In a large randomized clinical trial, 952 euthyroid women (normal TSH level; range, 0.44 to 3.63 mIU/L and free thyroxine level; range, 10 to 21 pmol/L) who were planning on conceiving and had elevated thyroid peroxidase antibodies were randomized prior to conception to receive either thyroxine (50 µg) or placebo.7 After 12 months, outcomes were similar for women treated with thyroxine or placebo, including live birth rate (37.4% vs 37.9%), miscarriage rate for those who became pregnant (28.2% vs 29.6%), and preterm birth ≤ 34 weeks of gestation (3.8% vs 3.6%, respectively).7 The investigators concluded that the use of low-dose thyroxine in euthyroid women with thyroid peroxidase antibodies was not effective for increasing the rate of live birth or reducing the rate of miscarriage or early preterm birth.
Thyroid antibodies and the rate of IVF pregnancy and miscarriage
Some observational studies suggest that the presence of antithyroid antibodies may be associated with an increased rate of miscarriage.8 To test the effects of thyroxine treatment on the rate of miscarriage in euthyroid women with antithyroid antibodies, 600 euthyroid infertile women with antithyroid antibodies (antithyroid peroxidase levels ≥ 60 IU/mL) scheduled to have in vitro fertilization (IVF) were randomly assigned to receive thyroxine (dose adjustment to keep TSH levels in the range of 0.1 to 2.5 mIU/L) or no treatment.9 The thyroxine treatment was initiated 2 to 4 weeks before initiation of ovarian stimulation. In this study, treatment with thyroxine or no treatment resulted in similar rates of clinical pregnancy (35.7% vs 37.7%) and live birth (31.7% vs 32.3%).9 Among the women who achieved a clinical pregnancy, miscarriage rates were similar in the thyroxine and no treatment groups (10.3% vs 10.6%).9
Let’s focus on more serious problems that affect pregnancy
There is a clear consensus that women with overt hypothyroidism should be treated with thyroxine prior to attempting pregnancy.2,6 There is no clear consensus about how to treat women considering pregnancy who have one isolated laboratory finding, such as mild subclinical hypothyroidism, mild isolated hypothyroxinemia, or antithyroid antibodies. Given the lack of evidence from randomized trials that thyroxine improves pregnancy outcomes in these cases, obstetrician-gynecologists may want to either refer women with these problems to an endocrinologist for consultation or sequentially measure laboratory values to assess whether the patient’s laboratory abnormality is transient, stable, or worsening.
Obstetrician-gynecologists and their patients are confronted by many serious problems that adversely affect pregnancy and deserve priority attention, including iron deficiency anemia, excess gestational weight gain, peripartum depression, intimate partner violence, housing insecurity, cigarette smoking, substance misuse, chronic hypertension, morbid obesity, diabetes, gestational diabetes, preeclampsia, venous thromboembolism, obstetrical hemorrhage, sepsis, and infectious diseases. Given limited resources our expertise should be focused on these major obstetric public health problems rather than screening for mild subclinical hypothyroidism.