Background: Low thyroid hormone concentrations in the first few weeks of life in preterm infants may be linked with poor neurodevelopment. We conducted a multi-centred randomised controlled trial of thyroxine (T4) supplementation in babies born under 28-weeks gestation (TIPIT study). A post hoc subgroup analysis was undertaken to examine brain growth and development.
Methods: Seventy-eight infants received T4 supplementation and 75 received placebo. There was no difference in outcome between these groups. Longitudinal free T4 (FT4) plasma levels were measured weekly during the initial 4 weeks after birth. Comparisons were made between the 2 subgroups with FT4 values in the lowest and highest quartiles. A multivariate analysis of independent factors affecting width of subarachnoid space as a measure of brain volume at 36 weeks CGA was undertaken. In addition, 38 babies had MRI brain scans at term equivalent using diffusion tensor imaging (DTI). DTI enables quantitative assessment of white matter development, before myelination is visible on conventional MRI, by measurement of apparent diffusion coefficient (ADC).
Results: Among placebo infants, the lowest quartile of FT4 was associated with lower gestational age, lower birth weight and larger subarachnoid space (P=0.04) at 36 weeks CGA. These infants also had a higher mortality compared with those infants in the highest quartile (P=0.01). These associations were not seen in infants given supplemental T4. Among scanned infants ADC values in posterior corpus callosum and posterior limb of the internal capsule were lower among babies in the highest quartile of FT4.
Conclusion: Our findings suggest that longitudinal FT4 levels in the first 4 weeks of life are an independent factor affecting brain growth at 36 weeks CGA. A lower ADC suggests myelination was more advanced in the highest quartile of FT4 compared to the lowest quartile.