Behavioural disorders in adolescents with early-treated congenital hypothyroidism

Francesca TinelliCaterina CostagliStefania BargagnaMara MarcheschiBarbara ParriniValentina Perelli

Stella Maris Scientific Institute, Institute of Developmen-tal Neurology, Psychiatric and Educational Psychology,University of Pisa, Italy

Reprint requests to: Dr Stefania Bargagna, Stella Maris Scientific Institute, Institute of Developmental Neurology, Psychiatric and Educational Psychology, University of Pisa, Viale del Tirreno, 56018 Calambrone, Pisa, ItalyE-mail: s.bargagna@inpe.unipi.it

Accepted for publication: May 23, 2003

Summary

This study analyses the possible risk factors for the on-set of behavioural disorders and psychiatric distur-bances in a group of 30 early-treated congenital hy-pothyroidism (CH) subjects (12 children and 18 adoles-cents) compared with a control group of 116 age-matched normal subjects (58 children and 58 adoles-cents). The study also allowed us to evaluate the possi-ble age at onset of behavioural disorders. Both the sample’s and the controls’ behaviours wereassessed using a specific diagnostic instrument:Achenbach’s and Edelbrock’s Child Behaviour Check-list (CBCL). A clinical structured interview, the Diagnos-tic Interview for Children and Adolescents - Revised (DI-CA-R) was also administered to 18 adolescents withearly-treated CH, in order to determine the presence ofpsychopathological disturbances.In accordance with literature data, the children and ado-lescents with early-treated CH showed more behaviour-al problems than age-matched, normal controls. In thechildren, a statistically significant difference versus thecontrols emerged only in their higher delinquent behav-iour score, while the adolescents gave, on the CBCL,significantly higher scores compared with controls inthe withdrawal, anxiety/depression, thought problems,attention problems and aggressive behaviour scales. Inthe DICA-R, 44% of adolescents with early-treated CHshowed symptoms of anxiety disorder, in particular,separation anxiety disorder with phobic components;16% showed mood disorder and depression and 11%showed behavioural disorders with attention deficit.

KEY WORDS: adolescence, behaviour, congenital hypothyroidism,

psychiatric disorder.

I n t r o d u c t i o n

Neonatal screening for and early treatment of congenitalhypothyroidism (CH) have successfully reduced the mostsevere effects of this disorder on neuropsychological de-velopment (1,2). However, several studies (3,4) show that subtle learningdifficulties, defective language abilities, motor disorders(delayed achievement of the main milestones), and be-havioural disorders (including attention deficit and hyper-activity) may still occur, as may non specific neurologicalsigns (hypotonia, hyperlaxity), which present with greaterfrequency in early-treated CH children and adolescentsthan in the rest of the age-matched, unaffected population(5-8). It appears that high T4 and TSH levels contribute to poor-er attention and it is very likely that thyroid hormone is in-volved in the regulation of attention and motor skills.CH is a chronic disease, which has a negative impactboth on the psychological development of children and ontheir relationship with their parents and with their socialenvironment. Indeed, affective disorders, which are oftenpresent in CH children, are attributable more to difficultfamily relationships than to the subject’s neurobiologicaldeficit itself, because congenital diseases such as CHcan generate, in parents, a sense of guilt and feelings offrustration and disappointment over the “loss” of their ide-alised “perfect child” (5,9).Unlike other chronic diseases, CH is a latent condition,and while it can be treated, it cannot be cured. Further-more, its onset cannot be prevented, and it can occurabruptly at any moment in life. The main fear of the par-ents of subjects with early-treated CH, and of the childrenthemselves, is that the thyroid deficiency will result in im-paired growth. Parents view this disease as a threat, acondition that will sooner or later reveal their child to bedifferent from his peers (10). And however effective treat-ment may be, chronic diseases are associated with anongoing need for medical therapy and monitoring that,while protecting the patient against the manifestations ofthe disease, also involves the prospect of endless depen-dence (11). The present study analyses the behaviour of 30 early-treated CH subjects (treatment begun within the firstmonth of life) and of a group of 116 age-matched normalcontrols in order to identify possible risk factors for psychi-atric disturbances in subjects with early-treated CH. Thisstudy also allowed us to evaluate the possible age at on-set of behavioural disorders in these subjects.

Materials and methods

Subjects

Our study group was composed of 12 CH children (4males and 8 females) and 18 CH adolescents (6 males,

Functional Neurology 2003; 18(3): 161-164 161

12 females), who had started treatment between the20th and the 30 th day of life and who underwent regularfollow ups. They were compared, respectively, with 58healthy children (27 males and 31 females) and with 58healthy adolescents (30 males, 28 females). Recruitedfrom schools, these normal control subjects presentedno chronic illness (e.g., diabetes) or history of perinataldistress.The CH children had a mean age of 10 years 4 monthsand the control group children of 10 years 1 month. Thecorresponding values in the CH adolescents and ado-lescent controls were 13 years 5 months and 13 years3 months. The Wechsler Intelligence Scale for Children-Revised(WISC-R) was used to assess IQ both in the CH sub-jects and in the normal controls. The mean WISC-Rscore in the CH children was 92, versus 96 in the con-trol group, while the mean WISC-R score recorded inthe CH adolescents was 94, as opposed to 99 in thecontrol group. The pharmacological treatment regimens of the CHsubjects were determined by the Department of En-docrinology of the University of Pisa, while neuropsy-chological aspects were investigated by the Division ofChild Neurology and Psychiatry of the IRCCS StellaMaris Institute, University of Pisa.

Evaluation

The behaviour of both the patients and the normal con-trols was evaluated using a specific diagnostic instru-ment: Achenbach’s and Edelbrock’s Child BehaviourChecklist (CBCL) (12). The CBCL, which is designed tobe filled in by parents, is a structured questionnaire bro-ken down into 8 behavioural scales exploring 8 cross-informant syndromes: I = withdrawal; II = somatic com-plaints; III = anxiety/depression; IV= social problems, V = thought problems, VI = attention problems; VII =delinquent behaviour; and VIII = aggressive behaviour.(The questionnaire asks, for example, whether thechild/adolescent requires too much attention, shows off,is silent and irritable, and disobeys parents and teach-ers).The sum of the first three scales gives the InternalisingScore (IS) and that of the last two scales the Externalis-ing Score (ES), while the sum of the scores reported inall the scales is the Total Score (TS). Pathological scores do not constitute the basis for apsychiatric diagnosis, rather they indicate the behav-ioural areas in which parents experience the greatestproblems. In order to establish an overall level, the CBCL was ap-plied to all the participants. Moreover, a structured clinical interview, the Diagnos-tic Interview for Children and Adolescents - Revised (DICA-R) was administered to the 18 adolescents withearly-treated CH (aged>12), in order to determine thepresence of psychopathological disturbances (13,14).The DICA-R (15) is organised in such a way as to gath-er information about the onset of the disease, its dura-tion, and the severity of the symptoms in six differentselected areas: relational disorders, school difficulties,learning difficulties, somatic symptoms, neurotic symp-toms and psychotic symptoms.

Results

The data were analysed using Student’s T test (withp<0.05 taken as the level of significance).The mean scores recorded in the CBCL scales and inthe IS, ES and TS by the CH children and normal con-trols were compared.A significant group interaction emerged only in thedelinquent behaviour scale (T=1,927; p<0.05), in whichthe children with early-treated CH gave higher scores. No significant differences (p>0.05) were found in theother seven scales: withdrawal (T=1,067), somaticcomplaints (T=0,118), anxiety/depression (T=0,878),social problems (T=0,440), thought problems(T=0,165), attention problems (T=0,007), and aggres-sive behaviour (T=0,924).No significant differences (p>0.05) were noted in IS(T=0,254), ES (T=1,277) or TS (T=0,749) (Fig.s 1 and2).The mean scores reported in the CBCL scales and inthe IS, ES and TS by the early-treated CH adolescentsand normal controls were then compared.Significant group interactions were found in the with-drawal (T=1,913; p<0.05), anxiety/depression(T=2,516; p<0.02), thought problems (T=1,750;p<0.05), attention problems (T=2,665; p<0.01), and ag-gressive behaviour (T=2,096; p<0.05) scales, the ado-lescents with early-treated CH recording higher scoresin these scales. In addition, both the IS (T=2,774;

162 Functional Neurology 2003; 18(3): 161-164

F. Tinelli et al.

F i g u r e 1 - Description of behaviour in children with early-treatedCH and controls.

F i g u r e 2 - Description of the Internalising Score (IS), Externalisingscore (ES) andT o t a l Score (TS) in adolescents and children withearly-treated CH and controls.

Internalisingscore

Externalisingscore

Total score

p<0.01) and the TS (T=2,581; p<0.02) were higher inthe study group versus the controls. No significant dif-ferences were noted in the following scales: somaticcomplaints (T=1,738; p>0.05), social problems(T=0,914; p>0.05), delinquent behaviour (T=1,501;p>0.05) (Fig. 3) and ES (T=1,512; p>0.05). Both groups recorded IS, ES and TS that were withinthe normal range (Fig. 2).The DICA-R interview was administered to 18 adoles-cents (aged >12) with early-treated CH; 5 out of the 18subjects did not meet the inclusion criteria for any psy-chopathological disturbance, while 13 did: 7 femalesand 1 male showed anxiety disorder symptoms, 3 fe-males showed mood disorders, and 2 females showedbehavioural disorders (Fig. 4). Figure 4 summarises thepsychopathological disturbances emerging, on DICA-R,in the 18 early-treated CH adolescents, the percentagevalues representing the ratio of affected CH adoles-cents to the total number of CH adolescents: anxietydisorders, 44%; mood disorders, 16%; behavioural dis-orders, 11%; and no disorders, 29%).

Discussion

In accordance with literature data, the children and ado-lescents with early-treated CH investigated in this studyshowed more behavioural problems than their age-matched controls. In the children (< 12 years), a statisti-cally significant difference versus the controls emergedonly in their higher delinquent behaviour score, whilethe adolescents (> 12 years) gave, on the CBCL, signif-

icantly higher scores compared with controls in thewithdrawal, anxiety/depression, thought problems, at-tention problems and aggressive behaviour scales. Inthe DICA-R, 44% of adolescents with early-treated CHshowed symptoms of anxiety disorder, in particular,separation anxiety disorder with phobic components;16% showed mood disorder and depression and 11%showed behavioural disorders with attention deficit. A previous study (16) of chronic disease also reportedhow some affected subjects showed marked psy-chopathological traits.It can be hypothesised that these problems are attribut-able to family mechanisms created by the chronic ill-ness.As Rigardetto reports (17), patients with early-treatedCH are characterised by aggressive behaviour, instabil-ity, inhibition, excessive emotionality and anxiety crises.Rigardetto also asserts that affective disorders depend,to an extent, on excessive parental control and on theinferiority complex developed by the child himself, as aresult of his illness.In this study, children with early-treated CH showed be-havioural problems in the externalising area, especiallyin the area of delinquency (telling lies, stealing, callingothers names, playing truant, refusing to accept blame).Adolescents with early-treated CH showed different ex-ternalising disorders, such as aggressiveness (arguing,demanding too much attention, showing off, being silentand irritable, disobeying parents and teachers) and, assome other authors have reported (18,19), serious at-tention difficulties.Moreover, the higher scores in the CBCL withdrawal,anxiety/depression, and social problems scales indicatethe presence of mental suffering in these patients. TheDICA-R interview, showing 13 adolescents with early-treated CH to have anxiety disorders and/or mood dis-orders and/or behavioural disorders, confirmed thesedata. Our findings indicate that adolescents with early-treatedCH have more behavioural problems than the unaffect-ed, age-matched population. Several studies have sug-gested that an early (pre- and neonatal) hormonaldeficit (endogenous and/or exogenous) may influencethe psychological development of subjects with CH: el-evated levels of LT4 at birth lead more commonly to be-havioural problems reflecting anxiety, social withdrawaland poor concentration (20,21). In our sample, giventhat no hormone alterations emerged on frequentdosage during endocrinological follow up, it is very like-ly that the environmental component was the more im-portant. It can be hypothesised that in most cases thebehavioural difficulties identified result from the mecha-nisms that become established within the family when achronic disease is diagnosed.CH is a chronic disease associated with the need forcontinuous medical intervention and frequent monitor-ing; replacement therapy protects children and adoles-cents from the manifestations of the disease, but cre-ates a state of dependence. In fact, chronic illness in children and adolescents rep-resents a risk factor for the individual’s psychologicaldevelopment, because it can affect both the self imageand the social responses of the patient. Studies of otherchronic diseases claim that chronicity, while not neces-sarily a risk factor for mental retardation, might increase

Functional Neurology 2003; 18(3): 161-164 163

Behavioural disorders in early-treated CH

F i g u r e 3 - Description of behaviour in adolescents with early-treated CH and controls.

F i g u r e 4 - Description of psychopathological syndromes at DICA-R in18 early-treated CH adolescents.

the risk of psychopathological problems instead. In chil-dren and adolescents with insulin-dependent diabetesmellitus, for example, a clear and direct relationship be-tween psychopathological disorders (anxiety disorder,mood disorder and behaviour disorder) and chronic dis-eases, significantly more frequent than in the unaffect-ed population, has been reported (16).CH is also a subclinical disease whose few clinicalsigns are controlled by replacement therapy, but thedisease can nevertheless result in impaired growth,neurological deficits or mental retardation later in life.The worries and anxiety of the parents of a child oradolescent with a congenital, chronic and possibly in-validating disease could generate, in the affected indi-vidual, emotional distress, a lack of self-confidence,and anxious or depressed behaviour. Some parentsmay even ignore the problem, paying little attention tothe emotional well-being of their child and showing poorcompliance with treatment.It is also likely that young patients, aware of their condi-tion, are more burdened than their normal peers byworries over their future, as well as by the need fortherapy and medical checks. In particular, while the behavioural problems of youngerchildren might be interpreted as a reaction, adoles-cents, older and more sensitive, show more elaborateand complex internalising problems, linked to their“poor” self image.It seems crucial, therefore, to underline the importanceof considering behavioural aspects during the follow-upof children and adolescents with early-treated CH, in or-der to identify early psychopathological problems andpossibly to provide the patient and his family with earlypsychological support.

Acknowledgments

The authors acknowledge and thank Prof. A. Pinchera,Prof. L. Chiovato and all their colleagues at the Depart-ment of Endocrinology of the University of Pisa for theircontribution to this paper.

References

11. Dussault JH, Letarte J, Glorieux J et al. Psychological de-velopment of hypothyroid infants at age 12 and 18months: experience after neonatal screening. In: BurrowGN, Dussault JH eds Neonatal Thyroid Screening. NewYork; Raven Press 1980:271-276

12. Berenbaum SA. Neuropsychological follow-up in neonatalscreening: issues, methods and findings. Acta PaediatrSuppl 1999;432:83-87

13. Fuggle PW, Grant DB, Smith I, Murphy G. Intelligence,motor skills and behaviour at 5 years in early-treated con-genital hypothyroidism. Eur J Pediatr 1991;150:570-574

14. Kooistra L, van der Meere JJ, Vulsma T, Kalverboer AF.Sustained attention problems in children with early treated

congenital hypothyroidism. Acta Paediatr 1995;85:425-429

15. Marcheschi M, Bargagna S, Dinetti D, Giachetti C, Romoli-ni E. Dinamiche familiari e rischi evolutivi nell’ipotiroidismocongenito precocemente trattato. Psichiatria dell’infanziae dell’adolescenza1996;63:9-18

16. Spencer T, Biederman J, Wilens T, Guite J, Harding M.ADHD and thyroid abnormalities: a research note. J ChildPsychol Psychiatry 1995; 36:879-885

17. Rovet J, Alvarez M. Thyroid hormone and attention incongenital hypothyroidism. J Pediatr Endocrinol Metab1996;9:63-66

18. Rovet JF, Ehrlich RM, Sorbara DL. Neurodevelopmentaloutcome in infants and preschool children following new-born screening for congenital hypothyroidism. J PediatrPsychol 1992;17:187-213

19. Stawski M, Auerbach JG, Barash M, Lerner Y, Zimin R.Behavioural problems of children with chronic physical ill-ness and their siblings. European Journal of Childhoodand Adolescent Psychiatry1997;6:20-25

10. Rondanini GF, Cerabolini R, Ongari B, Focarile F, BollatiA, Chiumello G. Ipotiroidismo congenito: nuove problem-atiche psicologiche legate alla precocità della diagnosi.Minerva Pediatr 1987;39:457-461

11. Salerno MC, Militerni R, Di Maio S, Bravaccio C, GaspariniN, Tenore A. Intellectual outcome at 12 years of age in con-genital hypothyroidism. Eur J Endocrinol 1999;141:105-110

12. Achenbach TM, Edelbrock C. Manual for the Child Behav-ior Checklist and Revised Child Behavior Profile. Burling-ton, VT; University of Vermont, Department of Psychiatry,1983

13. Welma Z, Reich MD, Hervanic B et al. Reliability, validityand parent-child agreement studies of the Diagnostic In-terview for Children and Adolescents (DICA). J Am AcadChild Adolesc Psychiatry 1987;26:649-653

14. Ezpeleta L, de la Osa N, Domenech JM, Navarro JB,Losilla JM, Judez J. Diagnostic agreement between clini-cians and the Diagnostic Interview for Children and Ado-lescents -DICA-R- in an outpatient sample. J Child Psy-chol Psychiatry 1997;38:431-440

15. Shaffer D, Lucas CP, Richters JE, Richters J eds Diag-nostic Assessement in Child and Adolescent Psy-chopathology. New York; Guilford Press 1999

16. Bargagna S, Tosi B, Calisti L, Crespini L. Caratteristichepsicopatologiche in un gruppo di bambini e adolescenticon diabete mellito insulino dipendente. Minerva Pediatr1997;49:71-77

17. Rigardetto R. Lo sviluppo neuropsichico degli ipotiroidei.Minerva Pediatr 1980;32:1325-1330

18. Rovet J, Alvarez M. Thyroid hormone and attention inschool-age children with congenital hypothyroidism. JChild Psychol Psychiatry 1996;37:579-585

19. Rovet J, Hepworth S. Attention problems in adolescentswith congenital hypothyroidism: a multicomponentialanalysis. J Int Neuropsychol Soc 2001;7:734-744

20. Dattani M, Brook CG. Outcomes of neonatal screening forcongenital hypothyroidism. Curr Opin Pediatr 1996;8:389-395

21. Rovet JF. Long-term neuropsychological sequelae of ear-ly-treated congenital hypothyroidism: effects in adoles-cence. Acta Paediatr Suppl 1999; 432: 88-95

164 Functional Neurology 2003; 18(3): 161-164

F. Tinelli et al.