Levels of neonatal thyroid hormone in preterm infants and neurodevelopmental outcome at 5 1/2 years: Millennium cohort study

Context: Transient hypothyroxinemia is the commonest thyroid dysfunction of premature infants, and recent studies have found adverse associations with neurodevelopment. The validity of these associations is unclear because the studies adjusted for a differing range of factors likely to influence neurodevelopment. Objective: The aim was to describe the association of transient hypothyroxinemia with neurodevelopment at 5.5 yr corrected age. Design: We conducted a follow-up study of a cohort of infants born in Scotland from 1999 to 2001 ≤34 wk gestation. Main Outcome Measures: We measured scores on the McCarthy scale adjusted for 26 influences of neurodevelopment including parental intellect, home environment, breast or formula fed, growth retardation, and use of postnatal drugs. Results: A total of 442 infants ≤34 wk gestation who had serum T₄ measurements on postnatal d 7, 14, or 28 and 100 term infants who had serum T₄ measured in cord blood were followed up at 5.5 yr. Infants with hypothyroxinemia (T₄ level ≤ 10th percentile on d 7, 14, or 28 corrected for gestational age) scored significantly lower than euthyroid infants (T₄ level greater than the 10th percentile and less than the 90th percentile on all days) on all McCarthy scales, except the quantitative. After adjustment for confounders of neurodevelopment, hypothyroxinemic infants scored significantly lower than euthyroid infants on the general cognitive and verbal scales. Conclusions: Our findings do not support the view that the hypothyroxinemic state, in the context of this analysis, is harmless in preterm infants. Many factors contribute both to the etiology of hypothyroxinemia and neurodevelopment; strategies for correction of hypothyroxinemia should acknowledge its complex etiology and not rely solely on one approach.