Craniofacial malformations and psychiatric disorders from ...

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1688

Craniofacial malformations andpsychiatric disorders from aneurodevelopmental perspective

KARIN K. TILLMAN

ISSN 1651-6206ISBN 978-91-513-1028-2urn:nbn:se:uu:diva-421456

Dissertation presented at Uppsala University to be publicly examined in H:son Holmdahlsalen, Uppsala University Hospital, ing 100/101, 2nd floor, Uppsala, Friday,4 December 2020 at 09:15 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination will be conducted in Swedish. Faculty examiner: Professor Clara Hellner (Karolinska Institute).

AbstractTillman, K. K. 2020. Craniofacial malformations and psychiatric disorders from a neurodevelopmental perspective. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1688. 58 pp. Uppsala: Acta Universitatis Upsaliensis.ISBN 978-91-513-1028-2.

Orofacial clefts (OFC) and craniosynostosis (CS) are the two most common craniofacial malformations. Of note, craniofacial abnormalities share some overlapping risk factors with psychiatric disorders. Thus, this thesis aimed to study psychiatric and educational outcomes in this group.

In study I and III we examined psychiatric outcomes among children with nonsyndromic OFC stratified on cleft lip (CL), cleft lip and palate (CLP), cleft palate only (CPO), unilateral and bilateral CL and CLP. In study II we studied associations between nonsyndromic CS (NSCS) and psychiatric disorders. Study IV assessed national standardised tests in Swedish and mathematics, school grades and university degrees in children with CL, CLP and CPO. Children with craniofacial malformations were identified through the Swedish National Patient Register and compared to a cohort from the general population that was matched for month and year of birth, sex and county of birth. In addition, children with craniofacial malformations were compared to their unaffected siblings.

Individuals with OFC presented risk increases for intellectual disability, language disorders, psychosis, autism spectrum disorder, attention-deficit/hyperactivity disorder and behavioural disorders in childhood. CPO showed the most robust associations, followed in descending order by CLP and CL. Nonaffected siblings had a lower risk of psychiatric disorders. Females generally had higher risks for psychiatric comorbidity (Study I).

Children with bilateral clefts had higher risk increases for psychiatric disorders compared to children with unilateral clefts. We also found that females with bilateral CLP showed higher risks for intellectual disability and neurodevelopmental disorders compared to males with bilateral CLP (Study III).

Risk increases for any psychiatric disorder including intellectual disability, language disorders, other neurodevelopmental disorders and other psychiatric disorders, were seen in individuals with NSCS. In the crude analyses full siblings with NSCS, as compared to nonaffected siblings, were more likely to be diagnosed with any psychiatric disorder, intellectual disability, language disorders and other neurodevelopmental disorders. The higher risk for any psychiatric disorder and intellectual disability remained after adjusting for confounders. Females displayed borderline higher risk increases than males (Study II).

Finally, children with OFC had lower school performance almost throughout the educational years, especially in mathematics. Lower academic achievement was most evident in children with OFC without a concurrent psychiatric disorder. In the ninth school year and upper secondary school female academic outcomes were more negatively affected than male academic outcomes (Study IV).

In summary, craniofacial malformations were associated with increased risks for multiple psychiatric disorders and lower academic achievement.

Keywords: Nonsyndromic orofacial clefts, Nonsyndromic craniosynostosis, Neurodevelopmental disorders, Psychiatric disorders, Educational achievements

Karin K. Tillman, Department of Neuroscience, Child and Adolescent Psychiatry, Akademiska sjukhuset, Uppsala University, SE-75185 Uppsala, Sweden.

© Karin K. Tillman 2020

ISSN 1651-6206ISBN 978-91-513-1028-2urn:nbn:se:uu:diva-421456 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-421456)

http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-421456

Supervisors Fotios Papadopoulos, MD, PhD Department of Neuroscience, Psychiatry Uppsala University Lisa Ekselius, Professor, MD, PhD Department of Neuroscience, Psychiatry Uppsala University Mia Ramklint, MD, PhD Department of Neuroscience, Child and Adolescent Psychiatry Uppsala University Daniel Nowinski, MD, PhD Department of Surgical Sciences, Plastic Surgery Uppsala University

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I. Tillman, K.K., Hakelius, M., Höijer, J., Ramklint, M., Ekselius, L.,

Nowinski, D., Papadopoulos, F.C. Increased risk for neurodevelop-mental disorders in children with orofacial clefts. (2018) JAACAP

II. Tillman, K.K., Nowinski, D., Höijer, J., Ramklint, M., Ekselius, L., Papadopoulos, F.C. Non-syndromic Craniosynostosis is associated with increased risk for psychiatric disorders. (2020) Plastic and Re-constructive Surgery

III. Tillman, K.K., Hakelius, M., Höijer, J., Ramklint, M., Ekselius, L., Nowinski, D., Papadopoulos, F.C. Bilateral orofacial clefts confer increased risk of psychiatric morbidity relative to unilateral clefts Submitted Manuscript

IV. Tillman, K.K., Nowinski, D., Höijer, J., Ramklint, M., Ekselius, L., Papadopoulos, F.C. Academic Achievement in children with orofa-cial clefts - a nationwide study in Sweden. Manuscript

Reprints were made with permission from the respective publishers.

Contents

Introduction ................................................................................................... 11 Craniofacial malformations ...................................................................... 11

Orofacial clefts ..................................................................................... 11 Craniosynostosis .................................................................................. 15

Neurodevelopment and the origins of psychiatric disorders .................... 18 Risk factors ............................................................................................... 20

Orofacial clefts ..................................................................................... 20 Craniosynostosis .................................................................................. 20 Neurodevelopmental disorders ............................................................ 20

Aim and scope of this thesis .......................................................................... 22

Methods ......................................................................................................... 23 Study design ............................................................................................. 23 Registers ................................................................................................... 23 Outcome measures .................................................................................... 23 Diagnostic classification systems ............................................................. 23

ICD ....................................................................................................... 23 DSM ..................................................................................................... 24 Study populations study I-IV ............................................................... 25 Statistical analyses ............................................................................... 27

Results ........................................................................................................... 29 Study I ....................................................................................................... 29 Study II ..................................................................................................... 31 Study III .................................................................................................... 33 Study IV .................................................................................................... 34

Discussion ..................................................................................................... 35 Study I ....................................................................................................... 35 Study II ..................................................................................................... 36 Study III .................................................................................................... 37 Study IV .................................................................................................... 37 General discussion .................................................................................... 38 Ethical considerations ............................................................................... 40 Methodological considerations ................................................................. 40 Implications and future directions ............................................................ 44

Conclusions ................................................................................................... 45

Sammanfattning på svenska .......................................................................... 47 Bakgrund och frågeställning ..................................................................... 47 Metod ........................................................................................................ 47 Resultat ..................................................................................................... 48 Slutsats ...................................................................................................... 49

Acknowledgements ....................................................................................... 50

References ..................................................................................................... 53

Abbreviations

ADHD Attention-deficit/hyperactivity disorder aHR Adjusted hazard ratio ASD Autism spectrum disorder BCL Bilateral cleft lip BCLP Bilateral cleft lip and palate CL Cleft lip CLP Cleft lip and palate CPO Cleft palate only CS Craniosynostosis DSM Diagnostic and statistical manual of mental disorders ESSENCE Early symptomatic syndromes eliciting neurodevelop-

mental clinical examinations HR Hazard ratio ICD International statistical classification of diseases and re-

lated health problems MBR Swedish medical birth register NPR Swedish national patient register NSCS Nonsyndromic craniosynostosis NSOFC Nonsyndromic orofacial clefts OFC Orofacial clefts SGA Small for gestational age SGR Swedish school grade register UCL Unilateral cleft lip UCLP Unilateral cleft lip and palate

11

Introduction

Craniofacial malformations Craniofacial malformations are congenital conditions in which the develop-ment of cranial or facial structures has been disturbed during early embryo-genesis, leading to various forms of anatomical defects and deformities. Cra-niofacial malformations may be grossly categorised into four major groups: orofacial clefts (OFCs), craniosynostosis (CS), microsomia and dysostosis. These types of malformation may occur as an isolated phenomenon or as part of a genetic syndrome. The two most prevalent types of craniofacial malfor-mation are nonsyndromic OFC (NSOFC) and nonsyndromic CS (NSCS). Children born with craniofacial disorders need multidisciplinary care that ex-tends from birth to adulthood and includes, apart from surgical treatment, close follow-up and treatment for speech, hearing, facial growth and appear-ance. Specialized craniofacial teams provide this kind of care.

Orofacial clefts OFC is the most common craniofacial malformation and affects approxi-mately two in 1000 live births in Sweden, which results in about 150-200 new-borns per year1. The prevalence varies across geographic origin and racial and ethnic groups but is considered stable over time2. OFCs occur at gestational week four to eight when the neural crest cells differentiate from the neural tube and migrate to the facial region. The facial structures are formed from the parietal maxillary and mandibular prominences and the central frontonasal prominence. The middle face is formed by the closure of the maxillary and frontonasal prominences during the fourth to eighth gestational weeks. In ges-tational week 10 the facial structures are fully developed (Figure 1). Facial clefts occur when the prominences do not merge and the location of the cleft depends on involved prominences.

12

Figure 1. Facial embryological development gestational week four to eight By Kicko26 - Own work, CC BY-SA 4.0. https://commons.wikimedia.org/w/in-dex.php?curid=484930392018.

https://commons.wikimedia.org/w/in-dex.12



13

Children with OFC can either have a cleft lip (CL), cleft palate (CPO) or cleft lip and palate (CLP) (Figure 2-3). The lip and primary palate (anterior to the incisive foramen) have distinct developmental origins from the secondary palate; clefts of these areas are divided into CL, CLP and isolated CPO, in which the lip is not affected. Incomplete fusion between the medial nasal and maxillary prominences (between the primary and secondary palate) causes CL. Cleft of the palate behind the incisive foramen is caused by incomplete fusion between the lateral palatal shelves (secondary palate). The cleft anterior to the incisive foramen can be either unilateral or bilateral. Because of the difference in developmental origin, patients with CPO are believed to consti-tute an aetiologically different group from patients with a CL. Approximately 70% of cases with clefts have no associated malformations or known syn-dromes (nonsyndromic OFC).

Children with OFC need multidisciplinary care from birth to young adult-hood, including staged surgical treatment as well as monitoring of the various cleft-associated symptoms. Multidisciplinary care for OFC with long-term follow-up until the age of 19 was established in Sweden in the 1960s and is today conducted through six regional OFC centres. Surgical protocols have gradually evolved over time and with some variations in surgical techniques, or the timing of different surgical treatments, between centres. Individuals with OFC receive staged surgical treatment beginning already in infancy ac-cording to specific protocols. In Sweden, children with CLP most often re-ceive surgery at the age of three months (lip and nose), six months (soft pal-ate), two years (hard palate) and at nine years (bone transplant from hip to palate). Children with CPO receive surgery at six months (soft palate) and two years (hard palate). Some centres operate the soft and hard palate in one ses-sion at the age of one year instead of staged surgeries for the soft and hard palate. Additional surgery may be needed to enhance speech and is often per-formed from the age of six. Secondary corrections for the nose and lip, usually performed in adolescence, are sometimes needed. At 19-20 years of age, cor-rective jaw surgery is occasionally required to correct malocclusion and en-hance facial appearance. The longitudinal assessments of facial growth and speech are today largely harmonised nationally through the professional net-work Swedecleft. Outcomes are openly reported through the National Cleft Lip and Palate Register founded in 1999 and with a coverage rate above 90 per cent for individuals born 2009 or later3,4.

Children with OFC have been shown to have an increased risk for worse educational achievement in school5-7, increased risk of psychiatric disor-ders2,8,9, increased use of psychotropic medications10 and increased all-cause mortality and suicide11. Of note, some studies have shown an association with children with OFC and structural brain anomalies, including midline anoma-lies and differences in the sizes of the cerebrum and cerebellum associated with mild cognitive impairments12,13.

14

Figure 2. Baby with a cleft lip.

Figure 3. Baby with a cleft palate involving both soft and hard palate. Both figures published with permission from Centers for Disease Control and Prevention, Na-tional Center on Birth Defects and Developmental Disabilities, 2020. https://www.cdc.gov/ncbddd/birthdefects/cleftlip.html

https://www.cdc.gov/ncbddd/birthdefects/cleftlip.html

15

Craniosynostosis Craniosynostosis (CS) is the second most common craniofacial disorder after orofacial clefts, affecting nearly 1 in 2000 live births and approximately 80-90 children per year receive surgical treatment for CS in Sweden14,15. The prevalence varies across geographic origin and racial and ethnic groups. A progressive change in the distribution of the different subtypes of NSCS has been observed across different geographic centres, with the subtype metopic synostosis presenting the most dramatic increase16-19. CS is a congenital con-dition entailing premature fusion of one or more of the cranial sutures. About 85 per cent of cases are NSCS while in 15 per cent of the patients CS is part of a syndrome with multiple manifestations. CS constitutes a very heteroge-neous group with different embryological origins, symptomatology, surgical treatment protocols and follow-up. The embryonic cellular origin differs, where the frontal bone of the neurocranium originates from the neural crest and the more posterior cranial vault bones derive from the paraxial meso-derm20. An infant’s skull comprises six cranial bones (i.e. the frontal bone, the occipital bone, two parietal and two temporal bones) held together by strong fibrous connection tissues, called sutures, enabling cranial expansion through intramembranous ossification. The expansion is driven by the growing brain pushing the cranial vault bones apart and osteoprogenitor cells then differen-tiate to osteoblastic cells causing ossification and growth of the skull during the first two years of life. From age two to eight years, the growth of the skull is mostly driven by resorption of endocranial bone and epicranial ossification. In typically developed children the metopic suture fuses physiologically in infancy, but the other sutures normally fuse in adulthood21,22.

CS is divided anatomically into simple (single suture involvement) or com-plex (multiple suture involvement) synostosis and then subdivided depending on suture or sutures involved. The most common CS is the sagittal, followed by metopic, unilateral coronal and lambdoid synostosis as the rarest type of CS23(Figures 4-8).

Fusion of cranial sutures leads to an altered cranial shape in which the type of cranial deformity depends on the pattern of sutural fusion. The growth of the skull is typically restricted in a direction perpendicular to and increased in parallel with the fused suture. Moreover, in syndromic CS the restricted growth of the skull base and facial skeleton may lead to severe functional im-pairment of the eyes and upper aero-digestive tract.

16

Figure 4. Sagittal craniosynostosis

Figure 5. Unilateral coronal craniosynostosis. Both figures published with permission from the Centers for Disease Control and Prevention, National Center on Birth Defects and Developmental Disabilities. https://www.cdc.gov/ncbddd/birthdefects/craniosynostosis.html

https://www.cdc.gov/ncbddd/birthdefects/craniosynostosis.html

% ;J%

%

%%%%%%%%%%%%%%='-"3*%<@%T,-'((,)%0$160(60'0%% % ='-"3*%J@%Q'-4(%"1'),(*3,)%]0A,/46A*/4,)$^% % A6361,)%0$160(60'0%]/),-'6A*/4,)$^%

%='-"3*%K@%M*(6/'A%0$160(60'0%](3'-616A*/4,)$^%

T$1836#'A% RT% '0% 65(*1% ,006A',(*8% S'(4% A6#/36#'0*8% 1*"36/0$A46)6-'A,)%5"1A('6109FL% 0"A4% ,0%-)6G,)% '1(*))'-*1A*% ]Vl^L% G"(%#60(% A4')83*1%S'(4%NTRT%4,2*%-)6G,)%8*2*)6/#*1(,)%63%Vl%0A63*0%S'(4'1%(4*%163#,)%3,1-*@%IG6"(%61*\(4'38% 65% (4*% A4')83*1% S'(4% NTRT% 8*#610(3,(*% 2,3'6"0% 1*"368*2*)6/#*1(,)%/36G)*#0% ]*@-@L% )*,31'1-% 8'0,G')'($L% ),1-",-*% '#/,'3#*1(% 63% G*4,2'6"3,)% 63%A6-1'('2*%8'55'A")('*0^9a@%%&4*%4$/6(4*0'0%4,0%G**1%(4,(%0"("3*%5"0'61%)*,80%(6%,G163#,)%G3,'1%8*2*)\

6/#*1(@%&4*%(S6%0"--*0(*8%A,"0,)%/,(4S,$0%,3*%*)*2,(*8%'1(3,A3,1',)%/3*00"3*%]VR+^% S'(4% 4$/62,0A"),3'($% ,18% 0*A618,3$% A*3*G3,)% 8*563#,('61% 3*0")('1-%

18

from a brain growth in an abnormally shaped skull26. Current reviews suggest, however, that increases in ICP occur only in a minority of cases, mostly in sagittal synostosis and very rarely in metopic or unilateral coronal synostosis. There is no agreed level for what is ‘normal’ versus ‘abnormal’ ICP in chil-dren. Studies have also failed to associate ICP with neurodevelopment27. The cerebral deformation hypothesis has only been studied to a limited extent. It has, however, been suggested that abnormally long or short distances between major brain structures could cause small perturbations in cortical connectivity, eventually producing functional changes in the way the brain processes infor-mation28. Cognitive neuroscience research indicates that small variations in neural organisation can ultimately lead to changes in cognitive function-ing20,21.

Children with CS need surgery for aesthetic and functional reasons and re-ceive multidisciplinary care with follow-up at least until the growth of the neurocranium is completed at eight years of age. The treatment aims to pre-vent deleterious effects on brain development, normalise appearance and re-store normal facial functions in syndromic CS. In NSCS a single transcranial surgical procedure aiming to correct cranial anatomy is normally performed during the first year of life. There are currently several surgical methods used: open calvarial reconstruction, extended craniectomies, minimally invasive strip craniectomy using a post-operative moulding helmet and minimally in-vasive strip craniectomy combined with cranial distraction using springs. The surgery is often undertaken before the age of one year among NSCS and more surgical complications have been reported with delayed surgical treatment29,30.

Neurodevelopment and the origins of psychiatric disorders Early neurodevelopmental disturbances are suspected of contributing to the pathophysiology of several psychiatric disorders. Hence, many studies em-phasise the importance of recognising the model of a neurodevelopmental ba-sis for psychiatric disorders31-34. Accumulating data suggest that both heritable and environmental factors influence neurodevelopment, establishing risk to later psychiatric disorders. Neurodevelopment extends from the early embry-onic period through adolescence to early adulthood and includes several crit-ical processes. The embryonic development of the brain and spinal cord starts with a phase called neurulation in the second gestational week where ectoder-mal tissue forms the neural plate. The neural folds begin to move together in the third gestational week and form the neural tube, which consequently, leads to cells separating from the surface ectoderm to form the neural crest. The neural crest will then give rise to the brain and several craniofacial structures with a distinct and predictable timetable35. The neurodevelopment starts with

19

the proliferation of neurons and radial glia, continues with the migration of neurons, differentiation, synapse formation in the last trimester and first two years of life and myelinisation and synapse pruning beginning in the second trimester and continuing through adolescence. The centre for executive func-tioning in the prefrontal cortex is not fully developed until after 2035. The tra-ditional division between childhood neurodevelopmental disorders and adult psychiatric disorders has been questioned and the neurodevelopmental hy-pothesis suggesting a neurobiological origin for psychotic disorder and to some extent also bipolar disorder (BD) has become the accepted conjecture. Data suggest that psychotic disorders, such as schizophrenia, have origins in disturbances of brain development, with several identified phenotypic simi-larities as well as shared genetic risk and pathogenic mechanisms with the early onset neurodevelopmental disorders34,36. The term ‘neurodevelopmental disorders’ often refers to a long and heterogeneous list of diagnoses, tradition-ally including intellectual disability (ID), autism spectrum disorder (ASD), at-tention-deficit/hyperactivity disorder (ADHD), speech and language disor-ders, learning disorders, developmental coordination disorder (DCD) and Tourette's syndrome (TS). Psychotic disorder and BD are also psychiatric dis-orders with a neurodevelopmental origin but are not traditionally regarded as neurodevelopmental disorders. The neurodevelopmental disorders are com-plex and highly heterogeneous and present symptoms early in childhood. Over the past decade, studies have indicated that developmental problems are often overlapping and not always distinct in the clusters of diagnostic criteria for a certain diagnosis36-38. A neurodevelopmental gradient hypothesis has also been introduced, suggesting that within the neurodevelopmental continuum stronger neurodevelopmental impairment and correlations are seen for ID and decreasing neurodevelopmental impairment following a gradient as follows: ID, ASD, ADHD, schizophrenia and BD34.

The acronym ESSENCE (Early Symptomatic Syndromes Eliciting Neuro-developmental Clinical Examinations) has been introduced and refers to chil-dren affected by impairing symptoms in general development, communication and language, social inter-relatedness, motor coordination, attention, activity, behaviour, mood or sleep before the age of five39. Significant problems in at least one ESSENCE domain before five years of age have proven to predict poor mental health, academic achievement and psychosocial outcomes later in life and multidisciplinary support is often needed in more than one area of overlapping symptoms or deficits38.

20

Risk factors Orofacial clefts Approximately 30% of OFCs are associated with a known genetic syndrome (syndromic OFC), but the remaining 70% of clefts occur without a known syndrome identified (nonsyndromic OFC)40,41.

Inheritance of orofacial clefts may be chromosomal, sporadic or related to environmental teratogens 42,43. Known maternal risk factors include smok-ing44-46, alcohol consumption45, diabetes, nutritional factors (e.g., vitamin A, folic acid insufficiency) and anticonvulsant medication47. Season of birth, and specifically winter and early spring, has been associated with increased risk for OFC48.

Craniosynostosis Approximately 15% of CSs are associated with a known genetic syndrome, but most are nonsyndromic cases23,49. Premature fusion of the cranial sutures occurs early prenatally but most patients are diagnosed postnatally50,51. Syn-dromic CS is caused by identified genetic mutations in genes coding for growth factor receptors or involved transcription factors, with a predominance of autosomal dominant inheritance. Well-known syndromes are the Apert, Crouzon, Pfeiffer, Muenke and Saethre-Chotzen syndromes, all caused by changes in the genes for fibroblast growth factor receptor (FGFR)52 2 and 3 and in the TWIST1 gene53.

Predisposing risk factors for NSCS are environmental factors such as intra-uterine fetal head constraint, abnormal position, maternal smoking46,54-56, ma-ternal thyroid disease57 and anticonvulsant medication such as valproic acid and phenytoin58,59.

Numerous studies have evaluated the association between NSCS and dif-ferent potential risk factors but there are many inconsistencies in the results. Some discrepancies may be attributed to the study design with small sample sizes, variations in the types of exposure assessed and lack of data to evaluate the potential effects modification and confounding factors16.

Neurodevelopmental disorders Neurodevelopmental disorders have a combination of predisposing genetic and specific environmental, gestational and perinatal risk factors affecting brain development during pregnancy, at birth or after birth. These risk factors include alcohol, medication or toxin exposure, nutritional deficiencies, infec-tion (herpes, toxoplasma, cytomegalovirus, influenza), hypoxia, extreme premature birth, neonatal seizures, trauma or metabolic changes60. Many of these risk factors are poorly understood and gene defects are often unknown.

21

Mutations in genes encoding for cell adhesion molecules and calcium channel subunits have been associated with some neurodevelopmental and psychiatric disorders61. The cell adhesion genes have crucial roles in early brain develop-ment, including neuronal migration, synaptogenesis and axonal growth but the exact mechanism for mutations in this gene family and their contribution to the pathogenesis of neurodevelopmental disorders is not known61.

Emerging evidence from studies of rare copy number variants has shown that childhood neurodevelopmental disorders such as ID, ASD and ADHD share specific genetic risk alleles and to some extent also with psychotic dis-orders and BD62,63

Season of birth, mainly during late winter and spring, has also been associ-ated with increased risk for psychotic disorders, bipolar and unipolar disorders and suicide, although suicide is affected by several other external factors64,65

Educational achievement There are several known risk factors for poor school performance, such as preterm birth66, small for gestational age (SGA)67, Apgar score below seven at five minutes after birth68, ID, specific learning disorders69, ASD70, ADHD71, psychiatric disorders72,73, low parental educational level74,75, parental psychi-atric disorders and a poor socio-cultural home environment76.

22

Aim and scope of this thesis

The general aim of this thesis was to examine the possible role of neurodevel-opmental disturbances in utero in psychiatric disorders by using craniofacial disorders as a model of developmental anomalies with a clear manifestation of disturbed embryonic development. It was hypothesised that children with craniofacial disorders would have an increased risk for psychiatric disorders.

The specific aims of the individual studies were:

I To investigate whether there is an increased risk for psychiatric dis-orders, suicide attempts or suicide in children with orofacial clefts compared to children without clefts and analyse whether there are any differences by sex and the different subtypes of OFC. The study also investigated possible familial influences (genetic or shared envi-ronmental factors) through sibling analyses.

II To investigate psychiatric disorders, suicide attempts or suicide in children with NSCS, to perform sibling analyses to address familial influences and sex-stratified analyses. We further studied the effect modifying role of the number of operations, age at first surgery, time at first admittance and surgery with springs.

III To investigate the risk of psychiatric disorders, suicide and suicide attempts in children with bilateral clefts compared to unilateral clefts and assess possible sex-related differences in psychiatric outcomes.

IV To investigate academic attainment in children born with nonsyn-dromic CL, CLP and CPO. As a secondary aim, we sought to assess possible sex differences and assess the academic outcomes in chil-dren with clefts, both with and without psychiatric disorders.

23

Methods

Study design Retrospective nationwide register cohort studies.

Registers We obtained data from the National Board of Health and Welfare in Sweden and Statistics Sweden. In the data we received anonymity was secured by re-placing identification numbers with arbitrary numbers. We used and linked the following registers: the Swedish Medical Birth Register (MBR), the Na-tional Patient Register (NPR), the Swedish Cause of Death Register, the Reg-ister of Total Population, the Multi-Generation Register, the Migration Regis-ter, the Census of the population and housing and the longitudinal integration database for health insurance and labour market studies (LISA).

Outcome measures Information on psychiatric diagnoses and suicide attempts was extracted from the NPR and data on suicide were extracted from the Swedish Cause of Death Register. Table 1 presents the codes used in the different versions of the Inter-national Statistical Classification of Diseases and Related Health Problems (ICD). Information on educational outcomes for study IV was retrieved from the Swedish School-Grade Register (SGR) (Figure 10-11).

Diagnostic classification systems ICD The ICD is the international standard health care classification system for clin-ical work, health management and epidemiology. The ICD is published by The World Health organisation (WHO) and promotes international compara-bility to classifying disorders. The ICD has been revised periodically. In our

24

longitudinal studies we use the Swedish versions of ICD-8 (1969-1986), ICD-9 (1987-1996) and ICD-10 (1997-2018).

DSM The Diagnostic and Statistical Manual of Mental Disorders (DSM) is another manual of classification of mental health problems maintained by the Ameri-can Psychiatric Association. The two classification systems strive to use the same diagnoses and criteria but are not identical and the revised versions are not automatically updated at the same time. WHO performed an evaluation comparing the two diagnostic systems throughout 66 countries and found that ICD-10 was more frequently used and valued for clinical diagnosis and train-ing and that DSM-IV was more valued for research77. A survey among mental health professionals, primarily psychiatrists, from 92 countries reported that both the ICD-10 and DSM IV were rated to be useful for assigning a diagnosis, communicating with other health care professionals and teaching, but more clinicians were using the ICD for administrative and billing purposes78.

Table 1. ICD-8, 9 and 10 codes for specific groups of psychiatric disorders used in the studies79,80

ICD-8 ICD-9 ICD-10 Any psychiatric disorder 290-315 290-319 F00-F99 Intellectual disability/ Mental retardation

310-315 317-319 F70- F79

Speech and language disor-ders

306,00 781,59 315D F80 R47

**Psychotic disorders

295 297-299 (except 298,00)

295 297 298 (ex-cept 298A)

F20-F29

**Bipolar disorders 296 (except 296,00)

296 (except 296B, 296X)

F30-F31

**Depression 298,00 300,40 790,20

298A 300E 311 F32-F33

**Eating disorders 306,50 307B 307F F50

*Autism spectrum disorders - 299 F84

*Attention-deficit hyperac-tivity disorder

- 314 F90

*Other behavioural/emo-tional disorders with onset in childhood

308,99 312-313 F91-F98

25

**Neurotic, stress-related or somatoform disorders

300 (except 300,40); 305 306,80 306,98 307,99

300 (except 300E) 306 307W 308 309

F40-F48

**Alcohol and substance use disorder

291 303 294,30 304 971

291 303 305A 292 304 305X

F10-F16 F18 F19

**Personality disorder 301 301 F60 Suicide attempt E950-E959 E950-E959 X60-X84 Suicide E950-E959 E950-E959 X60-X84 *’Other neurodevelopmental disorders’ in study II. **’Other psychiatric disorders’ in study II

Study populations study I-IV Study I We identified through the NPR 7900 children born in Sweden between 1973 and 2012 who received the diagnosis of OFC at birth or before five years of age; of these 7900 children, 7842 (99%) were included and 58 excluded in the analyses. These 58 children were excluded because they had no identified matched comparison individuals from the population. Children with OFC were identified using the ICD codes as indicated in the NPR using the follow-ing ICD codes: CL: 749.10-749.13(ICD-8), 749B(ICD-9) and Q36 (ICD-10); CLP: 749.20-749.24(ICD-8), 749C(ICD-9) and Q37 (ICD-10); and CPO: 749.00(ICD-8), 749A(ICD-9) and Q35 (ICD-10). Two comparison groups were used: (i) 78409 individuals without a diagnosis of OFC malformation from the general population, identified by Statistics Sweden and matched for month and year of birth, sex and county of birth, and (ii) 9637 full siblings to our individuals with OFC with less than 20 years age difference, identified in the Multi-Generation Register79.

Study II We identified through the inpatient NPR 1238 children born in Sweden be-tween 1973 and 2012 who received the diagnosis of NSCS at birth or before five years of age according to ICD-8 or ICD-10. We decided not to include children who received a diagnostic code of NSCS only from ICD-9 because unspecific diagnostic codes could also include other facial malformations such as hypertelorism and not representing exclusively NSCS. The NSCS cohort was compared to a matched community cohort (n=12380) and their unaffected full siblings (n=1487). Children with NSCS were identified using the ICD codes as indicated in the inpatient NPR by the following ICD codes: 756,00 (ICD-8) and Q75.0 (ICD-10) given at the same time as the respective opera-tion code 0202, AAK20 or AAK3080.

26

Study III We identified 2454 children born between January 1, 1973 and December 31, 2012 who received a diagnosis of unilateral or bilateral CL or CLP before 5 years of age. Each study participant was matched with 10 individuals without OFC (comparison cohort, n=24535) by month and year of birth, sex and county of birth.

Study IV We identified all children born with CL, CLP and CPO in Sweden between 1973 and 2004, with available records in the Swedish School Grade Register (SGR) (n=6286) and compared them to the matched community cohort (n=61352).

Figure 10 Timeline nonsyndromic OFC

Figure 11. Timeline nonsyndromic CS

Gestational week 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 birth

Exposure: Orofacial cleft

Risk factors: Smoking, diabetes, alcohol, folic acid deficiency, valproic acid

Preterm birthOutcome: Psychiatric disorders

Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Surgical protocol:Lip and nose 3 mSoft palate 6 m Hard palate 2 yBone graft 9 y

Sometimes additionalsurgery:To enhance speech 6 ySecondary correctionsOrthognatic from 20 y

Outcome: Educational achievement

Gestational week 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 birth

Exposure: Craniosynostosis from earlygestational to postnatally

Risk factors: Smoking, thyroid diseaseintrauterine compressionvalproic acid

Preterm birth

Outcome: Psychiatric disorders

Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Surgery duringfirst year

% 9J%

%='-"3*%;9@%=)6S%A4,3(%65%(4*%0("8$%/6/"),('610%'1%0("8$%V\VH%

T(,('0('A,)%,1,)$0*0%&4*%8,(,%S*3*%,1,)$0*8%"0'1-%(4*%0(,('0('A,)%065(S,3*%T(,(,%2@;aK;@%R6U%/36/63\('61,)%4,f,38%3*-3*00'61%#68*)0%S*3*%"0*8%563%0("8$%VL%VV%,18%VVV%(6%'12*0('-,(*%A3"8*%,18%,8Z"0(*8%4,f,38%3,('60%]XQ0^@%=63%0("8$%VHL%S*%"0*8%G'16#',)%3*\-3*00'61%S'(4%)6-'(%)'1B%5"1A('61%,18%A)"0(*3%36G"0(%0(,18,38%*33630%(6%'12*0('\-,(*%6880%3,('60%]`Q0^%563%(4*%1"#G*3%65%0"AA**8*8%1,('61,)%0(,18,38%0"G(*0(0@%7*%"0*8%638'1,)%)6-'0('A%3*-3*00'61%S'(4%A)"0(*3%36G"0(%0(,18,38%*33630%(6%'1\2*0('-,(*%`Q0%563%-3,8*0%S4*1%5'1'04'1-%A6#/")063$%0A466)%,18%,(%"//*3%0*A\618,3$%0A466)@%R618'('61,)%)6-'0('A%3*-3*00'61%S'(4%A)"0(*3%36G"0(%0(,18,38%*3\3630%S,0%/*3563#*8%(6%'12*0('-,(*%(4*%̀ Q%65%(4*%G'1,3$%6"(A6#*%65%,%"1'2*30'($%8*-3**%63%16(%'1%'18'2'8",)0%S'(4%`=R%A6#/,3*8%(6%(4*%"1,55*A(*8%'18'2'8",)0%]&,G)*%9^@%&6%*U,#'1*%,%/600'G)*%#68*3,('1-%*55*A(%65%0*U%S*%*0('#,(*8%0*/,3,(*%#68\

*)0%'1%S4'A4%,1%'1(*3,A('61%(*3#%65%0*U%,18%(4*%*U/60"3*%S,0%'1A)"8*8@%=36#%(4*0*%#68*)0L%(4*%/\2,)"*%65%(4*%'1(*3,A('61%(*3#%65%0*U%S,0%A,)A"),(*8@%7*%A61(36))*8%(4*%,1,)$0*0%563%3*)*2,1(%A6156"18*30>%/*3'1,(,)%A6#/)'A,('610%,18%06#,('A%'18'A,(630L%$*,3%65%G'3(4L%0*UL%A61-*1'(,)%#,)563#,('610%63%B16S1%-*\1*('A% 0$1836#*0L%/,3*1(,)%/0$A4',(3'A%#63G'8'($%,18%06A'68*#6-3,/4'A% 5,A\(630@%%I88'('61,))$L%'1%6"3%,1,)$0*0%'1%0("8$%VV%S*%*U,#'1*8%/600'G)*%#68*3,(630>%

('#*%65% A,3*% ,(% 5'30(% ,8#'00'61% ];\<%8,$0%,18%q<%8,$0^L%1"#G*3%65% 0"3-'A,)%(3*,(#*1(0%];%,18%9\F^L%0"3-*3$%,-*%]r:@a%$*,30%,18%:@a\a%$*,30^L%0"3-'A,)%(3*,(\#*1(%S'(4%0/3'1-%,18%3*#62,)%65%0/3'1-%E%#61(40%),(*3@%&6%'12*0('-,(*%/600'G)*%#68*3,('1-%*55*A(0%S*%/*3563#*8%0*/,3,(*%R6U%#68*)0%'1%S4'A4%'1(*3,A('61%(*3#0%65%(4*%,G62*\#*1('61*8%#68*3,(630%,18%(4*%*U/60"3*%NTRT%S*3*%'1\A)"8*8%,18%(4*%/\2,)"*0%65%(4*%'1(*3,A('61%(*3#%A,)A"),(*8@%

!"#$%

&

&&&

!"#$"%&'

(")*+,) -./

012131456

/+789:"'+% &:+,8

(")*+,) -./1

01216;1<=3

/*"$>:?%1(")* -./1

012161;<@

A%"B+:1#"$9)?:9$1-./1

%121@1<C<

/+789:"'+% &:+,8

(")*+,) -./1

0121@<1C5C

&'/*"$>:?%1(")* -./1(")*

9D9"$9#$? >9)91+%19E9>?7"E

+,)E+7?'

%12141@;4

/+789:"'+% &:+,8

(")*+,) -./1

01214F15C@

/*"$>:?%1(")* -./1"%1)*?1GHI1+:1

0JI1KF365B@=F@L1

0121613==

G"''"%& E+789:"'+% "%>"D">,9$'

%121C;

!"#$"%&'

(")*+,) /!1

0121F1<;C

/*"$>:?%1(")* /!1"%1)*?10JI1

KM/NB;19%>1BF=O1F365B@=F@L1

012F1@5;

/+789:"'+% &:+,8

(")*+,) /!1

0121F@15;=&&

28

In study IV we also stratified on CL, CLP and CPO with and without psy-chiatric disorders in that previous studies have reported an association be-tween psychiatric disorders and poorer academic achievement.

Table 2. Overview of exposure, unexposed population, outcome and statistical meth-ods for Study I-IV

Study Population with the ex-posure

Unexposed popu-lation

Outcome measu-res

Statistical methods

I CL, CLP, CPO

1. Matched indi-viduals without OFC 2. Siblings

Psychiatric disor-ders, suicide and suicide attempts

Cox proportional hazard regression models

II CS 1. Matched indi-viduals without CS 2. Siblings



III BCL, BCLP 1. UCL, UCLP 2. Matched indi-viduals without OFC



IV CL, CLP, CPO, stratified on psychiatric disorders

Matched individu-als without OFC, stratified on psy-chiatric disorders

National stand-ardised tests, school grades, university degree

Binomial regression, ordinal logistic re-gression and condi-tional logistic regres-sion

29

Results

Study I Children with OFC were more likely to receive a diagnosis of any psychiatric disorder, ID, language disorder, ASD, psychotic disorder, ADHD, and other behavioural and emotional disorders with onset in childhood and personality disorders. However, the outcomes differed somewhat between the different subtypes of OFC: Children with CL had a higher risk for any psychiatric dis-order, ID and language disorders. Children with CLP additionally had an in-creased risk for ASD. Children with CPO presented with higher HRs, and ad-ditionally for psychotic disorders, ADHD and behavioural or emotional dis-orders in childhood. Of note, children born with CL presented a risk decrease for depression.

No risk increase was found among any of the stratified cleft types for sui-cide-attempts, death by suicide and the diagnosis of affective disorders, anxi-ety disorders, eating disorders and alcohol or psychoactive substance abuse. A modest risk increase was seen for personality disorders in the OFC analysis but not in the subtype-stratified analyses.

In the sex-specific analyses we found within the OFC cohort higher risk increases in females than in males for the following: any psychiatric disorder, ID, language disorders, ASD and behavioural and emotional disorders with onset in childhood. These sex differences with generally higher risk increases in females persisted in some of the stratified-by-cleft-type analyses and were most evident in females born with CLP.

The unaffected full siblings without OFC were less likely to be diagnosed with any psychiatric disorder, ID, language disorders, ASD and ADHD, whereas no different risk pattern was observed for psychotic disorders, behav-ioural disorders in childhood and personality disorders. (Figure 13).

30

Figure 13: Adjusted hazard ratios (aHRs) for psychiatric disorders in children with OFC79

*

**

+*

+ * +

* p−value for sex*OFC interaction < 0.05+ p−value for sex*OFC interaction < 0.10

.1

.25

.5

1

2

4

8

12aH

R

Anypsychiatric

Intellectualdisability

Languagedisorders

Psychoticdisorders

Bipolar Depression Eatingdisorders

ASD ADHD Otherbehavioraldisorders

Neurotic,stressrelated

Alcoholand drug

use

Personalitydisorders

Suicideattempts

Suicide

OFC total OFC males OFC females non−OFC siblings

A. Orofacial clefts (OFC)

31

Figure 14: Kaplan-Meier curves with 95% CI for the detected associations for OFC

Study II Individuals with NSCS had a higher risk of any psychiatric disorder, including ID, language disorders, neurodevelopmental disorders and other psychiatric disorders.

In the crude analyses full siblings with NSCS were more likely to be diag-nosed with any psychiatric disorder, including ID, language disorders and neu-rodevelopmental disorders compared to their siblings without NSCS. The higher risk for any psychiatric disorder and ID remained after adjusting for confounders. Our sex interactions analyses showed only borderline risk in-creases in females for any psychiatric disorder, ID and a conglomerate of other neurodevelopmental disorders that included ADHD, ASD and other behav-ioral and emotional disorders with onset in childhood (Figure 15).

There was no observed difference in the risk increases when operated on before or after 6 months of age, except for a borderline risk increase for lan-guage disorders that needs to be investigated further before drawing any con-clusions. There were no detectable moderating effects for surgery with springs or without and the number of surgical procedures in our analyses.

32

Figure 15: Adjusted Cox derived hazard ratios (aHRs) for psychiatric disorders in children with NSCS80

+

+

+

+ p−value for sex interaction < 0.10

.1

.25

.5

1

2

4

8

16aH

R

Anypsychiatric


Languagedisorders

Neuro−develop.disorders

Learningdifficulties

Otherpsychiatric

Psychoticdisorders




Alcoholand drug

use


Suicideattempts

Suicide

Vs. controls Vs. controls − males Vs. controls − females Vs. Siblings

33

Figure 16: Kaplan-Meier curves with 95% CI for the detected associ-ations for NSCS

Study III Individuals with bilateral clefts demonstrated elevated risks of psychiatric dis-orders compared to unilateral clefts. Children with bilateral CL had an in-creased risk for any psychiatric disorder (including ID) than unilateral CL. Children with bilateral CLP demonstrated risk increases for any psychiatric disorder, including speech and language disorders, neurodevelopmental dis-orders and other psychiatric disorders relative to unilateral CLP. Females with bilateral CLP showed higher risks for ID and neurodevelopmental disorders than males (Figures 17-18).

Figure 17: Adjusted Cox derived hazard ratios (aHRs) for psychiatric diagnoses in children with BCL relative to UCL.

+

+

++ p−value for sex interaction < 0.10

.1

.25

.5

1

2

4

8

16

aHR

Anypsychiatric


Languagedisorders

Neurodevelop.disorders



Otherpsychiatric

Psychoticdisorders



Alcoholand drug

use


Suicideattempts

Suicide

Hazard ratios forBCL relative to UCL Males Females

34

Figure 18: Adjusted Cox derived hazard ratios (aHRs) for psychiatric diagnoses in children with BCLP relative to UCLP

Study IV Children with nonsyndromic CLP and CPO showed lower performance in mathematics as early as in the third school year in primary school; moreover children with CPO also had lower performance in Swedish in primary school national standardised tests. In their last year of compulsory school students with CLP and CPO had lower grades in mathematics; students with CLP also had lower grades in Swedish; and students with any clefts had lower av-erage percentile grades in all subjects compared to non-cleft peers. At upper secondary school, students with CLP and CPO had lower grades in mathe-matics but not in Swedish.

There were no detectable differences between boys and girls at national standardised tests in the third school year, but females’ academic outcomes were more negatively affected when leaving compulsory and upper secondary school.

At year three, lower academic achievement was seen on the national stand-ardised tests in children with clefts both with and without psychiatric comor-bidity. In the last year of compulsory school and upper secondary school lower academic achievement was only evident in children with OFC without psy-chiatric comorbidity. This lower academic achievement was not seen in chil-dren with OFC and who also have concurrent psychiatric comorbidity when compared to children without OFC and who also have concurrent psychiatric comorbidity.

+ + ++

++

+ p−value for sex interaction < 0.10

.1

.25

.5

1

2

4

8

16aH

R

Anypsychiatric


Languagedisorders

Neurodevelop.disorders



Otherpsychiatric

Psychoticdisorders



Alcoholand drug

use


Suicideattempts

Suicide

Hazard ratios forBCLP relative to UCLP Males Females

35

Discussion

Study I and II aimed to investigate psychiatric disorders, suicide and suicide attempts in children born with nonsyndromic craniofacial disorders using data from nationwide registers in Sweden. Study III investigated whether there were increased risks for psychiatric disorders in children with bilateral clefts compared to unilateral clefts and study IV continued on the explorative path of study I and assessed whether being born with OFC affected academic out-comes. In study IV we analysed academic attainment both in individuals with and without psychiatric disorders and stratified on sex.

Study I We showed that children with nonsyndromic OFC have an increased risk of being diagnosed with a neurodevelopmental disorder compared to children without clefts and full siblings. CPO showed the highest risk increase, fol-lowed in descending order by CLP and CL. There was no overall risk increase for affective disorders or anxiety disorders. The sizes of the risk increases var-ied for the different psychiatric diagnoses. Risk increases for ID and language disorders were approximately four times higher in OFC. In contrast, risk in-creases for ASD, ADHD, behavioural disorders, psychotic disorders and per-sonality disorders were less than half those of ID and language disorders, sug-gesting a stronger neurodevelopmental association for ID and language disor-ders. When stratifying on cleft type and sex, we did find a small risk increase for BDs (aHR 2.53, 95% CI 1.06-6.05) for males with nonsyndromic CPO, but not for any other subtypes of cleft or females. This finding suggests that there might also be an association for BDs. However, a larger study population may be needed or a more extended follow-up period to detect and reveal that risk increases because BDs are often diagnosed later in life (mean age 25.8 years in our study) compared to many of the neurodevelopmental disorders diagnosed early in childhood. Our findings support the neurodevelopmental gradient hypothesis of psychiatric disorders previously discussed in the intro-duction, but our results suggest a stronger neurodevelopmental association for psychotic disorders than proposed by Owen et al34. Our results support the traditional division between certain childhood neurodevelopmental disorders and adult psychiatric disorders. The present results point at the strongest neu-rodevelopmental associations in descending order for ID, language disorders,

36

ASD, psychotic disorders, other behavioural disorders, personality disorders and ADHD. The risk increase for BD was detectable only in a subgroup anal-ysis, but no other associations were seen for other general psychiatric disor-ders.

The findings in Study I are in line with those from a previous large popu-lation-based cohort study from Denmark, where a higher risk of psychiatric disorders emerged in individuals with nonsyndromic OFC, especially for CPO when stratifying for different cleft types. The Danish study also revealed a higher suicide mortality rate, which did not appear in our study9. Of note, our study revealed substantially higher risk increases for all neurodevelopmental disorders compared to the Danish study. A plausible explanation could be a higher detection rate of neurodevelopmental disorders in Sweden’s OFC pop-ulation and possibly more support services after diagnosis. This is particularly noteworthy when interpreting the higher risk of suicide and suicidal attempts in individuals with OFC in Denmark, but not in Sweden.

The results from the siblings’ analyses suggest that the strong associations with ID and language disorders and the modest associations with ASD and ADHD are unlikely to be fully explained by familial factors, supporting the theory that these neurodevelopmental disorders and OFC might be the result of the same abnormal neurodevelopment. Nevertheless, when interpreting and discussing these risk increases for neurodevelopmental disorders in individu-als with OFC, we need to emphasise that most children with OFC were not diagnosed with psychiatric or neurodevelopmental disorders. Even in the out-comes with the strongest detected associations with aHR above 4 (which means a risk increase of more than 400 per cent), most children were not di-agnosed with any psychiatric or neurodevelopmental disorder because the baseline risks of the observed outcomes were rare (Figure 15).

Study II Study II demonstrated risk increases for any psychiatric disorder, ID, language disorders and a conglomerate of other neurodevelopmental disorders (ASD, ADHD and behavioural disorders) in children born with NSCS compared to the matched children without CS. In the crude analyses we also detected risk increases for behavioural disorders in childhood, psychotic disorders and per-sonality disorders, but they did not remain in the adjusted analyses. OFC and CS are the two most common craniofacial malformations. However, important to bear in mind when interpreting the results is that they represent two entirely different clinical conditions of birth defects with different embryonic origin, developmental pathways, clinical phenotypes and treatment. All this may re-flect that the observed differences in psychiatric outcome may not be that eas-ily explained only by power issues or heterogeneity within the CS cohort. There may be a larger neurodevelopmental association for OFC than for CS.

37

However, the lack of different ICD codes for specific CS subtypes may have diluted our findings because we were unable to distinguish between different CS subtypes. Moreover, we cannot rule out that the higher risks seen for fe-males can be explained by different male to female ratios for different types of CS that pose different risks of neurodevelopmental disorders82,83.

While the causes for these modest associations remain to be clarified, the results are important for the clinical follow-up of these patients and the plan-ning of future studies.

To contextualise the risks for children with NSCS to be diagnosed with psychiatric and neurodevelopmental disorders the Kaplan-Meier curves illus-trate that most children were not diagnosed with psychiatric or neurodevelop-mental disorders, even though we detected important associations with re-vealed risk increases for these outcomes (Figure 16).

Study III New important knowledge on nonsyndromic bilateral and unilateral clefts and psychiatric comorbidity was revealed in study III. Children with bilateral clefts had increased risks for psychiatric disorders and specifically neurode-velopmental disorders compared to children with unilateral clefts, which sup-ports neuroimaging findings showing more structural brain anomalies in bi-lateral clefts.

Females with bilateral CLP showed higher risks for intellectual and neuro-developmental disorders than males.

Important comparisons within a cohort of individuals with OFC offer a unique opportunity to address psychiatric morbidity in a well-defined group of individuals with increased risks of psychiatric disorders and with otherwise similar anaesthesia exposures, which is relevant from a neurodevelopmental perspective.

Study IV Study IV showed that children with nonsyndromic CLP or CPO had poorer academic achievement than non-affected peers from the population already in primary school and, except for the national standardised tests that are per-formed in school year five or six, these differences were present throughout the educational years until adulthood, especially in mathematics. These asso-ciations were also evident in the sub-cohort of children with clefts but no psy-chiatric disorders.

Children with nonsyndromic clefts and concurrent psychiatric comorbidity had poorer academic achievement in only the third school year, but no such associations were evident in later school years. Conceivably, these results

38

suggest that children diagnosed with neurodevelopmental disorders in primary school have received the extra support needed to learn in school. At least our data suggest that their academic results are at a similar level to the comparison cohort of non-OFC children with psychiatric disorders.

These findings emphasise the need to screen children with clefts for neuro-developmental disorders or cognitive difficulties in order to provide additional support in school.

General discussion The general aim of this thesis was to examine the possible role of neurodevel-opmental disturbances in utero in psychiatric disorders by using craniofacial disorders as a model of developmental anomalies. In these studies we found a pattern of increased risks for neurodevelopmental disorders in children born with craniofacial disorders, higher risk increases for bilateral clefts than uni-lateral clefts and poorer educational achievement in children with clefts al-ready from primary school up until adulthood, especially in mathematics and also in the sub-cohort of children without psychiatric disorders.

The overall aim of this thesis was to address the neurodevelopmental per-spective in different psychiatric disorders. Our studies showed a distinct pat-tern by which no risk increases were revealed for psychiatric disorders not usually called neurodevelopmental disorders, i.e. no revealed increased risk of depression, anxiety disorders, eating disorders or substance use disorders in individuals with craniofacial disorders.

Sibling analyses and adjustment for socioeconomic status and psychiatric disorders in parents suggest that the higher risk of neurodevelopmental disor-ders and poorer educational attainment are unlikely to be explained by familial influences and point to supporting the neurodevelopmental hypothesis.

We found the same pattern for sex differences in all four studies, with higher risks for neurodevelopmental disorders in females than males, but only at a borderline level for females with NSCS. Academic outcomes for females with OFC were more negatively affected when leaving compulsory school and at upper secondary school, but not in previous school years. These sex differ-ences coincide with research revealing higher risks of being diagnosed with a neurodevelopmental disorder in females with additional comorbid features (e.g., congenital malformations but also epilepsy, cardiac malformation, growth abnormalities, dysmorphic features and other CNS malfor-mations)84,85. The role of sex-specific diagnostic bias has been discussed, with females being less often diagnosed with neurodevelopmental disorders such as ADHD and ASD due to different expressed phenotypic profiles of symp-toms or symptoms severity, compared to males with ADHD and ASD in the general population. However, another recent Swedish study found that educa-tional outcomes for females with cleft are more adversely affected than for

39

males with cleft when leaving compulsory school6, suggesting a possible ex-planation of a higher vulnerability of the female brain to the underlying neu-rodevelopmental mechanisms in craniofacial disorders. Another plausible ex-planation may be that girls with neurocognitive deficits are not always identi-fied and given enough special educational support at school.

The role of general anaesthetics during infancy and the risk of poor long-term neurodevelopmental outcomes have also been discussed. Preclinical studies have shown that anaesthesia and surgery in infancy have neurotoxic effects in neonatal animals, but there is limited relevant data on humans86-88. Clinical debate and conflicting evidence from cohort studies suggest that early childhood exposure to anaesthesia may increase the risk of poor long-term neurodevelopmental outcome, although the causality is uncertain89,90. Chil-dren exposed to anaesthesia in early childhood undergo surgery for a specific condition and most studies have not been able to control for the underlying comorbidity and possible causality for added risk of neurodevelopmental dis-orders91.

The findings in Study II demonstrate no increased risk for neurodevelop-mental disorders with increasing numbers of surgical treatments or surgery in infancy compared to later surgery at age 1-5 years. The difficulties investigat-ing the impact of early anaesthesia on neurodevelopment include finding a comparable cohort of patients with craniofacial malformations without surgi-cal treatment and preferably also a group of patients with anaesthesia at a young age but without face or brain surgery. In study III we were at least in part able to address the potential effect of early anaesthesia on brain develop-ment given that we made comparisons within a cohort of individuals with OFC. These comparisons gave us a unique opportunity to address psychiatric morbidity in individuals with similar anaesthesia exposures in the cohorts. The observed variations in psychiatric outcome indicate a link between pri-marily aberrant brain development and the abnormal facial development early in morphogenesis, with importance for the observed associations with neuro-developmental disorders in children with OFC.

The results from study II on CS need to be interpreted with caution as the information on type of CS or surgeries performed could not be obtained from the NPR. While the causes for these associations remain to be determined, our results are important for clinical follow-up.

The studies of this thesis indicate that it is beneficial to screen for early neurodevelopmental symptoms in children born with craniofacial malfor-mations, provide consistent long-term follow up of children with neurodevel-opmental disorders and identify those in need of special educational support at school.

40

Ethical considerations The studies were entirely register-based and the patients were not contacted during the study. Recognising possible long-term adverse outcomes for chil-dren with craniofacial malformations relative to psychiatric morbidity is a cru-cial step in primary and secondary prevention of these outcomes. This knowledge is necessary to plan the follow-up of these patients and address future research hypotheses. It is also important to advance our understanding of the neurodevelopmental pathophysiology of psychiatric disorders.

The studies were approved by the Uppsala Ethics Committee (Reg. No. 2012/363).

Methodological considerations These studies were register-based cohort studies that enabled us to study a uniquely large number of children born with craniofacial malformations over time. We were also able to explore possible different patterns in patients with various types of craniofacial malformations. Information gathered from regis-ters eliminates the risk of recall and selection bias. Another important strength is the availability of data used for our multivariate analyses to adjust for con-founders. Moreover, the sibling cohort in study I and II allowed us to match many unmeasured factors, including cultural background, parental character-istics, child-rearing practices and genetics92.

The diagnoses for identifying our study population and the psychiatric di-agnoses were obtained from the NPR. The validity of the psychiatric diagno-ses in the NPR is considered high as measured by moderate to high positive predictive values for psychiatric diagnoses93-95.

Still, there are some limitations to consider when interpreting the current results. It should be borne in mind that psychiatric symptoms (e.g., mild de-pressive and anxiety symptoms) that are not severe enough to require psychi-atric care are treated within the primary healthcare system in Sweden, and those diagnoses registered in primary care are not included in the NPR.

We have adjusted for several potential confounders, but there are always potential unknown or residual confounders that are not possible to adjust. So-cioeconomic status includes many hard-to-capture factors, but we used paren-tal country of birth, parental education level, parental age and parental psychi-atric disorders as a proxy for important sociodemographic factors. Relevant confounding variables were identified and visualised through directed acyclic graphs (DAGs), showing both adjusted confounders, unadjusted confounders and possible mediators based on available data and expert opinion96-98. The DAGs were performed using DAGitty.net v.2.0.(http://www.dagitty.net/his-tory/v1.1/dags.html)

http://www.dagitty.net/his-tory/v1.1/dags.html



% F;%

%='-"3*%;J@%!'3*A(*8%,A$A)'A%-3,/4%563%0("8$%V%,18%VVV%

%='-"3*%;K@%!'3*A(*8%,A$A)'A%-3,/4%563%0("8$%VV%

%%

%F9%

%='-"3*%;b@%!'3*A(*8%,A$A)'A%-3,/4%563%0("8$%VH%

7*%,)06%/*3563#*8%0'G)'1-%,1,)$0*0L%,1%'#/63(,1(%*/'8*#'6)6-'A,)%#*(468%(6%A61(36)%"1#*,0"3*8%A6156"18'1-% 5,A(630% 0"A4%,0%*12'361#*1(,)% 5,A(630%,18%-*1*0bJ@%7*%,1,)$f*8%5"))%0'G)'1-0%S'(4%)*00%(4,1%,%9:\$*,3%,-*%8'55*3*1A*@%=*S%0'G)'1-0%4,8%#63*%(4,1%,%;9\$*,3%,-*%8'55*3*1A*%,18%(4*%3,('61,)*%563%"0'1-%(4'0%),3-*%,-*%0/,1%S,0%(6%'1A)"8*%,)#60(%,))%5"))%0'G)'1-0@%&4*%1,('2'($%'1%TS*8*1%4,0%G**1%,G6"(%(S6%A4')83*1%8"3'1-%(4*%0("8$%/*3'68%,18%(4*%#*8',1%'1(*32,)%G*(S**1% 5'30(% ,18% 0*A618% A4')8% 4,0% 2,3'*8% 536#% E:\a:% #61(40% ,18% F:\K:%#61(40%G*(S**1%0*A618%,18%(4'38%A4')8L%S'(4%0463(*3%G'3(4%'1(*32,)0%'1%(4*%),(*3%$*,30%65%(4*%0("8$%/*3'68@%&4"0L%#60(%0'G)'1-0%'1%(4*%TS*8'04%/6/"),('61%S*3*%G631%S'(4'1%61*%8*A,8*k0%('#*bb@%%I1,)$0'1-%(4*%0'G)'1-0L%S*%/3'#,3')$%"0*8%G*(S**1\A)"0(*30%8,(,%,1,)$0*0%

,0%(4*%S'(4'1\A)"0(*3%,1,)$0*0%8*#,18%*16"-4%8,(,%(6%A,)A"),(*%,1%XQ%]'@*@%,(%)*,0(%61*%65%(4*%*U/60*8%,18%"1*U/60*8%'18'2'8",)0%'1%*,A4%A)"0(*3%1**80%(6%4,2*%(4*%0/*A'5'A%6"(A6#*%63%6G0*32*8%*2*1(%(6%G*%,G)*%(6%A,)A"),(*%,1%XQ^@%=63%,1,)$0*0%65%3,3*%6"(A6#*0L%(4*%S'(4'1\A)"0(*3%#*(468%S'))%4,2*%#,1$%#'00\'1-%'18'2'8",)0%,18%A)"0(*30@%&4*%G*(S**1\A)"0(*30%0'G)'1-%,1,)$0*0%3*0")(%'1%,%),3-*3%A6#/,3'061%G*(S**1%'18'2'8",)0%S'(4%A3,1'65,A',)%8'0638*30%,18%,))%0'G\)'1-0%S'(46"(% A3,1'65,A',)% 8'0638*30@% =63% 0("8$% VVL%S*% A6#/)*(*8% ,88'('61,)%S'(4'1\A)"0(*3%,1,)$0*0%563%0*10'('2'($%/"3/60*0%,18L%*UA*/(%563%#'00'1-%XQ0%'1%;:%65%;K%6G0*32*8%6"(A6#*0L%16%#,Z63%8'55*3*1A*0%'1%6"(A6#*%S*3*%3*2*,)*8%S4*1%A6#/,3*8%S'(4%(4*%3*0")(0%'1%(4*%G*(S**1\A)"0(*3%,1,)$0*0@%%7*%4,2*%/600'G)$%G**1%62*3\'1A)"0'2*%,18%,8Z"0(*8%16(%61)$%563%A6156"18\

*30%G"(%,)06%/600'G)*%#*8',(630%0"A4%,0%/*3'1,(,)%A6#/)'A,('610%])6S%I+cIQL%TcIL%)6S%G'3(4%S*'-4(%,18%/3*#,("3*%G'3(4^%,18%,))%A61-*1'(,)%#,)563#,('610L%

43

which may have led to over adjustment of the effect on our outcomes (i.e. a lower aHR). More children born in the earlier years of the study period may have had unknown syndromes later re-classified as syndromic: for example, for CS, the Muenke syndrome (FGFR3) was identified in 1996 and ERF-re-lated CS in 2013. We have adjusted for calendar year and all congenital mal-formations to address this potential issue. Some data of possible confounders in the registers are also not known to be valid, including self-reported amounts of alcohol consumption and smoking during pregnancy100,101. We could not measure data on smoking or alcohol consumption during pregnancy and be-cause the tobacco use data in the MBR are underreported, we decided not to use this information. Prenatal smoking, however, is associated with premature birth, SGA, low birth weight and low Apgar102, which we had solid data on from the MBR.

Surveillance bias could also be important in these studies. Children with craniofacial disorders are followed by craniofacial teams that may increase the likelihood to get referred to child and adolescent mental health services (CAMHS). Children who are not being brought into medical centres routinely might be more likely to go undetected. However, in clinical settings we often witness the opposite phenomenon, notably that if you are born with another severe medical condition or congenital malformation needing regular follow up, there is often a delay in referral to CAMHS for an assessment. People seem to explain behavioural and emotional difficulties from the view that the child’s medical condition could explain the behavioural or emotional difficulties. Moreover, the Swedish compulsory national screening programme with pub-lic health nurses screening for speech and language, motor, cognitive and so-cial development throughout the first six years of life and the school health nurses from primary school reduce the risk of not capturing children in the population with neurodevelopmental disorders.

Another limitation is the problem of multiple testing, given that we are an-alysing a set of statistical inferences simultaneously. In discussions with our statistician regarding compensating for multiple comparisons to reduce the false positive outcomes we decided to refrain as established methods (i.e. false discovery rate, Bonferroni-Holm method) decrease power and focus on indi-vidual tests. Thus, we decided to analyse the data without compensation for multiple testing in that we sought to investigate patterns and trends instead of each outcome in detail. Moreover, when interpreting the results, it is beneficial to look for patterns and trends in both crude and adjusted HRs and not focus on only specific p-values and confidence intervals.

Using data from large longitudinal registers over long periods with changes in the classification systems can be challenging when it comes to the validity of the specific diagnostic entity. We considered the definition of ADHD and ASD over the years problematic and excluded individuals that died before the introduction of ICD-9, as those diagnoses were not included in ICD-8, which had other classifications. Regarding the diagnosis of mental retardation and

44

ID, we included mental retardation from ICD-8 in the same definition as ID in the ICD-9 and 10.

Implications and future directions These studies hope to raise awareness in clinicians and stimulate researchers to investigate further the link between craniofacial disorders and psychiatric disorders. Congenital birth defects are an important window into new re-search. Accordingly, having better knowledge of how events during early em-bryogenesis could affect not only facial anatomical development but also neu-rodevelopment of the brain and the neurodevelopmental origin of different psychiatric disorders. Although these studies cannot explain the aetiology be-hind the disturbed craniofacial development and neurodevelopment, the sib-ling analyses have considered some of the familial factors that are otherwise hard to control for. Meanwhile, the neurodevelopmental impact on other gen-eral psychiatric disorders (e.g., depression, anxiety disorders, substance use disorders and eating disorders) seems to be absent in these studies. However, ID, speech and language disorders, ASD, ADHD, psychotic disorders and pos-sibly also BDs and personality disorders (the last two only in single sub-anal-yses) seem to be more affected by events of disturbed neurodevelopment.

The multifactorial aetiology and overlapping risk factors underlying both neurodevelopmental disorders and orofacial clefts would be highly relevant for future studies. In the future we have obtained ethical approval and are planning to link registry data to patient charts, so that the type of CS, type of surgery, anaesthesia and other important factors can be assessed. We also want to assess whether being born with NSCS affects subsequent academic out-comes later in life. Raising awareness of poorer educational achievement in children with OFC, even without concurrent psychiatric disorders, is crucial in the educational sector. Moreover, secondary prevention with additional spe-cial support in school needs to be addressed.

45

Conclusions

• In our studies, using craniofacial malformations as a group of individuals with disturbed development during early embryogenesis, we detected an association with several psychiatric disorders in OFC: ID, speech and lan-guage disorders, ASD, ADHD, psychotic disorders and possibly also BDs and personality disorders (the last two only in single sub-analyses), but these associations were more modest in NSCS. No such associations were seen for depression, anxiety disorders, eating disorders and substance use disorders, indicating a weak neurodevelopmental link to these psychiatric disorders.

• Individuals with OFC presented with risk increases for ID, language dis-orders, psychosis, ASD, ADHD and behavioural and emotional disorders in childhood. Individuals with CPO showed the largest risk increases and individuals with CL the least. Non-affected siblings presented with a lower risk for psychiatric disorders. Females presented with generally higher risks than males for psychiatric comorbidity.

• Individuals with NSCS had risk increases of any psychiatric disorder, in-cluding ID, language disorders and neurodevelopmental disorders. In the crude analyses full siblings with NSCS, compared to non-affected sib-lings, were more likely to be diagnosed with any psychiatric disorder, in-cluding ID, language disorders and neurodevelopmental disorders. The higher risk of any psychiatric disorder and ID remained after adjusting for confounders. Females presented with borderline higher risk increases than males for any psychiatric disorder, ID and a conglomerate of other neuro-developmental disorders, including ADHD and ASD.

• Individuals with bilateral clefts showed higher risks of psychiatric disor-ders compared to unilateral clefts. Individuals with BCL had an increased risk of any psychiatric disorder, including ID. Children with BCLP showed risk increases for any psychiatric disorder, including speech and language disorders, neurodevelopmental disorders and other psychiatric disorders. Females with BCLP, compared to their male coun-terparts, presented higher risks for ID and neurodevelopmental disorders.

• Children with nonsyndromic CLP and CPO showed lower performance in mathematics; children with CPO also had lower performance in Swedish as early as in the third school year in primary school. In their last year of compulsory school students with CLP or CPO had lower grades in math-ematics; students with CLP also had lower grades in Swedish; and

46

students with any clefts had lower average percentile grades in all subjects than non-cleft peers. At upper secondary school, students with CLP and CPO had lower grades in mathematics but not in Swedish. There were no detectable differences between males and females in national standardised tests in the third school year, but females’ academic outcomes were more negatively affected in the ninth school year and upper secondary school. After year 3, the lower academic achievements were only evident in OFC without a concurrent psychiatric disorder.

47

Sammanfattning på svenska

Bakgrund och frågeställning Läpp-, käk- och/eller gomspalt (LKG) är en av de vanligaste missbildning-arna. I Sverige föds 150–200 barn per år med någon form av LKG. Syftet med denna avhandling var att undersöka risken för neuropsykiatriska funktions-nedsättningar, psykiatriska sjukdomar och försämrade skolprestationer i lång-tidsuppföljningar av barn som föds med kraniofaciala missbildningar, såsom LKG och kraniosynostos (CS).

Dessa medfödda anläggningsdefekter kan anses som en modell för stör-ningar i hjärnutvecklingen eftersom fostrets hjärna, ansikte och kranium ut-vecklas mestadels från samma groddblad (ektodermet) i den tidiga embryo-nala utvecklingen. Enligt tidigare forskning anses störningar i hjärnutveckl-ingen under fosterstadiet vara en bakomliggande orsak till bl.a. autism, ADHD och intellektuell funktionsnedsättning, men även för vissa andra psykiatriska sjukdomar som psykoser.

Metod Avhandlingen bygger på fyra delarbeten, samtliga baserade på data från nat-ionella hälso- och sjukvårdsregister. Tre delarbeten studerar barn med LKG (Studie I, III och IV) och ett delarbete studerar barn med CS (Studie II). I studie I undersöktes förekomsten av neuropsykiatriska funktionsnedsätt-ningar, psykiatriska sjukdomar, självmordsförsök och självmord hos barn som fötts i Sverige med LKG mellan 1973 och 2012 utan känt genetiskt syndrom (cirka 70 procent av alla barn med LKG). De identifierades i patientregistret och kopplades sedan ihop med data från medicinska födelseregistret, flerge-nerationsregistret, dödsorsaksregistret, information om föräldrarnas födelse-land från registret över totalbefolkningen, information om emigrering från mi-grationsregistret och föräldrarnas utbildningsnivå från Longitudinella integ-rationsdatabasen för sjukförsäkrings och arbetsmarknadsstudier (LISA).

Barnen med LKG jämfördes med sina syskon samt för varje identifierat barn med LKG tog Statistiska centralbyrån (SCB) fram 10 barn utan LKG i den allmänna befolkningen som var födda i samma län, födelsemånad och kön. Totalt inkluderades 7842 barn med LKG från 1973 till 2012, 78409 barn utan spalt från den allmänna befolkningen samt 9637 syskon utan spalt.

48

I Studie II där barn med CS undersöktes med samma metod och register som vid första studien identifierades 1238 barn med CS, 12380 barn från den allmänna befolkningen och 1487 syskon utan CS. I studie III jämförde vi psy-kiatriskt utfall hos barn som fötts med dubbelsidiga spalter jämfört med ensi-diga spalter.

I Studie IV undersöktes skolprestationer hos barn som fötts med någon av LKG-subtyperna: läppspalt (CL), läpp-käk och gomspalt (CLP) samt enbart gomspalt (CPO). Vi använde oss av samma register som i tidigare studier med skolverkets register i tillägg och undersökte resultaten från nationella proven i matematik och svenska i åk 3, 5/6, betyg i åk 9, betyg i matematik och svenska på gymnasiet samt högskole- och universitetsexamen.

Resultat Första studien visade att barn med LKG hade ökad risk för intellektuell funkt-ionsnedsättning (ID), språkstörning, psykos, ASD, ADHD och andra beteen-destörningar i barndomen. Ingen förhöjd risk för depression, ätstörningar, missbruk, ångestsyndrom, suicid eller suicidförsök fanns, jämfört med barn utan LKG. Deras syskon utan LKG hade lägre risk för dessa utfall än barnen med LKG. Flickor hade generellt högre riskökningar än pojkar. Barn med LKG hade högre riskökningar än barn med bara läppspalter.

I tredje studien visades att barn med bilaterala spalter har högre risker för neuropsykiatriska funktionsnedsättningar än barn med unilaterala. Även i dessa analyser hade flickor högre riskökningar än pojkar.

I fjärde studien visades att barn med LKG har sämre skolprestationer, spe-ciellt i matematik, så tidigt som i tredje årskurs och med undantag av nation-ella proven i åk 5/6 fanns de försämrade skolprestationerna kvar genom hela skolgången. Bara individer med gomspalter hade lägre odds för att få univer-sitetsexamen.

Vi gjorde även subgruppsanalyser av barn med LKG utan psykiatrisk sam-sjuklighet och neuropsykiatriska funktionsnedsättningar och även i dessa ana-lyser fanns en tydlig association med försämrade skolprestationer.

Flickor uppvisade större risk än pojkar för försämrade skolprestationer från årskurs nio och på gymnasiet, men inte i de tidigare grundskoleåren.

Studie II visade att barn med icke-syndromal CS löper nästan fem gånger högre risk för ID och över fördubblad risk för språkstörning, jämfört med barn från den allmänna befolkningen och jämfört med sina syskon utan CS. Studien påvisade även en liten riskökning för ett konglomerat av andra neuropsykia-triska funktionsnedsättningar inklusive autism, ADHD och andra beteende-störningar samt en liten riskökning för övriga psykiatriska diagnoser när de neuropsykiatriska funktionsnedsättningarna exkluderades. Även här observe-rades könsskillnader där flickor med CS uppvisade gränsfall högre riskök-ningar än pojkar.

49

Slutsats Kunskapen om riskökningar för neuropsykiatriska funktionsnedsättningar och skolsvårigheter för personer med LKG och CS ger sjukvården och samhället en möjlighet till att tidigt identifiera stödbehov hos de individer som har be-hov. Dessa studier ger ytterligare stöd för att störningar under fostertiden kan leda till både neuropsykiatriska funktionsnedsättningar och kraniofaciala missbildningar såsom LKG och CS. Vetskapen om detta möjliggör prevention och tidig behandling för barnen och stimulerar till ytterligare forskning kring vilka riskfaktorer i den tidiga utvecklingen som kan påverka utvecklingen av neuropsykiatriska funktionsnedsättningar.

50

Acknowledgements

This work was done in close cooperation between the Department of Neuro-science, Child and Adolescent Psychiatry at Uppsala University and the De-partment of Surgical Sciences, Uppsala Craniofacial Center at Uppsala Uni-versity Hospital. Financial support was obtained from Uppsala University (ALF) and Lennanders foundation for medical research. I want to express my deepest gratitude to everyone involved in this thesis project and to everyone who has supported me. Special thanks to:

Fotios Papadopoulos, my main supervisor, for excellent tutoring and endless support during these years (especially with the register studies) and expertise in research integration and implementation. I’m particularly grateful for al-lowing me to be a part of this intriguing scientific project.

Lisa Ekselius, my co-supervisor, for always being positive, supportive and sharing her profound knowledge of psychiatry and research. Your hard work, wisdom and good judgement have created a stable and thriving research envi-ronment with high standards.

Mia Ramklint, my co-supervisor, for inspiring me as a person in clinical work, research and teaching. Your hard work in creating excellent medical education and a thriving teaching and research environment has set the highest standards.

Daniel Nowinski, my co-supervisor, for sharing your knowledge in craniofa-cial malformations, establishing the Uppsala Craniofacial Center, tutoring and excellent revising of the manuscripts, the superb support in research and in-viting me into your network of plastic surgery experts within the field of cra-niofacial surgery both in Sweden and internationally.

Jonas Höijer, for outstanding work with analysing data, statistical knowledge and explaining statistical problems through exceptional educational skills. I’m also thankful for the playdates with his wonderful children while he was help-ing me understand statistical methods.

Malin Hakelius, for expertise and clinical skills in plastic surgery and orofa-cial clefts. I’m immensely grateful for all our important discussions and your

51

input into our multidisciplinary research. I greatly appreciated the fact that you always remained positive and supportive.

A-C Fält, for all the support in practical matters regarding both research and teaching, your infinite enthusiasm, encouraging attitude and resilient approach to everyday problems.

Hans Arinell, for your invaluable support in technical matters in the depart-ment and always helping out in the best timely manner.

Eva-Lena Stattin, for expertise in genetics and invaluable genetic advice on study I.

Lena Brandt, for helping in the initial phase of study I with statistical analyses and introducing me to SAS programming and answering many of my initial statistical questions.

Janet Cunningham, for support and inspiring discussions, which profoundly contributed to the encouraging environment of PhD students in our depart-ment.

All colleagues at the Department of Neuroscience, for all your wisdom, sup-port, exciting seminars, discussions, “fika” and contributions of all kinds to enhance the research at our department and the best approaches and medical knowledge of the future.

All colleagues at the Child and Adolescent Mental Health at Uppsala Univer-sity Hospital, for encouraging and supporting me to start my research in 2013.

My fellow PhD students from the Department of Neuroscience at Uppsala University, for always being supportive and possessing the most amazing re-search skills and grit.

All colleagues at the Uppsala Craniofacial Center, for excellent multidiscipli-nary collaborations in seminars and discussions on children with craniofacial malformations.

All my colleagues at Cereb, for always supporting my clinical work and hav-ing a flexible approach that has enabled me to combine clinical work with research.

Jesper Unander Scharin, for help in creating beautiful 3D illustrations of CS.

52

All former colleagues at Te Whare o Te Rito, ICAMHS Gisborne Hospital in New Zealand, who supported me in enriching my cultural awareness and providing me invaluable knowledge and improved clinical skills as a consult-ant child and adolescent psychiatrist and researcher.

Special thanks to all the clinicians, Statistics Sweden and the National Board of Health and Welfare for the incredibly impressive work with our national registers. The registers contain invaluable data contributing to medical re-search, knowledge and improved evidence-based treatments for the future.

Lina, Lisa, Tina and Lotta, for always staying close, outstanding support and life-long friendship, even during periods of long-distance relationships or so-cial distancing.

All my friends and family, you know who you are and how important you are to me for always enriching my life and for your sincere interest in my life and work.

My mother and father, for the love, care and support, and for always believing in me.

And above all,

My wonderful husband, Tobias, and our beloved children, Elias, Filip and Juni, for your love and enduring support and to Cesar, for long walks in the forest.

53

References

1. Collander Farzaneh F. Cleft lip and palate : clinical studies regarding speech and facial growth. Malmö University Hospital, Lund University. 2009.

2. Paaske EB, Garne E. Epidemiology of orofacial clefts in a Danish county over 35 years - Before and after implementation of a prenatal screening programme for congenital anomalies. Eur J Med Genet. 2018.

3. Registry Centre South. The Swedish National Quality Registry for Cleft Lip and Palate. https://kvalitetsregister.se/englishpages/findaregistry/registerarkivenglish/nationalqualityregistryforcleftlipandpalateclp.2155.html/ Accessed 22.10.2020.

4. Klintö K, Karsten, A., Marcusson, A., Paganini, A., Rizell, S., Cajander, J., Brunnegård, K., Hakelius, M., Okhiria, Å., Peterson, P., Abdiu, A., Havstam, C., Mark, H., Hagberg, E., Björnström, L., Wiedel, A. P., & Becker, M. . Coverage, reporting degree and design of the Swedish quality registry for patients born with cleft lip and/or palate. . BMC health services research. 2020;20(1), 528.

5. Persson M, Becker M, Svensson H. Academic achievement in individuals with cleft: a population-based register study. Cleft Palate Craniofac J. 2012;49(2):153-159.

6. Persson M, Becker M, Conrad AL, Svensson H. Female and Male Differences in Academic Achievement in Individuals With Cleft: A Population-Based Register Study. Cleft Palate Craniofac J. 2018;55(2):196-203.

7. Richman LC, McCoy TE, Conrad AL, Nopoulos PC. Neuropsychological, behavioral, and academic sequelae of cleft: early developmental, school age, and adolescent/young adult outcomes. Cleft Palate Craniofac J. 2012;49(4):387-396.

8. Christensen K, Mortensen PB. Facial clefting and psychiatric diseases: a follow-up of the Danish 1936-1987 Facial Cleft cohort. Cleft Palate Craniofac J. 2002;39(4):392-396.

9. Pedersen DA, Wehby GL, Murray JC, Christensen K. Psychiatric Diagnoses in Individuals with Nonsyndromic Oral Clefts: A Danish Population-Based Cohort Study. PLoS One. 2016;11(5):e0156261.

10. Nilsson S, Merlo J, Lyberg-Ahlander V, Psouni E. Psychotropic drug use in adolescents born with an orofacial cleft: a population-based study. BMJ Open. 2015;5(4):e005306.

11. Christensen K, Juel K, Herskind AM, Murray JC. Long term follow up study of survival associated with cleft lip and palate at birth. BMJ. 2004;328(7453):1405.

12. Nopoulos P, Langbehn DR, Canady J, Magnotta V, Richman L. Abnormal brain structure in children with isolated clefts of the lip or palate. Arch Pediatr Adolesc Med. 2007;161(8):753-758.

https://kvalitetsregister.se/englishpages/findaregistry/registerarkivenglish/nati

54

13. Nopoulos P, Berg S, VanDemark D, Richman L, Canady J, Andreasen NC. Increased incidence of a midline brain anomaly in patients with nonsyndromic clefts of the lip and/or palate. J Neuroimaging. 2001;11(4):418-424.

14. Di Rocco F, Arnaud E, Renier D. Evolution in the frequency of nonsyndromic craniosynostosis. J Neurosurg Pediatr. 2009;4(1):21-25.

15. Nowinski DL, A.; Nilsson, P.; Enblad, P.;. Kraniosynostoser. https://www.internetmedicin.se/behandlingsoversikter/neurokirurgi/kraniosynostoser/ Published 2020. Accessed 24.09.2020.

16. Boulet SL, Rasmussen SA, Honein MA. A population-based study of craniosynostosis in metropolitan Atlanta, 1989-2003. Am J Med Genet A. 2008;146A(8):984-991.

17. Anderson IA, Goomany A, Bonthron DT, et al. Does patient ethnicity affect site of craniosynostosis? J Neurosurg Pediatr. 2014;14(6):682-687.

18. Selber J, Reid RR, Chike-Obi CJ, et al. The changing epidemiologic spectrum of single-suture synostoses. Plast Reconstr Surg. 2008;122(2):527-533.

19. van der Meulen J, van der Hulst R, van Adrichem L, et al. The increase of metopic synostosis: a pan-European observation. J Craniofac Surg. 2009;20(2):283-286.

20. Noden DM, Trainor PA. Relations and interactions between cranial mesoderm and neural crest populations. J Anat. 2005;207(5):575-601.

21. Slater BJ, Lenton, K. A. , Kwan, M. D. , Gupta, D. M. , Wan, D. C. & Longaker, M. T. . Cranial Sutures: A Brief Review. Plastic and Reconstructive Surgery. 2008;121(4), 170e-178e.

22. Morriss-Kay GM, Wilkie AO. Growth of the normal skull vault and its alteration in craniosynostosis: insights from human genetics and experimental studies. J Anat. 2005;207(5):637-653.

23. Cohen MM, Jr., MacLean RE. Craniosynostosis: Diagnosis, Evaluation, and Managment. 2nd ed. New York: Oxford University Press; 2 edition; 2000.

24. Speltz ML, Kapp-Simon KA, Cunningham M, Marsh J, Dawson G. Single-suture craniosynostosis: a review of neurobehavioral research and theory. J Pediatr Psychol. 2004;29(8):651-668.

25. Kapp-Simon KA, Speltz ML, Cunningham ML, Patel PK, Tomita T. Neurodevelopment of children with single suture craniosynostosis: a review. Childs Nerv Syst. 2007;23(3):269-281.

26. Fernbach SK, Feinstein KA. Radiologic evaluation of the child with craniosynostosis. Neurosurg Clin N Am. 1991;2(3):569-585.

27. Gewalli F, Guimaraes-Ferreira JP, Sahlin P, et al. Mental development after modified pi procedure: dynamic cranioplasty for sagittal synostosis. Ann Plast Surg. 2001;46(4):415-420.

28. Speltz ML, Kapp-Simon K, Collett B, et al. Neurodevelopment of infants with single-suture craniosynostosis: presurgery comparisons with case-matched controls. Plast Reconstr Surg. 2007;119(6):1874-1881.

29. McCarthy JG, Warren SM, Bernstein J, et al. Parameters of care for craniosynostosis. Cleft Palate Craniofac J. 2012;49 Suppl:1S-24S.

30. Mathijssen IM. Guideline for Care of Patients With the Diagnoses of Craniosynostosis: Working Group on Craniosynostosis. J Craniofac Surg. 2015;26(6):1735-1807.

31. Schmidt MJ, Mirnics K. Neurodevelopment, GABA system dysfunction, and schizophrenia. Neuropsychopharmacology. 2015;40(1):190-206.

32. Willsey AJ, Morris MT, Wang S, et al. The Psychiatric Cell Map Initiative: A Convergent Systems Biological Approach to Illuminating Key Molecular Pathways in Neuropsychiatric Disorders. Cell. 2018;174(3):505-520.

https://www.internetmedicin.se/behandlingsoversikter/neurokirurgi/kraniosyn

55

33. Sarrazin S, Cachia A, Hozer F, et al. Neurodevelopmental subtypes of bipolar disorder are related to cortical folding patterns: An international multicenter study. Bipolar Disord. 2018.

34. Owen MJ, O'Donovan MC. Schizophrenia and the neurodevelopmental continuum:evidence from genomics. World Psychiatry. 2017;16(3):227-235.

35. Rice D, Barone S, Jr. Critical periods of vulnerability for the developing nervous system: evidence from humans and animal models. Environmental health perspectives. 2000;108 Suppl 3(Suppl 3):511-533.

36. Doherty JL, Owen MJ. Genomic insights into the overlap between psychiatric disorders: implications for research and clinical practice. Genome Med. 2014;6(4):29.

37. Owen MJ. New approaches to psychiatric diagnostic classification. Neuron. 2014;84(3):564-571.

38. Gillberg C. [ESSENCE gathers the diagnoses into a whole]. Lakartidningen. 2014;111(39):1643-1646.

39. Gillberg C. The ESSENCE in child psychiatry: Early Symptomatic Syndromes Eliciting Neurodevelopmental Clinical Examinations. Res Dev Disabil. 2010;31(6):1543-1551.

40. Beriaghi S, Myers SL, Jensen SA, Kaimal S, Chan CM, Schaefer GB. Cleft lip and palate: association with other congenital malformations. J Clin Pediatr Dent. 2009;33(3):207-210.

41. Kallen B, Harris J, Robert E. The epidemiology of orofacial clefts. 2. Associated malformations. J Craniofac Genet Dev Biol. 1996;16(4):242-248.

42. Goodacre T, Swan M. Cleft lip and palate: current management. Paediatrics and Child Health. 2008;18(6):283-292.

43. Burdi AR. Cleft Lip and Palate. New York: Springer; 2006. 44. Chung KC, Kowalski CP, Kim HM, Buchman SR. Maternal cigarette smoking

during pregnancy and the risk of having a child with cleft lip/palate. Plast Reconstr Surg. 2000;105(2):485-491.

45. Little J, Cardy A, Arslan MT, Gilmour M, Mossey PA, United Kingdom-based case-control s. Smoking and orofacial clefts: a United Kingdom-based case-control study. Cleft Palate Craniofac J. 2004;41(4):381-386.

46. Hackshaw A, Rodeck C, Boniface S. Maternal smoking in pregnancy and birth defects: a systematic review based on 173 687 malformed cases and 11.7 million controls. Hum Reprod Update. 2011;17(5):589-604.

47. Yazdy MM, Honein MA, Rasmussen SA, Frias JL. Priorities for future public health research in orofacial clefts. Cleft Palate Craniofac J. 2007;44(4):351-357.

48. Chung MK, Lao TT, Ting YH, Hung Suen SS, Leung TY, Kin Lau T. Is there Seasonality in the Incidence of Oral-Facial Clefts? J Matern Fetal Neonatal Med. 2011.

49. Heuze Y, Holmes G, Peter I, Richtsmeier JT, Jabs EW. Closing the Gap: Genetic and Genomic Continuum from Syndromic to Nonsyndromic Craniosynostoses. Curr Genet Med Rep. 2014;2(3):135-145.

50. Delahaye S, Bernard JP, Renier D, Ville Y. Prenatal ultrasound diagnosis of fetal craniosynostosis. Ultrasound Obstet Gynecol. 2003;21(4):347-353.

51. Cornelissen MJ, Apon I, van der Meulen J, et al. Prenatal ultrasound parameters in single-suture craniosynostosis. J Matern Fetal Neonatal Med. 2018;31(15):2050-2057.

52. Lajeunie E, Crimmins DW, Arnaud E, Renier D. Genetic considerations in nonsyndromic midline craniosynostoses: a study of twins and their families. J Neurosurg. 2005;103(4 Suppl):353-356.

56

53. Twigg SR, Wilkie AO. A Genetic-Pathophysiological Framework for Craniosynostosis. Am J Hum Genet. 2015;97(3):359-377.

54. Ardalan M, Rafati A, Nejat F, Farazmand B, Majed M, El Khashab M. Risk factors associated with craniosynostosis: a case control study. Pediatr Neurosurg. 2012;48(3):152-156.

55. Carmichael SL, Ma C, Rasmussen SA, et al. Craniosynostosis and maternal smoking. Birth Defects Res A Clin Mol Teratol. 2008;82(2):78-85.

56. Singh RP, Dhariwal D, Bhujel N, et al. Role of parental risk factors in the aetiology of isolated nonsyndromic metopic craniosynostosis. Br J Oral Maxillofac Surg. 2010;48(6):438-442.

57. Rasmussen SA, Yazdy MM, Carmichael SL, Jamieson DJ, Canfield MA, Honein MA. Maternal thyroid disease as a risk factor for craniosynostosis. Obstet Gynecol. 2007;110(2 Pt 1):369-377.

58. Kini U, Hurst JA, Byren JC, Wall SA, Johnson D, Wilkie AO. Etiological heterogeneity and clinical characteristics of metopic synostosis: Evidence from a tertiary craniofacial unit. Am J Med Genet A. 2010;152A(6):1383-1389.

59. Lajeunie E, Barcik U, Thorne JA, El Ghouzzi V, Bourgeois M, Renier D. Craniosynostosis and fetal exposure to sodium valproate. J Neurosurg. 2001;95(5):778-782.

60. Huang J, Zhu T, Qu Y, Mu D. Prenatal, Perinatal and Neonatal Risk Factors for Intellectual Disability: A Systemic Review and Meta-Analysis. PLoS One. 2016;11(4):e0153655.

61. Corvin AP. Neuronal cell adhesion genes: Key players in risk for schizophrenia, bipolar disorder and other neurodevelopmental brain disorders? Cell Adh Migr. 2010;4(4):511-514.

62. Owen MJ, O'Donovan MC, Thapar A, Craddock N. Neurodevelopmental hypothesis of schizophrenia. Br J Psychiatry. 2011;198(3):173-175.

63. Khanzada NS, Butler MG, Manzardo AM. GeneAnalytics Pathway Analysis and Genetic Overlap among Autism Spectrum Disorder, Bipolar Disorder and Schizophrenia. Int J Mol Sci. 2017;18(3).

64. Hultman CM, Sparen P, Takei N, Murray RM, Cnattingius S. Prenatal and perinatal risk factors for schizophrenia, affective psychosis, and reactive psychosis of early onset: case-control study. BMJ. 1999;318(7181):421-426.

65. Salib E, Cortina-Borja M. Effect of month of birth on the risk of suicide. Br J Psychiatry. 2006;188:416-422.

66. Twilhaar ES, de Kieviet JF, Aarnoudse-Moens CS, van Elburg RM, Oosterlaan J. Academic performance of children born preterm: a meta-analysis and meta-regression. Arch Dis Child Fetal Neonatal Ed. 2018;103(4):F322-F330.

67. Perez-Roche T, Altemir I, Gimenez G, et al. Effect of prematurity and low birth weight in visual abilities and school performance. Res Dev Disabil. 2016;59:451-457.

68. Stuart A, Otterblad Olausson P, Kallen K. Apgar scores at 5 minutes after birth in relation to school performance at 16 years of age. Obstet Gynecol. 2011;118(2 Pt 1):201-208.

69. Morsanyi K, van Bers B, McCormack T, McGourty J. The prevalence of specific learning disorder in mathematics and comorbidity with other developmental disorders in primary school-age children. Br J Psychol. 2018;109(4):917-940.

70. Keen D, Webster A, Ridley G. How well are children with autism spectrum disorder doing academically at school? An overview of the literature. Autism. 2016;20(3):276-294.

57

71. Jangmo A, Stalhandske A, Chang Z, et al. Attention-Deficit/Hyperactivity Disorder, School Performance, and Effect of Medication. J Am Acad Child Adolesc Psychiatry. 2019;58(4):423-432.

72. Perez-Vigil A, Fernandez de la Cruz L, Brander G, et al. Association of Obsessive-Compulsive Disorder With Objective Indicators of Educational Attainment: A Nationwide Register-Based Sibling Control Study. JAMA Psychiatry. 2018;75(1):47-55.

73. Comes AL, Senner F, Budde M, et al. The genetic relationship between educational attainment and cognitive performance in major psychiatric disorders. Transl Psychiatry. 2019;9(1):210.

74. Assari S, Caldwell CH, Bazargan M. Association Between Parental Educational Attainment and Youth Outcomes and Role of Race/Ethnicity. JAMA Netw Open. 2019;2(11):e1916018.

75. Weinberg D, Stevens G, Finkenauer C, Brunekreef B, Smit HA, Wijga AH. The pathways from parental and neighbourhood socioeconomic status to adolescent educational attainment: An examination of the role of cognitive ability, teacher assessment, and educational expectations. PLoS One. 2019;14(5):e0216803.

76. Berg L, Back K, Vinnerljung B, Hjern A. Parental alcohol-related disorders and school performance in 16-year-olds-a Swedish national cohort study. Addiction. 2016;111(10):1795-1803.

77. Mezzich JE. International surveys on the use of ICD-10 and related diagnostic systems. Psychopathology. 2002;35(2-3):72-75.

78. First MB, Rebello TJ, Keeley JW, et al. Do mental health professionals use diagnostic classifications the way we think they do? A global survey. World Psychiatry. 2018;17(2):187-195.

79. Tillman KK, Hakelius M, Hoijer J, et al. Increased Risk for Neurodevelopmental Disorders in Children With Orofacial Clefts. J Am Acad Child Adolesc Psychiatry. 2018;57(11):876-883.

80. Tillman KK, Höijer J, Ramklint M, Ekselius L, Nowinski D, Papadopoulos FC. Nonsyndromic Craniosynostosis Is Associated with Increased Risk for Psychiatric Disorders. Plast Reconstr Surg. 2020;146(2):355-365.

81. StataCorp.2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC. 2017.

82. Boyadjiev SA, International Craniosynostosis C. Genetic analysis of nonsyndromic craniosynostosis. Orthod Craniofac Res. 2007;10(3):129-137.

83. Lajeunie E, Le Merrer M, Bonaiti-Pellie C, Marchac D, Renier D. Genetic study of nonsyndromic coronal craniosynostosis. Am J Med Genet. 1995;55(4):500-504.

84. Lai MC, Lombardo MV, Auyeung B, Chakrabarti B, Baron-Cohen S. Sex/gender differences and autism: setting the scene for future research. J Am Acad Child Adolesc Psychiatry. 2015;54(1):11-24.

85. Polyak A, Rosenfeld JA, Girirajan S. An assessment of sex bias in neurodevelopmental disorders. Genome Med. 2015;7:94.

86. Jevtovic-Todorovic V, Absalom AR, Blomgren K, et al. Anaesthetic neurotoxicity and neuroplasticity: an expert group report and statement based on the BJA Salzburg Seminar. Br J Anaesth. 2013;111(2):143-151.

87. Davidson AJ. Anesthesia and neurotoxicity to the developing brain: the clinical relevance. Paediatr Anaesth. 2011;21(7):716-721.

88. Stratmann G, Lee J, Sall JW, et al. Effect of general anesthesia in infancy on long-term recognition memory in humans and rats. Neuropsychopharmacology. 2014;39(10):2275-2287.

58

89. Ing C, DiMaggio C, Whitehouse A, et al. Long-term differences in language and cognitive function after childhood exposure to anesthesia. Pediatrics. 2012;130(3):e476-485.

90. Ing CH, DiMaggio CJ, Malacova E, et al. Comparative analysis of outcome measures used in examining neurodevelopmental effects of early childhood anesthesia exposure. Anesthesiology. 2014;120(6):1319-1332.

91. Lei SY, Hache M, Loepke AW. Clinical research into anesthetic neurotoxicity: does anesthesia cause neurological abnormalities in humans? J Neurosurg Anesthesiol. 2014;26(4):349-357.

92. Frisell T, Oberg S, Kuja-Halkola R, Sjolander A. Sibling comparison designs: bias from non-shared confounders and measurement error. Epidemiology. 2012;23(5):713-720.

93. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450.

94. Sellgren C, Landen M, Lichtenstein P, Hultman CM, Langstrom N. Validity of bipolar disorder hospital discharge diagnoses: file review and multiple register linkage in Sweden. Acta Psychiatr Scand. 2011;124(6):447-453.

95. Ruck C, Larsson KJ, Lind K, et al. Validity and reliability of chronic tic disorder and obsessive-compulsive disorder diagnoses in the Swedish National Patient Register. BMJ Open. 2015;5(6):e007520.

96. Textor J, Hardt J, Knuppel S. DAGitty: a graphical tool for analysing causal diagrams. Epidemiology. 2011;22(5):745.

97. Sjolander A, Zetterqvist J. Confounders, Mediators, or Colliders: What Types of Shared Covariates Does a Sibling Comparison Design Control For? Epidemiology. 2017;28(4):540-547.

98. Ferguson KD, McCann M, Katikireddi SV, et al. Evidence synthesis for constructing directed acyclic graphs (ESC-DAGs): a novel and systematic method for building directed acyclic graphs. Int J Epidemiol. 2020;49(1):322-329.

99. Statistics Sweden. Birth intervals- how long do parents wait before they have a second child? 2017. ISSN 1654-1510 (online)

100. Shipton D, Tappin DM, Vadiveloo T, Crossley JA, Aitken DA, Chalmers J. Reliability of self reported smoking status by pregnant women for estimating smoking prevalence: a retrospective, cross sectional study. BMJ. 2009;339:b4347.

101. Brooke HL, Talbäck M, Hörnblad J, et al. The Swedish cause of death register. Eur J Epidemiol. 2017;32(9):765-773. doi:10.1007/s10654-017-0316-1

102. Fifer WP, Fingers ST, Youngman M, Gomez-Gribben E, Myers MM. Effects of alcohol and smoking during pregnancy on infant autonomic control. Dev Psychobiol. 2009;51(3):234-242.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1688

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-421456

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2020

Craniofacial malformations and psychiatric disorders from ...

Documents

Transcript of Craniofacial malformations and psychiatric disorders from ...