Rochester, MN—Pregnant women with subclinical hypothyroidism have faced a difficult dilemma: On one hand, the condition can cause a number of health problems if left untreated. Yet, levothyroxine, the drug commonly prescribed for mildly underactive thyroids, has been associated with pregnancy-related adverse outcomes.
A new study published in The BMJ offers evidence that lack of treatment can have worse consequences than previously observed, including pregnancy loss.
The study’s results, along with recently released guidelines, are making the path clearer, according to a study team led by Mayo Clinic researchers. They point out that pregnant women typically produce a lower level than normal of thyroid-stimulating hormone (TSH)—0.4 to 4.0 milli-international units per liter (mIU/L).
When their TSH levels rise above that level, expectant mothers might experience subclinical hypothyroidism, and some international guidelines recommend levels be no higher than 2.5 to 3.0 mIU/L during pregnancy.
The new study finds that treating subclinical hypothyroidism, which is below the level that would require medication in nonpregnant women, can reduce pregnancy loss, especially for those with TSH levels on the upper end of normal or higher. Doing that could have a positive effect on as many as 15% of pregnancies, the researchers point out.
“A recent analysis of 18 studies showed that pregnant women with untreated subclinical hypothyroidism are at higher risk for pregnancy loss, placental abruption, premature rupture of membranes, and neonatal death,” explained lead author Spyridoula Maraka, MD.
“It seemed likely that treating subclinical hypothyroidism would reduce the chance of these deadly occurrences. But we know that treatment brings other risks, so we wanted to find the point at which benefits outweighed risks.”
Using the OptumLabs Data Warehouse, the researchers analyzed health information on 5,405 pregnant women diagnosed with subclinical hypothyroidism. Of these, 843 women, with an average pretreatment TSH concentration of 4.8 mIU/L, were treated with thyroid hormone, while the remaining 4,562, with an average pretreatment TSH concentration of 3.3 mIU/L, were left untreated.
Results indicate that, compared with the untreated group, women who received medication were 38% less likely to experience pregnancy loss. On the other hand, they were more likely to have a preterm delivery, or experience gestational diabetes or preeclampsia.
In light of that, the team examined the levels of pretreatment TSH to determine if there was a point when treatment would have the greatest benefit compared to risks.
“Unsurprisingly, we found that women with higher levels of pre-treatment TSH—between 4.1 and 10 milli-international units per liter—benefitted most from treatment,” Maraka noted.“ This group’s much lower likelihood of experiencing pregnancy loss was what brought the average down—and creates a good argument for updated clinical guidelines.”
Essentially, when treated, women with lower levels of TSH (2.5-4.0 mIU/L) were found to have a higher risk of gestational hypertension, which can lead to preeclampsia, than for untreated women. No difference was detected between treated women who had higher levels of TSH—i.e., 4.1 to 10 mIU/L—and untreated women.
“Our findings lead us to believe that overtreatment could be possible,” said co-author Juan Brito Campana, MBBS, “If the TSH levels are in that 2.5-4.0 [milli-international units per liter] range, it may be best to leave subclinical hypothyroidism untreated.”
Brito Campana emphasized that the association of levothyroxine therapy and the risk of pregnancy-related adverse outcomes such as preeclampsia or gestational diabetes should be seen as preliminary findings and called for additional studies evaluating the safety of levothyroxine therapy in pregnant women with subclinical hypothyroidism.
In January, the American Thyroid Association published updated care guidelines with treatment recommendations for women experiencing thyroid disease during pregnancy.
« Click here to return to Weekly News Update.