Clinical features and detection of congenital hypothyroidism

INTRODUCTION — Congenital primary hypothyroidism, occurring in approximately 1:2000 to 1:4000 newborns, is one of the most common preventable causes of intellectual disability worldwide. There is an inverse relationship between age at treatment initiation and intelligence quotient (IQ) later in life, so that the longer the condition goes undetected and untreated, the lower the IQ [1]. (See "Intellectual disability (ID) in children: Clinical features, evaluation, and diagnosis".)

Most newborn babies with congenital hypothyroidism have few or no clinical manifestations of thyroid hormone deficiency. In addition, the majority of cases are sporadic, so it is not possible to predict which infants are likely to be affected. For these reasons, newborn screening programs were developed to detect this condition as early as possible by measuring either thyroxine (T4) or thyrotropin (thyroid-stimulating hormone [TSH]) in heel-stick blood specimens [2]. These screening efforts were initiated in the mid-1970s and have been largely successful, although severely affected infants may still have a slightly reduced IQ and other neurologic deficits despite prompt diagnosis and initiation of therapy.

This topic will review the epidemiology, causes, clinical manifestations, and diagnosis of congenital hypothyroidism and its detection by newborn screening in full-term infants. Related content can be found in the following topic reviews:

●(See "Treatment and prognosis of congenital hypothyroidism".)

●(See "Thyroid physiology and screening in preterm infants".)

EPIDEMIOLOGY — Data obtained from national and regional screening programs indicate that the incidence of congenital primary hypothyroidism varies globally. The incidence varies by geographic location and by ethnicity. Reports from screening programs in the United States, Canada, European countries, Israel, Australia, New Zealand, and Japan note an incidence between 1:2000 to 1:4000 newborns [3,4]. The reported incidence varies among racial and ethnic groups, with rates of approximately 1:1200 in Southern Asian infants, 1:2380 in Eastern Asian (Chinese and Vietnamese) infants, 1:1600 in Hispanic infants, 1:3533 in non-Hispanic White infants, and 1:11,000 in non-Hispanic Black infants, as reported over a seven-year period from California [5]. Further, the incidence appears to be increased in twin births (1:900) and even higher in multiple births (1:600) [6]. The highest incidence, 1:581, was reported from the Markazi Province in Iran, likely related to consanguinity and a higher occurrence of autosomal recessive inborn errors of thyroid hormone synthesis [7].

There appears to be a trend of increasing incidence of congenital primary hypothyroidism detected by newborn screening programs, from 1:4000 in the mid-1970s to 1:2000 in 2010 [3]. This increase likely is the result of lowering of thyroid-stimulating hormone (TSH) cutoffs in TSH-based screening programs, leading to detection of milder cases of primary hypothyroidism [8], along with changes in population demographics (with more births in those populations with a higher incidence), rather than an actual increase in the incidence of the disorder itself.

Nearly all screening programs report a female preponderance, approaching a 2:1 female-to-male ratio. A report from Quebec shows that this female preponderance occurs mostly in cases of thyroid ectopy and less so in those with agenesis [9]. Another study from Quebec found that thyroid dysgenesis was more prevalent in White infants than in Black infants, whereas dyshormonogenesis occurred equally in both racial groups [10]. (See 'Thyroid dysgenesis' below.)

ETIOLOGY — Congenital hypothyroidism is most commonly caused by an embryologic defect in thyroid gland development (dysgenesis) or a defect in thyroid hormone synthesis (dyshormonogenesis). Most cases of thyroid dysgenesis are sporadic, while the dyshormonogenesis disorders are inherited in an autosomal recessive pattern. Defects in thyroid hormone transport or action are rare causes of congenital hypothyroidism (table 1). Worldwide, iodine deficiency remains the main cause of congenital hypothyroidism; in these cases, iodine replacement leads to normal thyroid function. Other causes of transient hypothyroidism include transfer of maternal antithyroid drugs, maternal thyroid-stimulating hormone (TSH) receptor-blocking antibodies, exposure to excess iodine, large hepatic hemangiomas, and some DUOX2 gene mutations (see 'Transient congenital hypothyroidism' below). Congenital central hypothyroidism is most commonly caused by a defect in the embryologic development of the pituitary gland or mutations in the genes responsible for pituitary hormone synthesis. (See 'Central hypothyroidism' below.)

Primary hypothyroidism — Primary hypothyroidism refers to inadequate thyroid hormone production in the gland itself. Following the inception of population-wide newborn screening programs in the 1970s, studies to determine the underlying etiology reported that approximately 85 percent were caused by thyroid dysgenesis, while 15 percent were caused by one of the inborn errors in thyroid hormone synthesis. On follow-up, the majority of these cases had permanent hypothyroidism. With lowering of the TSH screening cutoff and detection of milder cases of congenital hypothyroidism, more recent studies to examine the underlying etiology now report that a normal-sized or large thyroid gland in the normal location, so-called "gland in situ," is the most common finding [11]. While the underlying cause of gland in situ is often unknown, it likely represents a mild form of thyroid dyshormonogenesis. On follow-up, approximately one-half of cases of thyroid gland in situ have transient hypothyroidism.

Thyroid dysgenesis — The most common cause of permanent congenital hypothyroidism is thyroid dysgenesis (abnormal thyroid gland development) resulting from agenesis, hypoplasia, or ectopy. Thyroid ectopy accounts for two-thirds of the dysgenesis cases worldwide [12].

In a study of 230 infants with permanent primary hypothyroidism, representing 90 percent of all infants identified by newborn screening in Quebec from 1988 to 1997, 61 percent had ectopic thyroid tissue, 16 percent had thyroid agenesis, 4 percent had a normal-sized thyroid, and 18 percent had a goiter, as determined by radionuclide imaging [13]. More females than males had thyroid ectopy (104 versus 37), but there were similar numbers of females and males in other groups. Among these infants, 5 percent had other congenital abnormalities, mostly cardiac septal defects. (See 'Associated congenital malformations' below.)

Although most cases of thyroid dysgenesis are sporadic, there is evidence of a familial/genetic component in approximately 2 percent of cases [14,15]. Of note, most pairs of twins (both monozygotic and dizygotic) are discordant for congenital hypothyroidism [16]. The explanation for discordance in monozygotic twins is unknown, although it may be explained by differences in somatic autosomal monoallelic gene expression between the twins [17]. A long-term follow-up study of twins discordant for congenital hypothyroidism found that 15 percent became concordant for hypothyroidism, while 21 percent of the initially hypothyroid probands had transient hypothyroidism [18].

Mutations in several genes that regulate thyroid gland development have been reported as rare causes of thyroid dysgenesis [19]. These mutations may be associated with congenital anomalies in other tissues:

●PAX8 gene – Anomalies of the urogenital tract (horseshoe kidney, ureterocele, undescended testes, and hydrocele; MIM #218700) [20,21].

●TTF2 gene (also termed FOXE1) – Cleft palate, spiky hair, and bilateral choanal atresia (Bamforth-Lazarus syndrome; MIM #241850) [22,23].

●NKX2-1 gene (formerly termed TTF1) – Anomalies of the lungs (respiratory distress syndrome) and central nervous system (ataxia), including pituitary hormone deficiencies (MIM #610978) [24-26].

●NKX2-5 gene – Many different congenital heart defects (MIM #225250) [27].

●GLIS3 gene – Neonatal diabetes mellitus, congenital glaucoma, hepatic fibrosis, and polycystic kidneys (MIM #610199) [28].

●JAG1 gene – Alagille syndrome type 1 (MIM #118450) [29]. (See "Causes of cholestasis in neonates and young infants", section on 'Alagille syndrome' and "Alagille syndrome".)

●CDCA8 gene – No nonthyroidal anomalies identified [30].

●NTN1 gene – Thyroid ectopy, arthrogryposis [31].

●TUBB1 gene – Thyroid dysgenesis, abnormal platelets [32].

●TBX1 gene – DiGeorge syndrome, with parathyroid hypoplasia, congenital heart malformations, and thyroid in situ (MIM #602054) [33].

●TPO gene – Usually associated with dyshormonogenesis and goiter and also reported to cause thyroid hypoplasia, a form of thyroid dysgenesis (MIM #274500) [34].

Infants with trisomy 21 (Down syndrome) have a higher incidence of hypothyroidism detected by newborn screening programs, occurring in as many as 1:50 newborns [35]. Mild elevations of TSH with normal thyroxine (T4) levels are also relatively common in this population. Whereas older children with trisomy 21 have autoimmune thyroid disease, the hypothyroidism seen in neonates is not associated with antithyroid antibodies [36]. There is speculation that the extra chromosome 21 results in genomic dosage imbalance of dosage-sensitive genes interfering with thyroid hormone production. (See "Down syndrome: Clinical features and diagnosis", section on 'Thyroid disease'.)

Resistance to thyroid-stimulating hormone — Mutations in the thyrotropin (TSH) receptor gene will present as primary hypothyroidism, with an elevated serum TSH and low T4 level. Such defects are increasingly recognized [37]. In a report from Japan, TSH receptor mutations were present in 4.3 percent of patients with congenital hypothyroidism (1:118,000 of the general population) [38]. In a report from Great Britain, TSH receptor mutations were identified in 5 percent of children with congenital non-goitrous hypothyroidism born to consanguineous families [39].

In some forms of pseudohypoparathyroidism, which results from a GNAS gene defect, there is interference with TSH signaling, in addition to that of parathyroid hormone (PTH) [40]. TSH resistance can be associated with pseudohypoparathyroidism type 1A, which is due to a heterozygous mutation in GNAS, or, less commonly, with pseudohypoparathyroidism type 1B, which is caused by a defect in the methylation pattern of the GNAS locus (both maternal in origin). When such cases are detected by newborn screening, the hypothyroidism is mild and, despite thyroid hormone treatment, linear growth slows, accompanied by excessive weight gain. Such cases should then be screened for hypocalcemia and elevated PTH levels to confirm PTH resistance. A complex phenotype of congenital hypothyroidism and pseudohypoparathyroidism has been described in the presence of a combination of inactivating mutations of the TSH receptor gene and mutations of the downstream GNAS gene [41]. (See "Resistance to thyrotropin and thyrotropin-releasing hormone", section on 'Resistance to thyroid-stimulating hormone' and "Etiology of hypocalcemia in infants and children", section on 'Type 1 PHP'.)

Disorders of thyroid hormone synthesis and secretion — Hereditary defects in virtually all steps in thyroid hormone biosynthesis and secretion have been described, all of which are characterized by autosomal recessive inheritance, except for some cases caused by heterozygous DUOX2 mutations.

These defects are illustrated in the figure (figure 1) and include:

●Defects in iodide transport into thyroid follicular cells (step 1 in the figure), caused by a mutation in the SLC5A5 gene (MIM #274400) [42].

●Defects in transport across the apical membrane (step 3 in the figure), caused by a mutation in the Pendrin gene, SLC26A4 (also termed PDS), where it functions as a chloride/iodide pump and is responsible for transporting iodine out of the cell and into the follicular colloid, and in the cochlea, where it results in a sensorineural hearing loss (Pendred syndrome; MIM #274600) [43].

●A defect in thyroid peroxidase activity (MIM #606765) that results in impaired iodide oxidation and organification [44]; this is the most common defect reported in cases from the Netherlands. (See "Approach to congenital goiter in newborns and infants", section on 'Inborn errors of thyroid hormone production'.)

●Defects in the generation of hydrogen peroxide (step 4 in the figure), a substrate for thyroid peroxidase in the oxidation of iodide, caused by mutations in the DUOX2 gene (formerly called THOX2; MIM #607200) [45-47] or the related DUOXA2 gene (MIM #274900) [48]. DUOX2 mutations appear to cause a spectrum of thyroid disorders, from mild congenital hypothyroidism, to a large goiter causing respiratory distress associated with lifelong euthyroidism [49]. They are reported to cause both transient and permanent congenital hypothyroidism [50]. Defects in hydrogen peroxide generation (DUOX2) appear to be the most common inborn errors in the Japanese and Chinese populations [51,52].

●Production of abnormal thyroglobulin molecules (top of figure), caused by mutations in the TG gene (MIM #274700) [53]. Mutations in TG are among the most common inborn errors in Great Britain, European, and Middle Eastern populations [54].

●Iodotyrosine deiodinase deficiency (MIM #274800) due to homozygous mutations of the IYD (DEHAL1) gene (step 8 in the figure) (figure 1) [55].

Altogether, these disorders account for approximately 15 percent of cases of permanent congenital hypothyroidism. (See "Thyroid hormone synthesis and physiology".)

Defects in thyroid hormone transport — Passage of thyroid hormone into target organs is facilitated by plasma membrane transporters. A mutation in one such transporter gene, MCT8, located on the X chromosome, has been reported in more than 100 individuals with X-linked intellectual disability (Allen-Herndon-Dudley syndrome). The defective transporter appears to impair passage of triiodothyronine (T3) into neurons; this syndrome is characterized by decreased serum T4, associated with elevated T3, normal to mildly elevated TSH levels, and psychomotor retardation [56]. (See "Genetic defects in thyroid hormone transport and metabolism", section on 'Thyroid hormone cell membrane transport defect'.)

Defects in thyroid hormone metabolism — Inherited defects in thyroid hormone metabolism involve the SECISBP2 gene. In theory, mutations in this gene could cause congenital hypothyroidism, but to date, there are only reports in children (with short stature and delayed bone age) and adults (MIM #609698). (See "Genetic defects in thyroid hormone transport and metabolism", section on 'Thyroid hormone metabolism defect'.)

Defects in thyroid hormone action: Resistance to thyroid hormone — Resistance to thyroid hormone is caused by mutations in thyroid hormone receptors (primarily the THRB gene). The incidence is approximately 1:40,000. It is characterized by high serum T4, free T4, T3, and free T3 levels, with normal or slightly elevated serum TSH levels. Rarely, patients with high TSH levels may be detected by newborn screening programs. Typical findings in childhood include failure to thrive and attention deficit hyperactivity disorder, but patients may have no clinical manifestations of hyperthyroidism; for most patients, no treatment is indicated. (See "Resistance to thyroid hormone and other defects in thyroid hormone action", section on 'Resistance to thyroid hormone beta (RTH-beta and nonTR-RTH)'.)

Resistance to thyroid hormone caused by mutations in the THRA gene has been reported but is less common than that caused by mutations in THRB. Although the defect is present at birth, most cases are not discovered until infancy or childhood, when affected children present with growth retardation, variable motor and cognitive deficits, and macrocephaly. Thyroid function tests show a low free T4, elevated T3, and normal to slightly elevated TSH level [57]. (See "Resistance to thyroid hormone and other defects in thyroid hormone action", section on 'Resistance to thyroid hormone alpha (RTH-alpha)'.)

Central hypothyroidism — Central hypothyroidism refers to defects in the production of TSH due to either hypothalamic or pituitary dysfunction.

Newborn screening programs that employ the initial T4/follow-up TSH approach often detect central hypothyroidism but are not reliable for this purpose. Programs based on TSH screening alone generally will not identify these infants. Central hypothyroidism occurs in 1:16,404 to 1:29,000 newborns [58,59]. It may be associated with other congenital syndromes, particularly midline defects such as optic nerve hypoplasia/septo-optic dysplasia or midline cleft lip and palate defects and may follow birth trauma or asphyxia.

Mutations in several genes are responsible for isolated TSH deficiency [60]. Mutations in the IGSF1 gene are reported to be the most common cause of central hypothyroidism. This X-linked disorder is characterized by central hypothyroidism, delayed pubertal testosterone production, adult macro-orchidism, and variable prolactin and growth hormone deficiency (MIM #300888) [61]. Other rare genetic causes result from mutations in the genes for the thyrotropin-releasing hormone (TRH) receptor [62] or the beta subunit of TSH [63,64]. Mutations in the TBL1X gene, an X-linked gene, have been reported in a large pedigree with central hypothyroidism and hearing loss (MIM #301033) [65]. Mutations in the IRS4 gene have also been implicated in X-linked isolated TSH deficiency (MIM #301035); the pathogenesis is thought to be related to abnormal leptin signaling [66]. (See 'Newborn screening' below and "Resistance to thyrotropin and thyrotropin-releasing hormone".)

Most infants with central hypothyroidism, except those with mutations in the genes for the TSH beta subunit or TRH receptor, have other pituitary hormone deficiencies. Some cases of central hypothyroidism are present in infants with congenital hypopituitarism caused by mutations in transcription factors involved with pituitary development. A mutation in the POU1F1 gene caused central hypothyroidism in a mother and her infant (MIM #613038) [67]. In a large prospective study, all infants with delayed TSH response to TRH had multiple pituitary hormone deficiencies [68]. The presence of another pituitary hormone deficiency should be particularly suspected in infants with hypoglycemia or micropenis [58]. (See "Approach to hypoglycemia in infants and children".)

Congenital central hypothyroidism also can be caused by insufficient treatment of maternal Graves hyperthyroidism during pregnancy [69,70]. This form of central hypothyroidism may persist beyond six months of age, especially when maternal thyrotoxicosis occurred before 32 weeks gestation [70]. One study suggests that some of these infants also may have primary hypothyroidism with thyroid "disintegration," possibly because insufficient fetal TSH during the period of maternal hyperthyroidism inhibited the normal growth and development of the fetal thyroid [71].

Transient congenital hypothyroidism — Worldwide, the most common cause of congenital hypothyroidism that resolves during the first few months or years of life (transient hypothyroidism) is iodine deficiency. In iodine-sufficient countries, the most common causes are maternal antithyroid drugs during pregnancy, iodine exposure during gestation or in the postnatal period, and "gland in situ" with mild hypothyroidism (see below). Transient congenital hypothyroidism is more frequently detected in newborn screening programs with lower TSH thresholds. In a 20-year study from a French newborn screening program, congenital hypothyroidism was transient in 40 percent of patients [72].

The causes of transient hypothyroidism with goiter in newborn infants are:

●Iodine deficiency – Iodine deficiency, particularly in preterm infants, accounts for many cases in Europe and other worldwide areas where maternal dietary iodine intake is less than in the United States [73]. Data and a map indicating areas of iodine deficiency worldwide are available from the Iodine Global Network website.

●Iodine exposure – Exposure of the fetus or newborn to high doses of iodine can cause hypothyroidism. Examples include:

•Iodinated contrast agents given to infants – The US Food and Drug Administration suggests checking thyroid function in infants and young children within three weeks following exposure to iodinated contrast agents [74]

•Amniofetography with an iodinated radiographic contrast agent [75]

•Topical iodine-containing antiseptic compounds used in pregnant mothers or infants [76]

•Maternal treatment with amiodarone (for cardiac arrhythmias) [77]

•Maternal ingestion of excessive iodine from nutritional supplements during pregnancy [78]

Populations at particular risk include infants born prematurely and those with congenital heart defects or other anomalies, due to exposure to iodine through the skin and/or in contrast media used for cardiac catheterization or lymphangiography [79-82].

The risk of hypothyroidism appears to be related to the type and duration of iodine exposure. Among women exposed to an oil-soluble iodinated contrast agent used for hysterosalpingography to assess fertility who then became pregnant, approximately 2 percent of their offspring were detected with hypothyroidism [83]. Transient exposure of pregnant women to an iodinated contrast agent for a computed tomography study appears to have no effect on neonatal thyroid function [84,85]. (See "Iodine-induced thyroid dysfunction" and "Diagnostic imaging in pregnant and lactating patients", section on 'Use of iodinated contrast materials'.)

●Maternal blocking antibodies – Transplacental transfer of TSH-receptor blocking antibodies can occur in infants of mothers with autoimmune thyroid disease [86]. Such a diagnosis should also be considered if more than one infant born to the same mother (in the same or multiple pregnancies) is identified as having primary hypothyroidism by newborn screening. Studies using newborn screening specimens show TSH-receptor blocking antibodies in approximately 1:180,000 newborns [87]. This form of hypothyroidism usually subsides around three months of age (range one to six months) as the maternal antibodies are cleared [88,89]. (See "Iodine-induced thyroid dysfunction" and "Diagnostic imaging in pregnant and lactating patients", section on 'Use of iodinated contrast materials'.)

●Maternal antithyroid drugs – Antithyroid drugs given to mothers with hyperthyroidism also can cross the placenta. These drugs are cleared in days; as a result, many of these infants are euthyroid when restudied a few weeks after delivery.

●Large hepatic hemangiomas – Large hepatic hemangiomas, present from birth, may produce increased levels of type 3 deiodinase, resulting in "consumptive hypothyroidism" [90]. A case report describes an infant with a massive hepatic hemangioendothelioma that caused hypothyroidism, discovered at eight weeks of age [91]. Large doses of thyroid hormone were required to normalize serum TSH. The hypothyroidism resolved by 16 months of age as the hemangioendothelioma regressed.

●Mutations in the DUOX2 gene – Biallelic loss-of-function mutations in DUOX2 are rare causes of transient congenital hypothyroidism [92,93]. Mutations in the related gene, DUOXA2, also are reported to cause a more permanent, albeit mild, form of congenital hypothyroidism, as discussed above. (See 'Disorders of thyroid hormone synthesis and secretion' above.)

●Gland in situ – As noted above, approximately one-half of gland in situ cases have transient hypothyroidism. While the underlying etiology of gland in situ is unknown in most cases, most probably it represents a defect in thyroid hormone synthesis, of which the DUOX2 gene mutations described above are one example.

CLINICAL MANIFESTATIONS

Asymptomatic newborns — The vast majority (more than 95 percent) of infants with congenital hypothyroidism have few, if any, clinical manifestations of hypothyroidism at birth [94]. This is because some maternal thyroxine (T4) crosses the placenta, so that even in infants who cannot make any thyroid hormone, umbilical cord serum T4 concentrations are approximately 25 to 50 percent of those of normal infants [95]. In addition, many infants with congenital hypothyroidism have some, albeit inadequate, functioning thyroid tissue. Despite these mitigating influences, congenital hypothyroidism may have subtle effects in utero. One report described reduced variability in fetal heart rate tracings in nearly 50 percent of infants with congenital hypothyroidism [96].

Birth length and weight typically are within the normal range; birth weight often is at a relatively higher percentile than birth length, owing to myxedema [97]; head circumference also may be increased. The knee epiphyses often lack calcification, and this is more likely to occur in males than females (40 versus 28 percent, respectively) [98].

Symptomatic infants — Infants born in regions of the world that lack newborn screening programs typically present with symptoms and signs of hypothyroidism that develop over the first few months of life, which include lethargy, hoarse cry, feeding problems, often needing to be awakened to nurse, constipation, puffy (myxedematous) and/or coarse facies, macroglossia, umbilical hernia, large fontanels, hypotonia, dry skin, hypothermia, and prolonged jaundice (primarily unconjugated hyperbilirubinemia) [99]. Newborn infants with thyroid dyshormonogenesis may have a goiter detected on prenatal ultrasound or on clinical examination of the neonate, while in others, the goiter is discovered later in life. Palpable subcutaneous nodules (ossifications) may be a tip-off to congenital hypothyroidism caused by pseudohypoparathyroidism.

If an infant has central hypothyroidism, the clinical manifestations are often related to associated deficiencies of other pituitary hormones and include hypoglycemia (growth hormone and adrenocorticotropic hormone), micropenis (growth hormone and/or gonadotropins), undescended testes (gonadotropins), and, least commonly, features of arginine vasopressin deficiency (previously known as central diabetes insipidus). Hearing loss in an infant with central hypothyroidism may be a clue to a TBL1X mutation [65]. (See 'Central hypothyroidism' above and "Causes of hypopituitarism", section on 'Genetic diseases'.)

Associated congenital malformations — Congenital hypothyroidism is associated with a modestly increased risk of additional congenital malformations affecting the heart, kidneys, urinary system, gastrointestinal tract, and skeleton [100-104]. As an example, in a population-based study of 1420 infants with congenital hypothyroidism, the prevalence of other congenital malformations (mostly cardiac) was fourfold higher (8.4 percent) than in the control infant population (1 to 2 percent) [100]. In the New York State Congenital Malformation Registry (980 children), there was an increased risk of renal and urologic abnormalities (odds ratio 13.2) [104]. Infants with congenital hypothyroidism and cleft palate may have a TTF2 gene mutation [22]. Infants with pulmonary disease and/or persistent neurologic problems, including ataxia, are suspect for an NKX2-1 gene mutation (sometimes known as brain-lung-thyroid syndrome) [24]. GLIS3 mutations have been described in cases of congenital hypothyroidism associated with neonatal diabetes mellitus, congenital glaucoma, hepatic fibrosis, and polycystic kidneys [28]. JAG1 mutations most commonly are reported in patients with Alagille syndrome type 1, some cases of which are associated with thyroid hypoplasia [29]. (See 'Thyroid dysgenesis' above and "Genetic disorders of surfactant dysfunction", section on 'NKX2-1 sequence variants' and "Causes of cholestasis in neonates and young infants", section on 'Alagille syndrome' and "Alagille syndrome".)

NEWBORN SCREENING

Screening programs — Screening of all newborns is now routine in all 50 states of the United States and in Canada, Europe, Israel, Japan, Australia, and New Zealand; screening programs are also in place or under development in many countries in Eastern Europe, South America, Asia, and Africa [3]. As of 2014, many regions in south Central Asia, Southeast Asia, and sub-Saharan Africa had minimal rates of newborn screening. In the United States, approximately four million infants are screened annually, leading to the detection of 2000 infants per year with congenital hypothyroidism [105]. Out of a worldwide birth population of approximately 130 million infants annually, it is estimated that 37 million infants (29 percent) are screened and approximately 12,000 infants with hypothyroidism are detected annually [3].

Timing — Blood for screening is collected onto filter paper cards after heel prick, and the cards are then sent to a centralized laboratory for testing. For full-term infants, the sample is optimally collected between 24 and 72 hours after birth [4,106]. Some programs also routinely obtain a second specimen between one and three weeks after birth [4,58].

A second screen should be performed between two and four weeks of age for infants with the following characteristics [106]:

●Preterm or very low birth weight

●Acutely ill

●Down syndrome

●Monozygotic twin or multiple births

●Received a transfusion before the initial screen was obtained

The second screening is indicated for these infants because they have higher rates of false-positive and false-negative results on the initial screen [107]. As an example, a report from a California health plan described targeted screening at two weeks to detect delayed TSH rise in selected high-risk neonates; this approach found an additional 35 percent of hypothyroid cases despite only needing to test an additional 1.7 percent of infants [108]. For preterm infants, strategies and additional screening are discussed separately. (See "Thyroid physiology and screening in preterm infants", section on 'Newborn screening'.)

For infants with Down syndrome, a European guideline recommends serum tests instead of the second screening test, consisting of measurement of serum thyroid-stimulating hormone (TSH) and free thyroxine (fT4) at three to four weeks of age [4], with serum testing at 6 to 12 months and then annually thereafter [106]. (See "Down syndrome: Management", section on 'Thyroid function'.)

Technique — Three major screening strategies have evolved:

●Initial blood T4 assay, with reflex thyrotropin (TSH) assay if the blood T4 value is below a certain concentration (usually less than the 10^th percentile for a given day's run in the laboratory performing the assay).

●Initial blood TSH assay, with reflex T4 assay if TSH is elevated.

●Simultaneous T4 and TSH assays.

Historically, most programs in the United States started with the initial T4/follow-up TSH testing approach, but many have switched to initial TSH testing. Either approach detects the majority of infants with congenital primary hypothyroidism, and each has its advantages and disadvantages. Infants with a delayed rise in blood TSH concentration [109] and those with central hypothyroidism are detected more reliably by the initial T4/follow-up TSH assay method [58,59], whereas infants with subclinical hypothyroidism (high blood TSH, normal blood T4) are detected more reliably by TSH testing.

Interpretation and follow-up — Some newborn screening programs report results in serum units (eg, United States and some Canadian provinces), while others report results in whole blood units (eg, other Canadian provinces and Europe, except the Netherlands). TSH results in whole blood units are equivalent to approximately one-half of the corresponding serum value. An abnormally high hematocrit may result in modest reductions in the TSH concentration on the screening test, which may be important for infants with borderline results [110].

Infants with TSH values above certain levels on the initial newborn screen, usually >30 mIU/L in serum units (equivalent to >15 mIU/L in whole blood units), are recalled for clinical evaluation and serum testing as soon as possible (algorithm 1), typically by one week of age. If the TSH on the newborn screen is very elevated (>40 mIU/L), levothyroxine treatment should be initiated after drawing the confirmatory serum sample, without waiting for the results [106].

The results of the serum testing should be interpreted using a lower TSH cutoff (typically >10 mIU/L after one week of age). Unfortunately, some screening programs do not adjust for the infant's age, leading to false-negative results for infants with mild hypothyroidism [111]. (See 'False-negative and false-positive results' below.)

Using this screening technique, approximately 0.1 percent of infants in the population are recalled for further testing and approximately one-half of these are diagnosed with hypothyroidism. Thus, two infants are recalled for each infant who is ultimately diagnosed with congenital hypothyroidism. Programs that recall infants only on the basis of low blood T4 values have recall rates of approximately 0.3 percent. (See 'Serum tests of thyroid function' below.)

Infants who "pass" an initial screening test but are then detected with abnormal results on the second (and sometimes third) screen are recalled for further testing, consisting of a full set of thyroid function tests on a blood sample obtained by venipuncture. Infants with this unusual pattern of screening tests may have equivocal serum thyroid function tests; in selected cases, imaging may help make a management decision, as discussed below. (See 'Serum tests of thyroid function' below and 'Thyroid imaging' below.)

False-negative and false-positive results

●False-negative results – The likelihood of false-negative results (ie, the sensitivity of newborn screening) depends on the type of hypothyroidism and on the technique used by the screening program:

•Newborn screening using any of the above techniques is generally highly sensitive for detecting infants with primary hypothyroidism.

•Newborn screening has low sensitivity for detecting central hypothyroidism. Central hypothyroidism generally is not detected by newborn screening programs that use TSH screening alone. Moreover, even those programs that employ primary T4 testing fail to detect many cases of congenital central hypothyroidism because early on, these infants often have T4 levels above the cutoffs that are typically used by screening programs. As an example, a retrospective study from a program that employs simultaneous measurement of T4 and TSH reported that only 19 percent of infants subsequently diagnosed with central hypothyroidism had a newborn screening result of T4 <5 mcg/dL [112]. Thus, it is important to refer and evaluate infants who have clinical features suggestive of congenital hypopituitarism; clinicians should not be falsely reassured by a normal screening T4 result.

•Newborn screening will miss infants with subclinical hypothyroidism if the program uses the T4-based screening technique.

•If a screening sample is collected later than the usual age (one to four days of life), the test may fail to detect mild hypothyroidism unless the reference range is adjusted for the infant's age since the TSH values decrease sharply during the first week of life [111]. Many screening programs do not use age-specific cutoffs for TSH.

•The initial screening test in an affected monozygotic (monochorionic) twin may be normal. For unknown reasons, most monozygotic twins are discordant for congenital hypothyroidism, such that shared blood supply may temporarily normalize thyroid function in the affected twin. Most screening programs therefore recommend a second routine screening test in same-sex twins.

●False-positive results – The likelihood of false-positive results is influenced by the timing of the initial screening test as well as on the test cutoffs used by the screening program:

•Primary T4 with follow-up TSH screening technique – False-positive cases occur chiefly because the T4 cutoff is set relatively high (<10^th percentile), leading to capturing many normal babies. Further misidentification may occur if the specimen is obtained early in life (<24 hours of age) when the TSH is physiologically elevated as part of the postnatal TSH surge, which peaks at approximately 60 to 80 mIU/L, 30 minutes after delivery. For infants with positive results of the initial screen, a screening program may either request a second heel-prick screening specimen for retesting or request confirmatory serum testing. As noted above, there are approximately five false-positive cases for every true case of congenital hypothyroidism with this screening approach.

•Primary TSH screening technique – The false-positive rate is greatly influenced by the timing of collection of the specimen (following the postnatal TSH surge). To reduce the rate of false positives, many screening programs use age-related TSH cutoffs. As noted above, there is approximately one false-positive case for every true case of congenital hypothyroidism with this screening approach.

•Screening with multiple tests (eg, T4 and TSH; or T4, TSH, and T4-binding globulin) has been suggested as a way to improve detection of both primary and central hypothyroidism [113].

Lowering the TSH cutoff, while leading to improved detection of milder cases, typically is associated with an increased number of false-positive screens. False-positive screens may lead to unnecessary parental anxiety as well as a potential rise in overall costs because of unnecessary recall of patients and need for further testing.

DIAGNOSIS — In the majority of cases, the diagnosis of hypothyroidism can be confirmed or excluded by results of serum tests of thyroid function, informing the decision to start thyroid hormone treatment. (See 'Serum tests of thyroid function' below.)

If the diagnosis of hypothyroidism is confirmed, other studies (such as thyroid radionuclide uptake and scan, ultrasonography, serum thyroglobulin, tests for thyroid autoantibodies, or urinary iodine excretion) may be performed to identify the cause. Considerations and selection criteria for these tests are discussed below. (See 'Additional testing' below.)

Serum tests of thyroid function — When abnormal results are reported for the newborn screen, a blood sample should be obtained by venipuncture to confirm or exclude the diagnosis of hypothyroidism, measuring thyrotropin (thyroid-stimulating hormone [TSH]), as well as free thyroxine (free T4) or total T4 and triiodothyronine (T3) uptake. This strategy also applies if hypothyroidism is suspected because of clinical symptoms (eg, if the infant had not been identified by a screening program). Results of these tests help to determine the type of hypothyroidism and treatment approach and can be interpreted as follows (algorithm 1):

High TSH, low free T4 — These results on serum testing confirm the diagnosis of primary hypothyroidism (table 1). Treatment is indicated, beginning immediately [4,106]. (See "Treatment and prognosis of congenital hypothyroidism".)

One must keep in mind that serum T4 concentrations are higher in the first few weeks of life in normal infants than in adults because of the surge in TSH secretion that occurs soon after birth (figure 2). Serum TSH concentrations rise abruptly to 60 to 80 mIU/L, typically peaking 30 minutes after birth. The serum TSH concentration then decreases rapidly to approximately 20 mIU/L 24 hours after delivery and then more slowly to 6 to 10 mIU/L at one week of age. A serum TSH >10 mIU/L is elevated in infants after one week of age. Between one and four days of life, the normal range for serum total T4 is approximately 10 to 22 mcg/dL (129 to 283 nmol/L) and the normal range for serum free T4 is approximately 2 to 5 ng/dL (25 to 64 pmol/L). After four weeks of life, the normal range for serum total T4 is 7 to 16 mcg/dL (90 to 206 nmol/L) and the normal range for serum free T4 is 0.8 to 2.0 ng/dL (10 to 26 pmol/L). (See "Thyroid physiology and screening in preterm infants", section on 'Term infants'.)

High TSH, normal free T4 or total T4 — These results on serum testing define subclinical hypothyroidism.

If the TSH is significantly elevated (eg, >20 mIU/L), treatment should be initiated [4,106]. In cases where the serum TSH is marginally elevated (eg, 6 to 20 mIU/L), one option is to monitor carefully, repeating a serum TSH and free T4 in one week. Some infants will normalize TSH without treatment using this approach. However, if the serum TSH remains >10 mU/L by four weeks of age, we recommend starting thyroid hormone treatment because the development of the central nervous system is critically dependent on adequate amounts of T4, although there is minimal clinical evidence to guide this decision [106]. (See "Treatment and prognosis of congenital hypothyroidism".)

Low or normal TSH, low free T4 — These results on serum testing suggest the possibility of central hypothyroidism (table 1). These findings also are common in premature or acutely ill infants, results that may be compatible with nonthyroidal illness syndrome [114]. In these cases, we recommend repeating the serum free T4 by the more accurate equilibrium dialysis method. It is important that gestational and postnatal normative values be used (table 2) (see "Thyroid physiology and screening in preterm infants", section on 'Effects of nonthyroidal illness'). Rarely, infants with these findings are ultimately proven to have primary hypothyroidism (rather than central hypothyroidism) because the expected rise in serum TSH is delayed [115], but treatment is appropriate in either case.

Infants with central hypothyroidism may also have central adrenal insufficiency; if possible, tests to evaluate such infants for adrenal insufficiency should be undertaken before thyroid hormone treatment is started. If coexistent central adrenal insufficiency cannot be ruled out, glucocorticoid treatment should be initiated prior to levothyroxine treatment to prevent possible induction of an adrenal crisis.

Once central hypothyroidism has been confirmed, and the issue of possible associated central adrenal insufficiency has been addressed, thyroid hormone treatment should be promptly initiated.

Other results

●Low T4, normal free T4, and normal TSH – This combination of findings indicates the presence of a deficiency of binding proteins, the most common of which is T4-binding globulin deficiency, an X-linked recessive disorder that occurs in approximately 1:4000 newborns, predominantly males [116]. These infants are euthyroid and do not require treatment.

ADDITIONAL TESTING — Additional evaluation may be helpful for selected infants. In those cases where thyroid function appears to be improving and is almost normal before treatment, additional testing may facilitate a decision about whether to treat versus observe. Additional test results may also distinguish sporadic from hereditary etiologies, informing genetic counseling, or may distinguish permanent from transient cases, facilitating future management. However, in most cases, these tests do not change management, and we consider them to be optional.

Thyroid imaging — Thyroid ultrasonography or radionuclide uptake measurements and imaging ("thyroid scan") provide information about the underlying etiology, eg, thyroid dysgenesis or one of the types of dyshormonogenesis [117,118]. We do not recommend either test routinely, because for most cases, the results do not alter management. However, the tests may provide useful information in infants with the following characteristics:

●Infants with minor abnormalities in thyroid function, in whom it is not clear whether thyroid hormone treatment is indicated (eg, thyroid-stimulating hormone [TSH] 6 to 10 mIU/L, with free thyroxine (free T4) in the normal reference range for age). In such an infant, a finding consistent with some form of thyroid dysgenesis, eg, ectopically located thyroid tissue, would support a diagnosis of hypothyroidism.

●Infants with a goiter in whom an enzymatic defect in T4 synthesis is suspected; most of these infants have normal or high uptake values in normally located, if not enlarged, thyroid glands. As these defects are hereditary (autosomal recessive), this finding allows counseling about risk in future children.

●Infants suspected of having transient hypothyroidism (eg, due to iodine deficiency or exposure, maternal autoimmune thyroid disease, or antithyroid drugs during pregnancy). The presence of a normal thyroid gland on ultrasonography or scan supports a diagnosis of mild, possibly transient, hypothyroidism. Cases of excess iodine exposure likely will have reduced radionuclide uptake until the excess iodine is cleared. Cases of transient hypothyroidism due to transplacental passage of maternal TSH-receptor blocking antibodies typically do not have radionuclide uptake or image on scanning, but on ultrasonography, a normal thyroid gland may be visualized. (See 'Transient congenital hypothyroidism' above.)

Thyroid ultrasonography and color flow Doppler — If the clinician chooses to do thyroid imaging, we recommend starting with ultrasound to characterize thyroid shape, size, and location. If an ectopic thyroid gland is identified by ultrasound (typically found in the lingual, sublingual, or subhyoid areas), radionuclide imaging will not be necessary, thus avoiding radiation exposure and expense. However, ultrasound is not as reliable as radionuclide imaging in identifying cases of thyroid dysgenesis [119]. In studies comparing the two procedures, some infants had ectopic thyroid tissue detected by radionuclide imaging, but the ultrasound images revealed either no thyroid tissue or thyroid hypoplasia; in some of the other infants, radionuclide imaging revealed thyroid agenesis or hypoplasia, but the ultrasound images were normal [120,121]. Color flow Doppler ultrasonography may be superior to gray-scale ultrasonography. It detected ectopic thyroid tissue in 90 percent of infants with congenital hypothyroidism and ectopic tissue detected on radionuclide imaging [122].

Thyroid radionuclide uptake and scan — If the ultrasound study does not detect ectopic thyroid tissue, then a thyroid radionuclide uptake and scan may help to identify cases of thyroid dysgenesis or dyshormonogenesis. Either 99m-pertechnetate or iodine-123 should be used instead of iodine-131 because the radiation doses are much lower. 99m-pertechnetate may be more readily available and allows a good scan picture, but because it is not organified, there is no measure of uptake. Iodine-123 must be specially ordered due to its short half-life, but it will provide both a scan and a measure of uptake. A large gland in a normal location typically is seen with one of the enzymatic defects. A positive perchlorate discharge test is compatible with an organification defect [123]. Decreased uptake on scintiscan associated with a normally located thyroid gland on ultrasound suggests the possibility of a loss-of-function TSH receptor gene mutation, a sodium-iodide symporter gene variant, or maternally transmitted TSH receptor-blocking antibodies.

Thyroid autoantibodies — For infants born to mothers with known autoimmune thyroid disease, and in families where a previous sibling was detected with congenital hypothyroidism with a transient course, measurement of TSH-receptor blocking antibodies (in the mother and/or fetus) may be useful in diagnosing this form of transient congenital hypothyroidism.

Serum thyroglobulin concentration — Measurement of serum thyroglobulin has been advocated as a means to distinguish among the causes of congenital hypothyroidism. For infants with no thyroid gland seen by imaging, a low serum thyroglobulin level is consistent with thyroid aplasia, while in infants with a large gland on imaging, a low serum thyroglobulin level points to a thyroglobulin gene defect. As an example, one study reported low serum thyroglobulin concentrations in infants with thyroid agenesis on radionuclide imaging (mean 12 ng/mL, range 2 to 54 ng/mL), intermediate concentrations in those with ectopic thyroid tissue (mean 92 ng/mL, range 11 to 231 ng/mL), and high concentrations in those with goiters (mean 226 ng/mL, range 3 to 425 ng/mL); the normal range was 20 to 80 ng/mL [117]. Given the degree of overlap among these groups, we do not think that measurement of serum thyroglobulin alone can substitute for radionuclide imaging and do not recommend it routinely, although its measurement may be helpful in puzzling cases, when considered in association with additional findings.

Urinary iodine concentration — If there is a history of iodine exposure, or if the infant is born in an area of endemic goiter (ie, high rates of iodine deficiency, as shown on this map), measurement of urinary iodine can confirm an excess (or deficient) state. Usually, however, this information can be obtained by history. In infants thought to have iodine-induced hypothyroidism, any continuing iodine exposure should be discontinued. If thyroid function does not normalize within a few weeks, T4 therapy should be given for several months and then gradually withdrawn. (See "Iodine deficiency disorders", section on 'Geographic distribution' and "Approach to congenital goiter in newborns and infants", section on 'Iodine excess' and "Approach to acquired goiter in children and adolescents", section on 'Iodine-deficiency goiter'.)

Genetic testing — Existing evidence supports targeted genetic testing in cases of congenital hypothyroidism with familial or syndromic features that suggest a specific genetic etiology [106,124].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hypothyroidism" and "Society guideline links: Pediatric thyroid disorders".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Congenital hypothyroidism (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Congenital hypothyroidism is one of the most common preventable causes of intellectual disability, occurring in 1:2000 to 1:4000 newborns. (See 'Epidemiology' above.)

●Etiology

•Primary hypothyroidism refers to inadequate thyroid hormone production in the thyroid gland; causes include thyroid dysgenesis, a sporadic disorder, or one of the inborn errors in thyroid hormone synthesis, a hereditary (autosomal recessive) disorder (table 1). (See 'Primary hypothyroidism' above.)

•Central hypothyroidism refers to defects in the production of thyroid-stimulating hormone (TSH) due to hypothalamic or pituitary dysfunction. (See 'Central hypothyroidism' above.)

•Some cases of congenital hypothyroidism resolve during the first few months or years of life. Causes of this pattern include gestational exposure, such as iodine deficiency or excess, or maternal TSH receptor-blocking antibodies or antithyroid drugs, or a mild form of thyroid dyshormonogenesis (with thyroid gland in situ). (See 'Transient congenital hypothyroidism' above.)

●Clinical manifestations – Because some maternal thyroid hormone crosses the placenta, and because there is commonly some residual thyroid function in the neonate with congenital hypothyroidism, the vast majority (more than 95 percent) of affected infants lack recognizable clinical manifestations at birth. The following signs and symptoms develop over the first few weeks and months of life: lethargy, hoarse cry, feeding problems, often needing to be awakened to nurse, constipation, puffy (myxedematous) facies, macroglossia, umbilical hernia, large fontanels, hypotonia, dry skin, hypothermia, and prolonged jaundice. (See 'Clinical manifestations' above.)

●Newborn screening

•Technique – Newborn screening programs use different sequences of thyroxine (T4) and TSH testing. Each strategy detects the majority of infants with primary congenital hypothyroidism, but some do not detect central hypothyroidism or infants with delayed rise in TSH. (See 'Technique' above.)

•Timing – The first screening test is optimally performed between 24 and 72 hours after birth. A second screen should be performed between two and four weeks of age for selected infants (preterm, ill, monozygotic twins, or Down syndrome). (See 'Timing' above.)

•Units – Some newborn screening programs, including those in the United States, report results in serum units, while others report results in whole blood units. In general, results in whole blood units are equivalent to one-half of the serum value. The threshold for defining abnormal TSH is usually >30 mIU/L in serum units (equivalent to >15 mIU/L in whole blood units) (algorithm 1). (See 'Interpretation and follow-up' above.)

•Further evaluation – Infants with abnormal screening results are recalled for further testing. At recall, the infant should be examined and a serum sample obtained by venipuncture to confirm the diagnosis of hypothyroidism (algorithm 1).

●Diagnosis – The threshold for defining abnormal TSH is usually >30 mIU/L in serum units (equivalent to >15 mIU/L in whole blood units) (algorithm 1). Age-specific norms must be used for interpretation because TSH and thyroid hormone concentrations are higher in the first few days and weeks of life than in older infants, children, or adults (table 2). (See 'Serum tests of thyroid function' above.)

•Low serum free T4 (or total T4) with high serum TSH values confirm the diagnosis of primary hypothyroidism, which should be treated promptly.

•Low serum free T4 (or total T4) concentration with low or normal serum TSH concentrations suggest central hypothyroidism (or primary hypothyroidism with delayed rise in TSH).

•Newborn screening programs are increasingly detecting neonates with borderline thyroid function. In cases where the serum TSH is marginally elevated (eg, 6 to 20 mIU/L) on initial testing, one option is to monitor carefully, repeating a serum TSH and free T4 in one week. If the serum TSH does not normalize by four to six weeks of age, we recommend starting thyroid hormone treatment.

●Additional testing – If the diagnosis of hypothyroidism is confirmed, other studies, such as thyroid radionuclide uptake and scan, ultrasonography, serum thyroglobulin assay, tests for thyroid autoantibodies, urinary iodine excretion, or targeted genetic testing may be performed to identify the cause; these tests are optional. (See 'Additional testing' above.)