UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

Professor Esa Hohtola

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Director Sinikka Eskelinen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1125-1 (Paperback)ISBN 978-952-62-1126-8 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1345

ACTA

Fanni Päkkilä

OULU 2016

D 1345

Fanni Päkkilä

THYROID FUNCTION OF MOTHER AND CHILD AND THEIR IMPACT ON THE CHILD’S NEUROPSYCHOLOGICAL DEVELOPMENT

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF MEDICINE;MEDICAL RESEARCH CENTER OULU;NATIONAL INSTITUTE FOR HEALTH AND WELFARE

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 1 3 4 5

FANNI PÄKKILÄ

THYROID FUNCTION OF MOTHER AND CHILD AND THEIR IMPACT ONTHE CHILD’S NEUROPSYCHOLOGICAL DEVELOPMENT

Academic dissertation to be presented with the assentof the Doctoral Training Committee of Health andBiosciences of the University of Oulu for public defencein Auditorium 4 of Oulu University Hospital, on 8 April2016, at 12 noon

UNIVERSITY OF OULU, OULU 2016

Copyright © 2016Acta Univ. Oul. D 1345, 2016

Supervised byDoctor Eila SuvantoDocent Tuija MännistöDocent Anneli Pouta

Reviewed byDocent Saara MetsoProfessor Harri Niinikoski

ISBN 978-952-62-1125-1 (Paperback)ISBN 978-952-62-1126-8 (PDF)

ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)

Cover DesignRaimo Ahonen

JUVENES PRINTTAMPERE 2016

OpponentProfessor Alexander Stagnaro-Green

Päkkilä, Fanni, Thyroid function of mother and child and their impact on thechild’s neuropsychological development. University of Oulu Graduate School; University of Oulu, Faculty of Medicine; MedicalResearch Center Oulu; National Institute for Health and WelfareActa Univ. Oul. D 1345, 2016University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland

Abstract

Maternal gestational thyroid dysfunction has been associated with adverse neuropsychologicaldevelopment in children. This study investigated the effects of maternal thyroid dysfunction inearly pregnancy and/or antibodies on the thyroid function and antibody status of children, as wellas their association with the offspring’s ADHD symptoms, scholastic performance and sensorydevelopment.

The study population consisted of the Northern Finland Birth Cohort of 1986. The mothers’TSH, fT4 and TPO-Ab concentrations were evaluated in early pregnancy and in their offspring at16 years of age. Data on the mothers and their families, their child’s health, development, behaviorand scholastic performance were collected via parental questionnaires conducted in earlypregnancy and when the children were 7-8 and 16 years old. Their teachers evaluated thechildren’s behavior and scholastic performance at 8 years of age, and at 16 years old theadolescents evaluated themselves.

Maternal gestational thyroid dysfunction associated with adolescents’ increased odds ofhaving the same thyroid dysfunction type. Adolescents of TPO-Ab-positive mothers had increasedodds of being TPO-Ab-positive themselves. TPO-Ab-positive children had increased odds ofhaving thyroid dysfunction. Increasing maternal TSH concentrations increased a child’s odds ofhaving ADHD symptoms (OR 1.4 [95% CI 1.1-1.8]). Children of hypothyroxinemic mothers hadincreased odds of repeating a class at school (OR 3.5 [1.1-11.5]), and those of hyperthyroidmothers had increased odds of Finnish language learning difficulties (1.6 [1.03-2.4]).Furthermore, thyroid dysfunction in adolescents increased their odds of learning difficulties. Noassociation was observed between maternal thyroid dysfunction and a child’s diagnosedintellectual deficiency and sensory development.

Maternal thyroid dysfunction during pregnancy associated with thyroid dysfunction in theoffspring. Maternal thyroid dysfunction may have a mild impact on her offspring’sneuropsychological development, but it had no effect on a child’s risk of diagnosed intellectualdeficiency or sensory development. Children have compensatory mechanisms for overcomingearly developmental thyroid hormone insufficiencies. Randomized trials for screening andtreating maternal thyroid dysfunction during pregnancy are needed to evaluate the benefits tooffspring.

Keywords: ADHD, neuropsychological development, offspring, pregnancy, sensorydevelopment, Thyroid - pregnancy, Thyroid hormones, Thyroid – diseases

Päkkilä, Fanni, Äidin ja lapsen kilpirauhastoiminnan vaikutus lapsen neuro-psykologiseen kehitykseen. Oulun yliopiston tutkijakoulu; Oulun yliopisto, Lääketieteellinen tiedekunta; Medical ResearchCenter Oulu; Terveyden ja hyvinvoinnin laitosActa Univ. Oul. D 1345, 2016Oulun yliopisto, PL 8000, 90014 Oulun yliopisto

Tiivistelmä

Äidin raskauden aikaiset kilpirauhasen toimintahäiriöt on yhdistetty lapsen neuropsykologisenkehityksen ongelmiin, mutta aiempi tutkimustieto aiheesta on ristiriitaista. Tämän vuoksi tut-kimme äidin raskauden ajan kilpirauhasen toimintahäiriöiden ja/tai vasta-aineiden vaikutustanuoren kilpirauhastoimintaan ja vasta-ainestatukseen, ja näiden molempien vaikutusta lapsenADHD-oireisiin, koulumenestykseen ja aistien kehitykseen.

Tämän väitöskirjatyön aineistona oli väestöpohjainen Pohjois-Suomen syntymäkohortti1986, johon kuuluu yli 99 % alueen raskaana olevista naisista. Äitien TSH, T4-V ja TPO-Ab –mittaukset tehtiin alkuraskaudessa ja kohortin lasten mittaukset 16-vuotiaana. Molempien koh-dalla käytettiin väestöpohjaisia viitevälejä toimintahäiriön määrittämiseksi. Tietoja raskaudesta,äidin ja muun perheen sairastavuudesta, elintavoista ja sosioekonomisista tekijöistä ja lapsen ter-veydestä, kehityksestä, koulumenestyksestä ja käyttäytymisestä kerättiin kyselylomakkeilla ras-kauden aikana, 7-8-vuotiaana ja 16-vuotiaana. Myös luokanopettajat arvioivat lapsen koulume-nestystä ja käyttäytymistä, ja nuoret itse arvioivat koulumenestystään 16-vuotiaina.

Äidin raskauden aikainen kilpirauhasen toimintahäiriö nosti nuoren riskiä saada sama kilpi-rauhasen toimintahäiriö kuin äidillään. Äidin TPO-vasta-aine-positiivisuus nosti nuoren riskiävasta-ainepositiivisuuteen. Nuoren positiiviset vasta-ainepitoisuudet nostivat riskiä poikkeavillekilpirauhasarvoille. Äidin nouseva TSH-pitoisuus yhdistyi lapsen suurempaan riskiin saadaADHD oireita 8-vuotiaana, mutta selkeää raja-arvoa sille ei löytynyt. Äidin hypo- tai hyperty-reoosi eivät nostaneet lapsen ADHD-oireiden riskiä. Äidin kilpirauhastoimintahäiriöt nostivathieman nuoren riskiä oppimisvaikeuksille ja luokan kertaamiselle. Myös nuoren oma kilpirau-hastoiminta vaikutti vähäisessä määrin oppimiseen ja keskittymiseen. Äidin kilpirauhastoimin-nalla ei ollut vaikutusta lapsen matalaan älykkyysosamäärään tai aistien kehitykseen

Äidin raskaudenaikainen kilpirauhasen toimintahäiriö vaikutti lapsen neuropsykologiseenkehitykseen lievästi, mutta löydösten kliininen merkitys on vähäinen. Lasten keskushermostonkorjaavat mekanismit todennäköisesti kompensoivat varhaiskehityksen kilpirauhashormonienvajetta. Randomoidulla tutkimuksella voitaisiin selvittää, hyötyisivätkö lapset äidin kilpirauhas-sairauden seulomisesta ja hoitamisesta alkuraskaudessa.

Asiasanat: ADHD, aistien kehitys, kilpirauhanen – raskaus, kilpirauhanen – sairaudet,neuropsykologinen kehitys

To my beloved ones

8

9

Acknowledgements

This study was conducted in collaboration with the Department of Obstetrics and

Gynecology, the Institute of Health Sciences, the National Institute for Health and

Welfare and the Department of Child Psychiatry. A large amount of previous work

had been conducted using the 1986 Northern Finland Birth Cohort before I started

my research, and I am grateful to be able to use and learn from previous research.

I wish to especially thank our “cohort mothers,” professors Paula Rantakallio,

Anna-Liisa Hartikainen and Marjo-Riitta Järvelin for, firstly, developing the cohort,

and, secondly, for still being actively involved in cohort research and advising

young students.

I heartily thank my thesis supervisors, Dr. Eila Suvanto, M.D., Ph.D., and

docent Tuija Männistö for their continuous support, whether it was local or from

another continent. Their enthusiasm and kindness were my sources of inspiration

and encouraged me to combine research to early medical studies.

Committing to this project was easy because of our supportive research team.

Professor Anna-Liisa Hartikainen gave me valuable advice, and I always admired

her attitude towards the production of science. I thank docent Anneli Pouta for her

support and enabling this project in its early steps. I thank docent Marja Vääräsmäki

for her help and comments with the manuscripts, Professor Aimo Ruokonen for his

contribution from a laboratory medicine perspective, Aini Bloigu, B.Sc., for her

help in statistics and for always having a positive attitude and Heljä-Marja Surcel,

Ph.D., for her help in the laboratory analysis and for having me as a summer worker

in her laboratory. Finally, special thanks to Professor Irma Moilanen, who always

fitted my project into her busy schedule and with whom I had constructive

discussions on child psychiatry. She always read my work with care and improved

my manuscripts.

I thank the reviewers of this thesis, docent Saara Metso and Professor Harri

Niinikoski. Special thanks to my follow-up group, docent Laure Morin-Papunen

and docent Marjo Renko, for the early coffees and discussions on balancing

research and everyday life.

I also thank the laboratory staff of the National Institute for Health and Welfare,

Aljona Amelina and Jenna Aavavirta, for helping me with the adolescents’ serum

analyses. Thanks to Tuula Ylitalo and Sarianna Vaara for their help with cohort data.

In this study, I was able to use variables created by Tuula Hurtig, Ph.D., Tanja

Nordström, Ph.D., and Ulla Heikura, Ph.D., to whom I am grateful for their

valuable work.

10

I am blessed to have had many people help me along the way. Especially, thank

you, Erika, Jenni, Marja and Senja for just being there. Pekka, you gave me courage

to really get involved in research because together, new things just are not as scary

as alone. Thank you for your friendship. Thank you, Johanna, Hanne and Kreetta

for not talking about medicine. My love and gratitude to my sister, Jenni, and

mother, Paula. Thank you, Jussi for letting me have my peace and quiet at the

summer cabin, and father, Kari, and brother, Jarkko, for keeping me company while

picking berries and relaxing. My special love and gratitude to Petri, who kept me

sane while I finished this project.

This work was financially supported in part by grants from the Academy of

Finland, the Alma and K.A. Snellman Foundation (Oulu, Finland), the Jalmari and

Rauha Ahokas Foundation (Finland), the Northern Ostrobothnia Hospital District

(Finland), the Finnish Medical Foundation and the Finnish Medical Association of

Clinical Chemistry.

Oulu, April 2016 Fanni Päkkilä

11

Abbreviations

ADHD attention deficit hyperactivity disorder

ADHD-RS-IV ADHD rating scale IV

aOR adjusted odds ratio

ATA American Thyroid Association

BMI body mass index

CBCL Child Behavior Check List

CH congenital hypothyroidism

CNS central nervous system

DSM Diagnostic and Statistical Manual of Mental

Disorders

D1-3 type I-III deiodinase

ETA European Thyroid Association

fT4 free thyroxine

hCG human chorionic gonadotropin

ICD International Classification of Diseases

IQ intelligence quotient

NFBC 1986 Northern Finland Birth Cohort 1986

SD standard deviation

SDQ strengths and difficulties questionnaire

SES socioeconomic status

SWAN The strengths and weaknesses of ADHD symptoms

and normal behavior - scale

T3 triiodothyronine

T4 thyroxine

TBG thyroxine-binding globulin

TG thyroglobulin

TG-Ab thyroglobulin antibody

THOP neonatal hypothyroxinemia of prematurity

TPO thyroid peroxidase

TPO-Ab thyroid peroxidase antibody

TR thyroid hormone receptor

TRH thyroid-releasing hormone

TSH thyroid-stimulating hormone, thyrotropin

TSHRAb TSH-receptor antibodies

WHO World Health Organization

12

13

List of original articles

This thesis is based on the following publications, which are referred to throughout

the text by their Roman numerals:

I Päkkilä F, Männistö T, Bloigu A, Vääräsmäki M, Hartikainen A-L, Ruokonen A, Pouta A, Järvelin M-R, Surcel H-M & Suvanto E (2013) Maternal thyroid dysfunction during pregnancy and thyroid function of her child in adolescence. J Clin Endocrinol Metab 98(3): 965−972.

II Päkkilä F, Männistö T, Pouta A, Hartikainen A-L, Ruokonen A, Surcel H-M, Bloigu A, Vääräsmäki M, Järvelin M-R, Moilanen I & Suvanto E (2014) The impact of maternal thyroid hormone concentrations during pregnancy on the ADHD symptoms of the child. J Clin Endocrinol Metab 99(1): E1−8.

III Päkkilä F, Männistö T, Hartikainen A-L, Ruokonen A, Surcel H-M, Bloigu A, Vääräsmäki M, Järvelin M-R, Moilanen I & Suvanto E (2015) Maternal and child’s thyroid function and child’s intellect and scholastic performance. Thyroid 25(12):1363−74.

IV Päkkilä F, Männistö T, Hartikainen A-L & Suvanto E. The association between maternal thyroid function during pregnancy and a child’s linguistic and sensory development. Manuscript.

14

15

Contents

Abstract

Tiivistelmä

Acknowledgements 9 Abbreviations 11 List of original articles 13 Contents 15 1 Introduction 19 2 Review of the literature 21

2.1 Anatomy and physiology of the thyroid gland ........................................ 21 2.1.1 Thyroid hormone production and metabolism ............................. 21 2.1.2 Effects of thyroid hormones on tissue level ................................. 22 2.1.3 Fetal thyroid gland development .................................................. 23

2.2 Thyroid dysfunction ................................................................................ 23 2.2.1 Hypothyroidism ............................................................................ 23 2.2.2 Hyperthyroidism ........................................................................... 26 2.2.3 Thyroid dysfunction in the pediatric population .......................... 26

2.3 Maternal thyroid function during pregnancy .......................................... 28 2.3.1 Measuring thyroid function during pregnancy ............................. 31 2.3.2 Maternal hypothyroidism ............................................................. 32 2.3.3 Maternal hypothyroxinemia ......................................................... 33 2.3.4 Maternal hyperthyroidism ............................................................ 33

2.4 Child’s neuropsychological development and its relation to

maternal thyroid function ........................................................................ 34 2.4.1 Early development of the fetal central nervous system ................ 34 2.4.2 Thyroid hormones and brain development ................................... 37 2.4.3 Intellectual disability .................................................................... 38 2.4.4 Attention deficit hyperactivity disorder ........................................ 40

2.5 The effects of maternal thyroid dysfunction during pregnancy on

child neuropsychological and sensory development ............................... 42 2.5.1 Effects of maternal hypothyroidism on child

neuropsychological development ................................................. 42 2.5.2 Effects of maternal hypothyroidism on child sensory and

motor development ....................................................................... 45 2.5.3 Effects of maternal hypothyroxinemia on child

neuropsychological development ................................................. 46

16

2.5.4 Effects of maternal hypothyroxinemia on child sensory

development ................................................................................. 49 2.5.5 Effects of maternal hyperthyroidism on child

neuropsychological development ................................................. 50 2.5.6 Screening for maternal thyroid dysfunction during

pregnancy ..................................................................................... 50 3 Purpose of the present study 53 4 Subjects and methods 55

4.1 Subjects ................................................................................................... 55 4.1.1 The 1986 Northern Finland Birth Cohort ..................................... 55 4.1.2 Finnish Maternity Cohort ............................................................. 56

4.2 Methods ................................................................................................... 56 4.2.1 Thyroid function analysis of the NFBC 1986 mothers ................. 56 4.2.2 Thyroid function analysis of the NFBC 1986 children (I,

III) ................................................................................................. 58 4.2.3 Assessment of ADHD symptoms of the NFBC 1986

children (II) ................................................................................... 59 4.2.4 Scholastic performance of the NFBC 1986 children (III) ............ 60 4.2.5 Assessment of intellectual problems in the NFBC 1986

(III) ............................................................................................... 61 4.2.6 Assessment of sensory development (IV) .................................... 62 4.2.7 Statistical methods ........................................................................ 64

5 Results and comments 67 5.1 Maternal and adolescent thyroid function (I) .......................................... 67

5.1.1 Association of maternal and adolescent’s thyroid function .......... 67 5.1.2 Comments on maternal and adolescent thyroid function .............. 71 5.1.3 Reference intervals for serum TSH and fT4 concentrations

in the adolescent population and thyroid function in

adolescence ................................................................................... 72 5.1.4 Thyroid autoimmunity in the adolescent population .................... 73 5.1.5 Comments on the adolescent thyroid function analysis ................ 75

5.2 The association between maternal thyroid dysfunction and

ADHD symptoms in children (II) ........................................................... 75 5.2.1 Comments ..................................................................................... 78

5.3 Maternal and adolescent thyroid function and their association

on scholastic performance and intellect (III) ........................................... 79

17

5.3.1 Maternal thyroid dysfunction during early pregnancy and

the scholastic performance of children (III) ................................. 79 5.3.2 Maternal thyroid function and severe intellectual

deficiency/mild cognitive limitation in children .......................... 84 5.3.3 Abnormal adolescent thyroid function parameters at 16

years of age and self-evaluated scholastic performance ............... 85 5.3.4 Comments ..................................................................................... 87

5.4 The effect of maternal thyroid dysfunction on child sensory and

linguistic development (IV) .................................................................... 91 5.4.1 Comments ..................................................................................... 93

6 Conclusions 95 7 General discussion 97

7.1 Strengths and limitations of the present study ........................................ 97 7.2 Confounding factors ................................................................................ 98 7.3 Clinical significance of the results ........................................................ 100

References 103 Original publications 115

18

19

1 Introduction

Maternal thyroid hormone production during pregnancy is essential for a child’s

normal growth and development; a developing fetus needs maternal thyroid

hormones before its own thyroid hormone production begins. Maternal thyroid

hormone supply to the fetus may be insufficient because of an iodine deficiency in

the area or because of a limited thyroidal reserve due to a disease, for example

(Morreale de Escobar 2001).

Thyroid diseases are common in women of childbearing age. Up to

approximately 5-10% of non-pregnant and 5% of pregnant women have thyroid

dysfunction or thyroid autoantibodies (Stagnaro-Green et al. 2011). Thyroid

diseases have adverse effects on women’s overall reproductive health; for example,

women with untreated hypothyroidism (underactive thyroid) may not even become

pregnant. They also have an increased risk of pregnancy complications such as

miscarriages (Abalovich et al. 2002). The consequences of severe maternal thyroid

hormone deficiencies during pregnancy on children have long been known:

cretinism is a notorious condition, which includes intellectual deficiency, mutism,

muscle weakness, squint, thyroid dysfunction and growth delay (Man & Jones

1969).

Due to the drastic consequences of severe untreated hypothyroidism, nowadays

women with a known thyroid disease are treated with care during pregnancy

(Stagnaro-Green et al. 2011). In addition, almost all countries have a iodine

supplementation program and Finland has had on successfully since the 1940s

(Lamberg 1986). However, a growing number of studies suggest that even mild

maternal thyroid dysfunction during pregnancy or maternal iodine deficiency might

adversely affect a child’s neuropsychological development, for example by

increasing the child’s risk of attention-deficit and hyperactivity disorder (ADHD)

(Ghassabian et al. 2012, Modesto et al. 2015), cognitive functioning problems and

decreased intellect quotient (Haddow et al. 1999, Henrichs et al. 2010, Julvez et al. 2013).

The aim of this study was to evaluate how, in a large population-based birth

cohort in an iodine-sufficient area, maternal thyroid dysfunction and/or

autoantibodies during pregnancy affect a child’s thyroid function in adolescence

and how both affect a child’s behavior, scholastic performance and sensory

functions.

20

21

2 Review of the literature

2.1 Anatomy and physiology of the thyroid gland

The normal thyroid is a small 20-gram endocrine gland in front of the trachea. It

consists of two lobes, and two thyroid arteries and veins are responsible for its

circulation. The tissue is full of follicles, which are the main functional units of the

thyroid gland. The follicles consist of alveoli, a single layer of endothelial cells

surrounding colloid fluid. The colloid includes thyroglobulin (TG), which stores

thyroxin (T4) and triiodothyronine (T3) (Välimäki & Schalin-Jäntti 2010).

2.1.1 Thyroid hormone production and metabolism

The thyroid secretes two hormones, T4, which accounts for most of the daily

secretion, and T3, which is the biologically active hormone in the peripheral tissues.

Most of the peripheral T3 is deiodized from T4. The hormone production requires

iodine, which is obtained from the diet; main sources include milk, dairy products,

eggs, fish and iodized table salt. Iodine is converted to iodide and transported to the

thyroid gland. The thyroid peroxidase (TPO) enzyme incorporates the iodide with

TG in the follicle cell and through a chemical reaction with hydrogen peroxide,

forms monoiodotyrosine and diiodotyrosine, which couple to form T4 and T3. The

hormones are then stored in the colloid as part of the TG and/or excreted through

the circulation (Mullur et al. 2014, Abdalla & Bianco 2014).

In the circulation, thyroid hormones are mostly bound to thyroxine-binding

globulin (TBG). Only 0.02% of T4 and 0.2% of T3 are free and biologically active.

The concentration of TBG may be affected by pregnancy, familial conditions and

estrogen treatment (Glinoer 1997). The thyroid hormones are actively transported

into the target cells. In the cells, free T4 (fT4) is deiodized to active free T3 (fT3)

by deiodinase enzymes. Different tissues have their own deiodinase enzyme types,

and the amount of fT3 production depends on the deiodinase enzyme’s activity.

Type 1 deiodinase (D1) functions in the liver and kidneys and is also in the thyroid.

Its activity increases in hyperthyroidism and decreases in hypothyroidism. Fetal D1

activity is low. Type 2 deiodinase’s (D2) main function is in the central nervous

system (CNS); it is mainly responsible for fT3 production in the brain. Type 3

deiodinase (D3) is also situated in the CNS, but its main function is to inactivate

both fT3 and fT4 in order to maintain the equilibrium of hormone concentrations.

22

Its activity decreases in hypothyroidism and when there is an iodine deficiency

(Gereben et al. 2008). In the fetal and newborn child’s tissues, fT4 is the main

active hormone. In the CNS, however, the active hormone is fT3, which is locally

converted from fT4 in the glia cells by D2 enzymes (Rovet 2014).

The production of thyroid hormones is regulated by the hypothalamic-pituitary

axis. The main thyroid-regulating hormone is thyrotropin (thyroid-stimulating

hormone, TSH) secreted from the pituitary gland. TSH is comprised of two

subunits, and it shares a common alpha-unit with luteinizing hormone, follicle-

stimulating hormone and human chorionic gonadotropin (hCG) but has a specific

beta-unit that separates it from the other hormones. The function of the pituitary is

regulated by the hypothalamus, which excretes the thyrotropin-releasing hormone

(TRH). TRH increases the production of TSH, and its production is inhibited by

dopamine and somatostatin. TSH is excreted via a circadian and pulsatile rhythm;

the highest serum concentrations can be measured at midnight, and serum

concentrations decrease afterwards. The thyroid hormones have a negative

feedback effect on the pituitary and hypothalamus (Fekete & Lechan 2014).

2.1.2 Effects of thyroid hormones on tissue level

Thyroid hormones act via thyroid hormone receptors (TR) in the target cells. TRs

are located in the cell nucleus and mitochondria. Humans have two TR genes (alpha

and beta) that code several different active or inactive receptors. Different tissues

express different genes. The thyroid hormones have mainly anabolic effects on the

tissues and are essential for normal growth. When thyroid hormone concentrations

in the serum are high, catabolic effects appear because thyroid hormones increase

oxygen consumption, the rate of gluconeogenesis, glycolysis and lipolysis (Mullur et al. 2014).

The thyroid hormones play an important role in brain development and also in

the postnatal nervous system. Rodent studies show that thyroid hormones regulate

the genes of fundamental neurobiological processes such as neurogenesis, cell

migration and differentiation, synaptogenesis and myelination. A serious lack of

thyroid hormones during a child’s development leads to cretinism (Rovet 2014,

Schroeder & Privalsky 2014).

23

2.1.3 Fetal thyroid gland development

The development of the thyroid gland in humans begins at approximately day 24

of gestation, and its maturation is divided into two phases. The embryogenesis of

the thyroid gland and the hypothalamic-pituitary-thyroidal axis is the first phase. It

is then followed by the further development of the hypothalamic-pituitary-thyroidal

axis, as well as hormone production and regulation. The tiny thyroid gland first

appears at the base of the tongue, at approximately day 24 of gestation, and it

migrates to its final destination in the neck by 7 weeks. There, it continues to grow

through the rest of gestation. The colloid formation begins around 9-12 weeks, and

iodine uptake begins at the same time. Fetal TSH is produced from the 14th week

of gestation onwards. The synthesis of fetal thyroid hormones begins at about week

20. Regulation of the thyroid by the hypothalamic-pituitary-thyroidal axis begins

in the third trimester, but final adult-like concentrations of thyroid hormones are

not evident until parturition. Because fetal thyroid function begins slowly, the fetus

is dependent on maternal thyroid hormone production, especially in the first

trimester (Szinnai 2014).

2.2 Thyroid dysfunction

2.2.1 Hypothyroidism

Hypothyroidism is a condition in which thyroid underactivity causes fT4 deficiency.

Hypothyroidism is categorized as primary (thyroidal, 95% of cases) and secondary

(of hypothalamus-pituitary origin). Globally, the most important etiological factor

of hypothyroidism is iodine deficiency. In Finland, the most common reasons for

hypothyroidism are chronic autoimmune thyroiditis, previous radioiodine

treatment and thyroid surgery. Symptoms related to hypothyroidism are weight

gain and swelling, fatigue, decreased cold endurance, constipation and bradycardia.

Untreated hypothyroid women may suffer from menstrual irregularities and

infertility. Overt hypothyroidism is diagnosed when TSH is above its normal

reference limits and fT4 is below its limits. The disease is subclinical if TSH is

above reference limits but fT4 levels are still normal. Hypothyroidism is treated

with thyroxine substitution, levothyroxine (Garber et al 2012). In Finland, the usage

of levothyroxine has tripled during the past 15 years, from approximately 100000

users to over 300000 users (web page inquiry from the services of the Finnish

Social Insurance Institute).

24

Autoimmune hypothyroidism

Chronic autoimmune thyroiditis is the most common etiology of hypothyroidism.

In the US, the incidence of chronic autoimmune thyroiditis is about 3.5 per 1000

women and 0.8 per 1000 men (Garber et al. 2012). In the Finnish population, the

prevalence of thyroid peroxidase antibody positivity (TPO-Ab-positivity) is 14%

in the non-pregnant population aged 18 to 91 years (Schalin-Jäntti et al. 2011).

Elevated concentrations of TPO-Abs also exist without changes in TSH and fT4

concentrations. They are thought to increase the individual’s future risk of thyroid

dysfunction (Khan et al 2015). Increased concentrations of TPO-Abs or

thyroglobulin antibodies (TG-Abs) among subjects with hypothyroidism indicate

an autoimmune etiology. However, patients with Graves’ disease (autoimmune

hyperthyroidism) may also have increased levels of TPO-Abs and TG-Abs. The

diseases run in the same families, and a common human leukocyte antigen tissue

haplotype predisposes patients to both autoimmune hypo- and hyperthyroidism,

and sometimes a conversion of the diseases occur. Patients with autoimmune

thyroid diseases also have a higher risk of other autoimmune diseases like

Addison’s disease, rheumatoid arthritis, chronic hepatitis and Sjögren’s syndrome.

In autoimmune hypothyroidism, thyroid tissue is either atrophied and replaced with

connective tissue or enlarged to goiter (Hashimoto’s thyroiditis) (Khan et al. 2015).

Postpartum thyroiditis

Postpartum thyroiditis is a form of autoimmune thyroiditis, which often starts with

a thyrotoxicosis phase 1-3 months postpartum and is spontaneously followed by a

hypothyroid phase. Usually, the patient presents with elevated levels of TPO-Abs.

During the hypothyroid phase, the patient might need levothyroxine treatment. The

patient’s thyroid function should be examined before the next pregnancy because

she has an increased risk of permanent hypothyroidism (Stagnaro-Green and Pearce

2012).

Iodine deficiency

Iodine deficiency is a worldwide problem, which has multiple adverse effects on

growth and development (WHO report on the world wide iodine status in 2007,

http://apps.who.int/iris/bitstream/10665/43781/1/9789241595827_eng.pdf). The

problem has been previously thought to concern mostly developing countries, but

25

industrialized countries have recently been shown to be at risk, too (Andersson et al. 2012). Due to efficient iodine prophylaxis, Finland has been iodine sufficient

since the 1940s (iodine uptake was estimated at 300 µg/day for the whole

population) (Erkkola et al. 1998, Lamberg et al. 1981). Recent national FINRISK

(2012) research, however, suggested a mild decrease in nutritional iodine uptake in

women aged 25-64 years (mean of 190 µg/day [SD of 119 µg/day], when 150

µg/day is a recommended minimum for the non-pregnant population, 175 µg/day

for pregnant women and 200 µg/day for breastfeeding mothers) (Helldán et al. 2012). In the spring of 2015, the National Nutrition Council of Finland gave a

recommendation to use iodized salt in the bakery industry and in institutional

catering services such as in school kitchens (http://www.evira.fi/files/

attachments/fi/vrn/vrn_jodi_ toimenpide-suositus_10.2.2015_suomi.pdf).

Congenital hypothyroidism

Congenital hypothyroidism (CH) is a condition in which a newborn lacks thyroid

hormones. If untreated, CH leads to mental cretinism and severe growth and

development difficulties. The prevalence of CH is about 1/2500 children born alive.

Primary CH is due to ectopy, athyreosis or goiter. It seems that a family history of

unspecified adulthood onset thyroid diseases does not correlate with CH risk.

However, documented maternal thyroid autoimmune disease should be

acknowledged because transplacental passage of TSH receptor autoantibodies

accounts for about 2% of positive results in the CH screening of newborns.

Secondary CH is a central deficit, which almost always associates with other

pituitary hormone deficiencies. The diagnosis of CH is generally based on a

standardized universal screening method of measuring TSH from cord blood (in

Finland) or from dried blood drops, drawn usually at 2-5 days of life, in countries

where newborns are screened for phenylketonuria. A positive screening result leads

to confirmation of the diagnosis, etiological studies, finding optimal treatment and

outcome documentation. The careful monitoring of thyroid hormones is especially

important during the first three years of life because at that time the brain is the

most vulnerable to hypothyroidism. With proper treatment, the growth outcome of

the child is usually normal (Van Vliet & Deladoey 2014).

26

2.2.2 Hyperthyroidism

Thyrotoxicosis means thyroid hormone concentrations are over normal reference

limits in serum, irrespective of the cause. Its subcategory, hyperthyroidism, is a

condition in which thyroid hormone production exceeds normal physiological

needs. The most common etiology of hyperthyroidism is Graves’ disease

(autoimmune hyperthyroidism); other causes include toxic multinodular goiter (the

prevalence is diminishing because of iodine prophylaxis) and toxic adenoma. Other

causes of thyrotoxicosis are subacute thyroiditis and overuse of thyroid hormone

medication, TSH secreting pituitary adenoma and neonatal hyperthyroidism.

Symptoms of hyperthyroidism include nervousness, tachycardia, sweating,

elevated body temperature, diarrhea, weight loss, increased appetite, muscle

tremors and insulin resistance (Muldoon et al. 2014).

Graves’ disease (in the literature also known as Basedow’s disease) is an

autoimmune disease in which autoantibodies towards TSH receptors accelerate

thyroid function, imitating the normal function of TSH. Graves’ disease is a

syndrome that often includes eye and other autoimmune symptoms in the

connective tissue. Usually, the thyroid gland is enlarged. Graves’ disease and

hyperthyroidism from other causes are diagnosed through laboratory analyses:

concentrations of fT4 or fT3 are above normal reference limits, and TSH

concentrations are suppressed. The disease can also be subclinical if fT4

concentrations are still within normal limits. Hyperthyroidism is treated with

antithyroid drugs or other medical treatment, radioiodine therapy or surgery. Some

antithyroid drugs can be used during pregnancy even though they pass through the

placenta, and maternal fT4 levels must be carefully monitored to prevent fetal

hypothyroidism (Muldoon et al. 2014).

2.2.3 Thyroid dysfunction in the pediatric population

Sufficient thyroid function during childhood is essential for normal growth and

development. The most common reason for measuring TSH and fT4 values is

delayed growth and/or excessive weight gain. On the other hand, thyroid function

problems are more common in children with other autoimmune diseases such as

type 1 diabetes, alopecia, vitiligo and celiac disease, as well as in children with

genetic syndromes like Turner syndrome or Down syndrome. These children are

prone to develop Hashimoto’s thyroiditis, which is one of the most common

autoimmune endocrine diseases in the pediatric age group. However, data on its

27

prevalence in the normal pediatric population is scarce (Radetti 2014). One large

study in the USA reported that the prevalence of Hashimoto’s thyroiditis in school

children was 1.2%, but the data is rather old (Rallison et al. 1975). The prevalence

of autoimmune thyroid diseases is estimated to be lower in children than in adults;

for example, the prevalence of Graves’ disease is estimated to be 0.02% in children

versus 0.5-2.0% in adults (Radetti 2014). In Finland in 2013, a total of 2776

children used some form of thyroid medication, and 2741 of them included thyroid

hormones (2739 levothyroxine and 2 liothyronine users) (web inquiry from the

services of the Finnish Social Insurance Institute).

The prevalence of thyroid autoantibodies in different iodine sufficient regions

is presented in Table 1. The prevalence of thyroid autoantibodies in iodine sufficient

regions varies (Wiersinga 2014), but the percentage of children with TPO- or TG-

Abs is higher in girls than boys and seems to increase with age (García-García et al. 2012, Hollowell et al. 2002, Kaloumenou et al. 2008). Interestingly, the

presence of thyroid autoantibodies varied in two different socioeconomic

environments in a study of the neighboring areas of Russian Karelia and eastern

Finland (Kondrashova et al. 2008). The populations in these areas were genetically

relatively similar, but, for example, hygiene conditions differed between them. The

results of that study show that environmental factors also affect thyroid

autoimmunity (Kondrashova et al. 2008). Unfortunately, serum TSH

concentrations were measured only from the majority of thyroid antibody positive

children, and 92% of them were Finnish (Kondrashova et al. 2008). Therefore,

comparison of TSH concentration between countries was not possible in that stydy.

However, data on normal values for TSH and fT4 concentrations in children

without previously documented thyroid diseases, other autoimmune diseases or

symptoms of either, are scarce.

28

Table 1. The prevalence of thyroid autoantibodies in different iodine sufficient regions.

Author/Year Country Type of autoantibodies Children’s age Prevalence (%)

García-García et al.

2012 Spain TPO- and/or TG-Ab

1-6

6-12

12-16

0.6

4.6

6.2

Kaloumenou et al.

2008 Greece TPO- and TG-Ab 5-18 4.6 and 4.3

Kondrashova et al.

2008

Russian Karelia

and Finland TPO- and TG-Ab 7-15

2.6 vs. 0.4 for

TPO-Ab

3.4 vs. 0.6 for TG-

Ab

Hollowell et al. 2002 USA TPO-Ab 12-19 6.7 for girls and

2.9 for boys

2.3 Maternal thyroid function during pregnancy

Pregnancy is a physiological stress test for maternal thyroid function (Pearce 2015).

hCG is a direct stimulator of thyroidal TSH receptors, increasing fT4 production.

During high hCG concentrations at the end of the first trimester (there is a sharp

peak at 10-12 weeks), TSH concentrations are low (Pekonen et al. 1988). Thereafter,

serum TSH concentrations slowly rise in reflection of the increased thyroid

hormone need and decreasing hCG concentrations (Glinoer 1997). The relationship

of serum TSH and hCG concentrations is presented in Figure 1. Maternal serum

TBG concentrations increase due to a rise in estrogen levels from the first

gestational weeks onward. Maternal TBG concentrations reach a plateau around

mid-gestation (Laurell & Rannevik 1979). The increase in TBG concentrations, the

general increase in blood volume related to pregnancy, and maternal thyroid

hormone supply to the fetus lead to a transient drop in maternal free thyroid

hormone levels. This, in turn, leads to TSH-mediated thyroid stimulation in the first

trimester (Ain et al. 1987).

29

Fig. 1. During normal pregnancy, serum TSH and hCG concentrations act as each

other’s mirror images. The figure has been modified from Glinoer 1997.

The production of total circulating maternal thyroid hormones must increase up to

50% (peaking at gestational week 20) to ensure a constant amount of free thyroid

hormones for both the mother and the fetus. If the mother has normal thyroid

function before pregnancy, pregnancy-related physiological changes in thyroid

function lead only to a minor or no TSH increase (Glinoer 1997). Free T4 levels

are technically more difficult to measure during pregnancy than T4 levels because

changes in serum protein concentrations interact with the immunoassays. Free T4

levels, however, seem to rise in the first trimester above the levels of the non-

pregnant population. In the second and third trimesters they are approximately 20-

30% lower compared to the non-pregnant population (Glinoer 1997). However, not

all studies have reported fT4 concentration increases in pregnant populations

(Anckaert et al. 2010, Lee et al. 2009). In the 1986 Northern Finland Birth Cohort

(NFBC 1986), a small rise in maternal fT4 concentrations was seen in the early

gestational weeks and after the 9th gestational week, a small drop followed

(Männistö et al. 2011). As a result, TSH measurement is thought to be the most

reliable method for evaluating maternal thyroid status during pregnancy (Lazarus et al. 2014, Lee et al. 2009, Stagnaro-Green et al. 2011).

Maternal thyroid hormones have been directly measured within different

intrauterine fetal compartments during the first 24 weeks of pregnancy, and high

30

levels of thyroid hormones from the mother existed in both coelomic and amniotic

fetal cavities. These findings signified the transmission of thyroid hormones

through the placenta (Calvo et al. 2002). Maternal thyroxine, which is actively

transported through the placenta and in fetal tissues, is converted into T3 (Patel et al. 2011). Also, maternal TSH levels in early pregnancy and newborn TSH levels

tested from cord blood were correlated in a large population-based cohort study

(Medici et al. 2012). For these reasons, maternal thyroid hormones are important

for the fetus before its own thyroid hormone production begins. Sufficient thyroid

hormone availability for the fetus may have lasting consequences later in life,

which will be discussed in more detail.

The current professional recommendation for treating hypothyroidism during

pregnancy is monotherapy with levothyroxine and the treatment has clear

therapeutic targets (Lazarus et al. 2014, Lee et al. 2009, Stagnaro-Green et al. 2011).

However, a combination therapy of levothyroxine and liothyronine (T3) is

becoming more popular in the non-pregnant population, even though this goes

against formal recommendations (Foeller & Silver 2015). In some studies, this has

been demonstrated to be potentially more beneficial for psychological well-being

in special cases of hypothyroid non-pregnant adult individuals than levothyroxine

monotherapy (Biondi et al. 2012, Panicker et al. 2009). The safety of combination

therapy during pregnancy has not been studied in humans (Foeller & Silver 2015).

In rats with congenital hypothyroidism, maternal infusions of exogenous

liothyronine did not correct fetal brain T3 concentrations, whilst levothyroxine

infusion clearly did. Instead, maternal liothyronine infusions further decreased the

concentration of fetal brain T3 by lowering maternal T4 concentrations via a

negative feedback mechanism (Calvo et al 1990). An additional concern in

liothyronine use during pregnancy is that T3 has a short half-life of one day, and

its concentrations in serum fluctuate easily. Treatment with levothyroxine and

liothyronine also often provides excess amounts of liothyronine because the

commonly used combination drug is derived from porcine thyroid glands. Pigs

secrete T4 and T3 in a ratio of 4:1, which is high compared to the physiological

human ratio of 14:1. The over-supplementation of maternal T3 therefore may

actually lead to decreased maternal and fetal T4 supply (Foeller & Silver 2015).

31

Maternal thyroid function during pregnancy and thyroid function of the

offspring

Maternal autoimmune disease during pregnancy has been connected to transient

congenital hypothyroidism and hyperthyroidism via maternal transplacental TPO-

Abs and TSHR-Abs in several newborn screening program studies (Brown et al. 1996, Matsuura et al. 1980). There are also some studies suggesting that an

individual’s TSH levels are primarily genetically controlled, but levels of fT4 and

fT3 are more affected by environmental influences (Hansen et al. 2004, Panicker et al. 2008). Knowledge of the effects of maternal thyroid function during

pregnancy on the thyroid function of the offspring after newborn age is scarce. A

recent population-based cohort study reported a correlation between maternal TSH

and fT4 concentrations and a child’s thyroid function, assessed at birth and in 5-9

year old children (Korevaar et al. 2015b).

2.3.1 Measuring thyroid function during pregnancy

The international guidelines of the Endocrine Society, the American Thyroid

Association (ATA) and the European Thyroid Association (ETA) recommend using

population-based, assay-specific and trimester-specific reference intervals for

evaluating thyroid function in pregnant women (De Groot et al. 2012, Lazarus et al. 2014, Stagnaro-Green et al. 2011). Reference intervals should be based on the

2.5th and 97.5th percentiles of the respective iodine sufficient population. The

population should consist of healthy, non-selected women having singleton

pregnancies. In this regard, TPO-Ab-negative women with no pre-existing thyroid

disease and who are not taking thyroid medication are usually deemed healthy

(Stagnaro-Green et al. 2011).

If population-based reference intervals are not available, TSH reference

intervals of 0.1-2.5mU/L for the first trimester, 0.2-3.0mU/L for the second

trimester and 0.3-3.0mU/L for the third trimester are recommended by the

international guidelines, irrespective of laboratory methods (Stagnaro-Green et al. 2011). However, in most published studies, population-based upper reference limits

of TSH are higher than the fixed TSH cut-off concentrations of 2.5 and 3.0 mU/L

(Medici et al. 2015). Therefore, the use of population-based reference intervals is

of great clinical importance; the use of fixed TSH cut-offs would lead to the over-

diagnosis and treatment of euthyroid pregnant women. When measuring fT4, the

optimal method is the dialysate or ultrafiltrate of serum samples employing on-line

32

extraction, liquid chromatography or tandem mass spectrometry. When not

available, assay-, method- and trimester-specific reference intervals for fT4 must

be used (Stagnaro-Green et al. 2011).

In northern Finland, laboratories use the ATA’s recommended reference

intervals for the pregnant population (www.nordlab.fi). Our study group has

previously calculated the population-based reference intervals for the 1986

Northern Finland Birth Cohort (NFBC 1986 population) that are used in the current

study (Männistö et al. 2011). Reference intervals for TSH and fT4 in the first

trimester were 0.07-3.1 mU/L and 11.4-22.4 pmol/L and in the second trimester

were 0.10-3.5 mU/L and 11.1-18.9 pmol/L, respectively. Maternal TPO-Ab-

positivity was defined as a TPO-Ab concentration above the 95th percentile (>

167.7 IU/ml) (Männistö et al. 2011).

2.3.2 Maternal hypothyroidism

Approximately 2-3% of healthy, non-pregnant women in the US have an elevated

serum TSH concentration. Of them, 0.3-0.5% would be classified with overt

hypothyroidism and 2-2.5% with subclinical hypothyroidism (Stagnaro-Green et al. 2011). In Finland, using population-based reference intervals and maternal

thyroid function data from an early pregnancy serum sample of the national Finnish

Maternity Cohort, the respective prevalence of subclinical and overt

hypothyroidism during pregnancy is approximately 0.9-1% and 4.0-4.7%

(Gyllenberg et al. 2015, Männistö et al. 2009). Prevalence data on thyroid diseases

during pregnancy based on maternal clinical diagnoses has not been published in

Finland.

Women with TSH concentrations of 10.0 mIU/L or above are considered to

have overt hypothyroidism irrespective of their fT4 concentrations. If local

reference intervals are not available, subclinical hypothyroidism is defined as

having a serum TSH between 2.5 and 10 mIU/L with a normal fT4 concentration.

A euthyroid mother in the first trimester may develop hypothyroidism in later

stages of pregnancy if she has limited thyroidal reserve due to underlying

Hashimoto’s disease, another thyroid disease or a thyroidectomy. Approximately

10-20% of all pregnant women are TPO-Ab- or TG-Ab-positive and euthyroid in

the first trimester (depending on the population), and 16% of these women will

develop subclinical or clinical hypothyroidism by the third trimester. They are also

at risk of postpartum thyroiditis (Stagnaro-Green et al. 2011).

33

Effects on pregnancy

Overt hypothyroidism has been shown to be a risk factor of several pregnancy

complications including premature birth, low birth weight, miscarriage and fetal

death (Abalovich et al. 2002). Treated hypothyroidism seems to increase the risk

of cesarean section (Matalon et al. 2006).

The literature concerning subclinical hypothyroidism and pregnancy outcomes

is more variable because studies differ in populations, laboratory methods, timing

of the serum sampling and in their control of confounding factors. There is data

suggesting that subclinical hypothyroidism increases the risk of pregnancy

complications such as pregnancy loss and preterm delivery in both TPO-Ab-

positive and –negative pregnant women (Negro et al. 2010a, Negro et al. 2010b).

However, there is controversy regarding whether to treat these women (De Groot et al. 2012, Stagnaro-Green et al. 2011). There are also studies in which subclinical

hypothyroidism did not increase the risks of pregnancy complications (Cleary-

Goldman et al. 2008, Männistö et al. 2009, Männistö et al. 2010).

2.3.3 Maternal hypothyroxinemia

Isolated hypothyroxinemia during pregnancy is defined as a mother having normal

serum TSH concentrations with a low fT4 concentration. The exact limit of fT4

concentration varies between reference intervals and assays used. The ATA

recommends using the 5th (severe hypothyroxinemia) or the 10th percentile (mild

hypothyroxinemia) (Stagnaro-Green et al. 2011), and the European Thyroid

Association recommends using the 5th percentile of pregnancy-related reference

range for low fT4 limits (Lazarus et al. 2014). The cause of hypothyroxinemia is

not definite, but iodine deficiency is a probable etiology. Like subclinical

hypothyroidism, hypothyroxinemia is associated with adverse obstetric outcomes

in most studies (Lazarus et al. 2014).

2.3.4 Maternal hyperthyroidism

Maternal thyrotoxicosis during pregnancy is defined by the ATA as “the clinical

syndrome of hypermetabolism and hyperactivity that results when the serum

concentrations of free thyroxine hormone and/or free triiodothyronine are high.”

The most common etiology of autoimmune hyperthyroidism during pregnancy is

Graves’ disease with a prevalence of 0.1-1.0%. Less common non-autoimmune

34

diseases causing thyrotoxicosis are toxic multinodular goiter, toxic adenoma,

subacute painful or silent thyroiditis (rare) and struma ovarii (rare) (Stagnaro-Green et al. 2011).

The syndrome of gestational hyperthyroidism is defined as transient

hyperthyroidism and is limited to the first half of pregnancy. It is characterized by

elevated fT4 and suppressed TSH without serum markers of thyroid autoimmunity.

Its prevalence is about 1-3% in pregnancies, and it is secondary to elevated hCG.

It may associate with hyperemesis gravidarum (severe nausea and vomiting in early

pregnancy with dehydration, weight loss and ketonuria), multiple gestation,

hydatidiform mole or choriocarcinoma (Stagnaro-Green et al. 2011). A transient

increase in maternal thyroid function is physiological and needed for the

requirements of normal pregnancy. The physiological acceleration of thyroid

function, however, seldom changes TSH concentrations (Glinoer 1997).

Clinical hyperthyroidism during pregnancy has severe maternal and fetal

consequences and it must be treated. Treatment options for clinical

hyperthyroidism during pregnancy are controversial because the most commonly

used drugs (methimazole or carbimazole) are probably teratogenic, and

propylthiouracil can be hepatotoxic for the mother. The latter is, however, the drug

of choice in early pregnancy if treatment becomes necessary in severe

thyrotoxicosis cases. Methimazole or carbimazole may be considered after the first

trimester (Cooper & Laurberg 2013).

2.4 Child’s neuropsychological development and its relation to

maternal thyroid function

2.4.1 Early development of the fetal central nervous system

The development of the fetal central nervous system (CNS), which is composed of

the brain and spinal cord, begins early in the first trimester of the pregnancy during

a process called neurulation. The CNS is developed from ectodermal tissue, first

forming a neural plate at approximately two weeks of gestation. Then, at day 18, a

neural groove is formed. By the end of the third week, a neural tube is formed, and

it withdraws from the overlying ectoderm to create a neural crest. It will further

give rise to the sensory ganglia of spinal and ganglial nerves, Schwann cells, the

meninges and some skeletal and muscular components of the head. The neural tube

begins to close from the hindbrain and proceeds anteriorly and posteriorly. This

35

very delicate process is complete at day 26-28 of gestation. During the period in

which the neural tube closes, it is highly vulnerable to outside interruption; in a

worst case scenario, disturbance in this process might lead to spina bifida (divided

spine) (Rice & Barone 2000, Salminen 2015).

Early in the first month of gestation, the processes of neurogenesis and the

migration of cells in the forebrain, midbrain and hindbrain begins. This is followed

by further migration, cell proliferation, synaptogenesis (formation of neuronal

connections), gliogenesis (formation of numerous supportive cells such as

microglia, radial glia, astroglia, oligodendroglia and Schwann cells), myelination

(formation of a protective and conducting layer of glia cells on neurons) and

apoptosis (controlled cell death). Errors in this phase may lead to severe congenital

abnormalities of the CNS, the prevalence of which is about 0.7-1.9 per 1000 births.

The abnormalities include, at worst, anencephaly, hydrocephaly and spinal cord

herniation. The conditions associate with folic acid deficiency, high maternal age,

threatened interruption of the pregnancy, severe prematurity and intrauterine

hypotrophy (Rice & Barone 2000, Salminen 2015).

Different brain regions start to develop at their own pace after these early steps.

Differentiation, synaptogenesis and proliferation continue after birth.

Synaptogenesis even continues throughout adulthood. The major points of

progression are presented in Figure 2. Brain mass growth increases in early

pregnancy, peaking at approximately four months after birth. The mass growth of

the brain, however, is only a small part of the brain’s development (Rice & Barone

2000, Salminen 2015).

36

Fig. 2. Central nervous system development in the human fetus. The figure has been

modified from Rice & Barone 2000.

Neurogenesis and cell proliferation are tightly controlled processes, occurring in

varying amounts at different times in different brain regions. They are easily

interrupted by outside substances or exposures such as x-ray exposure or exposure

to ethanol, organophosphates and mercury. When proliferation is disturbed,

migration is also often affected. The differentiation of neurons and glia cells,

synaptogenesis, gliogenesis, myelination and apoptosis may also be affected by

various chemicals such as ethanol, nicotine, mercury, lead and drugs like warfarin,

tetracyclines, vitamin A and lithium. Outside factors such as viral infections

(Varicella Zoster, cytomegalovirus) and maternal diseases may also affect the

developing CNS of a fetus. Maternal ethanol use during pregnancy is suspected to

have a long-lasting effect on the child’s CNS via apoptosis regulation disorder.

Ethanol is the most common teratogenic agent used during pregnancy, and maternal

ethanol use may lead to fetal alcohol syndrome. All of the above mentioned factors

that may interrupt normal CNS development either affect cell function on a

molecular level directly (coding DNA or RNA damage) or disturb the genome in

such a way that the effects may not appear until later life (through, for example,

epigenetics and non-coding DNA damage) (Rice & Barone 2000, Salminen 2015).

37

Fetal exposure to maternal health problems and environmental factors in the

intrauterine environment and its long-lasting effects on a child’s later life has been

first described by Dr. Barker (1989) and the idea of fetal programming is called the

Barker’s hypothesis. Dr. Barker showed a correlation between fetal undernutrition

and low birth weight and later ischemic heart disease (Barker 1989). Afterwards,

the Barker’s hypothesis has been used as an etiological theory basis for various

adulthood diseases. The hypothesis can be applied also when assessing the effects

of maternal health during pregnancy on neurological development of the child.

2.4.2 Thyroid hormones and brain development

The fetal brain needs thyroid hormones long before the onset of its own thyroid

hormone production. Thyroid hormone receptors are present in fetal brain tissue in

all three trimesters of pregnancy (Rovet 2014). Evidence of the earliest stages of

neurodevelopment derives from rodent studies; thyroid hormones regulate genes

responsible for neurogenesis, cell migration and differentiation, synaptogenesis and

myelination (Anderson et al. 2003, Auso et al. 2004, Cuevas et al. 2005, Lavado-

Autric et al. 2003). T3 is the active hormone in the developing brain. It is mostly

derived from T4 deionization by T2D in glia cells and is then actively transported

into neurons (Santisteban & Bernal 2005).

Thyroid hormones in the fetal brain have been shown to be of maternal origin

(Contempre et al. 1993). Thyroid hormones have also been found in the brains of

aborted human fetuses, miscarriages and stillbirths, where the amount of thyroid

hormones differed in the brain regions at different stages of pregnancy termination.

This signifies that maternal thyroid hormones are needed throughout pregnancy for

normal brain development (Kester et al. 2004).

Children with transient neonatal hypothyroxinemia of prematurity (THOP)

provide interesting information on what happens when maternal-fetal thyroid

supply is disturbed. Prematurity is a multifactorial pregnancy complication with

many co-morbidities, suboptimal thyroid function being one of them. Compared

with full-term infants (≥ 37 weeks of gestation), infants born after 30 but prior to

37 weeks of gestation have subnormal thyroid hormone levels with a normal

hormone balance. Infants born before the 30th week have decreased thyroid

hormone levels with an abnormal T4/T3 profile (Fisher 2007). These children have

a higher risk of cognitive and perceptual impairment and IQ reduction assessed at

the age of eight (Lucas et al. 1996) and increased cerebral white matter

echoluminance implicating cell damage (Leviton et al. 1999).

38

Iodine deficiency has extensive adverse effects on brain and neurological

development, and in the worst case, it leads to cretinism: stunted growth and severe

mental retardation (Rovet 2014). The condition has been reported as early as the

middle of the 19th century, when maternal hypothyroidism due to iodine deficiency

was first associated with neurological cretinism in children (Fagge 1871). Also mild

maternal iodine deficiency has been connected to neurocognitive problems in

offspring (Bath et al. 2013). Children born to mothers with an iodine-to-creatinine

ratio of less than 150µg/g scored lower on verbal IQ, reading accuracy and reading

comprehension assessments than those of mothers with a ratio of 150µg/g or higher

(Bath et al. 2013).

2.4.3 Intellectual disability

Intellectual disability is a feared consequence of early CNS development defects.

Intellectual deficiency was first differentiated from mental illnesses in the early

19th century and is understood as a continuum of different degrees of lowered

intellect quotient (IQ). The most resent diagnostic criteria for intellectual disability

in Finland and Europe is based on the International Classification of Diseases (ICD)

by the World Health Organization and in the USA on the Diagnostic and Statistical

Manual of Mental Disorders (DSM) by the American Psychiatric Association

(Heikura 2008, Salvador-Carulla et al. 2011). ICD-9 and ICD-10 define

intellectual disability as the incomplete development of the mind, particularly

characterized by sub-normal intelligence. A diagnosis should be based on the

individual’s current level of functioning without regard to its nature or causation—

such as psychosis, cultural deprivation, Down syndrome, etc. The assessment of

intellectual level should base on all available information (ICD-9, WHO 1977, pp.

212-213; ICD-10, WHO 1996, pp. 2). ICD-9 and 10 classify intellectual deficiency

into mild (IQ 50-70), moderate (IQ 35-49), severe (IQ 20-34), profound (IQ < 20)

and unspecified. IQ assessment should be based on a test with a mean of 100 and a

standard deviation (SD) of 15, such as the Wechsler Scales (ICD-9, WHO 1977, pp.

212-213; ICD-10, WHO 1996, pp.2). In the Wechsler Scales, for example,

individuals with a moderate or severe intellectual deficiency score under 0.13%,

and those with a mild intellectual deficiency score under 2.14% of the normal

Gaussian curve area (Anastasi 1988). IQ measurements should only be used as a

guide, though, because intellectual deficiency frequently involves psychiatric

disturbances (Heikura 2008, Sheehan et al. 2015).

39

The etiology of intellectual deficiency varies depending on the population

studied. The lower the IQ, the more severe and early prenatal factors seem to be the

cause (Heikura 2008). If one wishes to study the environmental and intrauterine

factors possibly associated with a lowered IQ, a wider range of IQ changes should

be studied. Therefore, mild cognitive limitation is currently often defined in

research as an IQ of 50-85 (Heikura et al. 2013).

Rodent studies have shown that the fetal hippocampus and cortex particularly

suffer from maternal thyroid hormone insufficiency leading to abnormalities in size

and cell structure (Lavado-Autric et al. 2003, Opazo et al. 2008). Clinical signs of

cell damage are memory and learning problems in hippocampal injury and

impaired basic perceptual and higher-order cognitive skills in cortex injury (Opazo et al. 2008). The timing of thyroid hormones’ effect varies across different brain

areas: lower brain structures and visual processing need thyroid hormones earlier

in pregnancy than structures in the cortex (Bradley et al. 1992).

The importance of thyroid hormones for neuropsychological development is

emphasized in children with CH because maternal thyroid hormone supply during

pregnancy is the only thyroid hormone source for these children, and it may not

suffice. Their neurocognitive development is, on average, within normal limits. It

is, however, suspected that some children with CH are still at higher risk of learning

difficulties, but identifying those children early is still a challenge (Van Vliet &

Deladoey 2014). In a study in which children with CH were followed until 9-15

years of age, and some of them had MRI imagining, the children had an increased

risk of selective memory deficits accompanied by structural and functional

hippocampal abnormalities (Wheeler et al. 2011). In that study, the hippocampal

volume was smaller in children with CH compared to the control children, and this

was thought to correlate with the study subjects’ learning difficulties (Wheeler et al. 2011).

Cretinism is an extreme form of neuropsychological and developmental delay

caused by maternal iodine deficiency and subsequent thyroid hormone deficiency

during pregnancy. Due to worldwide iodine supplementation programs, severe

cretinism is rare nowadays, even in developing countries. Intellectual deficiency is

the most feared complication of cretinism (Rovet 2014). Mild iodine deficiency

during pregnancy has also been associated with reduced cognitive outcomes in

children (Bath et al. 2013, Hynes et al. 2013).

Cognitive and intelligence testing is usually conducted with a certified test

scale. The most commonly used pattern is the Wechsler Intelligence Scale for

Children, which includes subscales for intelligence, attention, language skills and

40

visual and motor performance. Abnormal test results are defined statistically

because general intelligence is a continuum in an ordinal scale (Wechsler 1991).

Other scales used include the Stanford-Binet test for school-aged children and the

Wechsler Preschool and Primary Scale of Intelligence for very young children.

Sensory development

Rat studies have shown that maternal thyroid hormones are important for the

sensory development of offspring. They affect several structures important for

normal vision such as opsin formation in cones in the retina (Lu et al. 2008, Roberts et al. 2006), thalamus (Guadano-Ferraz et al. 1991) and primary visual cortex

(Berbel et al. 1985). Preterm children with THOP are at risk of impaired visual

functioning (Rovet & Simic 2008) and psychomotor delays (Smit et al. 1999).

Children with CH often have weaknesses in their motor skills; however, their

language skills are better retained (Gottschalk et al. 1994, Kooistra et al. 1994).

2.4.4 Attention deficit hyperactivity disorder

Attention deficit hyperactivity disorder (ADHD) is a common multifactorial

neuropsychiatric disorder, the risk factors of which are being heavily researched.

ADHD is considered highly genetic, but many environmental factors such as

prematurity and maternal smoking are also associated with a child’s increased risk

of ADHD. Its prevalence is approximately 3-5% globally in children aged 6-18,

and it is typically diagnosed in school-aged children. ADHD is more common in

boys with a sex ratio of 3:1 in boys and girls, respectively (Moilanen et al. 2013,

Thapar & Cooper 2015).

Symptoms of ADHD include reduced control of attention and impulses and

increased hyperactivity. To be diagnosed with ADHD (based on the ICD-10 in

Finland), a child must have six or more inattention symptoms out of nine, three or

more hyperactivity symptoms out of five and three out of four impulse control

problem symptoms. All symptoms should have lasted for at least six months. The

symptoms must be noticeable at home and at school, and they must have started at

seven years of age, at the latest. Diagnostics in Finland are based on a clinical study

of the child and on a completed symptoms questionnaire by parents and teachers

(e.g., ADHD-RS-IV). Globally, the most commonly used symptoms questionnaire

for parents and teachers is the Conners Rating Scales-Revised, but it has not been

translated into Finnish. Researchers use overall psychiatric behavior questionnaires

41

such as the Child Behavior Check List (CBCL) and Strengths and Difficulties

Questionnaire (SDQ) to screen for ADHD symptoms, but these questionnaires are

not reliable enough for diagnostic purposes (Moilanen et al. 2013, Thapar &

Cooper 2015). For international research, the Diagnostic and Statistical Manual of

Mental Disorders (DSM) is used. The DSM-IV-TR version has the same principal

items as the ICD-10, but it classifies them into two groups: nine on inattention and

nine on hyperactivity-impulsivity. Accordingly, ADHD can be either combined

(having six or more inattention and six or more hyperactivity-impulsivity

symptoms), predominantly inattentive or predominantly hyperactive- impulsive

(American Psychiatric Association 2000).

The severity of ADHD symptoms varies significantly, and so does treatment.

Very difficult symptoms may be treated with psychostimulants such as

methylphenidate. Usually with milder symptoms, the child and family require

psychoeducation and increased learning support at school. Psychotherapy may also

be useful, as often other psychiatric or social problems coexist (Moilanen et al. 2013, Thapar & Cooper 2015). The effective treatment of ADHD is important. As

shown in the NFBC 1986 studies, ADHD often leads to a sustained negative

developmental trajectory, and a child’s ADHD symptoms are associated with

poorer scholastic success (Rodriguez et al. 2007).

The risk of ADHD was first associated with maternal thyroid hormone

insufficiency during pregnancy in a study from an iodine-deficient area (Vermiglio et al. 2004). Children with a generalized resistance to thyroid hormones have also

been reported to have an especially high prevalence of ADHD (Hauser et al. 1997).

Thyroid hormone resistance is a disease caused by mutations in the thyroid

hormone receptor-β gene and characterized by decreased responsiveness of the

peripheral and pituitary tissues to the activation of thyroid hormones (Hauser et al. 1997). Because thyroid hormones are so important for brain development, thyroid

hormone deficiency might also increase a child’s risk of ADHD (Vermiglio et al. 2004).

42

2.5 The effects of maternal thyroid dysfunction during pregnancy

on child neuropsychological and sensory development

2.5.1 Effects of maternal hypothyroidism on child

neuropsychological development

Several studies have evaluated the relationship between maternal overt and

subclinical hypothyroidism during pregnancy and neurodevelopment in children.

The most important studies are summarized in Table 2. The earliest findings date

back to 1969 when Man and Jones found that a precursor product for thyroid

hormones (butanol-extractable iodine) was correlated with the cognitive

performance of children aged 4-7 years. The children of mothers with the lowest

concentrations of serum butanol-extractable iodine had decreased IQ levels, more

visuospatial and locomotor disabilities, increased inactivity and slower than normal

reaction times than those of mothers with normal serum levels of thyroid hormone

precursor (Man & Jones 1969).

The first study to show an association between high maternal TSH during

pregnancy and adverse neuropsychological development in the child was published

in 1999 (Haddow et al. 1999). In that study, 64 children born to mothers with high

TSH during pregnancy were compared to 124 children born to mothers with normal

thyroid function. The study’s subjects were a part of a Down syndrome screening

program. Of the 64 women with high serum TSH concentrations, 48 women were

not treated with levothyroxine, and the IQ scores of the 48 women’s offspring were

seven points lower than those of the control mothers. The study did not separately

evaluate maternal overt and subclinical hypothyroidism (Haddow et al. 1999).

Thereafter, studies have also associated maternal overt or subclinical

hypothyroidism with a child’s decreased everyday memory function and smaller

hippocampal size (Willoughby KA et al. 2014), decreased IQ and motor

developmental scores (Li et al. 2010) and decreased sensory functions (Mirabella et al. 2005). ADHD symptoms in children have also been connected with high

maternal TSH concentrations and maternal hypothyroidism (Ghassabian et al. 2011,

Haddow et al. 1999). TPO-Ab concentrations during pregnancy have been

associated with children’s ADHD symptoms at three years of age, independent of

maternal TSH concentrations (Ghassabian et al. 2012). Diagnosed and treated

maternal hypothyroidism during pregnancy in a register-based dataset also

associated with increased risk of future autism spectrum disorders and psychiatric

drug use in children during late adolescence or young adulthood (Andersen et al.

43

2014a, Andersen et al. 2014b). These results were independent of maternal

psychiatric disease history.

However, not all studies have found a correlation between subclinical

hypothyroidism during pregnancy and impaired neuropsychological development

in children. In one Danish and one Spanish population-based cohort study, no

association between maternal subclinical hypothyroidism and development of the

child was found at 18, 24 or 30 months (Henrichs et al. 2010, Julvez et al. 2013).

A Scottish study among children born > 37 weeks found no neuropsychological

developmental delays at 5.5 years of age in relation to maternal TSH concentrations

(Williams et al. 2013).

In some studies, maternal levothyroxine treatment has improved children’s IQ

scores (Haddow et al. 1999, Momotani et al. 2012). However, in a randomized trial

where mothers with hypothyroidism or hypothyroxinemia were treated with

levothyroxine or had no treatment during pregnancy, the IQ scores of children did

not improve with treatment (Lazarus et al. 2012). The authors of the trial, however,

suggest that the timing of the screening (mean of 13 weeks of pregnancy, up to 20

weeks) may have been too late in order to observe an influence on brain

development. Also, the IQ assessment was conducted at three years of age, and a

longer follow-up would have been beneficial.

Although several studies have been conducted in this field, the ETA guidelines

for the management of subclinical hypothyroidism in pregnancy call for further

high-quality studies. The results of many previous studies of subclinical

hypothyroidism during pregnancy were affected by confounders such as the

country’s iodine status, differing IQ and memory function testing methods and

short follow-up times. Furthermore, not all studies differentiate between overt and

subclinical hypothyroidism (Lazarus et al. 2014). Countries also vary in their

routine maternal and child health care protocols.

44

Ta

ble

2. S

tud

ies

of

mate

rna

l h

yp

oth

yro

idis

m a

nd

th

e n

eu

rod

ev

elo

pm

en

t o

f th

eir

off

sp

rin

g

Au

tho

r, y

ear

Countr

y S

tudy

settin

g

Main

resu

lts

Ma

n &

Jo

ne

s 1

96

9

US

A

13

49

ch

ildre

n a

ge

d 4

-7.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m, ch

ild’s

low

IQ

sco

re a

nd

neuro

psy

cho

logic

al p

roble

ms

Ma

tsu

ura

et al.

19

90

Japan

23 in

fants

born

to 1

6 m

oth

ers

with

chro

nic

thyr

oid

itis.

A

ssoci

atio

n o

f m

ate

rnal o

vert

hyp

oth

yroid

ism

and s

eve

re

deve

lopm

enta

l pro

ble

ms

Haddow

et al.

19

99

U

SA

64 c

hild

ren a

ged 7

-9 b

orn

to m

oth

ers

with

hig

h T

SH

conce

ntr

atio

ns

vs.

124 c

hild

ren o

f euth

yroid

moth

ers

.

Ass

oci

atio

n o

f untr

eate

d m

ate

rnal h

ypoth

yroid

ism

and c

hild

’s

low

ere

d IQ

and m

ore

hyp

era

ctiv

ity s

ympto

ms.

Mirabella

et

al.

20

05

U

SA

13 in

fants

of h

ypoth

yroid

moth

ers

vs.

20 in

fants

born

to h

ealth

y m

oth

ers

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m a

nd in

fant’s

re

duce

d

visu

al c

ontr

ast

sen

sitiv

ity a

nd d

ecr

ea

sed a

ttentio

n.

Li e

t al.

2010

C

hin

a

1268 m

oth

ers

: 18 h

ad s

ubcl

inic

al h

ypoth

yroid

ism

and

142 c

ontr

ols

.

Ass

oci

atio

n o

f m

ate

rnal s

ubcl

inic

al h

ypoth

yroid

ism

and

child

’s

inte

llige

nce

and m

oto

r sc

ore

s at age o

f 25-3

0 m

onth

s.

Ghass

abia

n e

t al.

2011

Th

e

Neth

erlands

3736 m

oth

er-

child

pairs.

A

ssoci

atio

n b

etw

een T

SH

leve

ls a

nd

pare

nta

l report

ed

ext

ern

aliz

ing s

core

s of ch

ildre

n a

t 1.5

and 3

yrs

.

Anders

en

et al.

20

14

Denm

ark

30295 c

hild

ren w

ere

born

to m

oth

ers

with

thyr

oid

dys

funct

ion.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m a

nd c

hild

’s r

isk

of autis

m

spect

rum

dis

ord

ers

.

Anders

en

et al.

20

14

Denm

ark

3979 a

dole

scents

and y

oung a

dults

were

born

to

hyp

oth

yroid

moth

ers

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m a

nd c

hild

’s u

se o

f

psy

chia

tric

dru

gs

in la

te a

dole

scence

and y

oung a

dulth

ood.

Will

oughy

et

al.

20

14

US

A

24 c

hild

ren o

f m

oth

ers

tre

ate

d for

hyp

oth

yroid

ism

vs.

30 m

atc

hed c

ontr

ols

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m a

nd c

hild

’s s

malle

r ce

rtain

hip

poca

mpa

l are

as

and lo

w s

core

s o

n m

em

ory

funct

ion

test

s.

Sa

ma

di e

t al.

20

15

U

SA

22 c

hild

ren a

ged 9

-14 y

ears

of hyp

oth

yroid

moth

ers

vs. 22 m

atc

he

d c

ontr

ols

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

idis

m a

nd c

hang

es

in c

hild

’s

MR

I sc

ans

in c

orp

us

callo

sum

are

a.

45

2.5.2 Effects of maternal hypothyroidism on child sensory and motor

development

Limited research has been conducted on maternal thyroid dysfunction during

pregnancy and its association with a child’s sensory development. In the study by

Haddow et al. (1999), high maternal TSH concentrations during pregnancy did not

correlate with the offspring’s motor scores. Also, in a study by Oken et al. (2009),

no association was found between maternal thyroid function and a child’s visual

motor abilities. In these studies, motor abilities were identified as part of a larger

cognitive and motor ability survey and were not the main outcome. In a study by

Li et al. (2010), the mean motor scores of children aged 25-30 months from mothers

with either subclinical hypothyroidism and/or elevated TPO-Ab titres were lower

than those of the control subjects. A small study of eight mothers with subclinical

hypothyroidism and one with overt hypothyroidism examined children’s

psychomotor and audiological outcomes. The mothers received levothyroxine

treatment after their diagnosis. There was no impairment observed in their

children’s early development at nine months of age (Radetti et al. 2000). In adults,

it has been shown that treating hypothyroidism with levothyroxine improves

contrast sensitivity (Cakir et al. 2015). Additionally, rodent research data suggests

that maternal hypothyroidism might affect the sensory development of the

offspring (Roberts et al. 2006); more research in this area is needed.

Maternal TPO-Ab-positivity in the third trimester of pregnancy has been

associated with an increased risk of sensory neural hearing loss in either ear and in

at least one frequency in children. In that study, maternal thyroid function

parameters were not known, but 23 mothers reported having hypothyroidism and

using levothyroxine (Wasserman et al. 2008). The same authors later published

those results compared to controls; the children of mothers with high TPO-Ab titres

had a modest, but statistically significant, lowered IQ at the age of four (measured

with the Stanford-Binet). At the age of seven, the results were no longer statistically

significant (measured with the full-scale Wechsler). The authors suggested that a

child’s sensorineural hearing loss might mediate the correlation between maternal

TPO-Ab-positivity and the child’s IQ (Wasserman et al. 2012).

46

2.5.3 Effects of maternal hypothyroxinemia on child

neuropsychological development

Many studies have connected maternal hypothyroxinemia in early pregnancy to

adverse neuropsychological outcomes in children. These studies are presented in

Table 3. Studies from the Netherlands by Pop et al. (2003, 2009) linked maternal

hypothyroxinemia to a child’s delayed psychomotor development. Vermiglio et al. (2004) correlated maternal hypothyroxinemia due to iodine deficiency to a child’s

higher risk of diagnosed ADHD.

Recently, population-based cohort studies have been published from the

Netherlands and Spain. Maternal hypothyroxinemia has been associated with a

child’s expressive language delays at 18 and 30 months (Henrichs et al. 2010),

decreased mental scores at 18 months (Julvez et al. 2013), decreased response

speed in reaction time tests at 5-6 years of age (Finken et al. 2013), and small

differences in neurocognitive development (Ghassabian et al. 2014). Brain cortex

and gray matter volume measured by MRI imagining did not associate with

maternal hypothyroxinemia (TSH analyses were included in the model)

(Ghassabian et al. 2014). However, the same cohort later reported an inverted U-

shaped association between maternal early pregnancy fT4 concentrations and a

child’s lowered IQ assessed in five to nine year-old children and brain cortex and

gray matter volume assessed in six to eleven year-old children (Korevaar et al. 2015b).

Maternal hypothyroxinemia has also been associated with psychiatric

problems in children such as autism symptoms and schizophrenia (Gyllenberg et al. 2015, Roman et al. 2013). However, the risk evaluation for autism was based

on a parental questionnaire conducted when the children were six years old, which

is fairly young to evaluate autism (Roman et al. 2013, Gyllenberg et al. 2015).

47

Ta

ble

3. M

ate

rna

l h

yp

oth

yro

xin

em

ia d

uri

ng

pre

gn

an

cy

an

d its

eff

ects

on

off

sp

rin

g’s

ne

uro

ps

ych

olo

gic

al

de

ve

lop

me

nt.

Au

tho

r, y

ear

Countr

y S

tudy

settin

g

Main

resu

lts

Po

p e

t al.

1999

T

he N

eth

erland

s In

fants

of 220 m

oth

ers

were

follo

wed

up to 1

0

month

s.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s im

paired

psy

chom

oto

r de

velo

pm

ent.

Po

p e

t al.

2003

T

he N

eth

erland

s 135 h

ypoth

yro

xine

mic

moth

ers

and t

heir m

atc

hed

contr

ols

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s d

ela

yed

menta

l and m

oto

r deve

lopm

ent

at

ag

es

of

1 a

nd 2

years

.

Ve

rmig

lio e

t al.

20

04 I

taly

16 c

hild

ren b

orn

to

moth

ers

fro

m io

din

e-d

efic

ient

are

a (

A)

vs.

11

ch

ildre

n o

f m

oth

ers

fro

m io

din

e-

suff

icie

nt

are

a (

B).

Ass

oci

atio

n o

f m

ate

rnal i

odin

e d

efic

iency

during p

regn

ancy

,

child

’s A

DH

D a

nd I

Q in

10 y

ear

follo

w-u

p.

Be

rbe

l et al.

2009

S

pain

34

5 p

reg

na

nt

wom

en

an

d t

he

ir o

ffsp

rin

g,

mo

the

rs

iodin

e-s

upp

lem

ente

d a

t diff

ere

nt tim

e p

oin

ts o

f

pre

gnancy

.

Ass

oci

atio

n o

f m

ate

rnal i

odin

e d

efic

iency

and c

hild

’s

neuro

deve

lopm

enta

l and m

oto

r pro

ble

ms.

Li e

t al.

2010

C

hin

a

18 m

oth

ers

with

subcl

inic

al h

ypoth

yroid

ism

, 19

hyp

oth

yroxi

nem

ia a

nd 1

42 c

ontr

ols

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s lo

w m

ean

inte

llige

nce

and m

oto

r sc

ore

s th

an a

t 25-3

0 m

onth

s.

Henrich

s et

al.

201

0

The N

eth

erland

s 3659 m

oth

er-

child

pairs.

A

ssoci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s

exp

ress

ive la

ngua

ge d

ela

y at 1

8 a

nd

30 m

onth

s.

Julv

ez

et

al.

2013

S

pain

1761 m

oth

er-

child

pairs.

Low

mate

rnal f

T4 c

once

ntr

atio

n w

as

ass

oci

ate

d w

ith d

ecr

ease

d

menta

l sco

res

at a

ge o

f 18 m

onth

s.

Fin

ken

et al.

20

13

T

he N

eth

erland

s 1765 m

oth

er-

child

pairs.

A

ssoci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s

decr

ease

d r

esp

on

se s

peed a

t 5 t

o 6

years

of age.

Ghass

abia

n e

t al.

2014

T

he N

eth

erland

s 3727 m

oth

er-

child

pairs,

bra

in M

RI

in 6

52

child

ren a

t ag

e o

f 8.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s lo

w

nonve

rbal I

Q, b

ut

no a

ssoci

atio

n w

ith b

rain

morp

holo

gy.

Gyl

lenberg

et al.

2015

F

inla

nd

1010 d

iagno

sed s

chiz

ophre

nia

case

s and their

matc

hed c

ontr

ols

.

Ass

oci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s

schiz

ophre

nia

dia

gnosi

s up to a

ge o

f 26 y

ears

.

Note

n e

t al.

2015

T

he N

eth

erland

s 1196 m

oth

er-

child

pairs.

A

ssoci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s

subnorm

al a

rith

me

tic p

erf

orm

ance

.

48

Au

tho

r, y

ear

Countr

y S

tudy

settin

g

Main

resu

lts

Modest

o e

t al.

201

5

The N

eth

erland

s 3873 m

oth

er-

child

pairs.

A

ssoci

atio

n o

f m

ate

rnal h

ypoth

yro

xin

em

ia a

nd c

hild

’s A

DH

D

sym

pto

ms

at 6-1

1 y

ears

of age.

Ko

reva

ar

et al.

20

15

The N

eth

erland

s

3839 m

oth

er-

child

pairs

with

child

’s d

ata

on IQ

at

5-9

years

of age,

MR

I sc

ans

from

646 c

hild

ren

aged 6

-11.

An in

vert

ed U

-sha

ped a

ssoci

atio

n o

f m

ate

rnal f

T4

conce

ntr

atio

ns

an

d c

hild

’s IQ

, gre

y m

atter

volu

me a

nd

cort

ex

volu

me.

49

There are also studies in which no association was found between maternal

hypothyroxinemia and adverse neuropsychological outcomes in children. Oken and

colleagues from the USA found no association between maternal

hypothyroxinemia and a child’s verbal recognition memory, visuomotor skills or

language skills (Oken et al. 2009). In Spain, Grau et al. (2015) found no

associations between maternal thyroid status and a child’s development at one and

six to eight years of age in an observational double-blinded study. Craig et al. (2012)

showed no association between maternal thyroid status and a child’s development

at two years of age.

The results of the studies presented likely vary due to differences in populations,

country’s maternal and child health care protocols, study settings and in the ability

to control confounding factors. Furthermore, studying maternal hypothyroxinemia

during pregnancy poses many diagnostic and laboratory challenges (Lazarus et al. 2014). IQ scores and brain morphology are examples of measures that are easier to

compare between studies. The interpretation of different study findings, however,

remains difficult because the study settings and real life implications do not

necessarily correlate. For example, if a child’s slightly lowered IQ is measured at a

very young age, its longer term effects and meaning in the child’s life are not easy

to determine. Another example of the challenge of interpreting these results are

when study findings are based on single-source reporting, like parental reports on

a child’s behavior problems (for example, Henrichs et al. 2010, Modesto et al. 2015). Clinical diagnostics of ADHD are based on interviews and/or observations

from multiple sources. Therefore, studies that better meet the criteria for clinically

diagnosing ADHD, are needed.

2.5.4 Effects of maternal hypothyroxinemia on child sensory

development

Pop et al. (1999, 2003) have linked maternal hypothyroxinemia to impaired

psychomotor development (Bayley Scales of Infant development, including eye-

hand coordination) in children aged 10 months in 1999 and in children aged 1-2

years in 2003. Craig et al. (2012) found that second trimester maternal fT4 was

associated with around 3% lower motor scores than in control children, though only

in unadjusted data. Studies with a long follow-up do not exist.

50

2.5.5 Effects of maternal hyperthyroidism on child

neuropsychological development

There is only a limited number of studies concerning maternal hyperthyroidism

during pregnancy and its long-term effects on children. A Danish cohort study

(Andersen et al. 2014) has associated maternal hyperthyroidism that was diagnosed

and treated for the first time after delivery with diagnosed ADHD in children. There

was no connection between diagnosed and treated hyperthyroidism before birth or

between paternal thyroid status and ADHD in children. Therefore, the authors

suggest that the correlation is probably not caused by a genetic factor but rather is

due to an undiagnosed maternal hyperthyroidism during pregnancy (Andersen et al. 2014). The same study group also previously reported a connection between

maternal untreated hyperthyroidism during pregnancy and child’s epilepsy, where

the diagnosis and treatment of hyperthyroidism were conducted after the child’s

birth (Andersen et al. 2013).

2.5.6 Screening for maternal thyroid dysfunction during pregnancy

At the moment, a lot of research is being conducted on whether maternal thyroid

dysfunction during early pregnancy should be screened for in all pregnant women.

Universal screening for thyroid dysfunction during pregnancy require that thyroid

dysfunction is also prevalent in asymptomatic individuals and would require the

availability of a reliable testing method and a beneficial intervention with limited

adverse effects. In addition, the screening should be cost-effective (Lazarus et al. 2014, Stagnaro-Green et al. 2011).

Several factors pose obstacles to universal screening. First, fT4 measurements

during pregnancy may be unreliable, and measuring TSH only may be more

sensible. Secondly, study results concerning the relationship of subclinical

hypothyroidism and hypothyroxinemia during early pregnancy to adverse

pregnancy and neuropsychological outcomes in children are somewhat

controversial. Some studies also associate risk increases with TPO-Abs, the

significance of which is partly unknown. The treatment of TPO-Ab-positivity

without thyroid dysfunction is not advisable. Finally, there is not yet evidence that

levothyroxine treatment for individuals with increased TSH concentrations would

improve the outcomes for the pregnancy or the child (Lazarus et al. 2014, Stagnaro-

Green et al. 2011). It could also be speculated that treatment would not always

begin early enough. Even in countries such as Finland, where progressive maternal

51

healthcare is available to every pregnant woman (Koskela et al. 2000, Lehtinen et al. 2003) and women usually contact a healthcare provider after a positive

pregnancy test, the screening might occur too late to decrease the child’s risk of

adverse neuropsychological outcomes.

Factors that would favor universal screening are that TSH measurement is not

expensive, effective treatments for thyroid dysfunction exist and thyroid

dysfunction during pregnancy is prevalent enough to merit screening. Overt

hypothyroidism is clearly associated with adverse pregnancy and child

developmental outcomes, and treating overt maternal hypothyroidism is cost-

effective (Lazarus et al. 2014, Stagnaro-Green et al. 2011). There are also studies

suggesting that universal screening for subclinical hypothyroidism during

pregnancy is cost-effective if levothyroxine treatment prevented adverse

neuropsychological outcomes (Dosiou et al. 2008, Thung et al. 2009). A recent

Cochrane analysis on universal and/or case finding screening vs. no screening

found no harm or benefit to maternal and infant outcomes (Spencer et al. 2015).

The review lacked data on long-term developmental outcomes for the child.

The most recent recommendations from the ATA and ETA suggest that serum

TSH measurement should be done only in a targeted population; women with a

history of thyroid dysfunction and/or thyroid surgery, women with a family history

of thyroid diseases, those with having goiter, those with known thyroid antibodies,

those with clinical symptoms, women with type I diabetes or other autoimmune

disorders, women with a history of either miscarriage, preterm delivery, earlier

head or neck area radiation and women with morbid obesity (BMI ≥ 40 kg/m2)

should be screened for TSH abnormalities. Additionally, women treated with

amiodarone and lithium and those with recent exposure to iodinated radiological

contrast agents should be screened (Lazarus et al. 2014, Stagnaro-Green et al. 2011).

The only randomized study on levothyroxine treatment for maternal thyroid

dysfunction during pregnancy and its effects on a child’s cognitive function showed

negative results (Lazarus et al. 2012). The second part of the study is currently

underway; the children are aged between seven and ten years, and their cognitive

and motor functioning is now being assessed (Hales et al. 2014).

In conclusion, maternal thyroid dysfunction during pregnancy has been

associated with adverse neurodevelopmental outcomes in children. Previous

studies vary with regard to population, the country’s iodine status, methodology

and follow-up time. To date, the longest follow-ups have been conducted on eight-

year-old children, but most results are based on early childhood testing. Most

52

studies compare numerical values of differing neuropsychological tests or

questionnaires, but the impact of these findings on a child’s everyday life has not

been taken into account. None of the previous studies focus on a child’s scholastic

performance or sensory development. There are also no studies on how maternal

thyroid dysfunction during pregnancy potentially affects thyroid function in the

offspring and, furthermore, whether the offspring’s own thyroid dysfunction might

also have an effect on neuropsychological development. There is a need for a series

of studies using consistent, reliable methodology in an iodine-sufficient area.

53

3 Purpose of the present study

A population-based birth cohort study series with a long follow-up time, from early

pregnancy onwards until the age of 16, was conducted to contribute reliable data

on the association of maternal and child thyroid function and child

neuropsychological development. The specific study aims were:

1. To find out if maternal thyroid dysfunction and/or antibodies in early

pregnancy affects the thyroid function and antibody status of the offspring at

the age of 16. Reference intervals for TSH and fT4 concentrations of the

adolescents were also created.

2. To evaluate the possible effects of maternal thyroid dysfunction and/or

antibodies in early pregnancy on a child’s ADHD symptoms at the age of 8.

3. To evaluate whether maternal thyroid dysfunction and/or antibodies during

pregnancy affects a child’s scholastic performance at the ages of 8 and 16.

Whether an adolescent’s thyroid dysfunction affects scholastic performance

and ADHD symptoms at the age of 16 was also evaluated.

4. To find out whether maternal thyroid dysfunction and/or antibodies in early

pregnancy affects a child’s sensory and linguistic development up to the age of

8.

54

55

4 Subjects and methods

4.1 Subjects

4.1.1 The 1986 Northern Finland Birth Cohort

This study was carried out using prospective birth cohort material from the 1986 Northern Finland Birth Cohort (NFBC 1986). The NFBC 1986 covers 99% of all births with a calculated term between 1 July 1985 and 30 June 1986. The study subjects were drawn from the two northernmost provinces of Finland (9362 mothers, 9479 children). All mothers were recruited to the study by 24 weeks of gestation at the latest, and follow-up started from the first visit to a maternity welfare clinic at 8-12 weeks of gestation. Notably, all births took place in hospitals, and there were no private hospitals in the northern Finland area. Data were collected from 9479 pregnancies (99% of all births in the area). The study was originally organized by professors Paula Rantakallio, Anna-Liisa Hartikainen and Marjo-Riitta Järvelin from the University of Oulu’s Faculty of Medicine (Järvelin et al. 1993, Järvelin et al. 1997).

Data on cohort mothers and newborns were collected by local midwives in free-of-charge antenatal care units using three questionnaires designed specifically for this study. In the first questionnaire, maternal and family demographic, biologic and socioeconomic data were collected from all mothers at their first antenatal visit to the antenatal care unit. This questionnaire contained questions on the mothers’ previous diseases, including thyroid diseases. A previous dissertation study was conducted on maternal thyroid diseases in the NFBC 1986 in which 128 mothers reported a history of previous thyroid disease. The hospital records of these women were manually examined to verify the diagnosis (Männistö et al. 2009).

The second questionnaire concerning pregnancy and maternal health during pregnancy was completed by the local midwives during the mothers’ last visit to the antenatal care unit. Finally, information on the delivery, as well as the child’s date of birth, birth weight, death, illnesses and other data concerning the perinatal period was forwarded by the maternity clinics to the antenatal care units, where it was entered in the NFBC 1986 files (Järvelin et al. 1993, Järvelin et al. 1997).

After birth, data on the health of the children in the cohort and their familial demographic data were obtained through visits to communal child welfare clinics, cohort questionnaires and a clinical examination of each child at 16 years of age. At age 7-8 years, during their first school year, data on the children’s health, development and scholastic performance were collected via parent questionnaires.

56

A second questionnaire was sent to the children’s main teachers to assess their scholastic performance and behavior in the school environment. Both questionnaires included questions on whether the child had problems with attention, hyperactivity or with specific school subjects.

At age 16, in 2001-2002, 6798 children (74% of living and traceable NFBC 1986 children) participated in a clinical examination. This examination included clinical testing, blood sampling and a questionnaire concerning health, hobbies and scholastic performance. The adolescents’ usage of thyroid medication was also ascertained. This data has been further supplemented with register-based information up to the age of 16 years.

All data derived from cohort questionnaires and clinical examinations were directly saved into the university’s database after data collection. All data is therefore in electronical form and available for researchers to use (as in this study).

4.1.2 Finnish Maternity Cohort

The National Institute for Health and Welfare organizes routine infectious disease screening for hepatitis B, HIV and syphilis for all pregnant Finnish women in the first trimester of pregnancy. The screening is regulated by Finnish law, has been active since 1983 and covers over 98% of all pregnant women (Koskela et al. 2000, Lehtinen et al. 2003). Like all Finnish women, the NFBC 1986 mothers participated in this infectious disease screening in early pregnancy. Their blood samples were drawn at local healthcare centers and were sent to the prenatal serology laboratory of the National Institute for Health and Welfare in Oulu. Leftover serum samples were stored in the Finnish Maternity Cohort (FMC) serum bank at -25 °C. According to Finnish law (327/2001), these leftover samples can be used for public health research means without further subject consent.

4.2 Methods

4.2.1 Thyroid function analysis of the NFBC 1986 mothers

The NFBC 1986 maternal serum samples used in this study series were analyzed in the prenatal serology laboratory of the National Institute for Health and Welfare in Oulu. Samples were obtained from the FMC serum bank and analyzed for TSH, fT4, fT3, TPO-Abs and TG-Abs using the Abbott Architect i2000 method (Abbott Diagnostics, Abbott Park, IL) by means of chemiluminescent microparticle immunoassays in 2010. The maternal thyroid hormone laboratory analyses were

57

part of an earlier dissertation study by Dr. Tuija Männistö. Altogether, 5805 samples (61.2% of the whole NFBC 1986) were available from at least one of the previous analyses. Of them, 5791 samples (61.1% of the NFBC 1986) were analyzed for TSH or fT4 concentrations. When the sample volume was not sufficient for all analyses, TSH analyses were prioritized. Specific information on laboratory data collection and the effects of long-term storage on these laboratory parameters has been reported previously (Männistö et al. 2007). The population without laboratory analyses did not differ significantly from those with analyses when considering maternal demographic characteristics and birth outcomes (Männistö et al. 2009).

Categorization of the NFBC 1986 mothers (I, II, III, IV)

Mothers were categorized on the basis of their trimester and population-specific reference intervals for serum TSH and fT4 concentrations. Reference intervals for TSH and fT4 in the first trimester were 0.07-3.1 mU/L and 11.4-22.4 pmol/L and in the second trimester 0.10-3.5 mU/L and 11.1-18.9 pmol/L. Maternal TPO-Ab-positivity was defined as TPO-Ab concentration over the 95th percentile (> 167.7 IU/ml) (Männistö et al. 2011)

For the purposes of the current study, some additional categorization was used. In publication II, “high TSH” was defined as a TSH concentration above the upper reference limit of 3.1 mU/L in the first trimester or 3.5 mU/L in the second trimester. Maternal hypothyroxinemia was defined as having a serum TSH concentration within the reference limits and a low fT4 concentration, less than 11.4 pmol/L in the first trimester or less than 11.1 pmol/L in the second trimester.

In publications I, III and IV, mothers were divided into six thyroid function groups according to their serum fT4, TSH and TPO-Ab concentrations. We also separately analyzed the effects of overt (both TSH and fT4 outside reference intervals) and subclinical (TSH outside reference intervals, fT4 within reference intervals) maternal hypo- and hyperthyroidism.

1. Euthyroidism (reference group): maternal TSH and fT4 both within the reference intervals.

2. Hypothyroidism: TSH above the upper reference interval with low or normal fT4 concentrations.

3. Hypothyroxinemia: TSH within reference intervals with low fT4 concentrations.

4. Hyperthyroidism: TSH below the lower reference interval with high or normal fT4 concentrations.

5. TPO-Ab-negative: TPO-Ab concentration < 167.7 IU/mL.

58

6. TPO-Ab-positive: TPO-Ab concentration ≥ 167.7 IU/mL.

4.2.2 Thyroid function analysis of the NFBC 1986 children (I, III)

In 2001-2002, 6798 (74% of living and traceable NFBC 1986 children) children

attended a clinical examination during which blood samples were obtained. The

samples were drawn in the morning after an overnight fast. After primary analyses,

the samples were stored as serum at -80 °C. Quantitative analyses of TSH, fT4 and

TPO-Ab were performed for both maternal and adolescent serum samples by way

of chemiluminescent microparticle immunoassays using an Architect i2000

automatic analyzer (Abbott Diagnostics, Abbott Park, Illinois). The lower limits of

detection and intra- and inter-assay coefficients of variation were as follows: 0.0025

mU/L, 1.7 and 5.3% for TSH; 5.1 pmol/L, 3.6 and 7.8% for fT4; and 1.0I U/mL,

2.5 and 9.8% for TPO-Ab. Medians with 5th and 95th percentiles of the children’s

TSH, fT4 and TPO-Ab concentrations are presented in Table 4.

Table 4. Biochemical characteristics of the adolescent study population

Laboratory

parameter

n Median 5th percentile 95th percentile

TSH (mU/l) 3696 1.63 0.75 3.43

fT4 (pmol/l) 3703 13.42 11.43 15.92

TPO-Ab (IU/ml) 3686 0.11 0.00 11.41

The effects of repeated freezing and thawing

The children’s serum samples had been previously thawed 1-6 times for various

laboratory analyses. Fourteen percent of the samples were thawed for a second time,

and 84% were thawed for a third time for analyses connected with this study. We

tested the effect of repeated freezing and thawing for up to nine freeze-thaw cycles

using seven serum samples with an initial 5-year storage time at -80 °C from

healthy non-pregnant volunteers (provided by the National Institute for Health and

Welfare). Concentrations of TSH, fT4 and TPO-Ab were measured after every

other cycle (after thaws 1, 3, 5, 7 and 9). The concentrations of TSH, fT4 and TPO-

Abs did not change even after repeated freezing and thawing. This information was

personally given to Dr. Fanni Päkkilä by the head of the antenatal serology

laboratory (Dr. Heljä-Marja Surcel).

59

Categorization of the NFBC 1986 children (I, III)

In publication I and III, the 16-year-old children were categorized according to their

thyroid function status. However, the reference intervals given by the manufacturer

(Abbot Diagnostics) were based on adult population samples. Due to our large

number of children’s serum samples, we were able to create our own reference

intervals for TSH, fT4 and TPO-Abs. For reference interval calculation, TPO-Ab-

positive children were excluded, and data from boys and girls were analyzed

together and separately. Cases with notably high outlying TSH concentrations (>

5.28 mU/L) were identified and excluded from all analyses, as they were presumed

hypothyroid. Outliers were not identified among the TPO-Ab-positive children.

Separate reference intervals were then calculated with 2.5th and 97.5th percentiles

for TPO-Ab-negative and -positive children. For sensitivity analysis, all thyroid

function data and the prevalence of abnormal thyroid function parameters, thus

possible thyroid dysfunction and TPO-Ab-positivity, was assessed in children with

and without a maternal thyroid function analysis. The prevalence of abnormal

thyroid status was also evaluated using the manufacturer’s (Abbot Diagnostics)

reference intervals for the adult population for comparison.

4.2.3 Assessment of ADHD symptoms of the NFBC 1986 children (II)

The children’s ADHD symptoms were evaluated by their main teachers at eight years of age. The teachers were sent a questionnaire that included the official Finnish translations of the Rutter scale B2 with 29 questions (Almqvist et al. 1991). The scale has one question on inattention (“Child is not able to concentrate on anything for a longish period”) and two on hyperactivity (“Child is restless, does not have patience to sit down for a long period of time” and “Child wriggles and is restless”). The teachers answered the questions in a tripartite scale: “does not apply,” “applies sometimes” or “applies well” (rated 0-2 accordingly). The scale also includes questions on friends, general behavior during classes, mood swings, headaches and school absences. The total score from the Rutter B2 scale ranges from 0 to 52, where ≥ 9 indicates a probable psychiatric disturbance (Almqvist et al. 1991). Children were categorized as hyperactive if they scored 2 or more points combined on the questions on hyperactivity and as inattentive if they scored 1 or 2 points on inattention. Total Rutter B2 scores of ≥ 9 and a score of 3 or more points from the ADHD symptom questions indicated the highest probability of ADHD, and children meeting this criteria will be hence be referred to as having combined ADHD symptoms.

60

At the age of seven or eight, the children’s parents were also sent a postal

questionnaire on their learning and behavior (see also “assessment of scholastic

performance”). It included two questions on ADHD symptoms shared with the

teacher questionnaire: one on inattention (“Child is not able to concentrate on

anything for a longish period”) and one on hyperactivity (“Child is restless, does

not have patience to sit down for a long period of time”). The parents answered the

questions in the same tripartite way as the teachers. At the age of sixteen, parents answered another postal questionnaire, which

included an assessment instrument for ADHD symptoms, the Strengths and Weakness of ADHD Symptoms and Normal Behavior (SWAN) Scale (Swanson et al., 2001, available at http://www.adhd.net). It is an 18-item ADHD scale based on the 18 ADHD symptoms described in the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994). In the SWAN scale, ADHD symptoms are translated into statements and are rated with scores of 3, 2, or 1 (denoting difficulties); 0 (neutral behavior); and −1, −2, or −3 (denoting strengths). Previous scholars (Hurtig et al. 2007, Nyman et al. 2007, Smalley et al. 2007) have rated the SWAN scale ratings into summary scores. The 95th percentile of score distribution on a symptom scale was used as a cut-off point to define adolescents with probable ADHD symptoms (inattentive, hyperactive and combined type) (Smalley et al. 2007). Adolescents were categorized as having ADHD symptoms if one of those cut-off values was exceeded.

4.2.4 Scholastic performance of the NFBC 1986 children (III)

The Finnish school system

In Finland, all children attend a free-of-charge compulsory comprehensive school

for nine years between the ages of 7-16. All teachers in Finland are highly educated,

with master’s-level university degrees. The education system in different regions

highly accords with the Organization of Economic Cooperation and Development

Program in Secondary Assessment (OECD PISA) surveys (accessible at

http://dx.doi.org/10.1787/9789264091450-e). During the first six years of

compulsory education, children have a main class teacher (usually only one main

teacher per class) and after that, classes are taught by specialized subject teachers. Usually, students with minor learning or adjustment difficulties attend regular

schooling but are entitled to remedial teaching. Children who cannot follow the standard education due to an illness, disability or delayed development can be

61

admitted to or transferred to special-needs education. When possible, special-needs education must be integrated into regular education. Each student with special learning needs must be provided with an individual teaching and learning plan.

The school type attended by the NFBC 1986 children was asked in the parental questionnaire distributed when the children started elementary school (7-8 years of age). Data were received from 8416 children (89.9% of the total cohort living and traceable). Of these, 97.7% (n = 8219) of children attended public schools, 48 (0.8%) children went to private schools, 112 (1.1%) attended a special-needs school or a special-needs class and data was missing from 38 children (0.5%).

Assessment of scholastic performance in the NFBC 1986

First, the scholastic performance of the NFBC 1986 children was evaluated by their

main teachers at the age of eight, when the children were in the second year of

comprehensive schooling. The main teachers responded to whether a child had

difficulties in reading, writing and/or mathematics (yes/no). The second evaluation

was made during the final year at comprehensive school in 2001, when the 16-year-

old children self-evaluated their school performance in Finnish language and

mathematics. The children evaluated their school performance using a four-way

rating in comparison to their peers: “Better than average,” “Average,” “Worse than

average” or “Worse than most.” The children were also asked to report if they had

repeated a school year at any point during their compulsory education.

4.2.5 Assessment of intellectual problems in the NFBC 1986 (III)

The prevalence and etiology of intellectual disability in the NFBC 1986 was

previously studied in the dissertation of Dr. Ulla Heikura. In that study, the children

who potentially had intellectual problems were traced using maternal, peri- and

neonatal data collected via cohort questionnaires and during routine visits to free-

of-charge maternity and child welfare clinics. These data were linked with data

from Finnish national registers (Hospital Discharge Register, Cause-of-Death

Register, National Insurance and Medication Reimbursement Register) and

hospital, public health center and institutional health records. Data on psychometric

test results were collected from hospitals, institutions for children with intellectual

disabilities, family counselling centers and school psychologists. No separate

evaluation interventions were carried out for the purposes of the NFBC 1986

studies. The information collected was based on the routine Finnish clinical

62

practice of referring a child for further examinations as a result of developmental

or learning disorders, for example. Psychometric test results (82% of the testing

was conducted using the Wechsler Intelligence Scale for Children-Revised (WISC-

R)) and other relevant records were requested from all social/healthcare units in the

child’s original and/or current residential area (Heikura et al. 2005). A child was considered to have a severe intellectual disability if his/her IQ was

< 50 and was considered to have mild cognitive limitations if his/her IQ was ≥ 50 but ≤ 85. This assessment was based on either the most recent standardized psychometric test results or on a developmental assessment carried out on a clinical basis (Heikura et al. 2013). If no IQ estimate was available from psychometric testing, hospital records were sought to evaluate the assessment of a child’s intellectual level by a medical doctor or psychologist. If no assessment was found but it was obvious that the child had intellectual problems on the basis of a disorder or disease diagnosis (e.g., chromosomal disorders, specific syndromes, brain anomalies), then the classification was made by clinical estimation. This mainly pertained to cases of neonatal and infant deaths (Heikura et al. 2005).

4.2.6 Assessment of sensory development (IV)

Finnish child welfare clinic follow-up protocol

The NFBC 1986 children attended normal Finnish child welfare clinic follow-ups;

the results of these examinations were used in this study. In Finland, all children

participate in a follow-up protocol in local communal child welfare clinics. This

practice is regulated under the Finnish law (the Finnish Law on Public Health Care,

1326/2010 [Finlex]). Participating in the protocol is practically mandatory, and

social services are contacted if a family refuses to participate. From 1985-1986

onwards, all children in Finland have clinical examinations approximately once a

month up to one year during their first year of life, conducted either by a doctor or

a nurse. After the first year of life, the clinical examinations continue once or twice

a year depending on the family’s needs. The purpose of the child welfare clinic

follow-up program is to detect developmental delays in their early stages and to

enable their early treatment (Simell 1987).

The following protocol describes the follow-up program used in 1985-1986. In

order to detect childhood hearing impairments, a child’s hearing was examined as

follows: auditory localization response screening was conducted at eight months

by a doctor. A child’s speech production and understanding were assessed at the

63

ages of 12 months (by a nurse), 18 months (by a doctor) and at 2 and 3 years (by a

nurse). Audiometric screening for a hearing threshold of 20 dB for frequencies

0.25-4 kHz was conducted at the age of four or five by a nurse. If the child had an

infection at the time of auditory localization response or hearing threshold

screening, the infection was treated and the examination was repeated. If there were

still abnormal findings in the screenings, the child was sent for a further hearing

examination at the Oulu University Hospital in northern Finland. At about six years

old, a complete audiogram with hearing thresholds for frequencies 0.25-4 kHz was

conducted, and the audiogram was usually evaluated by the unit’s doctor.

Thresholds over 20dB were considered abnormal (Simell 1987).

A child’s vision was tested as follows: before the age of 2 years old, a child’s

pupillary reflexes were studied by a doctor at 1, 6, 8, 12 and 16 months. In addition,

strabismus was examined using the Hirschberg and cover tests. At the age of three,

a nurse showed a child pictures at a distance and asked the child to name them.

From the ages of four to seven, children saw a nurse once a year and were shown

Schnelle’s E-tables (Simell 1987).

Speech development assessment was routinely part of every child’s visit to a

nurse or doctor; all nurses in child welfare units are trained to observe children’s

communication with their parents. Speech disorders were specifically screened at

the age of two to three years when a child was asked to name objects pointed at by

a nurse. If the child did not pronounce any words correctly by the age of two, or

only pronounced a few words correctly at age three, the results were deemed

abnormal. Abnormal results led to speech therapy or hearing examinations.

Adequate production of speech sounds was evaluated by a nurse at the age of four

to five. If one or more sounds in certain words were incorrectly pronounced, the

test was repeated after a follow-up time of six months to one year. If the problem

continued, the child was directed to a speech therapist (one to three sounds incorrect)

or a phoniatrician (persistent problem) (Simell 1987).

Children were also additionally examined at any age if their parents expressed

any concerns. Children with diseases or syndromes affecting their overall

development such as Down syndrome were followed-up more intensively (Simell

1987).

Parental questionnaires in the NFBC 1986

In the child’s first autumn at school, at seven or eight years of age, parents were

sent the first of two postal questionnaires. The questionnaire included questions on

64

the school and family type, the social situation of the family and the child’s

development and health. The health and development data used in the current study

pertained to disturbances or delays for any reason in the development of sight (“Has

your child had any vision deficits?,” “Squinting?,” “Refractive errors?,” “If other,

specify”), in speech (“Is your child’s way of speech normal?,” “Any delay in speech

development or dysphasia?,” “Did your child require speech therapy before the age

of four?,” “If other, specify”) and in hearing (“Does your child have abnormal

hearing ability?,” “Has your child’s hearing required examination?,” “Abnormal

hearing ability in either ear?”). The parents answered on a three-way scale of “yes,”

“no” and “I do not know,” and the answers were further categorized as either “yes”

or “no” (“no” and “I don’t know” were categorized as “no”).

Welfare clinic data

To supplement the parental questionnaire data, a research assistant visited local

healthcare centers during the same year with a similar health and development

questionnaire form as the one sent to the children’s parents. The only addition to

the parental form was a question on any possible diagnose codes of the children.

Data were manually collected from case records. Data was categorized as “yes”

and “no” in the same way as the parental questionnaire data was categorized. Vision

defect diagnoses other than squint and refractive errors in children were categorized

as “other.”

4.2.7 Statistical methods

All statistical analyses were performed using SPSS software for Windows versions 18.0-20.0 (SPSS Inc., Chicago, Illinois, USA). Continuous variables with normal distributions (family and maternal characteristics) were compared with t-tests and variables with non-Gaussian distributions (TSH, fT4, TPO-Ab), using the Mann–Whitney U test. The prevalence of categorical variables was compared using chi-square tests. Pearson’s χ2 test and Fisher’s exact test were used to evaluate the prevalence differences. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated with logistic regression analysis.

In studies II, III and IV, first-line analyses were conducted amongst all children with an IQ ≥ 85, and in II and III, further adjustments were made for gender, number of children in the family at the time of the scholastic performance evaluation (≥ 2 vs. 1), maternal smoking (yes vs. no), socioeconomic status of the family

65

(professional, skilled, unskilled or farmers) and maternal age (< 20 or > 35 years vs. 20-35 years). All data in studies I-IV were also analyzed by including and excluding TPO-Ab-positive mothers and mothers with thyroid disease, currently or before pregnancy. The data was stratified to term and preterm (< 37 gestational weeks) children to see if preterm birth impacted the associations. All data was also stratified according to sex, because adjusting for a child’s sex significantly affected all results. In study IV, the previous adjustments were made when analyzing a child’s risk of vision defects.

Adolescents’ reference intervals (I)

For reference interval calculation, TPO-Ab-positive children were excluded, and

data from boys and girls were analyzed together and separately. Data was

logarithmically transformed to achieve normality. Outliers were identified using a

detection method described by Horn et al. (Horn et al. 2001). All outliers were

evaluated, and cases with notably high outlying TSH concentrations (> 5.28 mU/L)

were excluded from all analyses. Outliers were not identified among the TPO-Ab-

positive children. Separate reference intervals were then calculated with the 2.5th

and 97.5th percentiles for TPO-Ab-negative and -positive children. Bias-corrected

and accelerated 95% confidence intervals of the percentiles were calculated with

1000 bootstrap resamples.

66

67

5 Results and comments

5.1 Maternal and adolescent thyroid function (I)

5.1.1 Association of maternal and adolescent’s thyroid function

In publication I, the study population consisted of mother-child pairs for which

thyroid function analyses were available for both; thus the study population was

smaller. The categorization of the mothers in publication I is presented in Figure 3.

For a representativeness analysis, we tested the mothers’ TSH, fT4 and TPO-Ab

concentrations when the child’s serum samples were available and when they were

not; no difference was observed (Table 5). We found no statistically significant

differences in TSH, fT4 or TPO-Ab concentrations between these groups (Table 5).

Seventeen children were reported to use levothyroxine, and one was reported to use

a thyreostatic drug; they were all included in the analyses.

Table 5. Comparison of mothers with and without child’s laboratory data available

Thyroid function

parameter

Mothers with no available

child serum samples

Mothers with available child

serum samples

n concentration n concentration

TSH (mU/l) 2090 1.21 (0.21-3.88) 3689 1.20 (0.17-3.40)

fT4 (pmol/l) 2077 15.11 (11.91-20.22) 3649 15.10 (12.01-20.64)

TPO-Ab (IU/ml) 2086 4.18 (1.98-216.66) 3677 4.35 (2.04-155.31)

All thyroid function test values are median (5th-95th percentile). Comparison of the groups with and

without children’s laboratory data has been conducted with independent samples T-test with no

statistically significant findings.

The adolescent children of hypothyroid mothers had higher median TSH

concentrations (p-value < 0.005), and adolescent children of hyperthyroid mothers

had lower median TSH concentrations (p-value < 0.05) than those of euthyroid

mothers. The adolescent children of hypothyroid mothers also had statistically

significantly lower median fT4 and higher TPO-Ab concentrations than those of

euthyroid mothers. Adolescent children of hypothyroxinemic mothers had higher

TPO-Ab concentrations (p-value < 0.05), but their median TSH and fT4

concentrations did not differ from those of euthyroid mothers. Sons of hypothyroid

mothers had higher median TSH concentrations than sons of euthyroid mothers.

68

Fig. 3. Flow chart of the study population (Päkkilä et al. 2013, published with permission

of The Endocrine Society).

69

Daughters of hypothyroid mothers and sons of hypothyroxinemic mothers had

higher median TPO-Ab concentrations than those of euthyroid mothers. The results

remained similar after excluding mothers using thyroid medication and after

categorizing mothers by only using TSH data (Table 6).

Adolescent children of TPO-Ab-positive mothers had significantly higher

median TPO-Ab concentrations than those of TPO-Ab-negative mothers. TPO-Ab-

positive mothers were more likely to have TPO-Ab-positive offspring (Figure 4):

9.0% vs. 3.7% among boys and 22.7% vs. 7.5% among girls.

Hypo- and hyperthyroid mothers had a statistically significant increased risk

of having offspring with hypo- and hyperthyroidism, respectively, compared to

euthyroid mothers (Table 7). The results did not substantially change after

excluding mothers using thyroid medication and after categorizing mothers by only

using data on TSH. The children of hypothyroxinemic mothers did not have an

increased risk of having hypothyroxinemia: 5.1% vs. 2.4% (p=0.251) for all

children, 4.3% vs. 2.8% (p=0.488) for boys and 6.3% vs. 2.0% (p=0.289) for girls.

Fig. 4. Prevalence of TPO-Ab-positive children (%) grouped by maternal antibody status.

70

Ta

ble

6. C

om

pa

riso

n o

f th

e N

FB

C 1

986

ad

ole

sc

en

ts t

hy

roid

fu

nc

tio

n p

ara

me

ters

acc

ord

ing

to

mate

rna

l th

yro

id f

un

cti

on

sta

tus

.

Adole

scent th

yro

id

ho

rmo

ne

conce

ntr

atio

ns

Euth

yroid

H

ypoth

yroid

H

ypert

hyr

oid

H

ypoth

yro

xin

em

ic

TP

O-A

b-n

eg

ativ

e

TP

O-A

b-p

osi

tive

N (

moth

er-

child

pa

irs)

3178-3

189

203-2

07

81

45

3490-3

503

168-1

71

TS

H (

mU

/L)

1.6

(0

.8-3

.4)

1.8

(0

.8-3

.9)*

**

1.3

(0

.6-3

.5)*

**

1.5

(0

.7-5

.0)

1.6

(0

.8-3

.4)

1.7

(0

.8-3

.7)

fT4

(p

mo

l/L)

13

.4 (

11

.4-1

6.0

) 1

3.4

(1

1.3

-15

.6)*

1

3.6

(1

1.5

-15

.6)

13

.1 (

10

.6-1

5.6

) 1

4.4

(1

1.4

-15

.9)

14

.6 (

11

.6-1

6.2

)

TP

O-A

b (

IU/m

L)

0.1

(0

.0-9

.4)

0.2

4 (

0.0

-94

.4)*

* 0

.21

(0

.0-4

09

.9)

0.4

(0

.0-1

57

.1)*

* 0

.6 (

0.0

-7.5

) 1

.5 (

0.0

-24

6.0

)***

Boys

N (

moth

er-

child

pa

irs)

1589-1

591

107-1

08

32

27

1735

89

TS

H (

mU

/L)

1.7

(0

.8-3

.3)

2.0

(0

.9-4

.0)*

**

1.3

(0

.6-4

.2)*

* 1

.6 (

0.6

-5.1

) 1

.7 (

0.8

-3.4

) 1

.8 (

0.9

-3.4

)

fT4

(p

mo

l/L)

13

.4 (

11

.4-1

5.9

) 1

3.1

(1

1.4

-15

.7)

13

.0 (

10

.7-1

5.1

) 1

3.1

(1

0.8

-16

.9)

13

.4 (

11

.4-1

5.8

) 1

3.2

(1

1.7

-16

.2)

TP

O-A

b (

IU/m

L)

0.1

(0

.0-3

.9)

0.2

(0

.0-6

.5)

0.0

(0

.0-4

.3)

0.5

(0

.0-1

60

.6)

* 0

.1 (

0.0

-3.7

) 0

.2 (

0.0

-51

.1)*

*

Girls

N (

moth

er-

child

pa

irs)

1589-1

598

96-9

9

49

18

1746

75

TS

H (

mU

/L)

1.6

(0

.7-3

.4)

1.6

(0

.5-3

.9)

1.3

(0

.5-3

.5)*

1

.5 (

0.6

-3.9

) 1

.6 (

0.7

-3.5

) 1

.6 (

0.7

-3.8

)

fT4

(p

mo

l/L)

13

.5 (

11

.5-1

6.0

) 1

3.3

(1

1.2

-15

.5)

13

.6 (

11

.6-1

5.8

) 1

3.1

(1

0.1

-15

.1)

13

.5 (

11

.5-1

6.0

) 1

3.7

(1

1.5

-16

.2)

TP

O-A

b (

IU/m

L)

0.1

(0

.0-3

3.9

) 0

.3 (

0.0

-25

5.5

)*

0.3

(0

.0-5

06

.5)

0.4

(0

.0-7

2.4

) 0

.1 (

0.0

-32

.9)

0.6

(0.0

-38

7.1

)***

Num

bers

vary

due

to m

issi

ng s

am

ple

s. D

istr

ibutio

ns

are

exp

ress

ed a

s m

edia

n (

5–95th

perc

entil

es)

. P

-valu

es

were

obta

ined w

ith t

he

Ma

nn

–W

hitn

ey

U test

when c

om

paring in

div

idua

l gro

ups

with

the m

ate

rnal e

uth

yroid

gro

up,

or

an a

ntib

ody-

po

sitiv

e g

roup w

ith a

n a

ntib

ody-

negativ

e g

rou

p.

*P <

0.0

5,

**P

< 0

.01

,

***P

< 0

.00

1

71

Table 7. Adolescents’ estimated odds to have the same thyroid dysfunction as their

mothers.

Maternal

thyroid status

N of

mothers

OR (95% CIs)

N of all

mothers

ORa (95% CIs)

All

Children Boys Girls

All

children Boys Girls

Hypothyroid 207

(108/99)

2.7

(1.5-4.6)

2.2

(1.0-5.1)

3.1

(1.5-6.6)

118

(60/58)

3.4

(1.8-6.5)

2.9

(1.0-5.1)

3.9

(1.7-9.0)

Hyperthyroid 81

(32/49)

3.7

(1.5-8.8)

4.7

(1.0-20.8)

3.0

(1.0-8.8)

67 (23/44) 4.1

(1.7-9.8)

5.9

(1.3-26.7)

3.1

(1.1-9.0)

Number of mothers: mothers of all children (boys/girls). a: TPO-Ab-positive mothers (TPO-Ab

concentration ≥ 167.7 IU/mL) excluded from analysis.

5.1.2 Comments on maternal and adolescent thyroid function

No previous studies on the association between maternal thyroid function during

pregnancy and an adolescent child’s thyroid function later in life were found. There

was some evidence from a twin study that individual levels of TSH might be

genetically controlled and that fT4 and fT3 are more susceptible to environmental

influence than TSH (Panicker et al. 2008). In this study, maternal thyroid function

was assessed once in early pregnancy; unfortunately, possible clinical symptoms

and diagnoses of maternal thyroid disease were not available. However, mothers

who previously used thyroid medication were excluded. A cohort study from the

Netherlands associated maternal thyroid function and child’s thyroid hormone

parameters at birth and during childhood (measured from five to nine years of age);

the association persisted after adjusting for genetic risk scores (Korevaar et al. 2015a). Our results were the first to show that maternal abnormal thyroid function

during pregnancy, even without a known diagnosis, associates with a child’s risk

of having the same thyroid dysfunction in adolescence. Analyses were also

conducted using only maternal TSH concentrations for sensitivity analysis because

fT4 measurements may not be totally reliable during pregnancy. The results did not

change.

Daughters of mothers with positive TPO-Ab concentrations during early

pregnancy had approximately a three-fold risk of being TPO-Ab-positive

themselves in adolescence. Previous data shows that TPO-Ab-positivity is more

prevalent in girls with Tanner stages of II-V compared to Tanner stage I

(Kaloumenou et al. 2008). It is therefore possible that puberty has an impact on the

autoimmune status of adolescents. The current study population is 16 years old, and

72

most of them have probably gone through puberty. Therefore, it can be postulated

that this study identified most of the individuals that will be seropositive during

pregnancy and thus at risk of thyroid dysfunction and postpartum thyroiditis. The

results of the current study emphasize the importance of screening for thyroid

dysfunction in early pregnancy in those individuals who have known thyroid

diseases in the family. Screening for thyroid dysfunction is also important in the

non-pregnant population when an individual with a history of thyroid disease or

other autoimmune disease in the family presents clinical symptoms of thyroid

dysfunction.

5.1.3 Reference intervals for serum TSH and fT4 concentrations in

the adolescent population and thyroid function in adolescence

A total of 5962 (62.9% of the NFBC 1986) adolescents had data available on both

TSH and fT4 concentrations. These data were used to calculate reference intervals

for TSH and fT4 concentrations for adolescents. After excluding TPO-Ab-positive

children and children with high outlying values, 5765 (60.8% of the NFBC 1986)

adolescent samples were available for reference interval calculation. Adolescent

reference intervals are presented in Table 8; these are 0.64-3.74 mU/L for TSH and

11.01-16.63 pmol/L for fT4.

When using these reference values of TSH and fT4 concentrations to evaluate

the prevalence of thyroid dysfunction in adolescents, 219 (3.7%) adolescents had

overt or subclinical hypothyroidism, 149 (2.6%) had overt or subclinical

hyperthyroidism and 140 (2.4%) had hypothyroxinemia. The respective prevalence

of adolescents with maternal thyroid function data available (n = 3673) was 125

(3.4%) hypothyroid adolescents, of whom 60 were boys and 65 girls, 90 (2.5%)

hyperthyroid adolescents, of whom 29 were boys and 61 girls and 82 (2.2%)

hypothyroxinemic adolescents of whom 47 were boys and 35 girls.

When using reference intervals from the manufacturer (0.4-4.0 mU/L for TSH

and 9.0-19.0 pmol/L for fT4), 164 (2.8%) adolescents were hypothyroid, 31 (0.5%)

were hyperthyroid and no children had hypothyroxinemia.

Overall, adolescent boys had higher median TSH concentrations than girls (p

< 0.001). Girls had higher median TPO-Ab concentrations than boys (p = 0.002).

TPO-Ab-positive boys had higher median TSH and lower fT4 concentrations than

TPO-Ab-negative boys (TSH 1.84 vs. 1.67 mU/L, p = 0.04 and fT4 13.00 vs. 13.41

pmol/L, p = 0.03) and TPO-Ab-positive girls had higher median TSH

concentrations than TPO-Ab-negative girls (TSH 2.10 vs. 1.53 mU/L, p < 0.001).

73

5.1.4 Thyroid autoimmunity in the adolescent population

There were 381 (6.6%) adolescents who were TPO-Ab-positive with a TPO-Ab

concentration > 5.61 mU/L, of whom 248 (65.1%) were girls. Of hypothyroid

adolescents with TPO-Ab measurements available, 57/210 (27.1%) were TPO-Ab-

positive, whereas 15/147 (10.2%) of hyperthyroid adolescents were TPO-Ab-

positive. There was no difference in TPO seropositivity in subjects with and

without maternal thyroid function data.

The unadjusted odds of a TPO-Ab-positive adolescent having either hypo- or

hyperthyroidism was 4.3 (95% CI 3.3-5.7). Among TPO-Ab-positive boys, the

prevalence of hypo- or hyperthyroidism was 21.3%, and the odds of having hypo-

or hyperthyroidism was 5.6 (95% CI 3.5-8.9). Among TPO-positive girls, the

prevalence of thyroid dysfunction was 17.7%, and the odds of having thyroid

dysfunction was 3.5 (95% CI 2.5-5.1).

74

Ta

ble

8. R

efe

ren

ce

in

terv

als

wit

h 9

5%

CIs

se

pa

rate

ly f

or

TP

O-A

b-n

eg

ati

ve a

nd

-p

os

itiv

e a

do

les

cen

ts.

TS

H (

mU

/L)

fT

4 (

pm

ol/L

)

Ad

ole

sce

nt

TP

O-A

b

statu

s

n

2.5

th p

erc

entil

e (

95%

CI)

97.5

t h p

erc

entil

e (

95%

CI)

2

.5th

pe

rce

ntil

e (

95

% C

I)

97

.5t h

perc

entil

e (

95%

CI)

TP

O-A

b-n

eg

ativ

e

Bo

ys

27

64

0

.69

(0

.65

-0.7

0)

3.7

6 (

3.6

1-3

.95

)

11

.02

(1

0.8

6-1

1.1

4)

16

.44

(1

6.2

4-1

6.7

6)

Girls

2

63

3

0.5

9 (

0.5

7-0

.61

) 3

.66

(3

.56

-3.8

3)

1

1.0

8 (

10

.90

-11

.22)

16

.75

(1

6.5

1-1

6.9

8)

All

TP

O-A

b-n

eg

ativ

e

child

ren

57

64

0

.64

(0

.62

-0.6

5)

3.7

4 (

3.6

2-3

.84

)

11

.01

(1

0.9

1-1

1.1

2)

16

.63

(1

6.4

5-1

6.7

9)

TP

O-A

b-p

osi

tive

Bo

ys

12

2

0.3

5 (

0.0

0-0

.88

) 1

1.7

9 (

7.5

8-6

4.2

2)

1

0.0

4 (

9.2

6-1

0.9

5)

16

.23

(1

5.4

5-1

7.2

9)

Girls

2

46

0

.14

(0

.03

-0.6

8)

9.0

6 (

6.0

8-1

4.1

3)

1

0.5

7 (

9.4

4-1

1.2

1)

17

.92

(1

6.6

0-2

1.1

4)

75

5.1.5 Comments on the adolescent thyroid function analysis

Using population-based reference intervals among these adolescents identified a

statistically significant higher number of subjects with abnormal thyroid function

compared to the manufacturer’s reference intervals. Whether these children have

clinical symptoms of thyroid dysfunction is unknown. The use of the

manufacturer’s reference intervals for adults may fail to identify 55/219 (25.1%)

cases of possible hypothyroidism and 118/148 (79.7%) cases of possible

hyperthyroidism. The population-based adolescents’ reference interval for fT4 is

narrower than the manufacturer’s reference interval for the adult population. This

may be due to the adolescent population being healthier than the adult population.

Seventeen adolescents were known to use thyroxin, and one used a thyreostatic

drug. These subjects were included in the analysis because their thyroid function

values were assumed to be in a good balance.

Adolescents with positive TPO-Ab concentrations had significantly increased

odds of having abnormal thyroid function compared to TPO-Ab-negative

individuals. Our results are in line with one previous study from Finland showing

that one third of TPO-Ab-positive children had subclinical hypothyroidism

(Kondrashova et al. 2008). These findings signify the common autoimmune origin

of thyroid diseases. TG-Ab concentrations in adolescents were not analyzed in this

study, and TG-Ab-positivity might account for further cases of abnormal thyroid

function. In clinical practice, however, measuring TPO-Ab or TG-Ab

concentrations are not usually first line studies.

5.2 The association between maternal thyroid dysfunction and

ADHD symptoms in children (II)

The participation rate for the teacher questionnaire conducted when the NFBC 1986

children were 7-8 years old was 92% of the living and traceable NFBC 1986 cohort.

Among the families with data on thyroid function during pregnancy, 5089 teacher

questionnaires were completed, and a total of 5131 questionnaires had at least one

answered item on ADHD. A total of 1030 children (20.1%) had inattention

symptoms, 1095 children (21.3%) had hyperactivity symptoms, 686 children

(13.4%) had total Rutter B2 scores ≥ 9 and 461 children (9.0%) had combined

ADHD symptoms (inattention and hyperactivity symptoms with total Rutter B2

scores ≥ 9), based on the teachers’ evaluation. Overall, boys had a higher prevalence

76

of ADHD symptoms, Rutter B2 scores ≥ 9 and combined ADHD symptoms than

girls (Table 9). No statistically significant differences were seen in the prevalence

or odds of ADHD symptoms in children born to mothers with high and normal

serum TSH concentrations, or hypothyroxinemia and normal fT4 concentrations

(Table 9).

Table 10 presents the children’s estimated odds of ADHD symptoms by natural

logarithm increases in maternal serum TSH concentrations. In boys, no association

was observed between ADHD symptoms and log-increases in maternal TSH. Girls

had 1.2-fold odds of inattention and Rutter scores ≥ 9 and 1.4-fold odds of

combined ADHD symptoms with every natural log-increase in maternal TSH. As

sensitivity analyses, the exclusion of TPO-Ab-positive mothers and those using

thyroid medication affected our results only minimally. The number of preterm

children was too small to study the association of ADHD symptoms and maternal

thyroid function in preterm children.

The children’s inattention and hyperactivity symptoms were also evaluated by

their parents at eight years of age. According to parental evaluations, there were

797 (15.2%) children with inattention symptoms and 1166 (22.2%) children with

hyperactivity symptoms. According to the results of the parental evaluations, there

was no association between increasing maternal TSH concentrations, high TSH

concentrations, decreasing fT4 concentrations or maternal hypothyroxinemia and a

child’s inattention or hyperactivity symptoms. The results did not change after

categorizing children by sex. Parental inattention and hyperactivity evaluations

were also combined with teacher evaluations; this did not change the results.

77

Ta

ble

9. P

rev

ale

nc

e (

n [

%])

an

d e

sti

ma

ted

od

ds

of

AD

HD

sy

mp

tom

s i

n c

hil

dre

n o

f m

oth

ers

wit

h h

igh

an

d n

orm

al

se

rum

TS

H

co

nc

en

tra

tio

ns a

nd

wit

h o

r w

ith

ou

t h

yp

oth

yro

xin

em

ia.

Child

’s A

DH

D s

ympto

ms

Norm

al T

SH

(n =

4763)a

Hig

h T

SH

(n =

348)a

aO

R (

95%

CI)

b

No

rma

l fT

4

(n =

4370)a

Hyp

oth

yroxi

nem

ia

(n =

66)a

aO

R (

95%

CI)

c

Ina

tte

ntio

n,

n (

%)

Boys

7

00

(1

4.7

) 4

4 (

12

.6)

0.7

6 (

0.5

1-1

.12

) 6

41

(1

4.7

) 9

(9

.1)

0.7

5 (

0.3

3-1

.69

)

Girls

2

60

(5

.5)

19

(5

.5)

1.1

4 (

0.6

6-1

.97

) 2

42

(5

.5)

2 (

3.0

) 0

.66

(0

.15

-2.8

3)

Hyp

era

ctiv

ity,

n (

%)

Boys

7

45

(1

5.6

) 5

6 (

16

.1)

1.0

7 (

0.7

5-1

.52

) 6

80

(1

5.6

) 8

(1

2.1

) 0

.69

(0

.31

-1.5

6)

Girls

2

67

(5

.6)

21

(6

.0)

1.2

0 (

0.7

0-2

.03

) 2

47

(5

.7)

3 (

4.5

) 1

.07

(0

.31

-3.6

5)

Ru

tte

r B

2 s

core

s ≥ 9

, n

(%

)

Boys

4

62

(9

.7)

27

(7

.8)

0.7

7 (

0.4

9-1

.24

) 4

23

(9

.7)

7 (

10

.6)

0.7

5 (

0.2

8-1

.96

)

Girls

1

77

(3

.7)

14

(4

.0)

1.3

0 (

0.7

0-2

.42

) 1

62

(3

.7)

0

NA

Com

bin

ed A

DH

D,

n (

%)

Boys

3

30

(6

.9)

21

(6

.0)

0.8

9 (

0.5

2 -

1.5

0)

29

9 (

6.8

) 3

(4

.5)

0.6

1 (

0.1

8-2

.03

)

Girls

9

5 (

2.0

) 1

0 (

2.9

) 1

.69

(0

.79

-3.5

6)

86

(2

.0)

0

NA

a N

um

bers

may

vary

due t

o m

issi

ng m

ate

rnal T

SH

and

/or

free T

4 m

easu

rem

ents

or

ava

ilabili

ty o

f te

ach

er

eva

luatio

n o

f ch

ild’s

sym

pto

ms.

30

te

ach

er

quest

ionna

ires

we

re n

ot co

mple

tely

fill

ed o

ut.

b T

he o

dds

ratio

s w

ith 9

5%

CIs

for

child

’s A

DH

D s

ympto

ms

after

exp

osu

re t

o h

igh m

ate

rnal T

SH

conce

ntr

atio

ns

during e

arly

pre

gnancy

are

ca

lcula

ted b

y lo

gis

tic r

egre

ssio

n,

after

adju

stin

g f

or

havi

ng >

2 c

hild

ren in

fam

ily, m

ate

rnal s

moki

ng,

mate

rnal e

duca

tion a

nd m

ate

rnal a

ge. H

igh T

SH

conce

ntr

atio

n w

as

defin

ed a

s h

avi

ng T

SH

conce

ntr

atio

n >

3.1

mU

/L in

the f

irst

or

> 3

.5 m

U/L

in the s

eco

nd trim

est

er.

c T

he o

dds

ratio

s w

ith 9

5%

CIs

for

child

’s A

DH

D s

ympto

ms

after

exp

osu

re t

o m

ate

rnal h

ypo

thyr

oxi

nem

ia d

urin

g e

arly

pre

gnancy

are

calc

ula

ted b

y lo

gis

tic

regre

ssio

n, after

adju

stin

g for

ha

ving

>2 c

hild

ren in

fam

ily, m

ate

rnal s

moki

ng, m

ate

rnal e

duca

tion a

nd m

ate

rna

l age. M

ate

rnal h

ypoth

yroxi

nem

ia w

as

defin

ed

as

havi

ng n

orm

al T

SH

conce

ntr

atio

n a

nd fre

e T

4 c

once

ntr

atio

n <

11.4

pm

ol/L

in t

he f

irst

or

< 1

1.1

pm

ol/L

in t

he

seco

nd t

rim

est

er.

78

Table 10. Prevalence and odds of ADHD symptoms in children after exposure to

increasing maternal serum TSH concentration.

Child’s ADHD symptom n (%)a aOR (95% CI)b

Inattention

Boys 749 (14.6) 1.00 (0.90-1.12)

Girls 281 (5.5) 1.18 (1.02-1.37)

Hyperactivity

Boys 806 (15.7) 1.00 (0.90-1.10)

Girls 289 (5.6) 1.10 (0.95-1.25)

Rutter B2 scores ≥ 9

Boys 494 (9.6) 1.02 (0.90-1.15)

Girls 192 (3.7) 1.23 (1.03-1.48)

Combined ADHD

Boys 355 (6.9) 1.17 (1.00-1.36)

Girls 106 (2.1) 1.39 (1.07-1.80) a Compared to N of all mother-children pairs with sufficient maternal serum samples and children’s ADHD symptom evaluation available (n=5131). b The odds ratios with 95% CIs for ADHD symptoms in the child per one natural logarithmic unit increase in maternal early pregnancy thyroid stimulating hormone concentrations are calculated with logistic regression, after adjusting for having > 2 children in family, maternal smoking, maternal education and maternal age.

5.2.1 Comments

In the current study, a modest association was seen between increased maternal

early pregnancy TSH concentrations and girls’ high Rutter B2 scores and combined

ADHD symptoms when evaluated by teachers. In boys, no association was seen

between maternal thyroid function status and teacher evaluated ADHD symptoms.

Only one previous population-based study on the possible association of maternal

thyroid function in early pregnancy and a child’s ADHD symptoms or diagnosis at

school age has been conducted (Modesto et al. 2015). In that study, a child’s ADHD

symptoms were evaluated by his or her parents using the Conners’ Parent Rating

Scale – Revised Short Form with 12 items concerning ADHD symptoms. The

children were aged six to eleven years old. The children of hypothyroxinemic

mothers had 7% higher ADHD scores compared to children of euthyroid mothers.

However, the large age distribution must be acknowledged as a limitation of this

study (Modesto et al. 2015).

Previous literature indicates that teacher evaluations of ADHD symptoms

using the Rutter B2 form is more reliable than parental evaluations using the Rutter

A2 scale, and Rutter B2 scales can better discern psychiatric disturbances than

79

Rutter A2 and the Children’s Depression Inventory (Kresanov et al. 1998).

Therefore, in our study, more weight was given to the teachers’ evaluation of

ADHD symptoms. Teachers are shown to be more competent in evaluating a child’s

abnormal behavior because they observe more children in their working

environment than parents do. Furthermore, a child’s behavior might be different in

school than at home (De Los Reyes & Kazdin 2005, Kolko & Kazdin 1993).

Parental evaluation is affected by level of education and the amount of stress

associated with a child’s behavior (Collishaw et al. 2009).

Our current study results must be interpreted with care because the clinical

diagnosis of ADHD was not used in this study (diagnostic interviews were available

only for a low number of cases). Children with combined ADHD symptoms in

childhood tended to meet the inattentive subtype diagnosis in adolescence (when a

subpopulation of the NFBC 1986 had ADHD diagnostics conducted) (Hurtig et al. 2007).

The number of mothers with high TSH concentrations in our cohort was small.

Therefore, it was meaningful to also study maternal TSH concentrations as a

continuous variable. The clinical implication of the increased odds of an association

of a child’s high Rutter B2 scores and combined ADHD with logarithmically

increasing maternal TSH concentrations is, nevertheless, difficult to specify. A mild

association possibly exists, but no specific cut-off point for TSH concentrations can

be found in this population. ADHD is a relatively common multifactorial disorder.

According to our study, the effect of mild maternal thyroid dysfunction during

pregnancy in this equation is small.Maternal and adolescent thyroid function and

their association on scholastic performance and intellect (III)

5.2.2 Maternal thyroid dysfunction during early pregnancy and the

scholastic performance of children (III)

The study protocol is described in Figure 5. Teacher responses were available for

8525 children (92% of the living and traceable NFBC 1986 children). Teachers

evaluated the children’s performance at school at eight years of age. A total of 590

(11.6%) children had reading difficulties, 839 (16.5%) had writing difficulties, 416

(8.2%) had difficulties in mathematics and 1062 (20.9%) had difficulties in at least

one of these. No significant differences in scholastic problems at eight years of age

were observed when children of mothers with thyroid dysfunction were compared

to children of euthyroid mothers. Increased TSH did not increase a child’s risk of

scholastic problems.

80

Self-reported responses were available for 7344 children (80% of the living

and traceable NFBC 1986 children). At 16 years old, 374 (8.6%) adolescents

evaluated themselves as being worse than average in Finnish language, 1020

(23.4%) worse than average in mathematics, 1204 (27.7%) worse than average in

either Finnish or mathematics and 78 (1.8%) adolescents had repeated a class.

Adolescents of hypothyroid mothers had similar scholastic outcomes as those

of euthyroid mothers (Table 11). Adolescents of mothers with subclinical

hypothyroidism had a statistically significant increased prevalence and increased

unadjusted odds of repeating a class at school than those of euthyroid mothers (3.4%

vs. 1.6%, OR 2.14, 95% CI 1.01–4.53, adjusted OR 2.02, 95% CI 0.78-5.21).

Logarithmically increasing TSH did not increase adolescents’ odds of poor

scholastic performance.

Adolescents of hyperthyroid mothers reported difficulties in mathematics more

often than those of euthyroid mothers (32.0% vs. 23.2%, OR 1.56, 95% CI 1.03-

2.38, adjusted OR 1.61, 95% CI 1.01-2.49) (Table 11). The numbers were not

sufficient to analyze overt and subclinical hyperthyroidism separately. Adjustment

factors included the number of children in the family at the time of the scholastic

performance evaluation (≥ 2 vs. 1), maternal smoking (yes vs. no), socioeconomic

status of the family (professional, skilled, unskilled or farmers) and maternal age

(< 20 or > 35 years vs. 20-35 years).

Adolescents of hypothyroxinemic mothers had an increased prevalence and

increased unadjusted odds of having repeated a class at school than those of

euthyroid mothers (5.5% vs. 1.6%, OR 3.49, 95% CI 1.06–11.48) (Tables 12 and

13). However, after adjustments, the difference was not statistical significance (OR

3.50 95% CI 0.81-15.21). All of the adolescents of hypothyroxinemic mothers who

repeated a class were boys. They had significantly higher odds of repeating a class

at school than those of euthyroid mothers (10.0% vs. 2.3%, OR 4.74, 95% CI 1.38–

16.25, adjusted OR 5.46, 95% CI 1.19-25.06) (Table 11).

All results were similar after excluding preterm births, TPO-Ab-positive

mothers and mothers with diagnosed and/or treated thyroid disease during the index

pregnancy. Maternal hyperthyroidism associated with the self-evaluated

mathematical skills of the adolescents and hypothyroxinemia correlated with class

repetition. These associations were not mediated by the adolescents’ ADHD

symptoms at the age of 16 (SWAN-evaluations) because maternal thyroid

dysfunction did not associate with ADHD symptoms in adolescents at 16 years of

age

81

Fig. 5. Flow chart of the study population in scholastic performance evaluations of the

NFBC 1986 (Päkkilä F et al 2015, published with permission of The American Thyroid

Association)

82

Ta

ble

11

. T

he

pre

va

len

ce a

nd

OR

of

a c

hil

d’s

sch

ola

sti

c d

iffi

cu

ltie

s a

t a

ge

s e

igh

t (t

ea

ch

er

ev

alu

ati

on

) an

d s

ixte

en

(s

elf

-ev

alu

ati

on

)

by

mate

rna

l th

yro

id s

tatu

s

Sch

ola

stic

pe

rfo

rma

nce

Eu

thyr

oid

(n =

4831)

H

ypoth

yroid

n =

365 (

7.0

%)

H

ypoth

yroxi

nem

ic

n =

71 (

1.4

%)

H

ypert

hyr

oid

n =

124 (

2.5

%)

Pre

vale

nce

Pre

vale

nce

OR

s (9

5%

CIs

)

Pre

vale

nce

OR

s (9

5%

CIs

) P

reva

lence

O

Rs

(95%

CIs

)

8-y

ea

r

quest

ionna

ire:

Diff

icu

lties

in

readin

g

49

8/4

33

5 (

11

.5%

)

41

/32

8 (

12

.5%

) 1

.10

(0

.78

-1.5

5)

4

/66

(6

.1%

) 0

.50

(0

.18

-1.3

7)

12

/11

8 (

10

.2%

) 0

.87

(0

.48

-1.6

0)

Bo

ys

32

9 (

21

89

(1

5.0

%)

2

9/1

73

(1

6.8

%)

1.1

4 (

0.7

5-1

.73

)

3/3

9 (

7.7

%)

0.4

7 (

0.1

4-1

.54

) 8

/42

(1

9.0

%)

1.3

3 (

0.6

1-2

.90

)

Girls

1

69

/21

46

(7

.9%

)

12

/15

5 (

7.7

%)

0.9

8 (

0.5

3-1

.81

)

1/2

7 (

3.7

%)

0.4

5 (

0.0

6-3

.34

) 4

/76

(5

.3%

) 0

.65

(0

.24

-1.8

0)

Diff

icu

lties

in

writin

g

71

1/4

33

5 (

16

.4%

)

56

/32

9 (

17

.0%

) 1

.05

(0

.78

-1.4

1)

8

/66

(1

2.1

%)

0.7

0 (

0.3

3-1

.48

) 1

5/1

18

(1

2.7

%)

0.7

4 (

0.4

3-1

.28

)

Bo

ys

46

1/2

18

6 (

21

.1%

)

39

/17

3 (

22

.5%

) 1

.09

(0

.75

-1.5

8)

5

/39

(1

2.8

%)

0.5

5 (

0.2

1-1

.42

) 1

0/4

2 (

23

.8%

) 1

.17

(0

.57

-2.4

0)

Girls

2

50

/21

49

(1

1.6

%)

1

7/1

56

(1

0.9

%)

0.9

3 (

0.5

5-1

.56

)

3/2

7 (

11

.1%

) 0

.95

(0

.28

-3.1

8)

5/7

6 (

6.6

%)

0.5

4 (

0.2

1-1

.34

)

Diff

icu

lties

in

math

em

atic

s 3

53

/43

28

(8

.2%

)

26

/32

8 (

7.9

%)

0.9

7 (

0.6

4-1

.47

)

7/6

6 (

10

.6%

) 1

.34

(0

.61

-2.9

5)

7/1

17

(6

.0%

) 0

.72

(0

.33

-1.5

5)

Bo

ys

16

3/2

18

0 (

7.5

%)

1

5/1

72

(8

.7%

) 1

.18

(0

.68

-2.0

6)

4

/39

(1

0.3

%)

1.4

1 (

0.5

0-4

.03

) 2

/42

(4

.8%

) 0

.62

(0

.15

-2.5

8)

Girls

1

90

/21

48

(8

.8%

)

11

/15

6 (

7.1

%)

0.7

8 (

0.4

2-1

.47

)

3/2

7 (

11

.1%

) 1

.29

(0

.38

-4.3

2)

5/7

5 (

6.7

%)

0.7

4 (

0.2

9-1

.85

)

Diff

icu

lties

in a

t

least

one

9

00

/43

37

(2

0.8

%)

7

3/3

29

(2

2.2

%)

1.0

9 (

0.8

3-1

.43

)

12

/66

(1

8.2

%)

0.8

5 (

0.4

5-1

.59

) 2

3/1

17

(1

9.7

%)

0.9

3 (

0.5

9-1

.48

)

Bo

ys

54

0/2

18

4 (

24

.7%

)

50

/17

3 (

28

.9%

) 1

.24

(0

.88

-1.7

4)

6

/39

(1

5.4

%)

0.5

5 (

0.2

3-1

.33

) 1

3/4

2 (

31

.0%

) 1

.37

(0

.70

-2.6

4)

Girls

3

60

/21

53

(1

6.7

%)

2

3/1

56

(1

4.7

%)

0.8

6 (

0.5

5-1

.36

)

6/2

7 (

22

.2%

) 1

.42

(0

.57

-3.5

5)

10

/75

(1

3.3

%)

0.7

7 (

0.3

9-1

.51

)

83

Sch

ola

stic

pe

rfo

rma

nce

Eu

thyr

oid

(n =

4831)

H

ypoth

yroid

n =

365 (

7.0

%)

H

ypoth

yroxi

nem

ic

n =

71 (

1.4

%)

H

ypert

hyr

oid

n =

124 (

2.5

%)

Pre

vale

nce

Pre

vale

nce

OR

s (9

5%

CIs

)

Pre

vale

nce

OR

s (9

5%

CIs

) P

reva

lence

O

Rs

(95%

CIs

)

16

-ye

ar

quest

ionna

ire:

Re

pe

ate

d a

cla

ss

61

/37

50

(1

.6%

)

8/2

65

(3

.0%

) 1

.88

(0

.89

-3.9

8)

3

/55

(5

.5%

) 3

.49

(1

.06

-11

.48

)*

0/1

02

N

A

Bo

ys

41

/17

90

(2

.3%

)

5/1

29

(3

.9%

) 1

.72

(0

.67

-4.4

3)

3

/30

(1

0.0

%)*

4

.74

(1

.38

-16

.25

)*

0/3

4

NA

Girls

2

0/1

96

0 (

1.0

%)

3

/13

6 (

2.2

%)

2.1

9 (

0.6

4-7

.46

)

0/2

5

NA

0

/68

N

A

Diff

icu

lties

in

Fin

nis

h

32

7/3

75

2 (

8.7

%)

2

1/2

62

(8

.0%

) 0

.91

(0

.58

-1.4

5)

3

/54

(5

.6%

) 0

.62

(0

.19

-1.9

9)

6/1

03

(5

.8%

) 0

.65

(0

.28

-1.4

9)

Bo

ys

24

1/1

79

5 (

13

.4%

)

17

/12

7 (

13

.4%

) 1

.00

(0

.59

-1.6

9)

3

/29

(1

0.3

%)

0.7

4 (

0.2

2-2

.48

) 4

/34

(1

1.8

%)

0.8

6 (

0.3

0-2

.46

)

Girls

8

6/1

95

7 (

4.4

%)

4

/13

5 (

3.0

%)

0.6

6 (

0.2

4-1

.84

)

0/2

5

NA

2

/69

(2

.9%

) 0

.65

(0

.16

-2.7

0)

Diff

icu

lties

in

math

em

atic

s 8

66

/37

39

(2

3.2

%)

6

2/2

61

(2

3.8

%)

1.0

3 (

0.7

7-1

.39

)

14

/55

(2

5.5

%)

1.1

3 (

0.6

2-2

.09

) 33/1

03

(32

.0%

)*

1.5

6 (

1.0

3-

2.3

8)*

Bo

ys

34

9 (

19

.5%

)

20

/12

7 (

15

.7%

) 0

.77

(0

.47

-1.2

6)

7

/30

(2

3.3

%)

1.2

6 (

0.5

4-2

.96

) 9

/34

(2

6.5

%)

1.4

9 (

0.6

9-3

.22

)

Girls

5

17

/19

47

(2

6.6

%)

4

2/1

34

(3

1.3

%)

1.2

6 (

0.8

7-1

.84

)

7/2

5 (

28

.0%

) 1

.08

(0

.45

-2.5

9)

24

/69

(3

4.8

%)

1.4

8 (

0.8

9-2

.45

)

Diff

icu

lties

in

Fin

nis

h o

r

math

em

atic

s

1026/3

735

(27

.5%

)

75

/26

1 (

28

.7%

) 1

.07

(0

.81

-1.4

1)

1

5/5

4 (

27

.8%

) 1

.02

(0

.56

-1.8

5)

34

/10

3 (

33

.0%

) 1

.30

(0

.86

-1.9

7)

Bo

ys

47

6/1

79

0 (

26

.6%

)

33

/12

7 (

26

.0%

) 0

.97

(0

.64

-1.4

6)

8

/29

(2

7.6

%)

1.0

5 (

0.4

6-2

.39

) 1

0/3

4 (

29

.4%

) 1

.15

(0

.55

-2.4

2)

Girls

5

50

/19

45

(2

8.3

%)

4

2/1

34

(3

1.3

%)

1.1

6 (

0.7

9-1

.69

)

7/2

5 (

28

.0%

) 0

.99

(0

.41

-2.3

8)

24

/69

(3

4.8

%)

1.3

5 (

0.8

2-2

.24

)

The p

reva

lence

of

a s

chola

stic

outc

om

e h

as

been c

alc

ula

ted b

y th

e n

um

ber

of ch

ildre

n w

ith a

sch

ola

stic

pro

ble

m/tota

l num

ber

of

child

ren w

ith d

ata

on s

chola

stic

perf

orm

ance

ava

ilable

in the m

ate

rnal t

hyr

oid

funct

ion g

roup (

and p

erc

enta

ge).

*p <

0.0

5, w

hen th

e p

reva

lence

of sc

hola

stic

pro

ble

ms

were

com

pare

d b

y usi

ng c

hi s

quare

test

s.

In 8

-year

eva

luatio

ns,

10.0

–10.9

% o

f th

e d

ata

wa

s m

issi

ng a

nd in

16-y

ear

eva

luatio

ns,

22.5

–22.9

% w

as

mis

sin

g, va

ryin

g b

etw

ee

n q

uest

ions.

84

5.2.3 Maternal thyroid function and severe intellectual

deficiency/mild cognitive limitation in children

In the NFBC 1986, 147 children with maternal thyroid function data available had

IQs ≤ 85. This was based on IQ testing in 142 children and on a clinical diagnosis

in 5 children. A total of 25 (0.4% of the entire NFBC 1986) of these were

categorized as having a severe intellectual deficiency (IQ < 50) (14 boys and 11

girls), and 117 (2.0% of the NFBC 1986) were categorized as having a mild

cognitive limitation (50 ≤ IQ ≤ 85) (82 boys and 35 girls). Maternal thyroid

dysfunction was not associated with the prevalence or odds of a severe intellectual

deficiency or mild cognitive limitation in the child (Tables 12 and 13).

The adjusted results were similar but were attenuated due to small sample sizes.

Adjustment factors included the number of children in the family at the time of the

scholastic performance evaluation (≥ 2 vs. 1), maternal smoking (yes vs. no),

socioeconomic status of the family (professional, skilled, unskilled or farmers) and

maternal age (< 20 or > 35 years vs. 20-35 years).

Table 12. Prevalence of children’s intellectual problems in maternal thyroid dysfunction.

Child’s intellectual

problem

Euthyroidism

n = 4957

Hypothyroid

n = 375 (7.0%)

Hypothyroxinemic

n = 73 (1.5%)

Hyperthyroid

n = 127 (2.5%)

Mild cognitive

limitation (na/nb) 100/4931(2.0%) 8/373 (2.1%) 2/71 /2.7%) 2/126 (1.6%)

Boys 73/2527 (2.9%) 4/198 (2.0%) 2/44 (4.5%) 1/45 (2.2%)

Girls 27/2403 (1.1%) 4/175 (2.3%) 0/29 1/81 (1.2%)

Severe intellectual

deficiency (na/nb) 22/4853 (0.5%) 1/366 (0.3%) 0/71 1/125 (0.8%)

Boys 11/2465 (0.4%) 1/195 (0.5%) 0/42 1/45 (2.2%)

Girls 11/2387 (0.5%) 0/171 0/29 0/80

All children with IQ ≤85

(na/nb) 126/4831 (2.5%) 10/375 (2.7%) 2/73 (2.7%) 3/127 (2.4%)

Boys 87/2541 (3.4%) 6/200 (3.0%) 2/44 (4.5%) 2/46 (4.3%)

Girls 49/2415 (1.6%) 4/175 (2.3%) 0/29 1/81 (1.2%)

The groups include mothers with information on IQ available. Mild cognitive limitation: IQ ≥50 but ≤85.

Severe intellectual deficiency: IQ < 50. na/nb: Number of children with an intellectual problem/total number

of children in the maternal thyroid function group (and percentage).

85

Table 13. Maternal thyroid dysfunction and a child's risk of intellectual disability.

Child’s intellectual problem Hypothyroid

n = 375

Hypothyroxinemic

n = 73

Hyperthyroid

n = 127

Mild cognitive limitation 1.08 (0.52-2.24) 1.40 (0.34-5.78) 0.79 (0.19-3.25)

Boys 0.70 (0.25-1.94) 1.63 (0.39-6.88) 0.76 (0.10-5.61)

Girls 2.15 (0.74-6.23) NA 1.15 (0.15-8.57)

Severe intellectual deficiency 0.55 (0.07-4.07) NA 1.63 (0.22-12.15)

Boys 1.06 (0.14-8.19) NA 4.69 (0.50-36.79)

Girls NA NA NA

IQ ≤85 1.05 (0.55-2.02) 1.08 (0.26-4.45) 0.93 (0.29-2.96)

Boys 0.87 (0.38-2.02) 1.34 (0.32-5.64) 1.28 (0.31-5.37)

Girls 1.43 (0.50-4.04) NA 0.76 (0.10-5.61)

Mild cognitive limitation: IQ ≥50 but ≤85. Severe intellectual deficiency: IQ < 50.

5.2.4 Abnormal adolescent thyroid function parameters at 16 years

of age and self-evaluated scholastic performance

Hyperthyroid adolescent girls had more self-reported difficulties in Finnish than

euthyroid girls (11.1% vs. 4.2%, OR 2.82, 95% CI 1.42-5.61) (Table 15).

Adolescent boys with hypothyroxinemia more often self-reported having

difficulties in Finnish and/or mathematics than euthyroid boys (43.1% vs. 26.2%,

OR 2.13, 95% CI 1.26-3.62) (Table 14).

Because hyperthyroid girls had more problems in learning Finnish and

hypothyroxinemic boys had more difficulties with Finnish and/or mathematics, the

association of adolescent thyroid function and ADHD symptoms in 16-year-old

adolescents was evaluated. The prevalence of ADHD symptoms increased in boys

with hypothyroxinemic thyroid function parameters compared to euthyroid boys

(10.2% vs. 5.2%, OR 2.26 1.01-5.06) (Table 15).

86

Ta

ble

14

. A

do

les

cen

ts’ t

hy

roid

fu

nc

tio

n a

nd

pre

va

len

ce

an

d O

Rs

(9

5%

CIs

) o

f se

lf-e

va

lua

ted

po

or

sc

ho

las

tic

pe

rfo

rma

nc

e.

Adole

scent ha

s

diff

iculti

es

in

Eu

thyr

oid

n =

5256 (

89.3

%)

H

ypoth

yroid

n =

216 (

3.7

%)

H

ypoth

yroxi

nem

ic

n =

134 (

2.3

%)

H

ypert

hyr

oid

n =

146 (

2.5

%)

P

reva

lence

Pre

vale

nce

O

Rs

(95%

CIs

)

Pre

vale

nce

O

Rs

(95%

CIs

) P

reva

lence

O

Rs

(95%

CIs

)

Fin

nis

h

42

2/4

80

1 (

8.8

%)

1

4/2

04

(6

.9%

) 0

.77

(0

.44

-1.3

3)

1

2/1

13

(1

0.6

%)

1.2

3 (

0.6

7-2

.26

) 1

7/1

33

(1

2.8

%)

1.5

2 (

0.9

1-2

.56

)

Bo

ys

31

8/2

34

7 (

13

.5%

)

9/1

02

(8

.8%

) 0

.62

(0

.31

-1.2

4)

1

1/5

8 (

19

.0%

) 1

.49

(0

.77

-2.9

1)

7

/43

(1

6.3

%)

1.2

4 (

0.5

5-2

.81

)

Girls

1

04

/24

53

(4

.2%

)

5/1

02

(4

.9%

) 1

.16

(0

.46

-2.9

2)

1

/55

(1

.8%

) 0

.42

(0

.06

-3.0

5)

1

0/9

0 (

11

.1%

) 2

.82

(1

.42

-5.6

1)

*

Ma

the

ma

tics

10

68

/47

79

(2

2.3

%)

4

1/2

04

(2

0.1

%)

0.8

7 (

0.6

2-1

.24

)

25

/11

3 (

22

.1%

) 0

.99

(0

.63

-1.5

5)

3

0/1

31

(2

2.9

%)

1.0

3 (

0.6

8-1

.56

)

Bo

ys

45

1/2

33

6 (

19

.3%

)

15

/10

2 (

14

.7%

) 0

.72

(0

.41

-1.2

6)

1

3/4

3 (

30

.2%

) 1

.73

(0

.98

-3.0

8)

7

/43

(1

6.3

%)

0.8

1 (

0.3

6-1

.84

)

Girls

6

16

/24

42

(2

5.2

%)

2

6/1

02

(2

5.5

%)

1.0

1 (

0.6

4-1

.60

)

8/5

5 (

14

.5%

) 0

.51

(0

.24

-1.0

7)

2

3/8

8 (

26

.1%

) 1

.05

(0

.65

-1.7

0)

Fin

nis

h a

nd

/or

ma

th

12

70

/48

12

(2

6.4

%)

4

8/2

04

(2

3.5

%)

0.8

6 (

0.6

2-1

.19

)

34

/11

3 (

30

.1%

) 1

.20

(0

.80

-1.8

0)

4

0/1

33

(3

0.1

%)

1.2

0(0

.82

-1.7

5)

Bo

ys

61

7/2

35

4 (

26

.2%

)

20

/10

2 (

19

.6%

) 0

.69

(0

.42

-1.1

3)

2

5/5

8 (

43

.1%

) 2

.13

(1

.26

-3.6

2)

* 1

3/4

3 (

30

.2%

) 1

.22

(0

.63

-2.3

5)

Girls

6

52

/24

57

(2

6.5

%)

2

8/1

02

(2

7.5

%)

1.0

5 (

0.6

7-1

.63

)

9/5

5 (

16

.4%

) 0

.54

(0

.26

-1.1

1)

2

7/9

0 (

30

.0%

) 1

.19

(0

.75

-1.8

8)

Has

repeate

d a

class

8

6/4

91

2 (

1.8

%)

2

/20

5 (

1.0

%)

0.5

5 (

0.1

4-2

.26

)

2/1

23

(1

.6%

) 0

.93

(0

.23

-3.8

1)

1

/13

8 (

0.7

%)

0.4

1 (

0.0

6-2

.96

)

Bo

ys

58

/23

45

(2

.5%

)

2/1

01

(2

.0%

) 0

.80

(0

.19

-3.3

1)

0

/58

(0

.0%

) N

A

1

/43

(2

.3%

) 0

.94

(0

.13

-6.9

4)

Girls

2

6/2

45

2 (

1.1

%)

0

/10

2 (

0.0

%)

NA

1/5

6 (

1.8

%)

1.6

9 (

0.2

3-1

2.7

3)

0

/91

(0

.0%

) N

A

87

Table 15. The association between adolescents’ thyroid function status and ADHD

symptoms.

ADHD symptom Euthyroid Hypothyroid Hypothyroxinemic Hyperthyroid

Inattentive type 177/4404 (4.0%) 7/182 (3.8%) 7/118 (5.9%) 1/121 (0.8%)

Boys 121/2171 (5.6%) 5/91 (5.5%) 5/59 (8.5%) 1/37 (2.7%)

Girls 53/2193 (2.4%) 2/90 (2.2%) 2/55 (3.6%) 0/82 (0%)

Hyperactive type 188/4404 (4.3%) 5/182 (2.7%) 10/118 (8.5%) 8/121 (6.6%)

Boys 118/2171 (5.4%) 3/91 (3.3%) 8/59 (13.6%) 3/37 (8.1%)

Girls 66/2193 (3.0%) 2/90 (2.2%) 2/55 (3.6%) 5/82 (6.1%)

ADHD-case 171/4404 (3.9%) 6/182 (3.3%) 8/118 (7.6%) 5/121 (4.1%)

Boys 112/2171 (5.2%) 5/91 (6.5%) 6/59 (10.2%)* 3/37 (7.7%)

Girls 54/2193 (2.5%) 1/90 (3.9%) 2/55 (3.6%) 2/82 (2.4%)

Children with IQ≤85 excluded from analysis.* p-value <0.05 when compared to euthyroid children.

5.2.5 Comments

Maternal thyroid dysfunction had a small effect on a child’s self-evaluated

scholastic performance at the age of 16. Adolescent children of hyperthyroid

mothers had more problems in mathematics, and boys of hypothyroxinemic

mothers were more likely to repeat a class at school. The present data is unique,

and no similar data were found in the literature. Cognitive functioning has been

studied in very young and school-aged children, and adverse associations have been

found between maternal hypothyroidism and hypothyroxinemia during pregnancy

and children’s neuropsychological development (Haddow et al. 1999, Henrichs et al. 2010, Julvez et al. 2013, Ghassabian et al. 2014, Korevaar 2015b). A study by

Korevaar et al. associated both low and high maternal fT4 concentrations with

neuropsychiatric changes; that finding is somewhat in line with the current study

results (Korevaar et al. 2015b). However, children in that study were between the

ages of five and nine years in IQ testing and between six and 11 years in MRI

imagining (Korevaar et al. 2015b).

In the current study, boys of hypothyroxinemic mothers more often repeated a

class at school. It can be postulated that this may relate directly to low maternal fT4

concentrations, or to some other maternal disease, which may cause maternal

hypothyroxinemia and may or may not be known.

Adolescent girls with hyperthyroidism reported more problems in Finnish

language. The possible mechanism behind the association is unclear, but it probably

relates to the other symptoms of hyperthyroidism. Hypothyroxinemic boys more

88

often reported problems in Finnish language and/or mathematics.

Hypothyroxinemia in a non-pregnant population might be a sign of some other

disease such as obesity or asthma, which might affect child’s learning more than

the hypothyroxinemia itself.

It is plausible that ADHD symptoms and specific learning difficulties account

for a vast majority of learning difficulties in adolescents with a normal IQ (Taanila et al. 2014). Lowered IQ naturally brings its own learning difficulties because the

etiology of the lowered IQ might be a specific disease or a syndrome. According to

the current results, the impact of maternal thyroid dysfunction on IQ is small.

However, when examining a child with ADHD symptoms or learning difficulties,

TSH and fT4 measurements would reasonably be used to exclude possible thyroid

diseases, especially if the child has other symptoms of thyroid dysfunction.

Maternal thyroid function did not have an effect on a child’s intellectual

disability. The prevalence and etiology of intellectual disability in the NFBC 1986

have been previously studied in a dissertation by Dr. Ulla Heikura. According to

her studies, the majority of intellectual disabilities were caused by biomedical

factors (93.7 % in severe and 47.9% in mild cases). When studying the timing of

the etiology, prenatal factors accounted for 58.8% in both categories. Down

syndrome was the most common genetic etiology, and a few cases of single and

multifactorial gene mutations were identified. Malformations accounted for 9.2%,

and external prenatal disorders (caused by maternal infection, medication, toxins,

etc.) accounted for 16.0% of intellectual disability cases. Paranatal factors (factors

occurring between one week before delivery and four weeks after) such as

infections, delivery complications and other neonatal complications were found in

3.4% of cases and postnatal factors in 4.2% of cases. Unknown causes applied in

33.6% of the cases. The lower the IQ, the more severe and early the prenatal factor

seemed to be that caused it (Heikura et al. 2005).

Maternal thyroid diseases during pregnancy are diagnosed and treated in most

cases. Women with overt thyroid diseases may not even become pregnant

(Stagnaro-Green et al. 2011). When studying the effects of milder thyroid function

abnormalities on a child’s neuropsychological development, the multifactorial

etiology of neuropsychological disorders must be remembered.

It is known, however, that severe maternal thyroid dysfunction during

pregnancy increases a child’s risk of adverse cognitive outcomes, in the worst case

leading to cretinism (Man & Jones 1969). The present study, in combination with

the previous literature, suggests that an association between mild maternal thyroid

dysfunction during pregnancy and a child’s adverse neuropsychological

89

development also exists. Previous literature, however, varies with regard to how

strong this association is; this is probably due to differences in populations, settings

and the severity of maternal thyroid diseases included in these studies. Also, some

studies used accurate IQ testing, and some relied on parental questionnaire data on

learning and behavior symptoms (Ghassabian et al. 2014, Henrichs et al. 2010,

Julvez et al. 2013, Modesto et al. 2015, Korevaar et al. 2015b). The uncertain

clinical implication of numerical IQ testing must be remembered: even if a child of

a hypothyroid or hypothyroxinemic mother scored a few points less in IQ testing

than the average score of a child of a euthyroid mother, the significance in the

child’s real life might be minimal.

In conclusion, this study measured the actual scholastic performance and

behavior of children in a normal school environment. Children with an IQ ≤85 were

excluded due to previous cohort research (Heikura 2008). Our cohort mothers

mostly had mild thyroid dysfunction during pregnancy, and those with a known

previous thyroid disease were excluded from our analyses. Although a small

association between mild maternal thyroid dysfunction during pregnancy and some

scholastic and behavioral outcomes was seen in our study (Table 16), our results

were mostly relieving. The children of mothers with thyroid dysfunction performed

as well as the children of euthyroid mothers in practically all of the

neuropsychological sectors that were studied. We conclude that a child’s

compensatory mechanisms in the CNS during growth enable normal learning even

after exposure to mild maternal thyroid underactivity during pregnancy.

90

Ta

ble

16

. S

um

ma

ry o

f th

e c

hil

d’s

AD

HD

s

ym

pto

ms a

nd

s

ch

ola

sti

c p

erf

orm

an

ce re

su

lts

(s

tud

ies

II

an

d III)

b

y m

ate

rna

l an

d

ad

ole

sc

en

t th

yro

id f

un

cti

on

sta

tus

.

Ch

ild o

utc

om

e

Ma

tern

al H

YP

O

Ma

tern

al

hyp

oth

yroxi

nem

ia

Ma

tern

al H

YP

ER

A

do

lesc

en

t H

YP

O

Ad

ole

sce

nt

hyp

oth

yroxi

nem

ia

Ad

ole

sce

nt

HY

PE

R

8-y

ear

eva

luatio

ns

Inattentio

n o

r

hyp

era

ctiv

ity

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

- -

-

Ru

tte

r B

2 s

core

s ≥9

Girls

: O

R o

f ln

TS

H 1

.23

non-s

ignifi

cant

non-s

ignifi

cant

- -

-

Com

bin

ed A

DH

D

Girls

: O

R o

f ln

TS

H 1

.39

non-s

ignifi

cant

non-s

ignifi

cant

- -

-

16-y

ear

eva

luatio

ns

Has

repeate

d a

class

non-s

ignifi

cant

OR

3.4

9 (

boys

4.7

4)

not applic

aple

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

Pro

ble

ms

in F

innis

h

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

Girls

: O

R 2

.82

Pro

ble

ms

in m

ath

s non-s

ignifi

cant

non-s

ignifi

cant

OR

1.5

6

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

Pro

ble

ms

in F

innis

h

an

d/o

r m

ath

s

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

Boys

: O

R 2

.13

non-s

ignifi

cant

AD

HD

sym

pto

m

eva

luatio

n (

SW

AN

)

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

non-s

ignifi

cant

Boys

: O

R 2

.26

non-s

ignifi

cant

-: n

ot ca

lcula

ted b

eca

use

of diff

ere

nt

time p

oin

t. L

n T

SH

: H

YP

O n

on-s

ign

ifica

nt, b

ut lo

garith

mic

ally

incr

easi

ng T

SH

conce

ntr

atio

n r

esu

lt diff

ere

nt.

91

5.3 The effect of maternal thyroid dysfunction on child sensory

and linguistic development (IV)

There were 9326 living and traceable NFBC 1986 children when the cohort was

seven to eight years old. The parental response rate was 90.2% for the living and

traceable NFBC 1986. The percentage of completed parental questionnaires

regarding the sensory and linguistic development of children with maternal thyroid

function data available varied from 55.8% to 100%, depending on the questions.

Welfare clinic data were obtained from 7014 children (75.2% of the total NFBC

1986). The data were, however, available only from 0-2329/5391 (43.2%) children

with an IQ >85 and available maternal thyroid function data.

Of the whole NFBC 1986, 9.4% of children were reported to have a vision

impairment at some point in their life, 6.5% had a squint, 5.4% had a refractive

error and 1.2% of the children had other impairments such as amblyopia, myopia,

hyperopia, esophoria, color sensitivity deficit, unpaired pupils, artery malformation,

a cataract or esophoria. Of the NFBC 1986, 7.6% were reported to have speech

abnormalities, 4.0% had delayed speech development or dysphasia, 5.4% had other

speech deficits (cannot speak [0.5%], developmental delay [1.7%], hoarse voice

[0.7%], sound deficits [93.9%], stuttering [3.8%]), 3.1% needed speech therapy and

9.1% altogether were reported to have had any of the previous conditions. Of the

entire cohort, 1.5% had lowered hearing ability (parental reporting), 3.9% had been

examined for abnormal hearing, 2.9% were reported to have lowered hearing ability

only in one ear and 5.5% altogether were reported to have any of the previous

hearing abnormalities.

Among mothers with laboratory data available, 1024/5391 (19.0%) of children

had an abnormal sensory outcome. The prevalence of overall child outcomes

among mothers with thyroid dysfunction did not differ compared to the overall

prevalence of euthyroid mothers’ child outcomes (Table 17). Children of

hypothyroid, and especially hypothyroxinemic mothers, however, had a higher

prevalence of vision deficits (10.8% and 11.7%, respectively) than those of

euthyroid mothers (6.5%), although this difference was not statistically significant.

92

Table 17. Maternal thyroid function parameters during pregnancy and child’s sensory

and linguistic development (n of abnormal values/n of answered questions [%])

Child outcome Euthyroid

(n = 4831)

Hypothyroid

(n = 365)

Hypothyroxinemic

(n = 71)

Hyperthyroid

(n = 124)

Vision

Squint 152/3564 (4.3%) 16/282 (5.7%) 4/60 (6.7%) 2/79 (2.5%)

Refractive error 135/3564 (3.8%) 14/282 (5.0%) 4/60 (6.7%) 3/79 (3.8%)

Other 20/3564 (0.6%) 4/282 (1.4%) 1/60 (1.7%) 1/79 (1.3%)

Any previous vision

deficits

233/3564 (6.5%) 25/282 (10.8%) 7/60 (11.7%) 4/79 (5.1%)

Speech

Is your child's speech

normal

385/4529 (8.5%) 24/345 (7.0%) 8/64 (12.5%) 10/116 (8.6%)

Delayed development

of speech or

dysphasia

102/4317 (2.4%) 7/331 (2.1%) 1/59 (1.7%) 1/113 (0.9%)

Other) 212/4311 (4.9%) 10/330 (3.0%) 4/59 (6.8%) 7/113 (6.2%)

Needed speech

therapy before 4 years

old

110/4294 (2.6%) 6/329 (1.8%) 1/59 (1.7%) 0

Any previous speech

deficits

477/4538 (10.5%) 30/345 (8.7%) 9/64 (14.1%) 10/116 (8.6%)

Hearing

Lowered hearing

ability

87/4527 (1.9%) 4/345 (1.2%) 1/64 (1.6%) 2/116 (1.7%)

Child has been

examined for

abnormal hearing

92/2698 (3.4%) 12/217 (5.5%) 0 3/56 (5.4%)

Abnormal hearing in

either ear

120/4831 (2.5%) 10/365 (2.7%) 1/71 (1.4%) 2/124 (1.6%)

Any previous hearing

deficits

252/4831 (5.2%) 22/365 (6.0%) 2/71 (2.8%) 4/124 (3.2%)

93

The median maternal TSH, fT4 and TPO-Ab concentrations did not differ

between mothers with and without the sensory development data of their children.

The results did not substantially change after excluding children born at < 37 weeks

or mothers with positive TPO-Ab concentrations.

5.3.1 Comments

In this population of children who participated in an excellent follow-up program,

no association was seen between mild maternal thyroid dysfunction in early

pregnancy and the child’s sensory and linguistic development outcomes. This may

be partly due to the fact that outcomes in this study were rare, and although our

dataset was large, the number of abnormal outcomes was not sufficient to study

their association with maternal hypothyroidism and hypothyroxinemia. Our data

only included mothers with relatively mild thyroid dysfunction. Furthermore, in

Finland children receive rehabilitation quickly if sensory problems are detected in

child welfare clinics. Therefore, potential outcomes may have been halted through

early intervention.

Missing data may have affected our study results. Although the parental

response rate to the cohort questionnaire was excellent (90%), some parents did not

fully complete it even though they were given an opportunity to answer “I do not

know.” Questions concerning their child’s vision were particularly answered less

often, but they were still answered in over 70% of questionnaires. Parental

evaluation may not be as reliable as a professional evaluation, but the Finnish

healthcare system treats all children similarly, and parents usually remember if their

child has needed additional examinations. We were able to partially supplement

questionnaire data from the welfare clinics. However, because sensory impairments

are examined and treated in hospitals in Finland, the communal welfare clinics

probably did not have hospital data in their records.

Only a few studies on the association of maternal thyroid dysfunction during

pregnancy and a child’s sensory and linguistic development have been published

previously (Craig et al. 2012, Haddow et al. 1999, Oken et al. 2009, Pop et al. 1999). In the previous studies, the primary outcome was a child’s

neuropsychological development, and the follow-up varied from newborn age to

three years. Henrichs et al. associated maternal hypothyroxinemia to a child’s

increased risk of expressive language delay at the ages of 18 and 30 months

(Henrichs et al. 2010). The only previous study with a longer follow-up time

associated a child’s increased risk of sensorineural hearing loss at eight years of

94

age with positive maternal TPO-Ab concentrations (Wasserman et al. 2008). In our

study, no association was found between maternal thyroid dysfunction and/or TPO-

Ab-positivity and linguistic or hearing impairments in children.

95

6 Conclusions

The following conclusions can be drawn on the basis of studies I, II, III and IV:

1. Abnormal maternal thyroid function during early pregnancy associates with a

child’s thyroid function at the age of 16 years. This association may be

explained by the child’s increased odds of having positive TPO-Ab

concentrations if the mother is TPO-Ab-positive as well. TPO-Ab-positive

adolescents have increased odds of abnormal thyroid function. The reference

intervals for adults may not be sufficient for detecting thyroid dysfunction

among adolescents.

2. Maternal thyroid dysfunction is not clearly associated with a child’s risk of

teacher- or parent-evaluated ADHD symptoms at the age of 7-8.

3. Maternal thyroid dysfunction during early pregnancy had a mild effect on the

scholastic performance of adolescents. Furthermore, the thyroid function of the

adolescents themselves seems to affect their scholastic performance. Therefore,

when examining children or adolescents with learning difficulties or behavioral

problems, it may be beneficial to analyze their thyroid function.

4. Abnormal maternal thyroid function during pregnancy did not associate with a

child’s sensory and linguistic outcomes.

96

97

7 General discussion

7.1 Strengths and limitations of the present study

The current study is one of the first to evaluate the effects of maternal thyroid

dysfunction in early pregnancy on the thyroid function and neuropsychological and

sensory development of the offspring on a population level. In addition, this study

evaluated the effects of an adolescent’s own thyroid function on

neuropsychological outcomes. Furthermore, the long follow-up time distinguished

this study from previous ones. The NFBC 1986 covers over 99% of the pregnant

population during one year in northern Finland where no private hospitals exist,

communal healthcare, child welfare clinics and schooling systems are alike and the

population is homogenous (Järvelin et al. 1993).

The maternal serum samples were drawn mostly during the first trimester,

which is important when evaluating maternal effects on the child’s

neurodevelopment. It was assured that mothers with thyroid function data

represented the whole cohort. In addition, thyroid hormone reference intervals for

the pregnant population were used.

Cohort researchers have collected a tremendous amount of data during follow-

up concerning maternal, family and children’s medical, social and economic factors;

therefore, it was possible to adjust the results for the most important confounding

factors. In addition, the cohort researchers have been able to collaborate with each

other, and also with other Finnish health care institutions and registers, in order to

complement questionnaire and clinical intervention data. Collaboration has been

especially important when gathering data on the children’s growth, possible

illnesses, scholastic performance and factors affecting their intellect. Confirmed

clinical diagnosis of the children improved the assessment of their sensory

development.

Excellent participation led to over 5000 mother-child pairs and the availability

of parental evaluations of the children’s growth and sensory development, teacher

evaluations of the children’s ADHD symptoms and scholastic performance, the

adolescents’ self-evaluations of their scholastic performance at age 16 and maternal

thyroid function test results.

Sensory and neuropsychological development outcomes and intellectual

disabilities were rare in our cohort, which limited the power of the study with

regards to the association of maternal thyroid dysfunction and these disabilities.

98

Because some of the maternal or adolescent serum samples were unavailable and

some questionnaires were incompletely answered, some results must be interpreted

with care. However, the parental questionnaire data was supplemented with welfare

clinic data in the sensory and linguistic development study. In Finland, children are

freely provided with speech or physiotherapy and additional learning assistance;

therefore, sensory and linguistic deficits and specific learning difficulties are rare

in older children. In the NFBC 1986, most of the children’s intellectual problems

had a known etiology and did not associate with maternal thyroid dysfunction.

Serum fT4 measurements during pregnancy may not be totally reliable when

collected through immunoassay, although they are used clinically in addition to

TSH measurements. Fortunately, it was possible to use previously calculated

population- and trimester-specific reference intervals to define maternal thyroid

function, as currently recommended when gold-standard methods are not possible

(Stagnaro-Green et al. 2011).

This study did not incorporate children’s thyroid disease and ADHD diagnoses.

However, at the age of 16 years and in this iodine sufficient population, clinically

manifest thyroid diseases would probably be rare; it was therefore more worthwhile

to study associations of maternal and adolescent thyroid hormone concentrations.

Only a small subcohort of the adolescents went through diagnostic interviews for

ADHD. Therefore, ADHD symptoms were studied at both ages for consistency and

in order to obtain a representative sample of subjects. The ADHD symptom

questionnaires have been validated in the Finnish population.

7.2 Confounding factors

Some confounding factors, which affect both maternal thyroid function and child

outcomes, need to be acknowledged when evaluating the results of this and other

studies. Maternal smoking during pregnancy is a risk factor for ADHD symptoms

in children (Kotimaa et al. 2003), and advanced maternal age is associated with a

higher risk of adverse child outcomes (Kirkinen & Ryynänen 2015), which may

affect a child’s intellect. Our cohort included a comprehensive amount of maternal

health data, and therefore we had information on maternal smoking and age. These

factors did not alter our results.

Data were also adjusted or stratified for sex because ADHD and scholastic

problems are not equally representative in boys and girls (Moilanen et al. 2013),

and thyroid diseases are more common among girls (Kondrashova et al. 2007,

Kaloumenou et al. 2008). Data was adjusted for a family’s socioeconomic status

99

(calculated with maternal and paternal employment status) because low family

educational levels adversely affect a child’s learning abilities (Taanila et al. 2011).

Data was also adjusted for the number of children in the family because a child’s

behavior might be different if there are siblings (Taanila A et al. 2004). All results

were analyzed by including and excluded prematurely born children because

preterm and early term birth increases a child’s risk of ADHD by the degree of

immaturity, and many other comorbidities also correlate with prematurity

(Lindström et al. 2011).

There is some data suggesting that high maternal pre-pregnancy body mass

index (BMI) would increase a child’s risk of ADHD and other neuropsychiatric

problems (Jo et al. 2015). The data in this study was not adjusted for maternal pre-

pregnancy BMI because the association of thyroid dysfunction and maternal BMI

applies both ways: high maternal BMI increases a mother’s risk of hypothyroidism,

and hypothyroidism might also increase a mother’s weight.

The effect of the country’s iodine status must also be acknowledged because

even in some countries previously considered iodine-sufficient, iodine deficiency

has been identified (Andersson et al 2012). Iodine insufficiency during pregnancy

should not confound our results because in Finland iodine supplementation has

been in use since the 1940s (Lamberg et al. 1981, Erkkola et al. 1998). In 1986,

Finland had the highest iodine intake of all European countries (approximately 300

µg/day), and at that time Finland was considered iodine-sufficient (Lamberg 1986).

In conclusion, were able to able to account for the most important maternal and

family factors in adjustments: maternal age, maternal education or family SES, the

number of children in the family (ADHD and learning difficulty analyses),

maternal smoking and the child’s sex. Most of our study findings remained similar

after adjustments.

100

7.3 Clinical significance of the results

Abnormal maternal thyroid function parameters during pregnancy associated with

the thyroid function of the offspring. Positive maternal TPO-Ab concentrations

increased adolescents’ risk of being TPO-ab-positive themselves. The risk increase

was especially noticeable in daughters of TPO-Ab-positive mothers, whose

increased risk was three-fold compared to those of TPO-Ab-negative mothers.

Adolescents with positive TPO-Ab concentrations had significantly increased risk

of having abnormal thyroid function themselves compared to TPO-Ab-negative

individuals. This finding signifies the autoimmune origin of thyroid diseases and

their prevalence difference between sexes. The reference intervals of adults may

not be sufficient for identifying thyroid dysfunction in adolescents. TSH and fT4

values are narrowly distributed within the normal reference limits. Therefore, the

TSH, fT4 and TPO-Ab concentrations of children with thyroid dysfunction

symptoms should be controlled.

The study results did not clearly associate maternal thyroid dysfunction during

pregnancy with a child’s intellectual disability, ADHD symptoms (evaluated at

eight and sixteen years old) or sensory and linguistic outcomes. However,

logarithmically increasing maternal TSH concentrations mildly increased girls’

odds of having high Rutter B2 scores and combined ADHD symptoms. This finding

probably demonstrates that maternal thyroid dysfunction during pregnancy has an

effect on the developing CNS, but ADHD is a multifactorial disorder, and maternal

thyroid dysfunction is not its most important etiological factor. Learning difficulties

are also multifactorial problems, and they are probably even more affected by social

factors than ADHD. However, adolescent boys’ hypothyroxinemia associated with

an increased prevalence of ADHD symptoms. Hyperthyroid girls also exhibited

more ADHD symptoms before the exclusion of those with an IQ ≤85. Therefore,

the thyroid function of children and adolescents should be studied as part of ADHD

and learning difficulty diagnostics. This is especially true if the subject presents

other symptoms related to thyroid dysfunction.

Maternal hypothyroxinemia increased boys’ odds of repeating a class, and

maternal hyperthyroidism increased an adolescent’s risk of mathematical problems.

Adolescent boys with hypothyroxinemic thyroid function had increased odds of

having problems in Finnish and/or mathematics, and they also had increased odds

of ADHD symptoms. These ADHD symptoms may partly explain their learning

difficulties. Hyperthyroid adolescent girls had increased odds of Finnish language

101

difficulties. Therefore, we conclude that a child’s own thyroid dysfunction might

affect their learning and attention. There are no previous studies on whether

maternal thyroid function during pregnancy affects a child’s everyday learning

abilities, and additionally, no studies investigate both maternal and child thyroid

function in relation to cognitive abilities. It is possible that in some cases, a change

in a child’s scholastic success and/or attention is a first symptom of thyroid

dysfunction. This finding is important because the cause is treatable.

At the moment, there is active discussion on whether all pregnant women

should be screened for thyroid dysfunction during pregnancy. The results of this

study contribute to current literature by incorporating long-term

neuropsychological outcomes. However, whether children would benefit from

screening (and treatment) of maternal thyroid dysfunction during pregnancy would

require an intervention study. Answering that question requires a randomized trial,

and the one trial previously conducted had a negative result (Lazarus et al. 2012).

The authors of that study reported some limitations concerning the loss of study

subjects and the timing of screening and treatment (possibly too late at up to 20

weeks of gestation). However, the situation in real life may be similar; not all

pregnant women will wish to participate and the timing of screening and treatment

would vary.

102

103

References

Abalovich M, Gutierrez S, Alcaraz G, Maccallini G, Garcia A & Levalle O (2002) Overt and subclinical hypothyroidism complicating pregnancy. Thyroid 12(1): 63−68.

Abdalla SM & Bianco AC (2014) Defending plasma T3 is a biological priority. Clin Endocrinol (Oxf) 81(5): 633−641.

Ain K, Mori Y & Refetoff S (1987) Reduced clearance rate of thyroxine-binding globulin (TBG) with increased sialylation: a mechanism for estrogen-induced elevation of serum TBG concentration. J Clin Endocrinol Metab 65(4): 689−696.

Almqvist F, Tuompo-Johansson E, Panelius E, Aronen E & Kairemo A (1991) Screening for psychiatric symptoms on the basis of the Rutter Teacher questionnaire. Eur Child Adolesc Psychiatry 91: 35−46.

American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC.

Anastasi A (1988) Psychological testing. New York: Macmillan Publishing Company. Anckaert E, Poppe K, Van Uytfanghe K, Schiettecatte J, Foulon W & Thienpont L (2010)

FT4 immunoassays may display a pattern during pregnancy similar to the equilibrium dialysis ID-LC/tandem MS candidate reference measurement procedure in spite of susceptibility towards binding protein alterations. Clin Chim Acta 411(17−18): 1348−1353. DOI: 10.1016/j.cca. 2010.05.032.

Andersen SL, Laurberg P, Wu CS & Olsen J (2014) Attention deficit hyperactivity disorder and autism spectrum disorder in children born to mothers with thyroid dysfunction: a Danish nationwide cohort study. BJOG 121(11): 1365−1374.

Andersen S, Laurberg P, Wu C & Olsen J (2013) Maternal thyroid dysfunction and risk of seizure in the child: a Danish nationwide cohort study. J Pregnancy: 636705. DOI: 10.1155/2013/636705.

Andersen S, Laurberg P, Wu C & Olsen J (2014a) Attention deficit hyperactivity disorder and autism spectrum disorder in children born to mothers with thyroid dysfunction: a Danish nationwide cohort study. BJOG 121(11): 1365−1374. DOI: 10.1111/1471−0528.12681.

Andersen S, Olsen J, Wu C & Laurberg P (2014b) Psychiatric disease in late adolescence and young adulthood. Foetal programming by maternal hypothyroidism? Clin Endocrinol (Oxf) 81(1): 126−133. DOI: 10.1111/cen.12415. Epub 2014 Feb 17.

Anderson G, Schoonover C & Jones S (2003) Control of thyroid hormone action in the developing rat brain. Thyroid 13(11): 1039−1056.

Andersson M, Karumbunathan V & Zimmermann MB (2012) Global iodine status in 2011 and trends over the past decade. J Nutr 142(4): 744−750.

Auso E, Lavado-Autric R, Cuevas E, Del Rey FE, Morreale De Escobar G & Berbel P (2004) A moderate and transient deficiency of maternal thyroid function at the beginning of fetal neocorticogenesis alters neuronal migration. Endocrinology 145(9): 4037−4047.

Barker DJ, Winter PD, Osmond C, Margetts B, Simmonds SJ (1989) Weight in infancy and death from inchemic heart disease. Lancet 2(8663): 577−580.

104

Bath SC, Steer CD, Golding J, Emmett P & Rayman MP (2013) Effect of inadequate iodine status in UK pregnant women on cognitive outcomes in their children: results from the Avon Longitudinal Study of Parents and Children (ALSPAC). Lancet 382(9889): 331−337.

Berbel P, Escobar del Rey F, Morreale de Escobar G & Ruiz-Marcos A (1985) Effect of hypothyroidism on the size of spines of pyramidal neurons of the cerebral cortex. Brain Res 337(2): 217−223.

Biondi B & Wartofsky L (2012) Combination treatment with T4 and T3: toward personalized replacement therapy in hypothyroidism? J Clin Endocrinol Metab 97: 2256−2271.

Bradley D, Towle H & Young W (1992) Spatial and temporal expression of alpha- and beta-thyroid hormone receptor mRNAs, including the beta 2-subtype, in the developing mammalian nervous system. J Neurosci 12(6): 2288−2302.

Brown R, Bellisario R, Botero D, Fournier L, Abrams C, Cowger M, David R, Fort P & Richman R (1996) Incidence of transient congenital hypothyroidism due to maternal thyrotropin receptor-blocking antibodies in over one million babies. J Clin Endocrinol Metab 81(3): 1147−1151.

Cakir M, Turgut Ozturk B, Turan E, Gonulalan G, Polat I & Gunduz K (2015) The effect of hypothyroidism on color contrast sensitivity: a prospective study. Eur Thyroid J 4(1): 43−47.

Calvo RM, Jauniaux E, Gulbis B, Asuncion M, Gervy C, Contempre B & Morreale de Escobar G (2002) Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early pregnancy. J Clin Endocrinol Metab 87(4): 1768−1777.

Calvo R, Obregon MJ, Ruiz de Ona C, Escobar del Rey F, Morreale de Escobar G (1990) Congenital hypothyroidism, as studied in rats. Crucial role of maternal thyroxine but not of 3,5,3-triiodothyronine in the protection of the fetal brain. J Clin Invest 86: 889−899.

Cleary-Goldman J, Malone F, Lambert-Messerlian G, Sullivan L, Canick J, Porter T, Luthy D, Gross S, Bianchi D & D'Alton M (2008) Maternal thyroid hypofunction and pregnancy outcome. Obstet Gynecol 112(1): 85−92.

Collishaw S, Goodman R, Ford T, Rabe-Hesketh S & Pickles A (2009) How far are associations between child, family and community factors and child psychopathology informant-specific and informant-general? J Child Psychol Psychiatry 50(5): 571−580.

Contempre B, Jauniaux E, Calvo R, Jurkovic D, Campbell S & De Escobar G (1993) Detection of thyroid hormones in human embryonic cavities during the first trimester of pregnancy. J Clin Endocrinol Metab 77(6): 1719−1722.

Cooper D & Laurberg P (2013) Hyperthyroidism in pregnancy. Lancet Diabetes Endocrinol 1(3): 238−249.

Craig WY, Allan WC, Kloza EM, Pulkkinen AJ, Waisbren S, Spratt DI, Palomaki GE, Neveux LM & Haddow JE (2012) Mid-gestational maternal free thyroxine concentration and offspring neurocognitive development at age two years. J Clin Endocrinol Metab 97(1): E22−E28.

105

Cuevas E, Auso E, Telefont M, Morreale de Escobar G, Sotelo C & Berbel P (2005) Transient maternal hypothyroxinemia at onset of corticogenesis alters tangential migration of medial ganglionic eminence-derived neurons. Eur J Neurosci 22(3): 541−551.

De Groot L, Abalovich M, Alexander E, Amino N, Barbour L, Cobin R, Eastman C, Lazarus J, Luton D, Mandel S, Mestman J, Rovet J & Sullivan S (2012) Management of thyroid dysfunction during pregnancy and postpartum: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 97(8): 2543−2565.

De Los Reyes A & Kazdin AE (2005) Informant discrepancies in the assessment of childhood psychopathology: a critical review, theoretical framework, and recommendations for further study. Psychol Bull 131(4): 483−509.

Dosiou C, Sanders G, Araki S & Crapo L (2008) Screening pregnant women for autoimmune thyroid disease: a cost-effectiveness analysis. Eur J Endocrinol 158(6): 841−851.

Erkkola M, Karppinen M, Jarvinen A, Knip M & Virtanen SM (1998) Folate, vitamin D, and iron intakes are low among pregnant Finnish women. Eur J Clin Nutr 52(10): 742−748.

Fagge C (1871) On Sporadic Cretinism, occurring in England. Med Chir Trans 54: 155−170. Fekete C & Lechan RM (2014) Central regulation of hypothalamic-pituitary-thyroid axis

under physiological and pathophysiological conditions. Endocr Rev 35(2): 159−194. Finken M, Van Eijsden M, Loomans E, Vrijkotte T & Rotteveel J (2013) Maternal

hypothyroxinemia in early pregnancy predicts reduced performance in reaction time tests in 5- to 6-year-old offspring. J Clin Endocrinol Metab 98(4): 1417−1426.

Fisher D (2007) Thyroid function and dysfunction in premature infants. Pediatr Endocrinol Rev 4(4): 317−328.

Foeller M & Silver R (2015) Combination levothyroxine + liothyronine treatment in pregnancy. Obstet Gynecol Surv 70(9): 584−586.

Garber JR, Cobin RH, Gharib H, Hennessey JV, Klein I, Mechanick JI, Pessah-Pollack R, Singer PA, Woeber KA (2012) Clinical practice guidelines for hypothyroidism in adults: cosponsored by the American Association of Clinical Endocrinologists and the American Thyroid Association. Endocr Pract 18(6): 988−1028.

García-García E, Vazquez-Lopez M, Garcia-Fuentes E, Rodriguez-Sanchez F, Munoz F, Bonillo-Perales A & Soriguer F (2012) Iodine intake and prevalence of thyroid autoimmunity and autoimmune thyroiditis in children and adolescents aged between 1 and 16 years. Eur J Endocrinol 167(3): 387−392.

Gereben B, Zavacki AM, Ribich S, Kim BW, Huang SA, Simonides WS, Zeöld A, Bianco AC (2008) Cellular and molecular basis of deiodinase-regulated thyroid hormone signaling. Endocrin Rev 29(7): 898−938.

Ghassabian A, El Marroun H, Peeters RP, Jaddoe VW, Hofman A, Verhulst FC, Tiemeier H & White T (2014) Downstream effects of maternal hypothyroxinemia in early pregnancy: nonverbal IQ and brain morphology in school-age children. J Clin Endocrinol Metab 99(7): 2383−2390.

106

Ghassabian A, Bongers-Schokking JJ, de Rijke YB, van Mil N, Jaddoe VW, de Muinck Keizer-Schrama SM, Hooijkaas H, Hofman A, Visser W, Roman GC, Visser TJ, Verhulst FC & Tiemeier H (2012) Maternal thyroid autoimmunity during pregnancy and the risk of attention deficit/hyperactivity problems in children: the Generation R Study. Thyroid 22(2): 178−186.

Ghassabian A, Bongers-Schokking JJ, Henrichs J, Jaddoe VW, Visser TJ, Visser W, DE Muinck Keizer-Schrama SM, Hooijkaas H, Steegers EA, Hofman A, Verhulst FC, Van Der Ende J, DE Rijke YB & Tiemeier H (2011) Maternal thyroid function during pregnancy and behavioral problems in the offspring: the Generation R study. Pediatr Res 69(5 Pt 1): 454−459.

Glinoer D (1997) The regulation of thyroid function in pregnancy: pathways of endocrine adaptation from physiology to pathology. Endocr Rev 18(3): 404−433.

Gottschalk B, Richman R & Lewandowski L (1994) Subtle speech and motor deficits of children with congenital hypothyroid treated early. Dev Med Child Neurol 36(3): 216−220.

Grau G, Aguayo A, Vela A, Aniel-Quiroga A, Espada M, Miranda G, Martinez-Indart L, Martul P, Castano L & Rica I (2015) Normal intellectual development in children born from women with hypothyroxinemia during their pregnancy. J Trace Elem Med Biol 31: 18−24.

Guadano-Ferraz A, Escamez M, Rausell E & Bernal J (1991) Expression of type 2 iodothyronine deiodinase in hypothyroid rat brain indicates an important role of thyroid hormone in the development of specific primary sensory systems. J Neurosci 19(9): 3430−3439.

Gyllenberg D, Sourander A, Surcel H, Hinkka-Yli-Salomäki S, McKeague I & Brown A (2015) Hypothyroxinemia During Gestation and Offspring Schizophrenia in a National Birth Cohort. Biol Psychiatry 19. pii: S0006-3223(15)00520-X.

Haddow JE, Palomaki GE, Allan WC, Williams JR, Knight GJ, Gagnon J, O'Heir CE, Mitchell ML, Hermos RJ, Waisbren SE, Faix JD & Klein RZ (1999) Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med 341(8): 549−555.

Hales C, Channon S, Taylor P, Draman M, Muller I, Lazarus J, Paradice R, Rees A, Shillabeer D, Gregory J, Dayan C & Ludgate M (2014) The second wave of the Controlled Antenatal Thyroid Screening (CATS II) study: the cognitive assessment protocol. BMC Endocr Disord 14: 95.

Hansen P, Brix T, Sorensen T, Kyvik K & Hegedus L (2004) Major genetic influence on the regulation of the pituitary-thyroid axis: a study of healthy Danish twins. J Clin Endocrinol Metab 89(3): 1181−1197.

Hauser P, Soler R, Brucker-Davis F & Weintraub B (1997) Thyroid hormones correlate with symptoms of hyperactivity but not inattention in attention deficit hyperactivity disorder. Psychoneuroendocrinology 22(2): 107−114.

Heikura U (2008) Intellectual disability in the Northern Finland Birth Cohort 1986. Oulu, University of Oulu.

107

Heikura U, Hartikainen A, Nordström T, Pouta A, Taanila A & Järvelin M (2013) Maternal hypertensive disorders during pregnancy and mild cognitive limitations in the offspring. Paediatr Perinat Epidemiol 27(2): 188−198.

Heikura U, Linna SL, Olsen P, Hartikainen AL, Taanila A & Järvelin MR (2005) Etiological survey on intellectual disability in the northern Finland birth cohort 1986. Am J Ment Retard 110(3): 171−180.

Helldán A, Raulio S, Kosola M, Tapanainen H, Ovaskainen M & Virtanen S (2012) The National FINDIET 2012 Survey. URI: https://www.julkari.fi/ bitstream/handle/10024/110839/THL_RAP2013_016_%26sliitteet.pdf?sequence=1.

Henrichs J, Bongers-Schokking JJ, Schenk JJ, Ghassabian A, Schmidt HG, Visser TJ, Hooijkaas H, de Muinck Keizer-Schrama SM, Hofman A, Jaddoe VV, Visser W, Steegers EA, Verhulst FC, de Rijke YB & Tiemeier H (2010) Maternal thyroid function during early pregnancy and cognitive functioning in early childhood: the generation R study. J Clin Endocrinol Metab 95(9): 4227−4234.

Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer CA & Braverman LE (2002) Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 87(2): 489−499.

Horn PS, Feng L, Li Y & Pesce AJ (2001) Effect of outliers and nonhealthy individuals on reference interval estimation. Clin Chem 47(12): 2137−2145.

Hurtig T, Ebeling H, Taanila A, Miettunen J, Smalley SL, McGough JJ, Loo SK, Järvelin MR & Moilanen IK (2007) ADHD symptoms and subtypes: relationship between childhood and adolescent symptoms. J Am Acad Child Adolesc Psychiatry 46(12): 1605−1613.

Hynes K, Otahal P, Hay I & Burgess J (2013) Mild iodine deficiency during pregnancy is associated with reduced educational outcomes in the offspring: 9-year follow-up of the gestational iodine cohort. J Clin Endocrinol Metab 98(5): 1954−1962. DOI: 10.1210/jc.2012−4249.

Järvelin MR, Hartikainen-Sorri AL & Rantakallio P (1993) Labour induction policy in hospitals of different levels of specialisation. BJOG 100(4): 310−315.

Järvelin M, Elliott P, Kleinschmidt I, Martuzzi M, Grundy C, Hartikainen AL & Rantakallio P (1997) Ecological and individual predictors of birthweight in a northern Finland birth cohort 1986. Paediatr Perinat Epidemiol 11(3): 298−312.

Jo H, Schieve L, Sharma A, Hinkle S, Li R & Lind J (2015) Maternal prepregnancy body mass index and child psychosocial development at 6 years of age. Pediatrics. 135(5): e1198−e1209.

Julvez J, Alvarez-Pedrerol M, Rebagliato M, Murcia M, Forns J, Garcia-Esteban R, Lertxundi N, Espada M, Tardon A, Riano Galan I & Sunyer J (2013) Thyroxine levels during pregnancy in healthy women and early child neurodevelopment. Epidemiology 24(1): 150−157.

108

Kaloumenou I, Mastorakos G, Alevizaki M, Duntas LH, Mantzou E, Ladopoulos C, Antoniou A, Chiotis D, Papassotiriou I, Chrousos GP & Dacou-Voutetakis C (2008) Thyroid autoimmunity in schoolchildren in an area with long-standing iodine sufficiency: correlation with gender, pubertal stage, and maternal thyroid autoimmunity. Thyroid 18(7): 747−754.

Kester M, Martinez de Mena R, Obregon M, Marinkovic D, Howatson A, Visser T, Hume R & Morreale de Escobar G (2004) Iodothyronine levels in the human developing brain: major regulatory roles of iodothyronine deiodinases in different areas. J Clin Endocrinol Metab 89(7): 3117−3128.

Khan FA, Al-Jameil N, Khan MF, Al-Rashid M, Tabassum H (2015) Thyroid dysfunction: an autoimmune aspect. Int J Clin Exp Med 8(5): 6677−6681.

Kirkinen P & Ryynänen M (2015) Prenataalidiagnostiikka. In: Naistentaudit ja synnytykset. Kustannus Oy Duodecim.

Kolko DJ & Kazdin AE (1993) Emotional/behavioral problems in clinic and nonclinic children: correspondence among child, parent and teacher reports. J Child Psychol Psychiatry 34(6): 991−1006.

Kondrashova A, Viskari H, Haapala AM, Seiskari T, Kulmala P, Ilonen J, Knip M & Hyoty H (2008) Serological evidence of thyroid autoimmunity among schoolchildren in two different socioeconomic environments. J Clin Endocrinol Metab 93(3): 729−734.

Kooistra L, Laane C, Vulsma T, Schellekens J, Van der Meere J & Kalverboer A (1994) Motor and cognitive development in children with congenital hypothyroidism: a long-term evaluation of the effects of neonatal treatment. J Pediatr 124(6): 903−909.

Korevaar T, Muetzel R, Medici M, Chaker L, Jaddoe VW, De Rijke Y, Steegers EA, Visser TJ, White T, Tiemeier H, Peeters RP (2015) Association of maternal thyroid function during early pregnancy with off spring IQ and brain morphology in childhood: a population-based prospective cohort study. Lancet Diabetes Endocrinol 2015 Oct 19. DOI: 10.1016/S2213-8587(15)00327-7.

Korevaar T, Chaker L, Jaddoe VW, Visser TJ, Medici M, Peeters RP (2015) Maternal and Birth Characteristics are Determinants of Offspring Thyroid Function. J Clin Endocrinol Metab 2015 Nov 19:jc20153559.

Koskela P, Anttila T, Bjorge T, Brunsvig A, Dillner J, Hakama M, Hakulinen T, Jellum E, Lehtinen M, Lenner P, Luostarinen T, Pukkala E, Saikku P, Thoresen S, Youngman L & Paavonen J (2000) Chlamydia trachomatis infection as a risk factor for invasive cervical cancer. International Journal of Cancer 85(1): 35−39.

Kotimaa AJ, Moilanen I, Taanila A, Ebeling H, Smalley SL, McGough JJ, Hartikainen AL & Jarvelin MR (2003) Maternal smoking and hyperactivity in 8-year-old children. J Am Acad Child Adolesc Psychiatry 42(7): 826−833.

Kresanov K, Tuominen J, Piha J & Almqvist F (1998) Validity of child psychiatric screening methods. Eur Child Adolesc Psychiatry 7(2): 85−95.

Lamberg BA (1986) Endemic goitre in Finland and changes during 30 years of iodine prophylaxis. Endocrinol Exp 20(1): 35−47.

109

Lamberg BA, Haikonen M, Makela M, Jukkara A, Axelson E & Welin MG (1981) Further decrease in thyroidal uptake and disappearance of endemic goiter in children after 30 years of iodine prophylaxis in the east of Finland. Acta Endocrinol 98(2): 205−209.

Laurell C & Rannevik G (1979) A comparison of plasma protein changes induced by danazol, pregnancy, and estrogens. J Clin Endocrinol Metab 49(5): 719−725.

Lavado-Autric R, Auso E, Garcia-Velasco J, Arufe Mdel C, Escobar del Rey F, Berbel P & Morreale de Escobar G (2003) Early maternal hypothyroxinemia alters histogenesis and cerebral cortex cytoarchitecture of the progeny. J Clin Invest 111(7): 1073−1082.

Lazarus J, Brown R, Daumerie C, Hubalewska-Dydejczyk A, Negro R & Vaidya B (2014) 2014 European thyroid association guidelines for the management of subclinical hypothyroidism in pregnancy and in children. Eur Thyroid J 3(2): 76−94. DOI: 10.1159/000362597.

Lazarus JH, Bestwick JP, Channon S, Paradice R, Maina A, Rees R, Chiusano E, John R, Guaraldo V, George LM, Perona M, Dall'Amico D, Parkes AB, Joomun M & Wald NJ (2012) Antenatal thyroid screening and childhood cognitive function. N Engl J Med 366(6): 493−501.

Lee R, Spencer C, Mestman J, Miller E, Petrovic I, Braverman L & Goodwin T (2009) Free T4 immunoassays are flawed during pregnancy. Am J Obstet Gynecol 200(3): 260.e1-260.e6.

Lehtinen M, Koskela P, Ogmundsdottir HM, Bloigu A, Dillner J, Gudnadottir M, Hakulinen T, Kjartansdottir A, Kvarnung M, Pukkala E, Tulinius H & Lehtinen T (2003) Maternal herpesvirus infections and risk of acute lymphoblastic leukemia in the offspring. Am J Epidemiol 158(3): 207−213.

Leviton A, Paneth N, Reuss M, Susser M, Allred E, Dammann O, Kuban K, Van Marter L & Pagano M (1999) Hypothyroxinemia of prematurity and the risk of cerebral white matter damage. J Pediatr 134(6): 706−711.

Li Y, Shan Z, Teng W, Yu X, Li Y, Fan C, Teng X, Guo R, Wang H, Li J, Chen Y, Wang W, Chawinga M, Zhang L, Yang L, Zhao Y & Hua T (2010) Abnormalities of maternal thyroid function during pregnancy affect neuropsychological development of their children at 25−30 months. Clin Endocrinol (Oxf) 72(6): 825−829.

Lindström K, Lindblad F & Hjern A (2011) Preterm Birth and Attention-Deficit/Hyperactivity Disorder in Schoolchildren. Pediatrics 127(5): 858−865.

Lu A, Ng L, Ma M, Kefas B, Davies T, Hernandez A, Chan C & Forrest D (2008) Retarded developmental expression and patterning of retinal cone opsins in hypothyroid mice. Endocrinology 150(3): 1536−1544.

Lucas A, Morley R & Fewtrell M (1996) Low triiodothyronine concentration in preterm infants and subsequent intelligence quotient (IQ) at 8 year follow up. BMJ 312(7039):1132−1133.

Man E & Jones W (1969) Thyroid function in human pregnancy. V. Incidence of maternal serum low butanol-extractable iodines and of normal gestational TBG and TBPA capacities; retardation of 8-month-old infants. Am J Obstet Gynecol 104(6): 898−908.

Matalon S, Sheiner E, Levy A, Mazor M & Wiznitzer A (2006) Relationship of treated maternal hypothyroidism and perinatal outcome. J Reprod Med 51(1): 59−63.

110

Männistö T, Surcel H, Bloigu A, Ruokonen A, Hartikainen A, Järvelin M, Pouta A, Vääräsmäki M & Suvanto E (2007) The effect of freezing, thawing, and short- and long-term storage on serum thyrotropin, thyroid hormones, and thyroid autoantibodies: implications for analyzing samples stored in serum banks. Clin Chem 53(11): 1986−1987.

Männistö T, Surcel H, Ruokonen A, Vääräsmäki M, Pouta A, Bloigu A, Järvelin M, Hartikainen A & Suvanto E (2011) Early pregnancy reference intervals of thyroid hormone concentrations in a thyroid antibody-negative pregnant population. Thyroid 21(3): 291−298.

Männistö T, Vääräsmäki M, Pouta A, Hartikainen A, Ruokonen A, Surcel H, Bloigu A, Järvelin M & Suvanto E (2009) Perinatal outcome of children born to mothers with thyroid dysfunction or antibodies: a prospective population-based cohort study. J Clin Endocrinol Metab 94(3): 772−779.

Männistö T, Vääräsmäki M, Pouta A, Hartikainen A, Ruokonen A, Surcel H, Bloigu A, Järvelin M & Suvanto E (2010) Thyroid dysfunction and autoantibodies during pregnancy as predictive factors of pregnancy complications and maternal morbidity in later life. J Clin Endocrinol Metab 95(3): 1084−1094.

Matsuura N, Yamada Y, Nohara Y, Konishi J, Kasagi K, Endo K, Kojima H & Wataya K (1980) Familial neonatal transient hypothyroidism due to maternal TSH-binding inhibitor immunoglobulins. N Engl J Med 303(13): 738−741.

Medici M, de Rijke Y, Peeters R, Visser W, de Muinck Keizer-Schrama S, Jaddoe V, Hofman A, Hooijkaas H, Steegers E, Tiemeier H, Bongers-Schokking J & Visser T (2012) Maternal early pregnancy and newborn thyroid hormone parameters: the Generation R study. J Clin Endocrinol Metab 97(2): 646−652.

Medici M, Korevaar T, Visser W, Visser T & Peeters R (2015) Thyroid function in pregnancy: what is normal? Clin Chem 61(5): 704−713.

Mirabella G, Westall C, Asztalos E, Perlman K, Koren G & Rovet J (2005) Development of contrast sensitivity in infants with prenatal and neonatal thyroid hormone insufficiencies. Pediatr Res 57(6): 902−907.

Modesto T, Tiemeier H, Peeters R, Jaddoe V, Hofman A, Verhulst F & Ghassabian A (2015) Maternal Mild Thyroid Hormone Insufficiency in Early Pregnancy and Attention-Deficit/Hyperactivity Disorder Symptoms in Children. JAMA Pediatr. 2015 Jul 6. DOI: 10.1001/jamapediatrics.2015.0498.

Moilanen I, Puustjärvi A, Laukkala T, Närhi V, Olsén P, Pihlakoski L, Raevuori A, Virkkunen L & Voutilainen A (2013) Käypä hoito, ADHD (aktiivisuuden ja tarkkaavuuden häiriö, nuoret ja lapset). URI: http://kaypahoito.fi/web/kh/ suositukset/suositus?id=hoi50061#NaN.

Momotani N, Iwama S & Momotani K (2012) Neurodevelopment in children born to hypothyroid mothers restored to normal thyroxine (T4) concentration by late pregnancy in Japan: no apparent influence of maternal T4 deficiency. J Clin Endocrinol Metab 97(4): 1104−1108.

Morreale de Escobar G (2001) The role of thyroid hormone in fetal neurodevelopment. J Pediatr Endocrinol Metab 14 Suppl 6: 1453−1462.

111

Muldoon BT, Mai VQ, Burch HB (2014) Management of Graves' disease: an overview and comparison of clinical practice guidelines with actual practice trends. Endocrinol Metab Clin North Am 43(2): 495−516.

Mullur R, Liu YY, Brent GA (2014) Thyroid hormone regulation of metabolism. Physiol Rev 94(2): 355−382.

Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T & Stagnaro-Green A (2010a) Increased pregnancy loss rate in thyroid antibody negative women with TSH levels between 2.5 and 5.0 in the first trimester of pregnancy. J Clin Endocrinol Metab 95(9): E44-E48.

Negro R, Schwartz A, Gismondi R, Tinelli A, Mangieri T & Stagnaro-Green A (2010b) Universal screening versus case finding for detection and treatment of thyroid hormonal dysfunction during pregnancy. J Clin Endocrinol Metab 95(4): 1699−1707.

Negro R, Soldin O, Obregon M & Stagnaro-Green A (2011) Hypothyroxinemia and pregnancy. Endocr Pract 17(3): 422−429.

Nyman ES, Ogdie MN, Loukola A, Varilo T, Taanila A, Hurtig T, Moilanen IK, Loo SK, McGough JJ, Jarvelin MR, Smalley SL, Nelson SF & Peltonen L (2007) ADHD candidate gene study in a population-based birth cohort: association with DBH and DRD2. J Am Acad Child Adolesc Psychiatry 46(12): 1614−1621.

Oken E, Braverman LE, Platek D, Mitchell ML, Lee SL & Pearce EN (2009) Neonatal thyroxine, maternal thyroid function, and child cognition. J Clin Endocrinol Metab 94(2): 497−503.

Opazo M, Gianini A, Pancetti F, Azkcona G, Alarcon L, Lizana R, Noches V, Gonzalez P, Marassi M, Mora S, Rosenthal D, Eugenin E, Naranjo D, Bueno S, Kalergis A & Riedel C (2008) Maternal hypothyroxinemia impairs spatial learning and synaptic nature and function in the offspring. Endocrinology 149(10): 5097−5106.

Panicker V, Wilson SG, Spector TD, Brown SJ, Falchi M, Richards JB, Surdulescu GL, Lim EM, Fletcher SJ & Walsh JP (2008) Heritability of serum TSH, free T4 and free T3 concentrations: a study of a large UK twin cohort. Clin Endocrinol (Oxf) 68(4): 652−659.

Panicker V, Saravan P, Vaidya B, Evans J, Hatterslay AT, Frayling TM, Dayan CM (2009) Common variation in the DIO2 gene predicts baseline psychological well-being and response to combination thyroxine plus triiodothyronine therapy in hypothyroid patients. J Clin Endocrinol Metab 94: 1623−1629.

Patel J, Landers K, Li H, Mortimer RH, Richard K (2011) Thyroid hormones and fetal neurological development. J Endocrinol 209: 1−8.

Pearce E (2015) Thyroid disorders during pregnancy and postpartum. Best Pract Res Clin Obstet Gynaecol 29(5): 700−706.

Pekonen F, Alfthan H, Stenman U & Ylikorkala O (1988) Human chorionic gonadotropin (hCG) and thyroid function in early human pregnancy: circadian variation and evidence for intrinsic thyrotropic activity of hCG. J Clin Endocrinol Metab 66(4): 853−856.

Pop VJ, Brouwers EP, Vader HL, Vulsma T, van Baar AL & de Vijlder JJ (2003) Maternal hypothyroxinaemia during early pregnancy and subsequent child development: a 3-year follow-up study. Clin Endocrinol (Oxf) 59(3): 282−288.

112

Pop VJ, Kuijpens JL, van Baar AL, Verkerk G, van Son MM, de Vijlder JJ, Vulsma T, Wiersinga WM, Drexhage HA & Vader HL (1999) Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin Endocrinol (Oxf) 50(2): 149−155.

Radetti G (2014) Clinical aspects of Hashimoto's thyroiditis. Endocr Dev 26: 158−70. Radetti G, Gentili L, Paganini C, Oberhofer R, Deluggi I & Delucca A (2000) Psychomotor

and audiological assessment of infants born to mothers with subclinical thyroid dysfunction in early pregnancy. Minerva Pediatr 52(12): 691−698.

Rallison M, Dobyns B, Keating F, Rall J & Tyler F (1975) Occurrence and natural history of chronic lymphocytic thyroiditis in childhood. J Pediatr 86(5): 675−682.

Rice D & Barone SJ (2000) Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models. Environ Health Perspect 108 Suppl 3: 511−533.

Roberts M, Srinivas M, Forrest D, Morreale de Escobar G & Reh T (2006) Making the gradient: thyroid hormone regulates cone opsin expression in the developing mouse retina. Proc Natl Acad Sci USA 103(16): 6218−6223.

Rodriguez A, Järvelin M, Obel C, Taanila A, Miettunen J, Moilanen I, Henriksen T, Pietiläinen K, Ebeling H, Kotimaa A, Linnet K & Olsen J (2007) Do inattention and hyperactivity symptoms equal scholastic impairment? Evidence from three European cohorts. BMC Public Health 7: 327.

Roman G, Ghassabian A, Bongers-Schokking J, Jaddoe V, Hofman A, de Rijke Y, Verhulst F & Tiemeier H (2013) Association of gestational maternal hypothyroxinemia and increased autism risk. Ann Neurol Aug 13. DOI: 10.1002/ana.23976.

Rovet J & Simic N (2008) The role of transient hypothyroxinemia of prematurity in development of visual abilities. Semin Perinatol 32(6): 431−437.

Rovet J (2014) The role of thyroid hormones for brain development and cognitive function. Endocr Dev 26: 26−43.

Salminen M (2015) Keskushermosto. In: Kehitysbiologia. Kustannus Oy Duodecim. Salvador-Carulla L, Reed G, Vaez-Azizi L, Cooper S, Martinez-Leal R, Bertelli M, Adnams

C, Cooray S, Deb S, Akoury-Dirani L, Girimaji S, Katz G, Kwok H, Luckasson R, Simeonsson R, Walsh C, Munir K & Saxena S (2011) Intellectual developmental disorders: towards a new name, definition and framework for "mental retardation/intellectual disability" in ICD-11. World Psychiatry 10(3): 175−180.

Santisteban P & Bernal J (2005) Thyroid development and effect on the nervous system. Rev Endocr Metab Disord 6(3): 217−228.

Schalin-Jäntti C, Tanner P, Välimäki M (2011) Serum TSH reference interval in healthy Finnish adults using the Abbot Architect 2000i Analyzer. Scand J Clin Lab Invest 71(4): 344−349.

Sheehan R, Hassiotis A, Walters K, Osborn D, Strydom A & Horsfall L (2015) Mental illness, challenging behaviour, and psychotropic drug prescribing in people with intellectual disability: UK population based cohort study. BMJ 351: h4326. DOI: 10.1136/bmj.h4326.

Simell O (1987) Neuvolakirja. Espoo, Orion-yhtymä Oy/ Lääketehdas Remeda.

113

Smalley SL, McGough JJ, Moilanen IK, Loo SK, Taanila A, Ebeling H, Hurtig T, Kaakinen M, Humphrey LA, McCracken JT, Varilo T, Yang MH, Nelson SF, Peltonen L & Järvelin MR (2007) Prevalence and psychiatric comorbidity of attention-deficit/hyperactivity disorder in an adolescent Finnish population. J Am Acad Child Adolesc Psychiatry 46(12): 1575−1583.

Smit B, Kok J, De Vries L, Dekker F & Ongerboer de Visser B (1999) Motor nerve conduction velocity in very preterm infants. Muscle Nerve 22(3): 372−377.

Spencer L, Bubner T, Bain E & Middleton P (2015) Screening and subsequent management for thyroid dysfunction pre-pregnancy and during pregnancy for improving maternal and infant health. Cochrane Database Syst Rev 9: CD011263. DOI: 10.1002/14651858.CD011263.pub2.

Stagnaro-Green A, Abalovich M, Alexander E, Azizi F, Mestman J, Negro R, Nixon A, Pearce EN, Soldin OP, Sullivan S, Wiersinga W & American Thyroid Association Taskforce on Thyroid Disease During Pregnancy and,Postpartum (2011) Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and postpartum. Thyroid 21(10): 1081−1125.

Szinnai G (2014) Clinical genetics of congenital hypothyroidism. Endocr Dev 26: 60−78. Taanila A, Ebeling H, Kotimaa A, Moilanen I & Järvelin MR (2004) Is a large family a

protective factor against behavioural and emotional problems at the age of 8 years? Acta Paediatrica 93(4): 508−517.

Taanila A, Ebeling H, Tiihala M, Kaakinen M, Moilanen I, Hurtig T & Yliherva A (2014) Association between childhood specific learning difficulties and school performance in adolescents with and without ADHD symptoms: a 16-year follow-up. J Atten Disord 18(1): 61−72.

Taanila A, Yliherva A, Kaakinen M, Moilanen I & Ebeling H (2011) An epidemiological study on Finnish school-aged children with learning difficulties and behavioural problems. Int J Circumpolar Health 70(1): 59−71.

Thapar A & Cooper M (2015) Attention deficit hyperactivity disorder. Lancet 16. pii: S0140-6736(15)00238-X. DOI: 10.1016/S0140-6736(15)00238-X.

Thung S, Funai E & Grobman W (2009) The cost-effectiveness of universal screening in pregnancy for subclinical hypothyroidism. Am J Obstet Gynecol 200(3): 267.e1-267.e7.

Välimäki, M. & Schalin-Jäntti, C. (2010) Kilpirauhanen. In: Endokrinologia, Kustannus Oy Duodecim: 13.

Van Vliet G & Deladoey J (2014) Diagnosis, treatment and outcome of congenital hypothyroidism. Endocr Dev 26: 50−59.

Vermiglio F, Lo Presti VP, Moleti M, Sidoti M, Tortorella G, Scaffidi G, Castagna MG, Mattina F, Violi MA, Crisa A, Artemisia A & Trimarchi F (2004) Attention deficit and hyperactivity disorders in the offspring of mothers exposed to mild-moderate iodine deficiency: a possible novel iodine deficiency disorder in developed countries. J Clin Endocrinol Metab 89(12): 6054−6060.

114

Wasserman E, Nelson K, Rose N, Eaton W, Pillion J, Seaberg E, Talor M, Burek L, Duggan A & Yolken R (2008) Maternal thyroid autoantibodies during the third trimester and hearing deficits in children: an epidemiologic assessment. Am J Epidemiol 167(6): 701−710.

Wasserman E, Pillion J, Duggan A, Nelson K, Rohde C, Seaberg E, Talor M, Yolken R & Rose N (2012) Childhood IQ, hearing loss, and maternal thyroid autoimmunity in the Baltimore Collaborative Perinatal Project. Pediatr Res 72(5): 525−530.

Wechsler D (ed) (1991) Wechsler intelligence scale for children. San Antonio, TX: The Psychological Corporation.

Wheeler S, Willoughby K, McAndrews M & Rovet J (2011) Hippocampal size and memory functioning in children and adolescents with congenital hypothyroidism. J Clin Endocrinol Metab 96(9): E1427- E1434.

Wiersinga W (2014) Thyroid autoimmunity. Endocr Dev 26: 139−57. Williams FL, Watson J, Ogston SA, Visser TJ, Hume R & Willatts P (2013) Maternal and

umbilical cord levels of T4, FT4, TSH, TPOAb, and TgAb in term infants and neurodevelopmental outcome at 5.5 years J Clin Endocrinol Metab 98(2): 829−838.

Willoughby KA, McAndrews MP & Rovet JF (2014) Effects of maternal hypothyroidism on offspring hippocampus and memory. Thyroid 24(3): 576−584.

World Health Organization, ICD-9 and ICD-10 Guides for Mental Retardation. Geneva, 1972 and 1996.

www.evira.fi/files/attachments/fi/vrn/vrn_jodi_toimenpidesuositus_10.2.2015_su omi.pdf Yang C, Wang L, Mo X, Li Y, Xiao H & Cao X (2013) Perinatal complications and higher

risks of offspring thyroid dysfunction in early childhood of Graves' disease mothers with euthyroidism. J Pediatr Endocrinol Metab 26(11−12): 1035−1040.

115

Original publications

This thesis is based on the following articles, which are referred to in the text by

their Roman numerals:

I Päkkilä F, Männistö T, Bloigu A, Vääräsmäki M, Hartikainen A-L, Ruokonen A, Pouta A, Järvelin M-R, Surcel H-M & Suvanto E (2013) Maternal thyroid dysfunction during pregnancy and thyroid function of her child in adolescence. J Clin Endocrinol Metab 98(3): 965−972.

II Päkkilä F, Männistö T, Pouta A, Hartikainen A-L, Ruokonen A, Surcel H-M, Bloigu A, Vääräsmäki M, Järvelin M-R, Moilanen I & Suvanto E (2014) The impact of maternal thyroid hormone concentrations during pregnancy on the ADHD symptoms of the child. J Clin Endocrinol Metab 99(1): E1−8.

III Päkkilä F, Männistö T, Hartikainen A-L, Ruokonen A, Surcel H-M, Bloigu A, Vääräsmäki M, Järvelin M-R, Moilanen I & Suvanto E (2015) Maternal and child’s thyroid function and child’s intellect and scholastic performance. Thyroid 25(12):1363−74.

IV Päkkilä F, Männistö T, Hartikainen A-L & Suvanto E. The association between maternal thyroid function during pregnancy and a child’s linguistic and sensory development. Manuscript.

Reprinted with permission from The Endocrine Society (I, II) and The American

Thyroid Association (III).

Original publications are not included in the electronic version of the dissertation.

116

A C T A U N I V E R S I T A T I S O U L U E N S I S

Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/

S E R I E S D M E D I C A

1328. Uusitalo, Jouko (2015) The role of drug metabolism in drug discovery anddevelopment : case ospemifene

1329. Lantto, Marjo (2015) Childhood mortality in Finland

1330. Kerimaa, Pekka (2015) Magnetic resonance imaging-guided percutaneousmusculoskeletal biopsies and therapeutic bone drillings

1331. Holma, Tuomas (2015) Hearing among Finnish professional soldiers :epidemiological study

1332. Petrov, Petar (2015) Leukocyte protein Trojan, as a candidate for apoptoticregulatory role

1333. Mattila, Riikka (2015) The roles of virulence factors Us3 and γ134.5 duringdifferent phases of HSV-1 life cycle

1334. Keskinen, Emmi (2015) Parental psychosis, risk factors and protective factors forschizophrenia and other psychosis : the Northern Finland Birth Cohort 1966

1335. Kantola, Tiina (2016) Systemic inflammation in colorectal cancer : the role ofcytokines and endostatin

1336. Lithovius, Riitta (2015) Aspects of cleft lip and palate from Northern Finland :clefts in Northern Finland

1337. Kuusisto, Milla (2015) Translational research on challenges in the treatment ofdiffuse large B-cell lymphoma

1338. Sneck, Sami (2016) Sairaanhoitajien lääkehoidon osaaminen ja osaamisenvarmistaminen

1339. Lehto, Tiina (2016) Evaluation of new laboratory methods for routine use

1340. Kerätär, Raija (2016) Kun katsoo kauempaa, näkee enemmän : monialainentyökyvyn ja kuntoutustarpeen arviointi pitkäaikaistyöttömillä

1341. Kuoppala, Ritva (2016) Outcome of implant-supported overdenture treatment

1342. Nanekar, Rahul (2016) Biochemical and biophysical studies on adenosinereceptors and their interaction partners

1343. Sakko, Marjut (2016) Antimicrobial activity and suitability of 2-hydroxyisocaproicacid for the treatment of root canal infections

1344. Jukuri, Tuomas (2016) Resting state brain networks in young people with familialrisk for psychosis

UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

Professor Esa Hohtola

University Lecturer Santeri Palviainen

Postdoctoral research fellow Sanna Taskila

Professor Olli Vuolteenaho

University Lecturer Veli-Matti Ulvinen

Director Sinikka Eskelinen

Professor Jari Juga

University Lecturer Anu Soikkeli

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-952-62-1125-1 (Paperback)ISBN 978-952-62-1126-8 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1345

ACTA

Fanni Päkkilä

OULU 2016

D 1345

Fanni Päkkilä

THYROID FUNCTION OF MOTHER AND CHILD AND THEIR IMPACT ON THE CHILD’S NEUROPSYCHOLOGICAL DEVELOPMENT

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF MEDICINE;MEDICAL RESEARCH CENTER OULU;NATIONAL INSTITUTE FOR HEALTH AND WELFARE