Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

75A. Hypothyroidism in Infants and Children: Neonatal Screening

Guy van Vliet

Mass biochemical screening of newborn infants for congenital hypothyroidism became possible after the development of methods for assaying thyrotropin (TSH) and thyroxine (T4) in blood collected on filter paper in the 1970s [this type of blood sampling was originally introduced for the detection of phenylketonuria (1)]. The historical aspects of this important public health breakthrough have been recounted (2), and its results have been reviewed regularly (3,4). Screening for hypothyroidism among neonates has proven to be highly cost effective, because the societal burden of a person with mental deficiency caused by congenital hypothyroidism far exceeds the costs of the screening program (3).


Most infants with congenital hypothyroidism appear normal at birth, and the diagnosis is rarely suspected clinically in the first few weeks of life. At my institution, only 2 of 200 cases of congenital hypothyroidism diagnosed by screening since 1987 were suspected clinically. Yet, it is during the first few months if not weeks of life that irreversible brain damage occurs in infants with hypothyroidism who are not treated. Before biochemical screening of newborn infants for hypothyroidism was introduced, the mean IQ of children with congenital hypothyroidism was less than 80 (5), mainly because less than 20% of affected infants were diagnosed within 3 months after birth; even those with an IQ above 80 had deficits in fine motor control and learning disabilities (6).

When biochemical screening was implemented, it became clear that the disorder was present at birth, no matter how normal the infant looked. It has subsequently also become clear that most infants with hypothyroidism treated soon after birth with high doses of T4 can attain their full intellectual potential (7,8,9). However, some controversy remains as to whether the consequences of very severe congenital hypothyroidism can be entirely prevented (see section Congenital Hypothyroidism in this chapter).


Measurements of Thyroxine and Thyrotropin in Blood

Both T4 and TSH can be measured accurately in blood spotted on filter paper, the standard method for sample collection from newborn infants for screening tests. T4 refers to total T4, because free T4 cannot be measured reliably in blood spots (10). Initially, infants were screened by measurement of T4, with measurement of TSH as a second step if the blood-spot T4 value was below a certain cut-off point in the distribution of values measured that day (10), and this sequence is used now in many states in the United States and some European countries. In many places, however, TSH is measured first, with measurement of T4 as a second step if the blood-spot TSH value is only slightly elevated (Fig. 75A.1).


FIGURE 75A.1. Guidelines for biochemical screening for congenital hypothyroidism used in Quebec in 2003.

The T4-first approach misses some infants who have ectopic thyroid tissue that, when stimulated by TSH, produces enough T4 so that the blood T4concentration is above the cut-off value (3). On the other hand, this approach detects infants who have low blood-spot T4 values caused by prematurity, low serum thyroid hormone–binding proteins, central hypothyroidism, or severe nonthyroidal illness; these infants need second-step TSH measurements to exclude hypothyroidism (3). The TSH-first approach is designed to detect high blood-spot TSH values, and will miss infants with central hypothyroidism. Most programs using the latter approach consider a bloodspot TSH value of greater than or equal to 30 mU/L as indicative of hypothyroidism, and values of 15 to 29 mU/L as highly suspicious of it; all these infants must be studied further (Fig. 75.1). The TSH-first approach also has the advantage that it can be used to monitor the iodine intake of a population, because even borderline iodine deficiency increases TSH secretion in neonates (11).

Although most screening programs measure T4 or TSH in blood eluted from filter paper, a few use umbilical cord serum (12,13). One limitation of cord serum screening is that, in the case of twins discordant for congenital hypothyroidism, fetal blood mixing may result in a falsely normal serum TSH value in the hypothyroid twin (14).

Age at Screening for Congenital Hypothyroidism

As noted above, screening for congenital hypothyroidism started after screening for phenylketonuria, and this led to the use of the same filter paper blood spots, which were usually obtained several days after birth. As postnatal hospital stays have shortened, blood sampling has been done sooner after birth. TSH secretion rises abruptly immediately after delivery (see Chapter 74), and, although the surge usually lasts only hours, it results in blood-spot TSH values of 15 mU/L or higher in 2% to 3% of normal infants sampled within 24 hours after birth (15). Therefore, in Quebec, infants who have blood-spot TSH values of greater than or equal to 15 mU/L in a first sample obtained within 24 hours after birth are recalled for a second filter paper sample before any further action is taken.

Sample collection for screening may be delayed or overlooked in newborn infants who are severely ill (16), but there is no reason for sample collection to be delayed in any infant, and it should not be overlooked just because an infant is ill. Sick infants who are to be transferred should be tested before transfer, even if the infant is not 24 hours old.

Very-Low-Birth-Weight Newborn Infants

The amplitude of the neonatal surge in TSH secretion is less in very-low-birth-weight infants who are born prematurely than in term newborn infants, but most TSH-screening programs use the same blood-spot TSH value as a cut-off in both groups. This policy has not resulted in permanent congenital hypothyroidism being missed (16), indicating that in these very-low-birth-weight infants TSH secretion increases appropriately when thyroid secretion is decreased. It has been suggested that the increase in TSH secretion might be delayed in these infants, so that congenital hypothyroidism could be missed if only a single sample were taken soon after birth (17), but this must be rare, and multiple sampling is not routinely indicated. These infants often have low serum T
4 concentrations, due primarily to decreased serum thyroid hormone binding; repeated measurements of serum T4 are not indicated unless there is direct or indirect clinical evidence of hypothyroidism. Indeed, in a recent review of the large French database on congenital hypothyroidism, the disorder was associated with prolonged gestation and macrosomia, and not with low birth weight for gestational age (18). Furthermore, the lack of benefit of T4 administration in very-low-birth-weight infants (19) suggests that those who have hypothyroxinemia, but no increase in TSH secretion, have physiologic delay in maturation of the hypothalamic–pituitary–thyroid system rather than true central hypothyroidism.

Newborn Infants with Trisomy 21

As a group, newborn infants with trisomy 21 have slightly higher blood-spot TSH concentrations and slightly lower T4 concentrations than normal infants (20). At later ages, children with trisomy 21 have a propensity to develop autoimmune thyroid disease (see Chapter 75C), but congenital hypothyroidism is rare. There is not a single case of coexistence of permanent congenital hypothyroidism and of trisomy 21 in the Quebec database (21).


Infants found to have screening values indicative of hypothyroidism should be recalled immediately for further study. This is often done through the family's physician, but time is saved if the staff of the screening laboratory contacts the family directly and advises referral to a pediatric endocrinology center. Many screening programs have been able to achieve recall in 8 to 14 days, ensuring prompt evaluation and initiation of therapy when appropriate. On recall, serum TSH and free T4 should be measured. Other tests, for example, thyroid scinitigraphy, are useful as well, but the two serum measurements are critical to confirm the screening results. Thyroid scintigraphy is important because the demonstration of ectopically located thyroid tissue immediately establishes that hypothyroidism will be permanent regardless of the degree of TSH elevation on screening (see section Acquired Hypothyroidism in this chapter), and treatment should be started without waiting for the results of the serum studies. The same generally holds true when there is a goiter or when there is no detectable uptake on scintigraphy.

When scintigraphy shows a thyroid of normal shape, size, and position, the hypothyroidism may be transient. It is nevertheless prudent to begin T4 treatment immediately if the results of screening were frankly abnormal (blood-spot TSH concentration ≥30 mU/L), whereas treatment may be withheld pending the results of measurements of serum TSH and free T4 if the results of screening results were borderline (blood-spot TSH concentration 15 to 29 mU/L).

The prevalence and causes of transient primary congenital hypothyroidism vary according to the iodine intake of the population. In iodine-sufficient areas, the vast majority are infants born to mothers treated with an antithyroid drug or who have autoimmune thyroiditis. In areas of borderline or low iodine intake, they are most often infants exposed to acute iodine overload, whereas in areas of very low iodine intake they are a direct result of iodine deficiency compounded by thiocyanate overload (see section on iodine deficiency in Chapter 11 and Chapter 49).

False-negative results of blood-spot TSH screening are rare (3). Apart from the mixing of fetal blood between twins mentioned above, very few infants with blood-spot screening TSH values below the cut-off value have later proven to have severe hypothyroidism during infancy (at a time when it still has deleterious consequences for brain development). Most were infants with thyroid dyshormonogenesis in whom neonatal thyroid secretion was not severely impaired (16).

To minimize the risk for missing true congenital hypothyroidism while maintaining an acceptably low proportion of infants needing referral and retesting, most screening programs have adopted procedures for second-step screening, as described above and shown in Fig. 75.1. Nonetheless, clinical vigilance remains important, and serum TSH and free T4 should be measured in any infant with symptoms and signs suggestive of hypothyroidism, even if neonatal screening was reported to be normal.

The most dramatic result of biochemical screening for congenital hypothyroidism is that, because it allows treatment to be started much earlier than in infants diagnosed on the basis of clinical findings, it effectively prevents severe mental deficiency. Fig. 75A.2 shows the distribution of IQ values of children with congenital hypothyroidism in the prescreening era and the results in children who, thanks to screening, were treated at a median age of 14 days after delivery. The impact of disease severity, age at diagnosis, and treatment factors on developmental outcome are discussed in detail in the next section of this chapter.


FIGURE 75A.2. IQ values in children with congenital hypothyroidism in the past (pre-screening era) and present (screening era). Left panel: individual IQ values of children aged 3 to 17 years with congenital hypothyroidism diagnosed clinically in the pre-screening era, as a function of bone age at diagnosis and age at onset of therapy. There was no control group in that study; the theoretical population mean of 100 is indicated by the dotted line. Note that IQ values below 80 were most common in the children with a prenatal bone age at diagnosis. Right panel: individual IQ values of children aged 5 years 9 months with congenital hypothyroidism identified by newborn biochemical screening, subdivided by bone age at diagnosis and compared with a control group matched for socioeducational level of the family. Note that, even in the subgroup with a prenatal bone age, the mean and the distribution of IQ values are indistinguishable from those of the control group. (Adapted from Wolter R, Noel P, De Cock P, et al. Neuropsychological study in treated thyroid dysgenesis. Acta Pediatr Scand, Suppl 1979;277:41. and Simoneau-Roy J, Marti S, Deal C, et al. Cognition and behavior at school entry in children with congenital hypothyroidism treated early with high-dose levothyroxine.  2004;144:747.)J Pediatr

Another result of biochemical screening for congenital hypothyroidism is that it has revealed that the prevalence of permanent primary congenital hypothyroidism is remarkably similar around the world (1 in 3,500 newborn infants). The only exceptions are areas of severe iodine deficiency, in which the prevalence is higher (see section Iodine Deficiency in Chapter 11 and Chapter 49). Any differences in the proportions of infants in whom hypothyroidism was caused by thyroid ectopy and athyreosis likely reflect differences in the type and quality of the imaging tests done. The proportion of cases of dyshormonogenesis may vary slightly as a function of the background consanguinity of the population.


By allowing treatment before the onset of any clinical manifestations of hypothyroidism, biochemical screening of neonates for congenital hypothyroidism has resulted in the disappearance of mental deficiency caused by this condition. However, this true public health triumph should not lead to complacency. Continuous auditing of all steps of the screening program is necessary to ensure that infants with congenital hypothyroidism are identified and treated with the least possible delay.


1. McCabe LL, Therrell BL, McCabe ER. Newborn screening: rationale for a comprehensive, fully integrated public health system.  2002;77:267.Mol Genet Metab

2. Dussault JH. The anecdotal history of screening for congenital hypothyroidism.  1999;84:4332.J Clin Endocrinol Metab

3. Delange F. Neonatal screening for congenital hypothyroidism: results and perspectives.  1997;48:51.Horm Res

4. Van Vliet G, Czernichow P. Screening for neonatal endocrinopathies: rationale, methods and results.  2004;9:75.Semin Neonatol

5. Tillotson SL, Fuggle PW, Smith I, et al. Relation between biochemical severity and intelligence in early treated congenital hypothyroidism: a threshold effect.  1994;309:440.BMJ

6. Wolter R, Noel P, De Cock P, et al. Neuropsychological study in treated thyroid dysgenesis.  1979; 277:41.Acta Paediatr Scand Suppl

7. Dubuis JM, Glorieux J, Richer F, et al. Outcome of severe congenital hypothyroidism: closing the developmental gap with early high dose levothyroxine treatment.  1996;81:222.J Clin Endocrinol Metab

8. Bongers-Schokking JJ, Koot HM, Wiersma D, et al. Influence of timing and dose of thyroid hormone replacement on development in infants with congenital hypothyroidism.  2000; 136:292.J Pediatr

9. Simoneau-Roy J, Marti S, Deal C, et al. Cognition and behavior at school entry in children with congenital hypothyroidism treated early with high-dose levothyroxine.  2004;144: 747.J Pediatr

10. Screening for congenital hypothyroidism.  2003;13:87.Thyroid

11. Delange F. Screening for congenital hypothyroidism used as an indicator of the degree of iodine deficiency and of its control.  1998;8:1185.Thyroid

12. Alvarez MA, Guell R, Gonzalez J, et al. Neuropsychological development during the first two years of life in children with congenital hypothyroidism screened at birth: the Cuban experience.  1992;1:167.Screening

13. Virtanen M, Maenpaa J, Pikkarainen J, et al. Aetiology of congenital hypothyroidism in Finland.  1989; 78:67.Acta Paediatr Scand

14. Perry R, Heinrichs C, Bourdoux P, et al. Discordance of monozygotic twins for thyroid dysgenesis: implications for screening and for molecular pathophysiology.  2002;87:4072.J Clin Endocrinol Metab

15. Dussault JH, Grenier A, Morissette J, et al. Preliminary report on filter paper TSH levels in the first 24h of life and the following days in a program screening for congenital hypothyroidism. In: Pass KA, Levy HL, eds.  Atlanta: Council of Regional Networks for Genetics Services, Emory University School of Medicine, 1995:267.Early hospital discharge: impact on newborn screening.

16. Vincent MA, Rodd C, Dussault JH, et al. Very low birth weight newborns do not need repeat screening for congenital hypothyroidism.  2002;140:311.J Pediatr

17. Mandel SJ, Hermos RJ, Larson CA, et al. Atypical hypothyroidism and the very low birthweight infant.  2000; 10:693.Thyroid

18. Van Vliet G, Larroque B, Bubuteishvili L, et al. Sex-specific impact of congenital hypothyroidism due to thyroid dysgenesis on skeletal maturation in term newborns.  2003;88:2009.J Clin Endocrinol Metab

19. van Wassenaer AG, Kok JH, de Vijlder JJ, et al. Effects of thyroxine supplementation on neurologic development in infants born at less than 30 weeks' gestation.  1997; 336:21.N Engl J Med

20. van Trotsenburg AS, Vulsma T, Van Santen HM, et al. Lower neonatal screening thyroxine concentrations in Down syndrome newborns.  2003;88:1512.J Clin Endocrinol Metab

21. Devos H, Rodd C, Gagne N, et al. A search for the possible molecular mechanisms of thyroid dysgenesis: sex ratios and associated malformations.  1999;84:2502.