J Clin Res Pediatr En docrinol 2013;5(Suppl 1):45-49DO I: 10.4274/Jcrpe.855
Autoimmune Thyroiditis in Childhood
Rosalind S. Brown
Division of Endocrinology, Children’s Hospital Boston and Department of Pediatrics, Harvard Medical School, Boston, USA
In tro duc ti on
Autoimmune thyroiditis (AIT) is the most common thyroid
disorder in the pediatric age range. Both a goitrous (Hashimoto’sthyroiditis) and a nongoitrous (atrophic thyroiditis, also calledprimary myxedema) variant of AIT have been distinguished. Inpediatrics, the most common age at presentation is adolescence,but the disease may occur at any time, rarely even in children underone year of age (1). In the past decade, the application of molecularbiology has permitted an unparalleled insight into susceptibilitygenes that predispose to the development of AIT and into itscomplex immune pathophysiology. Approximately 70% of diseaserisk is related to genetic susceptibility with environmental factorsplaying a role in triggering disease in susceptible individuals. In thisbrief review, current concepts about the immuno-pathogenesis andmolecular genetics of AIT as well as putative environmentaltriggers will be discussed. In addition, clinical aspects unique to thepediatric age range will be reviewed.
Immuno-pathogenesis of Autoimmune Thyroiditis Cellular ?mmune Responses
Because of the importance of T- cells in immune regulation,much attention has focused on this lymphocyte subpopulation toexplain the breakdown in tolerance and the clinical manifestationsseen in autoimmune thyroid disease (AITD). Consistent with theirplaying a fundamental role, an increased proportion of activated T-helper (Th) (CD4+) cells can be demonstrated in the circulation of amajority of patients with AIT (2) and this is thought to lead to acascade of immune-mediated events. Lymphocytic infiltration ofthe thyroid gland and organization into lymphoid follicles occur dueto the secretion of numerous chemokines that direct leukocytemigration and to adhesion molecules that lead to cell attachment toextracellular matrix proteins. These processes are accompanied bythe secretion of complement, cytokines and other soluble
Autoimmune thyroiditis (AIT) is the most common thyroid disorder in thepediatric age range. The disease results from an as yet poorlycharacterized defect or defects in immunoregulation and a cascade ofevents progressing from lymphocyte infiltration of the thyroid, to T-cell-and cytokine-mediated thyroid follicular cell injury, and apoptotic celldeath. Approximately 70% of disease risk has been attributed togenetic background with environmental factors being important intriggering disease in susceptible individuals. The contribution ofindividual genes is small and probably polymorphisms in multiple genesplay a role. Some immunosusceptibility genes affect immune recognitionor response in general, while others are thyroid-specific. Environmentalagents may act through an epigenetic mechanism. Antibodies (Abs) toa variety of thyroid-specific antigens are detectable in a majority ofpatients, but the role of Abs in mediating cell injury and death is unclearand only thyrotropin (TSH) receptor Abs significantly affect thyroidfunction by interfering with (or stimulating) the action of TSH.Nonetheless, thyroid peroxidase (TPO) Abs and thyroglobulin (Tg) Abs,present in a majority of patients, are valuable diagnostically as markersof underlying autoimmune thyroid destruction. TSH receptor blockingAbs are found in ~18% of children and adolescents with severehypothyroidism and, when persistent, may identify an adolescent likelyto have a baby with TSH receptor blocking Ab-induced congenitalhypothyroidism. AIT may coexist with other organ-specific autoimmunediseases. Although the most common age at presentation isadolescence, the disease may occur rarely in children <1 year of life.Key words:Thyroiditis, hypothyroidism, children, autoimmunityConflict of interest:None declaredRe cei ved:25.09.2012
Ac cep ted:
Ad dress for Cor res pon den ce
Rosalind S. Brown MD, Division of Endocrinology, Children’s Hospital Boston and Department of Pediatrics, Harvard Medical School, Boston, USA
Phone: +617-355-7476 E-mail: Rosalind.Brown@childrens.harvard.edu
?Jo ur nal of Cli ni cal Re se arch in Pe di at ric En doc ri no logy, Pub lis hed by Ga le nos Pub lis hing.
Brown Rosalind S.
Autoimmune Thyroiditis in Childhood
mediators that damage the thyrocyte and lead ultimately to celldeath by apoptosis (3). It is of interest that not only ‘professional’antigen-presenting cells (e.g., macrophages, dendritic cells) butthyroid follicular cells themselves play a crucial role in determiningdisease progression by expressing major histocompatibilitycomplex (MHC) class II antigens (4) and secreting cytokines andchemokines. Goiter, present in approximately two-thirds ofchildren with AIT, results from both cellular infiltration and thyroidfollicular cell proliferation, a consequence of the compensatoryincrease in thyrotropin (TSH) that occurs in response to thyrocytedysfunction.
To add further to the complexity, there are actually 2 types ofTh cells: Th1 and Th2, each with a different pattern of cytokineproduction, effector function, chemokine receptors, and regulation(5). Th1 cells subserve cell-mediated immune responses, whileTh2 cells are involved in antibody production. It has beenpostulated that in AIT, a defect in a specific T cell subpopulation,termed regulatory T cells (Tregs), results in some way in a changein the thyroid microenvironment, leading to decreased inhibition ofTh1 cells and the overproduction of Th1 cytokines.
In addition to cell-mediated immune mechanisms, AIT ischaracterized by the secretion of antibodies (Abs) to a variety ofthyroid-specific antigens, most notably thyroglobulin (Tg), andthyroid peroxidase (TPO) but also to a lesser extent the TSHreceptor, the sodium iodide symporter (NIS) (6) and most recentlypendrin (7). Whether, and, if so, how antibody-mediated immunemechanisms contribute to the initiation, progression orpathogenesis of AIT remains unclear. Nonetheless, measurementof Abs to Tg and TPO is useful diagnostically as markers ofunderlying autoimmunity, and TSH receptor Abs may modulate theactivity of the TSH receptor, thereby affecting thyroid function in asubset of patients. The role of Abs to NIS or pendrin is not yet clear.
Antibodies to Thyroglobulin and Thyroid Peroxidase
Tg Abs are directed at Tg, the very large 660 kDA homodimericprotein that serves as the storage form and precursor of thyroidhormone. There is some evidence that Tg Abs in patients with AITare more restricted in their epitope specificity than in healthyindividuals, but the antigenic determinants recognized by Tg Absare not known (8). Post-translational modifications such asiodination and glycosylation may play a role in Tg antigenicity (9).TPO Abs recognize TPO, the key membrane-associated enzymeon the apical surface of the thyrocyte that mediates the iodinationand coupling of iodotyrosines to form thyroid hormone (10). Bothcytotoxicity (e.g., activation of the complement cascade andparticipation in Ab-dependent, cell-mediated cytotoxicity ) and TPOAb-mediated inhibition of TPO function have been reported in vitro,but these effects do not appear to be important in vivoas indicatedby the failure of these Abs to affect fetal or neonatal thyroidfunction when transmitted to the fetus. Like with Tg Abs, someepitopic specificity of TPO Abs has been reported in patients withAIT (11), the significance of which is unclear.
TSH Receptor Antibodies
Abs to the TSH receptor have been classified as stimulatory,blocking and neutral Abs (12). All varieties recognize specific linearepitopes presented in a 3-dimensional conformation , but theirinteraction differs subtly. Like other members of the G-coupledreceptor superfamily, the TSH receptor is composed of a large, N-terminal ectodomain, 7-transmembrane-spanning regions and asmall intracytoplasmic tail. Although initial studies suggestedseparate epitopes for stimulatory Abs on the N-terminus and forblocking Abs on the C-terminus of the ectodomain, recentevidence suggests more overlap than previously thought (12).Stimulatory Abs appear to be of limited heterogeneity, whereasblocking Abs are not similarly restricted.
Both blocking and stimulating TSH receptor Abs can be foundin some patients with AIT. The coexistence of stimulatory TSHreceptor Abs and AIT has been termed ‘Hashitoxicosis’. Incontrast, blocking Abs may contribute to the severity of thehypothyroidism by inhibiting TSH-induced cell proliferation andhormonogenesis. In a recent study of children and adolescents,TSH receptor blocking Abs were found in 17.8% of patients withsevere hypothyroidism (defined as a serum TSH concentration>20 mU/L). However, unlike in adults, they were found in goitrousas well as nongoitrous patients and, when present in highconcentrations, appeared to persist indefinitely (13). These datasuggest that the presence of potent TSH receptor blocking Abs inadolescent females may identify patients at risk of having babieswith transient congenital hypothyroidismin their future ages.
Like Graves disease (GD) with which it is closely associated,AIT is a complex immune disorder that occurs in individuals withan underlying genetic susceptibility. As many as 20-60immunosusceptibility genes, each with small effect, have beenpostulated (14). Some genes are involved in immune recognitionand/or response in general, while others are thyroid-specific;certain genes are common to both AIT and GD, while others tendto predominate only in GD (8,9). In this brief review, only those thataffect individuals with AIT will be discussed. These genes aresummarized in Table 1.
Immune Response Genes
Human leukocyte antigen (HLA):The MHC region encodes ahighly polymorphic region that includes the genes for the class I (A,
Brown Rosalind S.
Autoimmune Thyroiditis in Childhood
B and C) and class II (DR, DP and DQ) HLA. As peptide antigensare presented to T cells only when bound to HLA class IImolecules, the HLA haplotype plays a critical role in determiningwhich antigens will be recognized by the T-cell receptor and triggeran immune response. Unlike with GD, no consistent HLA-association has been identified in patients with AIT. Recently,Menconi et al (15) sequenced the polymorphic exon 2 of the HLA-DR gene in 94 patients with AIT and 149 controls. They identifieda four-amino acid haplotype (Tyr-26, Tyr-30, Gln-70, Lys-71) thatconferred an odds ratio of 3.73 irrespective of the HLA haplotype;the single amino acid Lys-71 conferred an odds ratio of 2.98. Theseinvestigators postulated that specific ‘pocket amino acidsignatures’ determine susceptibility to AIT by causing criticalstructural changes that influence antigenic peptide binding andpresentation.
Cytotoxic T Lymphocyte Antigen-4
Cytotoxic T lymphocyte (CTLA)-4, a transmembrane protein ofthe immunoglobulin superfamily, is an important costimulatorymolecule that downregulates T-cell activation by binding to the B7molecules present on antigen-presenting cells and preventing thesecond signal necessary for T-cell activation. Linkage of the CTLA-4 region with thyroid autoantibody production in patients with andwithout clinical thyroid disease has been demonstrated (16). Thissuggests that additional factors must be required for thedevelopment of a thyroid functional abnormality.
Protein Tyrosine Phosphatase-22
Protein tyrosine phosphatase 22 (PTPN22) encodes a protein,lymphoid tyrosine phosphatase (LYP), which is a potent inhibitor ofthe T-cell receptor signaling pathway . A tryptophan for argininesubstitution at codon 620 of the LYP molecule has beenassociated with both GD and AIT in some but not all ethnicpopulations (16).
The thyroid-specific genes that have been incriminated in AITDinclude those that code for Tg and the TSH receptor, but only theTg gene has been associated with AIT.
Tg is a very large 660kDA homodimeric protein that functionsas the storage form and precursor of thyroid hormone. Ingenetically susceptible mice, experimental AIT can be induced byTg immunization and there is evidence that Tg sensitization mayplay a role in patients with AITD as well. Detailed sequenceanalysis of the Tg gene has demonstrated significant associationsbetween specific polymorphisms of Tg SNPs and AIT (17,18).These findings suggest that specific Tg polymorphisms might bemore optimally presented to T cells by specific HLA haplotypes,thereby triggering an immune response. This is highly reminiscentof findings in experimental animals in which induction of AIT is alsoHLA-dependent.
In addition, the immunogenicity of Tg has been related to theextent of its iodination (19). The pathogenic mechanisms wherebyincreased iodine predisposes to AIT remain unclear.
The precise environmental trigger(s) leading to thedevelopment of disease is not known with certainty, but infection,drugs (lithium, amiodarone, interferon-alpha), hormones(estrogen), dietary substances (iodine, selenium), stress, smokingand, most recently, environmental toxins have all been implicated(20,21). An epigenetic mechanism has been postulated in somecases (21).
Incidence and Associations
AIT is more common in North America and Japan than insome iodine-deficient parts of Europe and there is a striking femalepreponderance, even prior to puberty. AIT may occur alone or maycoexist with other organ-specific autoimmune diseases,particularly type 1 diabetes mellitus (T1DM). The combination ofAIT with specific autoimmune disorders has been termed theautoimmune polyglandular syndromes (APS). APS 1, which tendsto present in the first decade of life and is caused by a mutation ina single (‘autoimmune regulator’) gene, is characterized by thesequential development of mucocutaneous candidiasis,hypoparathyroidism and adrenal deficiency (22). AIT is amongst amultiplicity of other organ-specific diseases that can also developin a minority of affected patients. In APS-2 which tends to presentlater in childhood or adolescence, Addison’s disease is found inassociation with AIT (Schmidt syndrome) or AIT plus T1DM(Carpenter syndrome) (23). Multiple other organ-specificautoimmune diseases may also be present. AIT has also beendescribed in children with immunodysregulationpolyendocrinopathy enteropathy X-linked (IPEX) syndrome, apolyglandular disorder characterized by early-onset diabetes andcolitis (24) . There is an increased incidence of AIT in patients withcertain chromosomal abnormalities [Down syndrome (25), Turnersyndrome (26), and, to a lesser extent, Klinefelter syndrome] aswell as in patients with Noonan syndrome. AIT may be associatedwith chronic urticaria (27) and rarely with immune-complexglomerulonephritis (28).
Patients with AIT may be euthyroid, or they may havesubclinical or overt hypothyroidism depending on the severity ofthe immunologic damage. Unexplained poor linear growth is aclassical initial finding in many hypothyroid children. In somepatients, an asymptomatic goiter may be noted on routineexamination. Occasionally, an initial thyrotoxic phase occurs dueto the discharge of preformed thyroid hormone from the damagedgland. In this case, the picture needs to be distinguished fromhyperthyroidism due to GD. Rarely, as noted previously, bothdiseases may coexist in the same patient (Hashitoxicosis).
The typical thyroid gland in AIT is diffusely enlarged and has arubbery consistency. Although the surface is classically describedas ‘pebbly’ or bosselated, occasionally asymmetric enlargementoccurs that must be distinguished from thyroid neoplasia. Apalpable lymph node superior to the isthmus (‘Delphian node’)may be noted.
Brown Rosalind S.
Autoimmune Thyroiditis in Childhood
Measurement of the serum TSH concentration is the bestinitial screening test for the presence of primary hypothyroidism. Ifthe TSH is elevated, then evaluation of the serum free thyroxine(fT4) concentration will distinguish whether the child hassubclinical (normal fT4) or overt (low fT4) hypothyroidism. Adiagnosis of AIT is made by the demonstration of an elevatedconcentration of Tg Abs and/or TPO Abs in serum. Measurementof TSH receptor blocking Abs should be considered in adolescentfemales with severe hypothyroidism because of the persistence ofthis Ab population in some patients and its association with anincreased risk of having offspring with TSH receptor blocking Ab-induced congenital hypothyroidism. Imaging studies (thyroidultrasonography and/or thyroid uptake and scan) may beperformed if thyroid Ab tests are negative or if a nodule is palpable,but are rarely necessary. Occasionally, the finding ofheterogeneous echogenicity on ultrasound examination has beendescribed prior to the appearance of Abs. However, the typicalpicture of spotty uptake of radioactive iodine that is seen in adultsis rare in children.
In patients who present with severe, longstandinghypothyroidism, slow correction with LT4 is advisable in order tominimize the potential development of unwanted side effects(deterioration in school performance, short attention span,hyperactivity, insomnia, and behavior difficulties) (29). In suchpatients, the replacement dose should be increased slowly overseveral weeks to months. Severely hypothyroid children shouldalso be observed closely for complaints of severe headache whentherapy is initiated because of the rare development ofpseudotumor cerebri. In contrast, full replacement can be initiatedat once without much risk of adverse consequences in childrenwith mild hypothyroidism.
Treatment of children and adolescents with subclinicalhypothyroidism (normal fT4, elevated TSH) is controversial. Inadults, particularly those > 60 years of age in whom the risk ofprogression to overt hypothyroidism is significant, treatment hasbeen recommended whenever the serum TSH concentration is>10 mU/L; if TSH is in the 6-10 mU/L range, treatment on a caseby case basis is suggested (30). Long-term follow-up studies ofchildren with subclinical hypothyroidism due to AIT havesuggested a significant likelihood of remission. Consequently, ifthere is not a strong family history of hypothyroidism and thepatient is not symptomatic, a reasonable option is to reassessthyroid function in 6 months. On the other hand, some initiallyeuthyroid patients will become hypothyroid over time. Therefore,regular follow-up is necessary.
The typical replacement dose of LT4 in childhood isapproximately 100 μg/m2or 4 to 6 μg/kg for children 1 to 5 yearsof age, 3 to 4 μg/kg for those aged 6 to 10 years, and 2 to 3 μg/kgfor those 11 years of age and older. In patients with a goiter, asomewhat higher LT4 dosage is used so as to keep the TSH in thelow normal range (0.3 to 1.0 mU/L in an ultrasensitive assay), and
thereby minimize its goitrogenic effect. T4 and TSH should bemeasured after the child has received the recommended dosagefor at least 6-8 weeks. Once a euthyroid state has been achieved,patients should be monitored every 6 to 12 months. Thyroidhormone replacement is not associated with significant weightloss in overweight children, unless the hypothyroidism is severe(31). Treatment is usually continued indefinitely.
1.Foley TP Jr, Abbassi V, Copeland KC, Draznin MB. Brief report:
hypothyroidism caused by chronic autoimmune thyroiditis invery young infants. N Engl J Med 1994;330:466-468.
2.Gessl A, Wilfing A, Agis H, Steiner G, Czernin S, Boltz-Nitulescu G, Vierhapper H, Waldh?usl W. Activated naiveCD4+ peripheral blood T cells in autoimmune thyroid disease.Thyroid 1995;5:117-125.
3.Wang SH, Baker JR. The role of apoptosis in thyroid
autoimmunity. Thyroid 2007;17:975-979.4.Bottazzo GF, Pujol-Borrell R, Hanafusa T, Feldmann M. Role of
aberrant HLA-DR expression and antigen presentation ininduction of endocrine autoimmunity. Lancet 1983;2:1115-1119.
5.Weetman AP. Cellular immune responses in autoimmune
thyroid disease. Clin Endocrinol (Oxf) 2004;61:405-413.6.Ajjan RA, Kemp EH, Waterman EA, Watson PF, Endo T, Onaya
T, Weetman AP. Detection of binding and blockingautoantibodies to the human sodium-iodide symporter inpatients with autoimmune thyroid disease. J Clin EndocrinolMetab 2000;85:2020-2027.
7.Yoshida A, Hisatome I, Taniguchi S, Shirayoshi Y, Yamamoto Y,
Miake J, Ohkura T, Akama T, Igawa O, Shigemasa C, Kamijo K,Ikuyama S, Caturegli P, Suzuki K. Pendrin is a novelautoantigen recognized by patients with autoimmune thyroiddiseases. J Clin Endocrinol Metab 2009;94:442-448. Epub2008 Dec 2
8.Brown RS. Autoimmune thyroid disease: unlocking a complex
puzzle. Curr Opin Pediatr 2009;21:523-528.
9.Jacobson EM, Tomer Y. The genetic basis of thyroid
autoimmunity. Thyroid 2007;17:949-961.
10.McLachlan SM, Rapoport B. Thyroid peroxidase as an
autoantigen. Thyroid 2007;17:939-948.11.Nielsen CH, Brix TH, Gardas A, Banga JP, Hegedüs L. Epitope
recognition patterns of thyroid peroxidase autoantibodies inhealthy individuals and patients with Hashimoto's thyroiditis*.Clin Endocrinol (Oxf) 2008;69:664-668. Epub 2008 Mar 1912.Smith BR, Sanders J, Furmaniak J. TSH receptor antibodies.
13.Feingold SB, Smith J, Houtz J, Popovsky E, Brown RS.
Prevalence and functional significance of thyrotropin receptorblocking antibodies in children and adolescents with chroniclymphocytic thyroiditis. J Clin Endocrinol Metab 2009;94:4742-4748. Epub 2009 Oct 2214.Davies TF. Really significant genes for autoimmune thyroid
disease do not exist--so how can we predict disease? Thyroid2007;17:1027-1029.15.Menconi F, Monti MC, Greenberg DA, Oashi T, Osman R,
Davies TF, Ban Y, Jacobson EM, Concepcion ES, Li CW, TomerY. Molecular amino acid signatures in the MHC class IIpeptide-binding pocket predispose to autoimmune thyroiditisin humans and in mice. Proc Natl Acad Sci USA2008;105:14034-14039.
Brown Rosalind S.
Autoimmune Thyroiditis in Childhood
16.Jacobson EM, Tomer Y. The CD40, CTLA-4, thyroglobulin, TSH
receptor, and PTPN22 gene quintet and its contribution tothyroid autoimmunity: back to the future. J Autoimmun2007;28:85-98. Epub 2007 Mar 21
17.Ban Y, Greenberg DA, Concepcion E, Skrabanek L, Villanueva
R, Tomer Y. Amino acid substitutions in the thyroglobulin geneare associated with susceptibility to human and murineautoimmune thyroid disease. Proc Natl Acad Sci USA2003;100:15119-15124. Epub 2003 Dec 1
18.Collins JE, Heward JM, Carr-Smith J, Daykin J, Franklyn JA,
Gough SC. Association of a rare thyroglobulin genemicrosatellite variant with autoimmune thyroid disease. J ClinEndocrinol Metab 2003;88:5039-5042.
19.Rose NR, Bonita R, Burek CL. Iodine: an environmental trigger
of thyroiditis. Autoimmun Rev 2002;1:97-103.
20.Brent GA. Environmental exposures and autoimmune thyroid
disease. Thyroid 2010;20:755-761.21.Eschler DC, Hasham A, Tomer Y. Cutting edge: the etiology of
autoimmune thyroid diseases. Clin Rev Allergy Immunol2011;41:190-197.
22.Betterle C, Greggio NA, Volpato M. Clinical review 93:
Autoimmune polyglandular syndrome type 1. J Clin EndocrinolMetab 1998;83:1049-1055.
23.Eisenbarth GS, Gottlieb PA. Autoimmune polyendocrinesyndromes. N Engl J Med 2004;350:2068-2079.
24.Wildin R, Smyk-Pearson S, Filipovich A. Clinical and molecular
features of the immunodysregulation, polyendocrinopathy,enteropathy, X linked (IPEX) syndrome. J Med Genet2002;39:537-545.25.Karlsson B, Gustafsson J, Hedov G, Ivarsson SA, Anneren G.
Thyroid dysfunction in Down's syndrome: relation to age andthyroid autoimmunity. Arch Dis Child 1998;79:242-245.
26.Elsheikh M, Wass JA, Conway GS. Autoimmune thyroid
syndrome in women with Turner's syndrome--the associationwith karyotype. Clin Endocrinol (Oxf) 2001;55:223-226.
27.Verneuil L, Leconte C, Ballet JJ, Coffin C, Laroche D, Izard JP,
Reznik Y, Leroy D. Association between chronic urticaria andthyroid autoimmunity: a prospective study involving 99patients. Dermatology 2004;208:98-103.
28.Gurkan S, Dikman S, Saland MJ. A case of autoimmune
thyroiditis and membranoproliferative glomerulonephritis.Pediatr Nephrol 2009;24:193-197.29.Rovet JF, Daneman D, Bailey JD. Psychologic and
psychoeducational consequences of thyroxine therapy forjuvenile acquired hypothyroidism. J Pediatr 1993;122:543-549.30.Surks MI, Ortiz E, Daniels GH, Sawin CT, Col NF, Cobin RH,
Franklyn JA, Hershman JM, Burman KD, Denke MA, GormanC, Cooper RS, Weissman NJ. Subclinical thyroid disease:scientific review and guidelines for diagnosis andmanagement. Jama 2004;291:228-238.
31.Carswell JM, Brown RS. Thyroid hormone in childhood
obesity--no "quick fix". Endocr Pract 2010;16:157-158.