Symptomatic hypothyroxinemia with normal TSH levels in preterm infants

Thyroid Function in Preterm/Low Birth Weight Infants: Impact on Diagnosis and Management of Thyroid Dysfunction

Maternal thyroid hormone crosses the placenta to the fetus beginning in the first trimester, likely playing an important role in fetal development. The fetal thyroid gland begins to produce thyroid hormone in the second trimester, with fetal serum T4 levels gradually rising to term. Full maturation of the hypothalamic-pituitary-thyroid (HPT) axis does not occur until term gestation or the early neonatal period. Postnatal thyroid function in preterm babies is qualitatively similar to term infants, but the TSH surge is reduced, with a corresponding decrease in the rise in T4 and T3 levels. Serum T4 levels are reduced in proportion to the degree of prematurity, representing both loss of the maternal contribution and immaturity of the HPT axis. Other factors, such as neonatal drugs, e.g., dopamine, and non-thyroidal illness syndrome (NTIS) related to co-morbidities contribute to the "hypothyroxinemia of prematurity". Iodine, both deficiency and excess, may impact thyroid function in infants born preterm. Overall, the incidence of permanent congenital hypothyroidism in preterm infants appears to be similar to term infants. However, in newborn screening (NBS) that employ a total T4-reflex TSH test approach, a higher proportion of preterm babies will have a T4 below the cutoff, associated with a non-elevated TSH level. In NBS programs with a primary TSH test combined with serial testing, there is a relatively high incidence of "delayed TSH elevation" in preterm neonates. On follow-up, the majority of these cases have transient hypothyroidism. Preterm/LBW infants have many clinical manifestations that might be ascribed to hypothyroidism. The question then arises whether the hypothyroxinemia of prematurity, with thyroid function tests compatible with either non-thyroidal illness syndrome or central hypothyroidism, is a physiologic or pathologic process. In particular, does hypothyroxinemia contribute to the neurodevelopmental impairment common to preterm infants? Results from multiple studies are mixed, with some randomized controlled trials in the most preterm infants born <28 weeks gestation appearing to show benefit. This review will summarize fetal and neonatal thyroid physiology, thyroid disorders specific to preterm/LBW infants and their impact on NBS for congenital hypothyroidism, examine treatment studies, and finish with comments on unresolved questions and areas of controversy.

Thyroid dysfunctions of prematurity and their impacts on neurodevelopmental outcome

BACKGROUND

Thyroid dysfunction is very common and is associated with neurodevelopmental impairments in preterm infants.

OBJECTIVES

This study was conducted to determine the incidence and natural course of various thyroid dysfunctions and their impacts on neurodevelopmental outcomes among premature infants.

METHODS

A total of 177 infants were enrolled who were born at <34 weeks or whose birth weight was <1500 g and who underwent repeat thyroid function tests. We analyzed how various thyroid dysfunctions affected neurodevelopmental outcomes at 18 months of corrected age.

RESULTS

Thyroid dysfunction was noted in 88 infants. Hypothyroxinemia was observed in 23 infants, and their thyroid function was influenced by variable clinical factors. Free T4 levels were all normalized without thyroxine medication, and neurodevelopmental outcomes were not affected. In contrast, hyperthyrotropinemia was not associated with other clinical factors. Among 58 subjects who had hyperthyrotropinemia, only 31 infants showed normal thyroid-stimulating hormone (TSH) levels at follow-up tests. The remaining 27 infants had persistently high TSH levels, which significantly and poorly influenced the neurodevelopmental outcomes.

CONCLUSIONS

Thyroid dysfunction is common among preterm infants. With the exception of persistent hyperthyrotropinemia, it generally does not affect neurodevelopmental outcomes. However, the beneficial effects of thyroid hormone therapy in patients with persistent hyperthyrotropinemia merits further study.

Thyroid function in the preterm infant

Thyroid gland function develops and matures during fetal life, with production of serum thyroxine (T4) concentrations beginning around 12 weeks gestation and increasing to term. Infants born prior to term have lower cord serum T4 concentrations that correlate with gestational age or birth weight. This is partially the result of lower thyroxine-binding globulin (TBG) concentrations. The cord serum free thyroxine (FT4) concentrations also correlate with gestational age, but they are not proportionately as low as the cord T4 concentration. Preterm infants have a postnatal thyrotropin (TSH) surge and rise in serum T4 and triiodothyronine (T3), which is qualitatively similar to, but quantitatively smaller than, term infants. In contrast to term infants, preterm infants often experience a fall in serum T4 and T3 in the first week of life to below birth levels. This drop appears to be the result of many factors, including nutritional problems and decreased hepatic TBG production, immaturity of hypothalamic-pituitary control of the thyroid gland, immaturity of the thyroid gland itself, and increased tissue utilization of T4. These changes are impacted by complications of prematurity, such as respiratory distress syndrome (RDS), which result in nonthyroidal illness-like changes. Again, serum FT4 seems less affected, and when measured by equilibrium dialysis may be in the normal range for age. Several studies have correlated different measures of morbidity and mortality in the preterm infant with lower serum T4 concentrations. However, as with adults, it may be that low serum T4 concentrations are a marker of the sickest preemies. Also, as with adults, this has led to speculation that T4 treatment might be beneficial in improving these complications of prematurity, in particular the neurological outcome. While some studies appear to show improvement in some facet of medical complications with T4 treatment, most show no effect. Regarding neurological outcome, the 2 best controlled trials do not show improvement in neuropsychiatric testing outcome assessed up to 2 years of age. One study, however, showed an IQ that was 18 points higher in the T4-treated subgroup less than 27 weeks gestational age. It may be that the most preterm infants, eg, those less than 27 weeks of age, are at a disadvantage compared with their intrauterine counterparts, in that they lack the maternal thyroid hormone contribution and are forced to adapt to extrauterine life before their hypothalamic-pituitary-thyroid axis is mature enough to deal with tissue thyroxine demands. Further controlled studies are needed to determine if this subgroup of infants indeed may benefit from transient thyroid hormone supplementation.

The relation of transient hypothyroxinemia in preterm infants to neurologic development at two years of age

BACKGROUND

Transient hypothyroxinemia, a common finding in premature infants, is not thought to have long-term sequelae or to require treatment. We investigated whether hypothyroxinemia in premature infants is a cause of subsequent motor and cognitive abnormalities.

METHODS

In this historical cohort study, we retrieved blood thyroxine values, obtained on routine screening in the first week of life, from state screening records on children who weighted 2000 g or less at birth, who were born at 33 weeks' gestation or earlier, and who were enrolled in a population-based study of the late sequelae of neonatal brain hemorrhage. We investigated the relation of these values to the odds for disabling cerebral palsy among 463 subjects for whom data were available and to the mental-development score on the Bayley Scales of Infant Development or the Stanford-Binet Intelligence Scales for Children at the age of two years in 400 subjects. The effects of severe hypothyroxinemia, defined as a blood thyroxine value more than 2.6 SD below the mean for New Jersey newborns, were assessed before and after adjustment for gestational age and potentially confounding variables.

RESULTS

In analyses adjusted for gestational age, infants with severe hypothyroxinemia had a risk of disabling cerebral palsy that was nearly 11 times that of infants without hypothyroxinemia (odds ratio, 10.8; 95 percent confidence interval, 3.0 to 39.3) and a mean mental-development score at the age of two that was 15.4 points lower (95 percent confidence interval, 8.1 to 22.6 points) than the mean score of children with normal neonatal blood thyroxine concentrations. After adjustment for gestational age and multiple prenatal, perinatal, and early and last neonatal variables, severe hypothyroxinemia was still associated with an increased risk of disabling cerebral palsy (odds ratio, 4.4; 95 percent confidence interval, 1.0 to 18.6) and a reduction of nearly 7 points (95 percent confidence interval, 0.3 to 13.2 points) in the mental-development score.

CONCLUSIONS

Severe hypothyroxinemia in preterm infants may be an important cause of problems in neurologic and mental development detected at the age of two years.

Low thyroxinaemia occurs in the majority of very preterm newborns

Transient hypothyroxinaemia with normal thyroid stimulating hormone (TSH) levels is a well-known condition in preterm neonates and is generally assumed to be a harmless epiphenomenon of prematurity. This assumption is, however, based on studies that included very few neonates with a gestational age (GA) below 30 weeks. We therefore measured serum free thyroxine (FT4) and serum TSH on days 1 and 14 in 263 neonates with a GA between 26 and 41 weeks. In 13 infants (5%), transient hypothyroidism (low FT4 and TSH >20 mU/l on day 14) was found. In the remaining 250 patients FT4 on days 1 and 14 but not TSH correlated positively with GA. In neonates with a GA of 35-41 weeks, FT4 increased postnatally to levels within or above the normal adult range. In contrast, in the very preterm group (26-31 weeks) the already low FT4 levels declined to values significantly below the range observed in term neonates. A significant proportion of these neonates had FT4 levels within the hypothyroid range. There was no difference in thyroid function between neonates treated with povidone-iodine or chlorhexidine.

CONCLUSION

Very preterm neonates have FT4 levels on day 14 that are much lower than is generally assumed while TSH remains in the normal range. We therefore propose to measure FT4 in all preterms with a GA below 33 weeks, during the 2nd week of life.