The Hypothyroxinemia of Prematurity

Gabriella Morreale de Escobar

Unidad de Endocrinología Molecular Instituto de Investigaciones Biomédicas 28029-Madrid, Spain Susana Ares Servicio de Neonatología Hospital Infantil La Paz 28046-Madrid, Spain

Dr. Delbert A. Fisher introduces his careful editorial on this subject, in the June issue of JCEM (1), with a summary of known alterations of thyroid function in premature newborns that supports the idea that "for premature infants the stresses of extrauterine transition are superimposed on an immature thyroid axis ... Thus the premature infant in the extrauterine environment, like the intrauterine fetus of comparable gestational age (GA), is relatively hypothyroxinemic with normal-to-low levels of serum TSH and T3 concentrations ... The prevailing view has been that the hypothyroxinemia of prematurity is physiologic, not requiring hormone supplementation."

After reviewing recent data of several lines of evidence suggesting that hypothyroxinemia during this premature period of development has a negative impact on later development, especially of the brain, the author concludes that the prevailing view quoted above is still valid.

We wish to question the statement that findings in the preterm neonates in the extra uterine environment can be extrapolated to the intrauterine fetus of comparable gestational age (GA). In the study by Ares et al. (2) appearing in the same issue, the free T4 (FT4) concentrations found in the premature infants were much lower than those reported for the intrauterine fetus at a comparable developmental stage by Thorpe-Beeston et al. (3), as shown in the upper panel of Fig. 1Go.Although Ares et al. (2) also indicated that the circulating levels of TSH of both preterm and term neonates were below the confidence intervals for the fetal TSH levels reported by Thorpe-Beeston et al. (3), the actual data were not shown. They now appear in the bottom panel of Fig. 1Go. As may be seen, except for the values at the earliest postmenstrual age, which were obtained within the first week after birth, and therefore likely to include the postnatal TSH surge, all other TSH values were lower than reported in utero. As already stressed by Thorpe-Beeston et al. (3), TSH is positively correlated with FT4 in the fetus, independently of GA, and continues to increase until term, when adult FT4 levels have been reached. These intrauterine TSH levels are much higher than those of young term infants after the postnatal TSH surge is over, or those of euthyroid adults.



View larger version (43K):
[in this window]
[in a new window]
 
Figure 1. Mean (± SEM) concentrations of FT4 and TSH corresponding to preterm and term neonates as a function of postmenstrual age are superimposed on data (open circles) published for fetuses in utero by Thorpe-Beeston et al. (3). The shaded areas correspond to the 95% confidence intervals for the fetal data.

 
Both observations indicate that the intrauterine fetus is developing in an environment quite different from that prevailing after premature birth, when left to its own resources. During intrauterine development fetal tissues are exposed to higher circulating FT4 and TSH levels than those available to the preterm neonate, thus overcoming the consequence of the developmental fetal hypothalamic-pituitary-thyroid immaturity. However, once the vascular connections between mother, placenta, and fetus are severed by premature birth, the consequences of this immaturity become apparent in the neonates. Their hypothyroxinemia is not "physiological," as in the truly physiological (intrauterine) condition their FT4 would have been higher and their high intrauterine circulating TSH would promote thyroid maturation. This latter effect would also be delayed after premature birth, as TSH levels drop markedly.

In the editorial (1), the very low FT4 levels found in premature neonates of the Madrid study (2) are questioned because they were not obtained by an equilibrium dialysis method. Neither were those reported by Thorpe-Beeston et al. (3), and we believe the comparison appearing in Fig. 1Go is valid, as the values obtained in the older fetuses (3) and in our term neonates (2) were comparable. Although the reference values recently reported in premature infants from California using a direct dialysis method are higher (4) than those of the Madrid study or those reported by Thorpe-Beeston for older fetuses (3), it should be pointed out that the reference samples submitted to the dialysis method were obtained quite soon after birth (after approximately 3.5 ± 2.5 days), when the surge in FT4 following parturition might still affect the findings, a possibility strongly supported by the very high TSH found in the same samples. The low FT4 levels from the Madrid study are influenced both by postmenstrual age and iodine intake, independently. By multiple regression analysis of the standardized variables (z-scores) it was found that two thirds of FT4 value was related to age and one third to the iodine intake. Even if the lowering effect of the iodine deficiency were taken into account, circulating FT4 of the premature infants remained lower than that of fetuses of comparable age. The marked differences in the circulating TSH of fetuses and preterm infants of comparable GA are not easily accounted for on the basis of methodological differences.

The other important point reviewed in Dr. Fisher’s editorial is whether or not the hypothyroxinemic premature infants would benefit from hormone replacement therapy. Evidence reviewed by Fisher (1) and Ares et al. (2) suggests that the critical period of central nervous system dependency on thyroid hormones in humans extends to intrauterine fetal life and to the extra uterine premature infant. Two double-blind studies have addressed this question, namely, those reported by Chowdry et al. (5) and van Wassenaer et al. (6). The first one was limited to a small number of infants and did not detect any beneficial effect of treatment with thyroxine (10–15 µg/kg per day) on neurodevelopment, as measured at 1 yr of age in 5 T4-treated as compared with 3 placebo-treated infants. The second study is based on 100 premature babies of less than 30 GA, who were treated daily with 8 µg/kg of birth weight for 6 weeks, and 100 placebo-treated controls. The latter study showed several important points, the first one being that treatment was not harmful. Although neurodevelopment, as measured at 2 yr of age, was not improved for the cohort of treated infants taken as a whole, it had a clear beneficial effect for those born at less than 27 weeks GA, whose IQ scored 18 points higher than that of placebo-treated infants (93 ± 26, compared with 75 ± 23, P < 0.01). These subgroups consisted of 13 T4-treated versus 18 placebo-treated infants. Van Wassenaer et al. (7) have also recently reported that mortality and morbidity of the preterm neonates born from mothers who had not received antenatal steroid treatment (32 T4-treated vs. 35 placebo-treated neonates) were significantly improved: 14 (40%) died in the placebo-treated group, compared with only 5 (16%) in the T4-treated group. Decreased mortality of early preterm neonates had also been reported by Schönberger et al. (8), who treated 45 infants daily with 25 µg + 5 µg T3 given orally: 7 (3%) died, as compared with 16 out of 55 (29%) untreated infants. They later reported a 9.5% mortality rate for the successive 126 prematures, born weighing less than 2200 g or before 37 weeks gestation, who received the treatment (10). Considering the publication dates, it is likely that the premature infants from these studies (9, 10) were born to mothers who had not received antenatal steroid treatment. Although the numbers are small, results appear to be of sufficient clinical relevance to encourage more attempts at treating the neonatal hypothyroxinemia of prematurity.

We again wish to point out that a low FT4 level is not the only main difference between the thyroid hormone status of the preterm infant and a fetus in utero at a comparable developmental age. None of the treatment schedules used so far have taken into consideration that TSH is higher in the fetus. The doses of T4 that have been used actually depressed circulating TSH even below the levels found in nontreated preterm infants (6). Possibly as a result of this, the thyroidal secretion of T3 would decrease. This would occur during a developmental period, when type I 5' iodothyronine deiodinases (5'D-I) are still low. Treatment with T4 also resulted in low circulating T3, and this would further compromise the T3-dependent expression of 5'D-I. Treatment schedules that would supply the thyroid hormone requirements of the premature infant without delaying thyroidal and 5'D-I maturation might improve the encouraging developmental results obtained so far.

In summary, we do not question that the hypothyroxinemia of the premature infant reflects an inherent hypothalamic-pituitary-thyroid immaturity. We do disagree, however, with the idea that this is a physiological condition for the premature neonate, because the intrauterine environment to which it would be exposed if still in utero would avoid the hypothyroxinemic state and, as a result of the high circulating TSH levels, would promote thyroidal maturation. The consequences of these contrasting points of view are quite different, with the first discouraging, and the latter encouraging therapeutic measures to avoid this period of hypothyroxinemia, which has been shown to be causally related to poor neurodevelopment (10, 11).

Footnotes

Received October 16, 1997. Address correspondence to: Gabriella Morreale de Escobar, Instituto de Investigaciones Biomédicas, Facultad de Medicina, UAM, Arzobispo Marcillo 4, 28029 Madrid, Spain.

References

  1. Fisher DA. 1997 The hypothyroxinemia of prematurity. J Clin Endocrinol Metab. 82:1701–1703.[Free Full Text]
  2. Ares S, Escobar-Morreale H, Quero J, Durán S, Presas MJ, Herruzo R, Morreale de Escobar G. 1997 Neonatal hypothyroxinemia: Effects of iodine intake and premature birth. J Clin Endocrinol Metab. 82:1704–1712.[Abstract/Free Full Text]
  3. Thorpe-Beeston JG, Nicolaides KH, Felton CV, Butler J, McGregor AM. 1991 Maturation of the secretion of thyroid hormone and thyroid stimulating hormone in the fetus. N Engl J Med. 324:532–536.[Abstract]
  4. Adams LM, Emery JR, Clark SG, Carlton EL, Nelson JC. 1995 Reference ranges for newer thyroid function tests in premature infants. J Pediatr. 126:122–127.[Medline]
  5. Chowdry P, Scanlon JW, Auerback R, Abbassi V. 1984 Results of controlled double-blind study of thyroid replacement in very low birth weight premature infants with hypothyroxinemia. Pediatrics. 73:301–305.[Abstract]
  6. van Wassenaer A, Kok JH, Briet JM, Smit BJ, Tamminga P, van Baar A, Dekker FW, Vulsma T. 1997 Effects of thyroxine supplementation on neurologic development in infants born at less than 30 weeks’ gestation. N Engl J Med. 336:21–26.[Abstract/Free Full Text]
  7. van Wassenaer A, Kok JH. 1997 Thyroxine supplementation to infants <30 weeks’ gestation may reduce mortality in the absence of antenatal steroids. 24th Meeting of the European Thyroid Association, Münich, 1997, J Clin Invest. 20:93 (Abstract 185).
  8. Schönberger W, Grimm W, Emmrich P, Gempp W. 1979 Thyroid administration lowers mortality in premature infants. Lancet. 2:1181.[CrossRef]
  9. Schönberger W, Grimm W, Emmrich P, Gempp W. 1981 Reduction of mortality rate in premature infants by substitution of thyroid hormones. Eur J Pediatr. 135:245–253.[Medline]
  10. den Ouden AL, Kok JH, Verkerk PH, Brand R, Verloove-Vanhorick SP. 1996 The relation between neonatal thyroxine levels and neurodevelopmental outcome at 5 and 9 years in a national cohort of very preterm and/or very low birth weight infants. Pediatr Res. 39:142–145.[Abstract]
  11. Reuss ML, Panetti N, Pinto-Martin JA, Lorenz JM, Susser M. 1996 The relation of transient hypothyroxinemia in preterm infants to neurologic development at two years of age. N Engl J Med. 334:821–826.[Abstract/Free Full Text]