The negative impact of even subtle maternal thyroid hormone deficiency on the pregnancy outcome and intellectual development of the progeny has been known for many years, but unfortunately the diagnosis and treatment of subclinical hypothyroidism in pregnant women still evokes controversies. Due to physiological changes in thyroid function and thyroid hormones metabolism during pregnancy, the trimester‑specific reference ranges for thyroid‑stimulating hormone (TSH) and free thyroid hormones should be established. However, because of interassay variability and other confounders including ethnicity and iodine intake, such norms are reliable only for local populations and a specific laboratory method. In turn, the fixed reference ranges suggested by endocrine societies may carry a risk of misclassificating some healthy pregnant women to be hypothyroid. The effect of levothyroxine treatment on pregnancy and children's cognitive outcomes remains unclear. Therapeutic benefits in decreasing miscarriage and preterm delivery rates were observed when intervention was held in the first trimester in women with a TSH level between 2.5 to 10 mU/l, mainly higher than or equal to 4 mU/l. The possible harmful effect of treatment includes preterm delivery, gestational diabetes, hypertension, and pre‑eclampsia. The only 3 prospective, randomized, placebo‑controlled trials evaluating the efficacy of levothyroxine therapy on children's intelligence quotient were started in the second trimester, which may be too late to demonstrate differences between treatment and placebo groups. Awaiting the results of future trials, clinicians should be aware of the fact that low‑dose levothyroxine at a daily dose of 25 to 50 µg is probably not harmful and may be beneficial, but the routine implementation of the therapy in each pregnant women with a TSH level exceeding 2.5 mU/l seems too premature.