Thyroid Function Throughout Pregnancy in Women With Subclinical Hypothyroidism or Hypothyroxinemia

OBJECTIVE: To evaluate thyroid status throughout pregnancy among women identified as having subclinical hypothyroidism or hypothyroxinemia before 20 weeks of gestation. METHODS: This is a secondary analysis of two parallel, multicenter, placebo-controlled randomized clinical trials in which participants with subclinical hypothyroidism (elevated thyroid-stimulating hormone [TSH], normal free thyroxine [FT4]) or hypothyroxinemia (normal TSH, low FT4) were identified before 20 weeks of gestation to evaluate whether treatment with levothyroxine would improve offspring intelligence at age 5. Thyroid function testing was performed monthly, and participants returned 1 year after delivery for follow-up testing. For this analysis, we included only participants in the placebo groups. Follow-up thyroid function tests were analyzed separately for the subclinical hypothyroidism and hypothyroxinemia groups. The Cochran-Armitage trend test in linear proportions was used to examine associations of subsequent test results with baseline thyroid peroxidase antibody status. Multivariate Cox regression modeling was used to examine time to first normal thyroid function test result and the association with thyroid peroxidase antibody status and other characteristics. RESULTS: A total of 326 of 338 (96.4%) participants in the subclinical hypothyroidism group and 254 of 261 (97.3%) in the hypothyroxinemia group underwent monthly testing. Most participants in both groups had at least one subsequent normal thyroid function test result in pregnancy (88.7% of participants in the subclinical hypothyroidism group and 61.4% of participants in the hypothyroxinemia group, P<.001). Development of overt hypothyroidism during the pregnancy was uncommon in either group (8/326 [2.5%] and 5/254 [2.0%] for subclinical hypothyroidism and hypothyroxinemia groups, respectively). There was no correlation between thyroid peroxidase antibody status and subsequent test results for either group (Ptrend=.18 and 1.0 for the subclinical hypothyroidism and hypothyroxinemia groups, respectively). CONCLUSION: Most untreated pregnant women who initially test positive for subclinical hypothyroidism or hypothyroxinemia do not have sustained abnormal results throughout the remainder of the pregnancy. Strategies to screen and treat these women in early pregnancy may result in the treatment of patients whose test results would spontaneously revert to normal. CLINICAL TRIAL REGISTRATION SECONDARY ANALYSIS: ClinicalTrials.gov, NCT00388297.