Eugenia Conti
Wiring the autistic brain: structural connectivity from early
developmental trajectories to predictive biomarkers
PhD in Basic and Developmental Neuroscience
Thesis, 2015
PhD Course in Basic and Developmental Neuroscience
Wiring the autistic brain:
structural connectivity from early developmental
trajectories to predictive biomarkers
Eugenia Conti, MD.
Supervisor: Andrea Guzzetta, MD, PhD
Ringraziamenti
Chapter 1 General Introduction and Outline of the Thesis
Part 1 The role of emerging aetiological factors associated to ASD: the example of assisted reproductive technologies.
Chapter 2 Are children born after assisted reproductive technology at increased risk of autism spectrum disorders? A systematic review. Hum Reprod. 2013; 28(12):3316-27.
Part 2 Structural connectivity in neurodevelopmental disabilities
Chapter 3 Reorganization of visual fields after periventricular haemorrhagic infarction: potentials and limitations.
Dev Med Child Neurol. 2013;55 Suppl 4:23-6.
Chapter 4 Wiring the preterm brain: contribution of new meta-analytic approaches.
Dev Med Child Neurol. 2015; 57(4):307-8.
Part 3 Structural connectivity in ASD
Chapter 5 The first 1000 days of the autistic brain: a systematic review of diffusion imaging studies.
Front Hum Neurosci. 2015; 9:159.
Chapter 6 High Angular Resolution Diffusion Imaging in a child with Autism Spectrum Disorder and comparison with his unaffected identical twin.
Functional Neurology (under revision)
Chapter 7 Lateralization of brain networks and clinical severity in toddlers with Autism Spectrum Disorder: a diffusion MRI study.
Autism Res (under revision)
Chapter 8 Network over-connectivity differentiates autism spectrum from other developmental disorders in toddlers: a diffusion MRI study. In preparation Chapter 9 Conclusions p.9 p.25 p.27 p.49 p.51 p.63 p.69 p.71 p.89 p.103 p.123 p.145
Fin dagli ultimi mesi della scuola superiore, quando il mondo della medicina e delle neuroscienze iniziava a diventare la mia “scelta”, l’Autismo e le Neuro-immagini hanno rappresentato per me l’oggetto di maggior fascino e interesse.
Ricordo quindi chiaramente come il mio pensiero si rivolse istantaneamente a questo, nel giugno del 2011, quando il Prof. Cioni mi propose di elaborare un progetto di ricerca da sottomettere al concorso per il Dottorato in Neuroscienze di Base e dello Sviluppo.
…e in un attimo mi ritrovo qui, con la stessa passione di allora, con qualche conoscenza in più nonché qualche dubbio in più, a scrivere i ringraziamenti della tesi, frutto dello studio della connettività cerebrale nei disturbi dello spettro autistico...
Chi conosce la mia storia sa che non è stato semplice intraprendere questo percorso, iniziato quasi come una sfida, come un percorso “alternativo” alla scuola di specializzazione, ma che si è rivelato nel suo sviluppo un’opportunità meravigliosa ed unica di crescita professionale e personale.
E chi conosce la mia storia sa anche quante energie sto mettendo nel chiudere questo capitolo a distanza, trovandomi, per casualità della vita, geograficamente lontana, ma mentalmente legata a una realtà che è riuscita ad insegnarmi davvero molto sulla metodologia della ricerca e sul neurosviluppo.
Questo Dottorato mi ha permesso, in primis, di vivere la meravigliosa esperienza della “Zerodue” a cui non posso che dedicare tutto questo lavoro. Senza i nostri piccoli pazienti, e le loro famiglie e senza le figure professionali che ho incontrato e con cui sono entrata in profondo legame, tutto questo non sarebbe stato possibile.
Grazie ad Adi, Elena, Simo, Viviana, Vittorio, per le idee brillanti e lo spiccato spirito critico; grazie ad Adria per la sua grande esperienza e competenza e grazie a Franceschina, Gessica, Marti, Sarina, Tiziana, poiché ognuno è riuscito a trasmettermi un pezzetto della propria professionalità. Un ringraziamento anche a Roberta B., per gli insegnamenti preziosi e la fiducia dimostrata nei miei confronti, grazie alla quale più volte ho toccato con mano il senso di responsabilità che si vive nel lavorare in questo ambito.
Un ringraziamento particolare va a Sara M., per avermi insegnato, con grande umanità e professionalità, ad “osservare” la prima infanzia da tutte le prospettive possibili, e per essere stata un’amicizia di riferimento in questi anni.
Grazie a Vale e Tere, per la stimolante condivisione di conquiste e delusioni nei diversi passaggi d’anno e grazie a Manu, per la pazienza e la costanza, in un confronto quotidiano e autentico sulla vita lavorativa e non solo.
Un ringraziamento va al Prof. Muratori, per avermi introdotto nel mondo dell’Autismo e per avermi sostenuto nel mio percorso successivo, e a Sara C., per la sua sincera e incoraggiante presenza come collega e amica, nonchè per il suo prezioso contributo nella stesura dei lavori. Un ringraziamento speciale va al Prof. Cioni, per avermi dato la possibilità di percorrere questa strada nelle migliori delle condizioni possibili e avermi motivato, gratificato e sostenuto sempre nel corso di questi anni.
Grazie al laboratorio di Risonanza Magnetica, e al supporto tecnico di Anna e Danilo, senza la cui mediazione sarebbe stato molto difficile entrare a pieno nel mondo tecnico della connettività cerebrale. Grazie anche ai collaboratori australiani Jhimli, Kerstin, Stephen e Ros per la serietà e professionalità, nonché per lo spiccato spirito di ospitalità. Il tempo trascorso
Un doveroso grazie ai miei genitori e alla Nina, per trasmettermi costantemente che “perseguire i sogni” dà un senso di compiutezza alla vita, e un grazie a Case Dipinte e dintorni, nei suoi significati più disparati.
Un ringraziamento unico va, infine, ad Andrea, poiché grazie alla sua professionalità, determinazione, instancabile curiosità e generosità, mi è stato possibile vivere in tutto e per tutto questa esperienza, trovando alternativamente lo spazio per “offrire” quando ne avevo le capacità, o “chiedere” quando avevo bisogno di aiuto.
Che questa conclusione possa essere un buon punto di partenza per poter restituire e ampliare quanto imparato.
Chapter 1
General introduction and outline of the thesis
Autism Spectrum Disorder: definition and epidemiology
Autism Spectrum Disorder (ASD) is a complex condition emerging in the first years of life characterized by core impairment in social interaction and social communication abilities, and the presence of repetitive, restrictive and stereotyped patterns of interests, activities and behaviours [1]. The disorder was firstly described by Leo Kanner in 1943, and subsequently received numerous changes in the diagnostic criteria through the different editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM), until the current inclusion within the wide group of neurodevelopmental disorders in DSM 5 [1]. Over the last several decades the dramatic rise in the diagnosis of ASD has been documented extensively around the world, reaching prevalence rates as high as 1/68 in the USA [2], thus indicating Autism as a highly impacting problem in our society.
Numerous factors have been identified as potential contributors to the interpretation of increasing prevalence, including the use of inappropriate referral statistics, changes in diagnostic approaches, such as the shift from the general diagnosis of developmental delays or language delays to autism [3,4], the diagnostic substitution from mental retardation to autism, or even greater ASD awareness [5]. Despite a general recognition that broader diagnostic criteria have increased prevalence, changes in diagnostic criteria and younger age at diagnosis were felt to explain only a fraction of the rising, and a great part of literature is focussed on environmental factors as possible explanation for the rising prevalence of autism [6]. Indeed, environmental factors have been widley implicated in the gene/environment interactions etiology theory of autism where also epigenetic factors play an important role [7]. Environmental factors/events can occur during early life and be involved in the regulation of neural development associated with synaptic plasticity and development of brain connectivity [8].
Autism Spectrum Disorder as a dysconnectivity syndrome
While the complex etiopathogenic mechanisms contributing to ASD remain elusive, the past decade has seen convergence across advanced neuroimaging research modalities pointing to the importance of altered brain connectivity in this disorder [9].
In 2004 Just, Cherkassky, Keller, and Minshew [10] firstly provided an influential formulation of what they dubbed ‘‘under-connectivity theory’’, arguing that
information at the neural and cognitive levels’’. In recent years, under-connectivity theory has attracted considerable attention as a number of studies have reported lower than expected between-region functional correlations on a range of tasks, and Diffusion imaging approaches have revealed markers of disordered structural connectivity in individuals with ASD [11]. Evidence of coesistence of overconnectivity and underconnectivity within the course of the disorder have led to a more recent definition of ASD as a Dysconnectivity Syndrome [12], interpreting these findings in a developmental perspective, thus reconciling the apparently divergent findings of under and over-connectivity [13]. As a matter of fact, emerging evidence of structural connectivity in young ASD subjects, and from longitudinal studies assessing ASD-siblings [14, 15], suggest very early atypicalities in the ASD brain, mainly in terms of over-connectivity, in different brain regions.
This observation is compatible with previously reported neuroanatomical early correlates of ASD [16-18] and with the early signs of enlarged head circumference and macrocephaly. Converging evidence from head-circumference, neuroimaging and post-mortem studies supports the view that the brains of ASD individuals are larger than controls from at least 2 years of age, particularly at the level of the frontal and temporal lobes. This finding tends to disappear within the preschool age and eventually results in significant undergrowth during adolescence [19] thus supporting the proposed interpretation of early overgrowth as an epiphenomenon of abnormal connectivity resulting from a lack of physiological synaptic pruning [20].
Advanced brain imaging to explore structural connectivity
In the last two decades, Diffusion Tensor Imaging MRI techniques have been extensively applied in neuroimaging studies of Autism Spectrum Disorders, in order to explore the structural connectivity of the disorder. Indeed, DTI [21] is a non-invasive, sensitive method to map and characterize the microstructural properties and macroscopic organization of WM tissues in the brain [22-24]. The most commonly investigated DTI measure is fractional anisotropy (FA), whose values range from 0 to 1 with lower FA in more isotropic tissues (e.g. gray matter and cerebral spinal fluid) and higher FA in regions of white matter (WM). FA is highly sensitive to microstructural changes or differences in WM including myelination and axonal density, and therefore it is often called a measure of WM integrity. However, FA does not provide a complete description
to better interpret the underlying changes in tissue microstructure [25]. Diffusion MRI also allows the delineation of white matter pathways by tractography technique, which was firstly introduced using the deterministic streamline algorithm and has subsequently evolved to use more sophisticated probabilistic approaches [26]. DTI is inadequate for describing white matter microstructure in regions with crossing WM fibers [27, 28]. New advanced techniques are now available to overcome this issue, such as for example the High Angular Resolution Diffusion Imaging and probabilistic tractography approaches [29].
Early detection: looking for a biomarker
The clinical and etiologic heterogeneity of children with ASD contribute to the complex challenges associated with developing a comprehensive early detection strategy. Indeed, early diagnosis is of utmost importance as it creates opportunities for children with ASD to benefit from early intervention as far to suggest the possibility of preventing the full manifestations of ASD by taking advantage of early brain plasticity [30]. Gains through early intervention can enhance adaptive and cognitive functioning [31] and may ultimately reduce the considerable family and societal costs related to ASD across the lifespan [32]. Earlier diagnosis also allows parents to be better informed about recurrence risk to later-born children, and better able to monitor for early signs of autism [33]. Although a wide consensus exists from both retrospective and prospective studies about the emergence of ASD symptoms in the first two years of life, the age at which clinical diagnosis is still generally made ranges between 3 and 4 years [34]. The rapid development of cognitive, social, and communication skills in children during their first years of life, does reflect on dynamic changes in symptoms of ASDs [35, 36] and requires deep knowledge of the earliest and most predictive markers of ASDs [37].
That’s the reason why great part of the literature is focussed on detecting an early sensitive and specific biological marker of the disorder which can add predictive power to early clinical behavioural evaluations of ASD in the wider field of neurodevelopmental disabilities. Reliable sets of autism biomarkers would be immensely useful in clinical practice, as they could: (a) provide risk estimates at birth in ‘baby siblings’ of children already diagnosed with autism, in order to design and pursue preventive health care strategies; (b) foster earlier and more reliable diagnoses, especially between the ages of 12 and 30 months; (c) predict spontaneous developmental trajectories; (d) predict treatment
psychoactive drugs [38].
Purposes of the Thesis
Increasing evidence suggests that the abnormalities of neural development in Autism Spectrum Disorders (ASD) already emerge during the first years of life, preceding the appearance of the first clinical signs. Detection of brain abnormalities at this time of development is therefore of utmost importance as it can give insights onto the basic neural defects of autism, help clinicians to improve prediction and allow a better planning for early intervention.
The application of new non-invasive techniques exploring brain connectivity non-invasively, such as Diffusion Imaging and tractography, has allowed the identification of reliable biomarkers of ASD. Still little is known however on the early developmental trajectories of these biomarkers and on their actual diagnostic and prognostic value.
The main aim of this PhD project has been to study structural connectivity applying advanced diffusion MRI protocol (HARDI) to those toddlers referred to the Infant Neurology Section of the Stella Maris Scientific Institute, who received a first clinical diagnosis of neurodevolpmental disorder, including ASD, in order to identify specific early neuro-biomarkers of ASD.
The present PhD thesis is composed of 3 parts. Part 1 consists of a literature review about the role of emerging aetiological epigenetic factors associated to ASD (i.e. assisted reproductive technologies); Part 2 consists of two review papers about structural connectivity in neurodevelopmental disabilities other than ASD; Part 3 focuses on structural connectivity in ASD, particularly including one review paper on early developmental trajectories of connectivity, and three original research papers exploring structural connectivity in children and toddlers with ASD in comparison with other neurodevelopmental disorders.
Outline of the thesis
Chapter 1 General Introduction and Outline of the Thesis
Chapter 2 Are children born after assisted reproductive technology at
development, given that during these periods epigenetic reprogramming occurs and the new epigenetic profile of the offspring is established. We then systematically queried existing literature about the association of Assisted Reproductive Technologies (ART) and ASD. We included studies in which i) subjects with either infantile autism or ASD could be identified according to international classification systems and ii) the diagnosis was obtained from hospital records. We obtained seven observational studies published since 2000 (2 cohort and 5 case-control) encompassing 9216 subjects diagnosed with ASD. Four out of seven studies, including the two with the best quality scores, did not show an association between Assisted Reproductive Technologies and ASD. The two papers supporting an increased risk of autism following ART had the lowest quality scores, due to major methodological limitations. Only one paper showed a protective role of ART. We finally conclude that there is no clear association between ASD and ART, although the large heterogeneity of available studies in terms of sample size, study design and cohort characterization, suggests that more studies are needed to better answer the research question.
Chapter 3 Reorganization of visual fields after periventricular haemorrhagic
infarction: potentials and limitations.
Considering structural connectivity and plasticity in the wide field of neurodevelopmental disorders, a systematic review focussing on visual function impairment and plastic reorganization in preterm infants with periventricular haemorrhagic infarction was performed. We in particular analysed those papers (four studies, 19 individuals) which reported an unexpected sparing of the visual fields related to the plasticity of thalamo-cortical afferents that are supposedly able to bypass the lesion occurring in the early third trimester of gestation. The involvement of the optic radiations was often associated with normal visual fields as only one of the four individuals with damaged optic radiations showed visual field defects. Overall, this review supports the existence of effective mechanisms of plastic reorganization that allow a rewiring of geniculo-calcarine connections with restoration of full field vision but which are hindered by the involvement of the basal ganglia and thalamus.
Chapter 4 Wiring the preterm brain: contribution of new meta-analytic approaches.
outcomes even in absence of obvious damage to the developing brain.
Recent advances in neuroimaging, such as Diffusion MRI techniques, have increased our capabilities to detect and quantitatively assess those abnormalities of the preterm brain affecting the white matter.
A recent study has systematically reviewed the existing literature on diffusion MRI in preterm infants, still emphasizing the need for further studies involving infants of all gestational ages to elucidate the relationship between gestational age and diffusion metrics, and to establish the utility of tractography as a predictive tool. An attempt to move forward in this direction has been undertaken by Li et al. 2014, who explored fractional anisotropy changes in preterm individuals by reviewing and quantitatively appraising current knowledge on brain diffusion in individuals born preterm. They used a coordinate-based meta-analytic method, the activation likelihood estimate (ALE), summarizing all the studies that provided the coordinates of peak activations. They found that changes in fractional anisotropy, and particularly in the corpus callosum, are correlated to the functional difficulties of preterm individuals. Indeed these results are of great relevance and support the use of more quantitative meta-analytic approaches to objectively summarize information from heterogeneous studies.
Chapter 5 The first thousands days of the autistic brain: a systematic review
of diffusion imaging studies.
In an attemp to better understand early developmental trajectories of structural connectivity in ASD, we performed a systematic review of studies exploring brain connectivity in ASD toddlers. We searched PubMed and Medline databases for all English language papers, published from year 2000, exploring structural connectivity in populations of toddlers whose mean age was below 30 months. Of the 264 papers extracted, four were found to be eligible and were reviewed. Three of the four selected studies reported higher fractional anisotropy values in subjects with ASD compared to controls within commissural fibers, projections fibers and association fibers, suggesting brain hyper-connectivity in the earliest phases of the disorder. Similar conclusions emerged from the other diffusion parameters assessed. These findings are reversed to what is generally found in studies exploring older patient groups and suggest a developmental course characterized by a shift towards hypo-connectivity starting at a time between two and four years of age.
Spectrum Disorder and comparison with his unaffected identical twin.
We used the model of disease-discordant identical twins to test the hypothesis that higher-order DWI protocols are able to detect abnormal connectivity at a single subject- level, assessing the brain structural connectivity of a child with ASD and his unaffected identical twin (5 years of age) using high-angular-resolution diffusion imaging (HARDI) probabilistic tractography. Cortical regions were automatically parcellated from high-resolution structural images, and HARDI based connection matrices were produced for statistical comparison. Differences in diffusion indexes between subjects were tested by Wilcoxon signed rank test. Tracts were defined as discordant when showing a between-subject difference of 10 percent or more. Around eleven percent of the discordant intra-hemispheric tracts were ASD underconnected, while only 1% were ASD overconnected. This difference was statistically significant. Our findings in a disease-discordant identical twin couple are consistent with the underconnectivity theory of ASD.
Chapter 7 Lateralization of brain networks and clinical severity in toddlers with Autism Spectrum Disorder: a diffusion MRI study.
Recent diffusion tensor imaging studies in adolescents and children with Autism Spectrum Disorder (ASD) have reported a loss or an inversion of the typical left–right lateralization in fronto-temporal regions crucial for socio-communicative skills. No studies explored atypical lateralization in toddlers and its correlation with clinical severity of Autism Spectrum Disorder.
We recruited a cohort of 20 subjects aged 36 months or less receiving a first clinical diagnosis of ASD (15 males; age range 20-36 months). Patients performed diffusion MRI (HARDI protocol). Data from cortical parcellation were combined with tractography to obtain a connection matrix and diffusion indexes (DI) including mean fractional anisotropy (DFA), number of tracts (DNUM) and total
tract length (DTTL) were obtained. A laterality index was generated for each
measure, and then correlated with the ADOS-G total score. Laterality indexes of DFA were significantly correlated with ADOS-G total scores only in two
intra-frontal connected areas (correlation was positive in 1 case and negative in the other). Laterality indexes of DTTL and DNUM showed significant negative
correlations (p<0.05) in six connected areas, mainly fronto-temporal. This study provides first evidence of a significant correlation between brain lateralization and clinical severity in toddlers with a first diagnosis of ASD. Significant
one were negative, suggesting an inversion of the typical left–right asymmetry in subjects with most severe clinical impairment.
Chapter 8 Network over-connectivity differentiates autism spectrum from other
developmental disorders in toddlers: a diffusion MRI study.
The majority of studies assessing brain structural connectivity compare Autism Spectrum Disorder (ASD) and typically developing subjects, thus providing little information on the specificity of the abnormalities detected as to other neurodevelopmental disorders (DD). To our knowledge, only one recent study explored brain structural differences between toddlers presenting with ASD and DD, through a voxel-based approach, reporting alterations within the temporal lobes, the corpus callosum, the posterior cingulate cortex, and limbic lobes. We performed a whole-brain tractography study in toddlers (39 ASD subjects; 20 DD subjects; mean age 28 months) following High angular resolution diffusion imaging (HARDI) diffusion protocol. Scans were acquired using 31 diffusion weighted images. Image post-processing consisted of cortical parcellation of the T1 images into 90 regions. Whole brain probabilistic tractography was done and network connectivity matrices (90x90) were built encoding the number of streamlines and fractional anisitropy connecting each pair of cortical regions. Network Based Statistics (NBS) was finally applied on the connectivity matrices to evaluate the network differences between the ASD and DD groups. The network differences revealed over-connectivity in the ASD group. The identified networks are mainly centred on superior temporal gyrus and caudate, regions known to be crucial for development of social skills and executive functions. Further more, our findings extend previous findings from voxel-based analysis showing increased volumes in comparable regions.
Association., Diagnostic and
statistical manual of mental disorders (5th ed.). 2013,
Arlington, VA: American Psychiatric Publishing. 2. CDC, Centers for Disease
Control and Prevention. Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2010. . MMWR
Surveill Summ 2012, 2014. 63: p. 1-21.
3. Jick, H. and J.A. Kaye,
Epidemiology and possible causes of autism.
Pharmacotherapy, 2003. 23(12): p. 1524-30.
4. Bishop, D.V., et al., Autism
and diagnostic substitution: evidence from a study of adults with a history of developmental language disorder. Dev Med Child
Neurol, 2008. 50(5): p. 341-5. 5. Posserud, M., et al., The
prevalence of autism
spectrum disorders: impact of diagnostic instrument and non-response bias. Soc
Psychiatry Psychiatr Epidemiol, 2010. 45(3): p. 319-27.
Tipping the balance of autism risk: potential mechanisms linking pesticides and autism.
Environ Health Perspect, 2012. 120(7): p. 944-51. 7. Tordjman, S., et al., Gene x
Environment interactions in autism spectrum disorders: role of epigenetic
mechanisms. Front
Psychiatry, 2014. 5: p. 53. 8. Bagot, R.C. and M.J.
Meaney, Epigenetics and the
biological basis of gene x environment interactions. J
Am Acad Child Adolesc Psychiatry, 2010. 49(8): p. 752-71.
9. Ameis, S.H. and P. Szatmari,
Imaging-genetics in autism spectrum disorder: advances, translational impact, and future directions. Front
Psychiatry, 2012. 3: p. 46. 10. Just, M.A., et al., Cortical
activation and synchronization during sentence comprehension in high-functioning autism: evidence of underconnectivity. Brain, 2004. 127(Pt 8): p. 1811-21. 11. Vissers, M.E., M.X. Cohen,
and H.M. Geurts, Brain
connectivity and high functioning autism: a
methodological convergence, and stronger behavioral links.
Neurosci Biobehav Rev, 2012. 36(1): p. 604-25. 12. Hoppenbrouwers, M., M.
Vandermosten, and B. Boets,
Autism as a disconnection syndrome: A qualitative and quantitative review of
diffusion tensor imaging studies. Research in Autism
Spectrum Disorders, 2014. 8(4): p. 387-412.
13. Uddin, L.Q., K. Supekar, and V. Menon, Reconceptualizing
functional brain connectivity in autism from a
developmental perspective.
Front Hum Neurosci, 2013. 7: p. 458.
14. Wolff, J.J., et al., Differences
in white matter fiber tract development present from 6 to 24 months in infants with autism. Am J Psychiatry,
2012. 169(6): p. 589-600. 15. Elison, J.T., et al., White
matter microstructure and atypical visual orienting in 7-month-olds at risk for autism.
Am J Psychiatry, 2013. 170(8): p. 899-908. 16. Courchesne, E. and K.
Pierce, Brain overgrowth in
autism during a critical time in development: implications for
connectivity. Int J Dev
Neurosci, 2005. 23(2-3): p. 153-70.
17. Shen, M.D., et al., Early brain
enlargement and elevated extra-axial fluid in infants who develop autism spectrum disorder. Brain, 2013. 136(Pt
9): p. 2825-35.
18. Xiao, Z., et al., Autism
Spectrum Disorder as Early Neurodevelopmental
Disorder: Evidence from the Brain Imaging Abnormalities in 2-3 Years Old Toddlers. J
Autism Dev Disord, 2014. 19. Courchesne, E., K. Campbell,
and S. Solso, Brain growth
across the life span in autism: age-specific changes in anatomical pathology. Brain
Res, 2011. 1380: p. 138-45. 20. Frith, C., What do imaging
studies tell us about the neural basis of autism?
Novartis Found Symp, 2003. 251: p. 149-66; discussion 166-76, 281-97.
21. Basser, P.J., J. Mattiello, and D. LeBihan, MR diffusion
tensor spectroscopy and imaging. Biophys J, 1994.
66(1): p. 259-67.
22. Catani, M. and M. Thiebaut de Schotten, A diffusion
44(8): p. 1105-32. 23. Jones, D.K., et al.,
Non-invasive assessment of axonal fiber connectivity in the human brain via diffusion tensor MRI. Magn Reson
Med, 1999. 42(1): p. 37-41. 24. Mori, S., et al., Imaging
cortical association tracts in the human brain using
diffusion-tensor-based axonal tracking. Magn Reson Med,
2002. 47(2): p. 215-23. 25. Alexander, A.L., et al., A
geometric analysis of diffusion tensor
measurements of the human brain. Magn Reson Med,
2000. 44(2): p. 283-91. 26. Yamada, K., Diffusion tensor
tractography should be used with caution. Proc Natl Acad
Sci U S A, 2009. 106(7): p. E14; author reply E15. 27. Alexander, A.L., et al.,
Analysis of partial volume effects in diffusion-tensor MRI. Magn Reson Med,
2001. 45(5): p. 770-80.
28. Wedeen, V.J., et al., Diffusion
spectrum magnetic resonance imaging (DSI) tractography of crossing fibers. Neuroimage, 2008.
41(4): p. 1267-77.
imaging and beyond. Magn
Reson Med, 2011. 65(6): p. 1532-56.
30. Dawson, G., Early behavioral
intervention, brain plasticity, and the prevention of autism spectrum disorder. Dev
Psychopathol, 2008. 20(3): p. 775-803.
31. Dawson, G., et al.,
Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. Pediatrics,
2010. 125(1): p. e17-23. 32. Jacobson, J.W. and J.A. Mulick, System and cost
research issues in treatments for people with autistic
disorders. J Autism Dev
Disord, 2000. 30(6): p. 585-93.
33. Zwaigenbaum, L., et al.,
Clinical assessment and management of toddlers with suspected autism spectrum disorder: insights from studies of high-risk infants.
Pediatrics, 2009. 123(5): p. 1383-91.
34. Zwaigenbaum, L., S. Bryson, and N. Garon, Early
identification of autism spectrum disorders. Behav
Brain Res, 2013. 251: p. 133-46.
atypical visual scanning and recognition of faces in 2 and 4-year-old children with autism spectrum disorder. J
Autism Dev Disord, 2009. 39(12): p. 1663-72.
36. Turner, L.M., et al., Follow-up
of children with autism
spectrum disorders from age 2 to age 9. Autism, 2006.
10(3): p. 243-65. 37. Barbaro, J. and C.
Dissanayake, Autism
the evidence on early signs, early identification tools, and early diagnosis. J Dev Behav
Pediatr, 2009. 30(5): p. 447-59.
38. Ruggeri, B., et al.,
Biomarkers in autism spectrum disorder: the old and the new.
Psychopharmacology (Berl), 2014. 231(6): p. 1201-16.
Part 1
The role of emerging aetiological factors associated to ASD:
the example of assisted reproductive technologies.
Chapter 2 Are children born after assisted reproductive technology at increased risk of autism spectrum disorders? A systematic review.
Chapter 2
Are children born after assisted reproductive technology at
increased risk of autism spectrum disorders?
A systematic review.
Eugenia Conti Sara Mazzotti Sara Calderoni Irene Saviozzi Andrea GuzzettaAbstract
Study question: Are children born after assisted reproductive technology
(ART) at increased risk for autism spectrum disorders (ASD)?
Summary answer: There is no evidence that ART significantly increases the risk of ASD in the offspring.
What is known already: A few systematic reviews have explored the
correlation between assisted conception and ASD with inconclusive results, partly due to the heterogeneity of diagnostic criteria and methodology in the different studies.
Study design, size, duration: Systematic review of seven observational
studies (2 cohort and 5 case-control) encompassing 9216 subjects diagnosed with ASD published since 2000.
Materials, setting, methods: Literature searches were conducted to retrieve
observational studies on the risk of ASD in ART population. Databases search included PubMed, EMBASE, and PsycINFO. In order to obtain more consistent results, we only included the studies in which i) subjects with either infantile autism or ASD could be identified according to international classification systems and ii) the diagnosis was obtained from hospital records. Seven studies matched the inclusion criteria.
Main results and the role of chance: Four out of seven studies, including the
two with the best quality scores, did not show an association between ART and ASD. The two papers supporting an increased rink of autism following ART had the lowest quality scores, due to major methodological limitations. Only one paper showed a protective role of ART.
Limitations, reasons for caution: In spite of the strict inclusion criteria applied
as to the diagnosis of ASD, the papers selected are heterogeneous in many aspects including study design, definitions of ART, data source and analysed confounders.
Wider implications of the findings: At present, there is no evidence that ART
is significantly associated with ASD and hence that current health policies should be modified. The divergent results of some of the studies suggest that further prospective, large and high quality studies are still needed.
Introduction
Autism Spectrum Disorders (ASD) are neurodevelopmental conditions characterized by impaired social interaction and communication, together with restricted and repetitive behaviour [1], due to abnormal brain development beginning early in life [2, 3]. Although the exact cause of ASD remains unknown, a strong genetic origin has been implicated [4]. Nevertheless, recent findings support a crucial role of environmental factors, which are essential in modulating the phenotypical expression of the disorder [5, 6]. Estimated prevalence of ASD has dramatically increased in the last decades, reaching values of one in 88 children in U.S. [7]. Broadening of diagnostic criteria as well as increased recognition due to higher awareness of symptomatology can be accounted for this prevalence increase. However, environmental factors are also suspected of contributing to this rise [8]. The trend in ASD prevalence parallels recent changes in pregnancy and birth factors, such as multiple pregnancies, high maternal age, low birth weight and preterm birth, suggesting a possible link between them [9]. However, a recent study that investigated through a rigorous mathematical model the role of several prenatal factors (prematurity, birth weight, multiple births, caesarean delivery, breech presentation and use of assisted reproductive technology) in the increase of ASD prevalence, revealed a minimal contribution of these factors in ASD increase [10].
Among the environmental factors that have been suggested as potentially associated with ASD, some authors reported the use of assisted reproduction technologies (ART), including ovulation induction (OI), In vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI). This relationship could be related to at least three shared factors: high parental age, hormonal disturbances, high maternal educational level or social class. However, this last association has been questioned. In fact, although several recent investigations reported an association of ASD diagnoses with higher socioeconomic status (SES) [11-13], the majority of studies concluded that results could be biased by differential access to services, as well as differential awareness by parents and providers. Also, a higher risk of ASD in children born after assisted conception might be related to the higher rates of multiplicity, preterm birth and low birth weight deliveries [14-17].
A few review papers have explored the correlation between assisted conception and ASD with inconclusive results, partly due to the heterogeneity of diagnostic criteria in the different studies [18-20]. In the most recent metanalysis specifically focussed on ART, the eight studies selected, as pointed out by the
authors, had very broad and heterogeneous diagnostic inclusion criteria (three used infantile autism, three used ASD and two used neurodevelopmental disorder), and differed in the source of the diagnostic information (questionnaires filled out by parents, hospital records, national registers)[18] [18].
The aim of this systematic review was to summarize the available published data regarding the association between ART and infantile autism or ASD (i.e. autism, Asperger’s disorder or pervasive developmental disorder not otherwise specified). Cohort and case-control studies were included.
Methods
Literature search
Studies were identified by searching multiple literature databases, including Pubmed, EMBASE and PsycINFO. The searches were limited to papers in English and included articles published between January 2000 and February 2013. References were exported into an endnote bibliographic management database and duplicates were removed. The following search strategy was performed: (‘intrauterine insemination’ OR ‘assisted conception’ OR ‘in vitro fertilization’ OR IVF OR ‘intracytoplasmic sperm injection’ OR ICSI OR ‘assisted reproductive technology’ OR ART OR ‘ovulation induction’ OR OI) AND (‘infantile autism”’ OR ‘autism spectrum disorder*’ OR ASD OR ‘asperger’ OR ‘pervasive development* disorder’ OR ‘neurological outcome’ OR ‘neurodevelopmental disorder*’).
Study selection
Criteria for inclusion in the study were established prior to the literature search. Inclusion was limited to studies that were either case-control or cohort studies. Reviews were not included in the analysis, but were used to collect original studies. We only selected the studies in which subjects were diagnosed on hospital records with either infantile autism or ASD, according to ICD or DSM classifications.
Validity assessment
Using the Newcastle-Ottawa Scale (NOS) for assessing quality of non-randomized studies in meta-analysis we assessed the quality and publication bias of included studies [21]. The NOS was developed using a Delphi process and was subsequently tested on systematic reviews. Different NOS scales exist for cohort and case-control studies. The NOS contains eight items, categorized into three dimensions including selection, comparability, and outcome (cohort
studies) or exposure (case-control studies). For each item a series of response options is provided. Differences were resolved by consensus. Publication bias was assessed at the outcome level by visual inspection of funnel plots.
Figure 1. PRISMA four-phase flow diagram of search yield, screening and
inclusion steps.
Results
Description of studies
PRISMA Flow Diagram [22, 23] of the review process is presented in Figure 1. The search strategy yielded 115 records. Additional records identified through other sources (journal index and handmade bibliographies of selected review found) yielded 12 more records. When duplicates had been removed the number was 110. All abstracts were independently reviewed by four of the authors (EC, IS, SC and SM) and conflicting judgements were solved by consensus. Eighty papers were excluded during this review based on clear
failure to meet the inclusion criteria. Thirty full-text papers were evaluated further and only seven were found to meet the inclusion criteria (Table 1). Of the 23 eliminated papers, six were excluded because the diagnosis of autism was not performed according to standardized criteria (ICD-10 or DSM IV-TR), or was included into broader categories of behavioural disorders [24-29], eight were excluded as they were review papers [10, 17-20, 30-32], two as they were commentaries [33, 34] and one conference proceedings [35]; six papers were excluded because autism was not included in the outcome measures [36-41]. Two independent blinded reviewers (EC and SM) assessed the quality of the 7 selected studies utilizing the NOS scoring system both for case control and cohort studies (Table 2). Conflicting judgements were solved by consensus. The average quality of the selected papers was moderately high (mean total score 5, range 3-8).
Cohort studies
The quality assessment of the two cohort studies selected is shown in table 2a. Both studies were population-based and analysed data extracted from the Danish National Birth Register, which contains information on all births in Denmark. The two studies were conducted in different time periods and overall covered a population born between 1995 and 2003. Although the use of a nationwide register minimizes the risk of a selection bias, generalizability of the findings is limited by the common source used by the two studies which makes them uniform in terms of demographic factors, socioeconomic status and ethnicity (mainly Caucasian).
ART definition and ascertainment
In both studies, data could be extracted according to the definition of assisted conception as In Vitro Fertilization (IVF) with or without intra-cytoplasmic sperm injection (ICSI). Ascertainment of exposure in these studies can be considered secure and inclusive, as children exposed to IVF were identified through the compulsory IVF Register holding data from all private and public fertility clinics in Denmark.
ASD definition and ascertainment
In both studies, children with a diagnosis of ASD were identified via the Danish Psychiatric Central Register, containing information on all Danish psychiatric inpatient and outpatient admissions since 1995 [42]. Diagnosis was based on ICD9 or ICD10 classification systems and all children with ASD were included
(codes F84.0, F84.1, F84.5, F84.8 and F84.9). Both cohort studies made no specific reference to the possible presence of subjects lost to follow-up, while Pinborg et al. [43] also used a non fully adequate follow-up, starting from 2 years of age, when the diagnosis of ASD is not yet stable.
Case-control studies
The quality assessment of the five case-control studies selected is shown in table 2b. Although all of the five studies can be technically viewed as case-control studies, in two of them case series have been compared to a more general population group [44, 45] , making them closer to case series than true case-control studies. Two of the five papers originated from Scandinavian countries, while the other three explored cohorts from Israel [44, 46] and from Japan [45]. This makes the findings potentially more generalizable in terms of geographic distribution than those from the two cohort studies.
ART definition and ascertainment
Definition of ART was heterogeneous in the different studies. Maimburg and Vaeth [47] included in their analysis all infants conceived both by technical treatment and hormonal therapy, while Zachor et al [44] excluded fertility drugs to induce ovulation. Stein et al. [46] used broad inclusion criteria encompassing all cases of “infertility requiring medical intervention ever”. Three of the five case-control studies used suboptimal means for the ascertainment of ART exposure, namely parental questionnaires or interviews not blinded to case/control status[45] [46]. In the remaining two studies, more reliable methods were used. In Lehti et al. [48] information on ART was retrieved by a national register, the Finnish Medical Birth Register, a nationwide register collecting data on fertilization treatments since October 1990. Mainburg et al. [47] extracted information from the medical birth records collected from the Danish maternity wards, in which information about fertility status was provided. Moreover, the ascertainment of exposure to ART was not homogeneous between cases and controls in three of the five studies. Zachor et al.[44] and Shimada et al. [45] did not apply the same method of ascertainment for the control group as they used the general population statistics as control parameter, thus making inapplicable the comparison of the non-respondents between the groups. Stein et al. [46]used the same ascertainment procedure but failed to control for non-response rate between cases and controls.
ASD definition and ascertainment
Although the inclusion criteria of our search were strict as to the definition of ASD, in two of the five papers case ascertainment could not be considered as the result of independent validation, as cases were identified through record linkage registers (the Danish Psychiatric Central Register in Mainburg et al.[47] and the Finnish Hospital Discharge Register, in Lehti et al.[48]). In all papers the diagnosis was based on clinical evaluation and international coding (ICD or DSM), however only in one paper diagnosis was supported and confirmed by standardized diagnostic tools such as ADI-R[49] and ADOS-G[50].
Findings
Of the seven studies included in this systematic review, only two reported evidence of an increased risk of ASD in children born after assisted conception [44, 45]. Interestingly, these are the two studies with the lowest quality scores at the NOS, mainly due to the fact that ASD subjects were compared to general population statistics thus making the ascertainment of both exposure and outcome different between cases and controls (strong selection bias). In contrast, one paper reported a protective effect of assisted conception on infantile autism after adjusting for several confounding factors [47]. The remaining four papers however did not confirm this result, as they found no evidence of a different risk of ASD in children born after assisted reproductive technology. The two cohort studies, which analysed partially overlapping nationwide populations, either found no difference [43] or found an increased risk of ASD after assisted conception, which did not hold true after adjusting for main confounders (i.e. maternal age, educational level, parity, smoking, birth weight and multiplicity) [51]. Stein et al. [46] used a loose definition of assisted reproduction including all cases of infertility requiring medical intervention, and found a higher risk of developing ASD, which was however not statistically significant. Finally, Lehti et al.[48], the most recent and the highest-ranked paper of this review, found no increased risk of ASD in children born after ART.
Discussion
This systematic review included studies assessing the risk of ASD in children born after assisted conception and revealed no evidence of a significant association between ASD and ART. Only one study showed a protective role of assisted conception, which did not change after adjusting for potential confounders (i.e. maternal age, multiplicity, parity and prematurity/birth weight). To explain their findings, the authors focused on the potential advantages related to pregnancies following assisted conception, including the closer contact with the health system or better promotion of good health behaviour before and during early pregnancy, such as for example folic acid intake. No other studies however found supporting evidence since their first observation. None of the other studies of this review in which the analysis was adjusted for potential confounders a significant association between ART and ASD was found, and the only two studies suggesting an increased risk of ASD in children born after assisted conception had major methodological limitations.
The quality of the studies, as assessed by NOS, ranged between very low (score of 2) and very good (score of 8). The two papers with the highest quality scores [48, 51] support the absence of any significant association between exposure and outcome. It is of interest that the only two papers showing a significant association between ART and ASD [44, 45] obtained the lowest quality scores (score=2) as their design was suboptimal in all domains explored (i.e. selection, comparability and exposure). A good quality score was obtained by the only paper favouring a protective role of ART [47] however, it has to be noted that it had one of the lowest sample sizes and number of events among the reviewed studies.
In spite of the strict inclusion criteria applied in relation to the diagnosis of ASD, the papers selected are heterogeneous in many aspects including definition of ART, analysed confounders and diagnostic procedure.
While there is no consensus on the definition of assisted reproduction technology, this is generally considered to include all fertility treatments in which both eggs and sperm are handled, thus excluding procedures limited to medical treatment to the woman [7]. Half of the papers analysed in this review have not followed this definition. Stein et al. [46] have used broad inclusion criteria encompassing all cases of “infertility requiring medical intervention ever” as determined by structured questionnaires for retrospective obstetric information. Maimburg and Vaeth [47] gathered information about fertility from birth records including in their analysis all infants conceived both by technical treatment and hormonal therapy. Similarly, Hvidtjorn et al. [51] includes both IVF, that can be
broadly assimilated to current definition of ART, and OI, with or without subsequent insemination. Although we were unable to extract separate data (IVF vs OI) for our analysis from this latter study, it is of interest that the authors calculated separate risk ratios founding no association between ASD and IVF. In addition to the variability in the definition of ART, papers did not use the same criteria for the ascertainment of exposure. In most cases structured interviews were used, while in some others the information was gathered from medical registers. Although both approaches are considered reliable methods of ascertainment [52, 53], getting full scores within the NOS quality scale, it has to be underlined that some authors have questioned the reliability of birth records, which can potentially result in underestimation and biased ascertainment [54]. Another aspect potentially contributing to the heterogeneity among studies is the year of birth, which ranged between 1970 and 2006, thus potentially reflecting the application of different techniques of assisted conception. It has to be noted however that the interpretation of the findings does not change even when limiting the analysis to those papers that only include subjects born after 1995.
All selected papers, with the exception of Stein et al. [46] take into account possible confounders. Hvidtjorn et al. [51], Maimburg and Vaeth [47] and Lehti et al. [48] consider covariates such as maternal age, gestational age, birth weight and multiplicity, using adjusted values for well-known risk factors for autism and assisted conception in analysis regarding measures of effect. Some authors dealt with the potential bias of prematurity and low birth weight, which are common in both ART and ASD, by considering multiplicity within the selection criteria, either excluding multiple births from the analysis [44] or just focussing on twin pregnancies [43]. In two of the seven papers, singletons and twins were analysed separately but no differences were detected, with the exception of a significant association between singletons and Asperger syndrome [48].
Shimada et al. [45] reported significantly higher paternal and maternal age in relation to ASD, however they did not put this finding in relation to ART. Overall, the heterogeneity of the studies in relation to the role of the different confounders, and in particular prematurity and multiplicity, does not allow for a clear evaluation of their contribution to the definition of ART-related risk in ASD. Although we limited our selection to studies in which ASD was identified according to international classification systems, some differences in the diagnostic procedure were present among the studies. Zachor and Ben Itzchak [44] were the only authors to systematically apply gold standard tools for ASD
diagnosis, such as the Autism Diagnostic Interview-Revised (ADI-R [49]), the Autism Diagnostic Observation Schedule-Generic (ADOS-G [50]) and ADOS Severity Scale [55] assessed by trained and reliability-tested experts. Lehti et al. [48] used clinical diagnoses based on the ICD, however their general validity was reassessed in a validation study using the ADI-R, demonstrating that 96% of the cases with registry diagnoses of childhood autism fulfil the ADI-R diagnostic criteria [56]. Notwithstanding the ADOS-G and the ADI-R are two internationally recognized and widely used diagnostic instruments for ASD, it has to be noted that their validity is questionable for disorders other than infantile autism. For example, the ADOS-G has had lower specificity and sometimes sensitivity for distinctions involving children with Pervasive developmental disorder not otherwise specified and Asperger’s Disorder [57]. Moreover, the ADOS-G tends to under-diagnose children with higher verbal and nonverbal skills [58]. In the remaining papers, less information is provided as to the diagnostic procedure; the ASD diagnosis is made according to different classification systems, such as DSM-IV-TR [45] and ICD 8, ICD 10, DSM III or DSM III-R in the less recent papers [43, 46, 51]. It is of note, that, at variance with the other analysed studies, Maimburg and Vaeth [47] only included patients who satisfied strict criteria of infantile autism according to ICD 8 and ICD 10, namely patients belonging to the most severe end of the spectrum and thus not reflecting the large heterogeneity of ASD. Conversely, Stein et al. [46], only included cases with idiopathic autism, in which all children with a diagnosis of secondary autism related to genetic or metabolic causes were excluded. Generally speaking, the use of different inclusion criteria for ASD might hinder the possibility to reliably compare results of different studies. It is of interest however that the only paper that reported a clear-cut protective effect of ART, is also the only one limiting the inclusion criteria to infantile autism [47].
In summary, our systematic review suggests that ART does not represent a risk factor for ASD, however, the divergent results of some of the studies suggest that further prospective, large and high quality studies are still needed. The main methodological limitation that has lead to this outcome is the heterogeneity of the selected studies, particularly in terms of study design (cohort vs case control), ART data recruitment strategy (registers, medical records, parental interviews), ASD clinical assessment (standardised tests vs clinical evaluation) and assessment of confounders. It is of interest however that the papers with the highest value in this review based on their methodological quality reached similar conclusions, supporting the absence of significant associations between ART and ASD.
In future studies it will be interesting to determine the risk of association between ART and ASD separately for each subtype of intervention, and to explore if children with ASD born with assisted reproduction show a different clinical phenotype if compared to ASD children born without ART, in terms of gender, severity of ASD symptoms, IQ level and psychiatric comorbidity. In particular, the identification of environmental exposure that represents a possible risk factor for the development of ASD could potentially be important for the surveillance of vulnerable subjects and, ultimately, to accelerate the diagnostic process. In wider terms, further research on larger, well-characterized samples of children born after ART, followed longitudinally, may allow for the identification of subgroups of subjects with different developmental profiles, and could ultimately contribute to a better understanding of the etiological underpinnings of ASD phenotypes.
References
1. American Psychiatric
Association., Diagnostic criteria from DSM-IV-TR. 2000,
Washington, D.C.: American Psychiatric Association. xii, 370 p.
2. Muratori, F., et al., Tracing back to the onset of abnormal head circumference growth in Italian children with autism spectrum disorder. Res Autism Spectr Disord, 2012. 6: p. 442-449. 3. Wolff, J.J., et al., Differences in
white matter fiber tract
development present from 6 to 24 months in infants with autism. Am J Psychiatry, 2012. 169(6): p. 589-600.
4. Bailey, A., et al., Autism as a strongly genetic disorder: evidence from a British twin study. Psychol Med, 1995. 25(1): p. 63-77.
5. Johnson, C.P., S.M. Myers, and D. American Academy of
Pediatrics Council on Children With, Identification and
evaluation of children with autism spectrum disorders. Pediatrics, 2007. 120(5): p. 1183-215.
6. Kogan, M.D., et al., Prevalence of parent-reported diagnosis of autism spectrum disorder
among children in the US, 2007. Pediatrics, 2009. 124(5): p. 1395-403.
7. CDC, Centers for Disease Control and Prevention.
Prevalence of autism spectrum disorders--Autism and
Developmental Disabilities Monitoring Network, 14 sites, United States, 2008. MMWR Surveill Summ, 2012. 61(3): p. 1-19.
8. Currenti, S.A., Understanding and determining the etiology of autism. Cell Mol Neurobiol, 2010. 30(2): p. 161-71. 9. Heron, M., et al., Annual
summary of vital statistics: 2007. Pediatrics, 2010. 125(1): p. 4-15.
10. Schieve, L.A., et al., Have
secular changes in perinatal risk factors contributed to the recent autism prevalence increase? Development and application of a mathematical assessment model. Ann Epidemiol, 2011. 21(12): p. 930-45.
11. Durkin, M.S., et al.,
Socioeconomic inequality in the prevalence of autism spectrum disorder: evidence from a U.S. cross-sectional study. PLoS One, 2010. 5(7): p. e11551. 12. Mandell, D.S., et al.,
Racial/ethnic disparities in the identification of children with autism spectrum disorders. Am J Public Health, 2009. 99(3): p. 493-8.
13. Thomas, P., et al., The
association of autism diagnosis with socioeconomic status. Autism, 2012. 16(2): p. 201-13. 14. Koyama, T., et al., Cognitive
and symptom profiles in
Asperger's syndrome and high-functioning autism. Psychiatry Clin Neurosci, 2007. 61(1): p. 99-104.
15. Parner, E.T., et al., Parental age and autism spectrum disorders. Ann Epidemiol, 2012. 22(3): p. 143-50.
16. Bhasin, T.K. and D. Schendel, Sociodemographic risk factors for autism in a US metropolitan area. J Autism Dev Disord, 2007. 37(4): p. 667-77.
17. Arpino, C., et al., Preterm birth and neurodevelopmental outcome: a review. Childs Nerv Syst, 2010. 26(9): p. 1139-49. 18. Hvidtjorn, D., et al., Cerebral
palsy, autism spectrum
disorders, and developmental delay in children born after assisted conception: a systematic review and meta-analysis. Arch Pediatr Adolesc Med, 2009. 163(1): p. 72-83. 19. Middelburg, K.J., et al.,
Neuromotor, cognitive, language and behavioural outcome in children born following IVF or ICSI-a systematic review. Hum Reprod Update, 2008. 14(3): p. 219-31.
20. Ludwig, A.K., et al., Post-neonatal health and
development of children born after assisted reproduction: a systematic review of controlled studies. Eur J Obstet Gynecol Reprod Biol, 2006. 127(1): p. 3-25.
21. Wells, G.A., et al. The
Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. [cited 2008; Available from:
http://www.ohri.ca/programs/clini cal_epidemiology/oxford.htm 22. Moher D, L.A., Tetzlaff J, Altman
DG, The PRISMA Group Preferred Reporting Items for Systematic Reviews and MetaAnalyses: The PRISMA Statement. PLoS Med 2009. 6(6): e1000097.
doi:10.1371/journal.pmed10000 97.
23. Liberati, A., et al., The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate health care
interventions: explanation and elaboration. J Clin Epidemiol, 2009. 62(10): p. e1-34. 24. Sanchez-Albisua, I., et al.,
Medical, psychological and intellectual development of 5-year-old children born after intracytoplasmic sperm injection.
Neuropediatrics, 2011. 42(3): p. 104-9.
25. Knoester, M., et al., Matched follow-up study of 5 8-year-old ICSI singletons: child behaviour, parenting stress and child
(health-related) quality of life. Hum Reprod, 2007. 22(12): p. 3098-107.
26. Klemetti, R., et al., Health of children born as a result of in vitro fertilization. Pediatrics, 2006. 118(5): p. 1819-27. 27. Lyall, K., et al., Fertility
therapies, infertility and autism spectrum disorders in the
Nurses' Health Study II. Paediatr Perinat Epidemiol, 2012. 26(4): p. 361-72.
28. Lidegaard, O., A. Pinborg, and A.N. Andersen, Imprinting diseases and IVF: Danish National IVF cohort study. Hum Reprod, 2005. 20(4): p. 950-4. 29. Middelburg, K.J., et al., Mental,
psychomotor, neurologic, and behavioral outcomes of 2-year-old children born after
preimplantation genetic screening: follow-up of a randomized controlled trial. Fertil Steril, 2011. 96(1): p. 165-9.
30. Wiener-Megnazi, Z., R. Auslender, and M. Dirnfeld, Advanced paternal age and reproductive outcome. Asian
Journal of Andrology, 2012. 14(1): p. 69-76.
31. Eisenberg, E., Long-term outcomes in children born after assisted conception. Semin Reprod Med, 2012. 30(2): p. 123-30.
32. Hediger, M.L., et al., Assisted reproductive technologies and children's neurodevelopmental outcomes. Fertil Steril, 2013. 99(2): p. 311-7.
33. Szatmari, P., Is autism, at least in part, a disorder of fetal programming? Arch Gen Psychiatry, 2011. 68(11): p. 1091-2.
34. Bhandari, A., J.I. Sandlow, and R.E. Brannigan, Risks to
offspring associated with advanced paternal age. J Androl, 2011. 32(2): p. 121-2. 35. Rao, A.N., 2- hydroxyisobutyric
aciduria, assisted reproductive technologies and its relation to autism- a pilot study.
Perinatology, 2008. 10: p. 47-53.
36. Pinborg, A., et al., Infant
outcome of 957 singletons born after frozen embryo
replacement: the Danish National Cohort Study 1995-2006. Fertil Steril, 2010. 94(4): p. 1320-7.
37. Steel, A.J. and A. Sutcliffe, Long-term health implications for children conceived by
IVF/ICSI. Hum Fertil (Camb), 2009. 12(1): p. 21-7.
38. Basatemur, E. and A. Sutcliffe, Follow-up of children born after ART. Placenta, 2008. 29 Suppl B: p. 135-40.
39. Sanchez-Albisua, I., et al., Increased frequency of severe major anomalies in children conceived by intracytoplasmic sperm injection. Dev Med Child Neurol, 2007. 49(2): p. 129-34. 40. Agarwal, P., et al., Two-year
neurodevelopmental outcome in children conceived by
intracytoplasmic sperm injection: prospective cohort study. Bjog-an International Journal of Obstetrics and Gynaecology, 2005. 112(10): p. 1376-1383. 41. Kondapalli, L.A., et al., A Pilot
Study of Neurodevelopment, Behavior and Obesity in Young Children Conceived by Assisted Reproductive Technology. Fertility and Sterility, 2011. 96(3): p. S17-S17.
42. Munk-Jorgensen, P. and P.B. Mortensen, The Danish Psychiatric Central Register. Dan Med Bull, 1997. 44(1): p. 82-4.
43. Pinborg, A., et al., Neurological sequelae in twins born after assisted conception: controlled national cohort study. BMJ, 2004. 329(7461): p. 311.
44. Zachor, D.A. and E. Ben Itzchak, Assisted reproductive technology and risk for autism spectrum disorder. Res Dev Disabil, 2011. 32(6): p. 2950-6. 45. Shimada, T., et al., Parental age
and assisted reproductive technology in autism spectrum disorders, attention deficit hyperactivity disorder, and Tourette syndrome in a Japanese population. Res Autism Spec Dis, 2012. 6(1): p. 500-507.
46. Stein, D., et al., Obstetric complications in individuals diagnosed with autism and in healthy controls. Compr Psychiatry, 2006. 47(1): p. 69-75.
47. Maimburg, R.D. and M. Vaeth, Do children born after assisted conception have less risk of developing infantile autism? Hum Reprod, 2007. 22(7): p. 1841-3.
48. Lehti, V., et al., Autism spectrum disorders in IVF children: a national case-control study in Finland. Hum Reprod, 2013. 28(3): p. 812-8.
49. Lord, C., M. Rutter, and A. Le Couteur, Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive
