Postnatal thyroid hormones for preterm infants with transient hypothyroxinaemia (Review)

David A. Osborn, R. W. Hunt

Research output: Contribution to journalReview ArticleResearchpeer-review

49 Citations (Scopus)


Background: Extremely premature infants are at risk of transient hypothyroxinaemia in the first weeks after birth. These low thyroid hormone levels are associated with an increased incidence of neonatal morbidity, mortality and longer term developmental impairments. Thyroid hormone therapy might prevent these problems. Objectives: To determine the evidence for thyroid hormone therapy in preterm infants with transient hypothyroxinaemia (low thyroid hormone level, normal TSH) for improvement of neonatal outcomes and neurodevelopment. Search strategy: Searches were performed of The Cochrane Central Register of Controlled (CENTRAL, The Cochrane Library, Issue 1, 2006), MEDLINE (1966 - March 2006), PREMEDLINE (March 2006), EMBASE (1980 - March 2006), previous reviews including cross references, abstracts and conference proceedings, supplemented by requests to expert informants. Selection criteria: Trials enrolling preterm infants with transient hypothyroxinaemia (low thyroid hormone level, normal TSH level) in the neonatal period, using random or quasi-random patient allocation to thyroid hormone therapy compared to control (placebo or no treatment). Data collection and analysis: Independent assessment of trial quality and data extraction by each review author. Synthesis of data using relative risk (RR) and weighted mean difference (WMD) using standard methods of the Cochrane Collaboration and its Neonatal Review Group. Main results: Only one study was eligible. Chowdhry (1984) enrolled 23 infants < 1250 g and 25 - 28 weeks gestation with transient hypothyroxinaemia (serum total T4 ≤ 4 μg/dl and TSH ≤ 20 IU/L). Infants were randomised to thyroxine 10 μg/kg/day or placebo beginning on day 15 and continuing daily for seven weeks. Chowdhry (1984) reported no neonatal mortality and one infant death in each group prior to discharge. No significant difference was reported in CLD at 28 days or 36 weeks, patent ductus arteriosus, necrotising enterocolitis, retinopathy or prematurity, weight gain, growth in head circumference or length. No significant difference was reported for mean T4 levels between thyroxine and placebo treated infants on day 21, 35, 49, 63 and 77 after birth. Free T4 was not measured. Neurodevelopmental follow up was inadequate to draw any conclusions from. Authors' conclusions: There is insufficient evidence to determine whether use of thyroid hormones for treatment of preterm infants with transient hypothyroxinaemia results in changes in neonatal morbidity and mortality, or reductions in neurodevelopmental impairments. Further research is required.

Original languageEnglish
Article numberCD005945
Number of pages18
JournalCochrane Database of Systematic Reviews
Issue number1
Publication statusPublished - 2007
Externally publishedYes


  • Humans
  • Infant, newborn
  • Infant, premature [*blood]
  • Thyroid diseases [blood; drug therapy]
  • Thyroid hormones [*therapeutic use]
  • Thyroxine [*blood]

Cite this