Subclinical hypothyroidism in early childhood: a frequent outcome of transient neonatal hyperthyrotropinemia.

Newborns with high TSH at birth and with normal free T(4) and normal or slightly elevated TSH at the confirmatory examination are considered false positive for congenital hypothyroidism. We evaluated thyroid function, thyroid antibodies, thyroid volume and morphology, thyroperoxidase and TSH receptor genes, and auxological data in 56 false positive children at 16-44 months of age. In these children thyroid function at confirmatory examination was fully normal in 33 (TSH, 0.8-4.9 mU/liter; group I) and nearly normal (borderline elevated TSH, 5.0-11.7 mU/liter) in the other 23 (group II). Compared with 65 control children with normal TSH at birth, false positive children had significantly higher basal serum TSH (mean +/- SD, 4.38 +/- 2.2 vs. 1.4 +/- 0.8 mU/liter; P < 0.01). Subclinical hypothyroidism, indicated by increased basal TSH and/or increased TSH response to TRH, was present in 36% children in group I and 70% in group II. Free T(4) was within the normal range in all children. Compared with the control group, false positive children had significantly higher free T(3) values (4.9 +/- 0.8 vs. 3.7 +/- 1.0 pmol/liter; P < 0.01) and a higher prevalence of antithyroid antibodies (25% vs. 1.5%; P < 0.001). Frequent thyroid morphology abnormalities and frequent thyroperoxidase and TSH receptor gene sequence variations were also observed. In conclusion, newborns classified false positive at congenital hypothyroidism screening have a very high risk of subclinical hypothyroidism in infancy and early childhood.