Initial and delayed thyroid-stimulating hormone elevation in extremely low-birth-weight infants

BACKGROUND: To determine the incidence, etiology, and outcomes of thyroid-stimulating hormone (TSH) elevation in extremely low-birth-weight infants (ELBWIs). METHODS: Newborn thyroid screening data of 584 ELBWIs (birth weight, < 1000 g; gestational age, ≥ 23 weeks) were retrospectively analyzed to identify initial (≤ 2 postnatal weeks) and delayed (> 2 weeks) TSH elevations. Growth and neurodevelopmental outcomes at 2 years’ corrected age (CA) were assessed according to levothyroxine replacement. RESULTS: Initial and delayed TSH elevations were detected at CAs of 27 and 30 weeks, respectively, with incidence rates of 0.9 and 7.2%, respectively. All infants with initial TSH elevations had perinatal asphyxia, and 95% of those with delayed TSH elevation were exposed to various stressors, including respiratory support, drugs, and surgery within 2 weeks before diagnosis of TSH elevation. Free thyroxine (T4) levels were simultaneously reduced in 80 and 57% of infants with initial and delayed TSH elevations, respectively. Both initial and delayed TSH elevations were transient, regardless of levothyroxine replacement. Infants receiving levothyroxine replacement therapy had significantly higher TSH elevations, significantly lower free T4 levels, and significantly reduced mortality, compared to untreated infants. However, levothyroxine replacement had no significant effect on long-term growth and neurodevelopmental outcomes. CONCLUSIONS: The timing of insult superimposition on hypothalamic–pituitary–thyroid axis maturation is a major determinant of initial or delayed TSH elevation in ELBWIs. Levothyroxine replacement did not affect growth or neurodevelopmental outcomes in this population.