk-based screening can miss women with abnormal thyroid function.
But the new study shows that the problem goes deeper.
Of the 41 women identified through universal screening, 14 did not meet any of the current risk criteria. The problem was therefore not simply whether the recommendations were being followed. Even if they had been followed perfectly, just over one in three would still have been missed.
“When we see that a third of these women would otherwise go undetected, it is worth considering whether we can do better than we do today. This is still a research project, and we have only carried it out in one region of Denmark, so we need to test it in more places and see whether the model works as intended,” says Stine Linding Andersen.
The study therefore shows that the model can be implemented in practice – and that it identifies women whom the risk criteria cannot pick out. What the study does not yet show is whether universal screening also leads to fewer pregnancy complications or better outcomes for children. Answering that question will require larger studies.
“But we have managed to identify these women and treat them early in pregnancy, and if we can measure a single thyroid marker in blood samples that are already being taken, we are talking about a relatively simple solution that could benefit both mother and child,” she says.
