Thyroid hormones are essential for brain maturation. Very preterm infants, who are at risk of neurodevelopmental disabilities also have low thyroxine (T4) and free thyroxine (FT4) values in the first weeks after birth. This transient hypothyroxinemia may in part be causal to the neurodevelopmental problems. We have carried out a randomized, double-blind, placebo-controlled trial with T4 in 200 infants less than 30 weeks gestation. T4 (or placebo) was given in fixed dose of 8 microg/kg birth weight per day during the first 6 weeks after birth. It resulted in a significant increase of T4, FT4, and reverse triiodothyronine (rT3). Thyrotropin (TSH) secretion was suppressed, and, probably as a result of TSH suppression, triodothyronine (T3) levels were decreased in the T4 group. Mortality was 14% in the T4 group and 21% in the placebo group (NS). No effect was found on morbidity. Heart rate was significantly higher in T4-treated infants less than 28 weeks gestation, but not in T4-treated infants 28 weeks or more, who had the highest FT4 levels. In the study groups as a whole, no clear effect of T4 administration was found on neurodevelopmental outcome. However, there was a strong trend toward improvement of adverse outcome, defined as death or abnormal developmental outcome at 2 years of age. In addition, mental outcome in a subgroup of T4-treated infants less than 27 weeks' gestation was significantly better than in placebo infants of the same age group. In conclusion, this trial does not clearly have conclusive results. New trials of thyroid hormone treatment should be carried out in preterm infants, in order to investigate whether indeed T4 supplementation is required in preterm infants less than 27 or 28 weeks gestation. Addition of T3 to the treatment schedule needs to be considered.