Autistic disorder (AD) is a disorder of brain development. Its neuropathogenesis and etiology are thought to be heterogeneous and are know in only a small fraction of cases. Genes have a complex indirect and, to date, undefined role in AD. This chronic neurologic disorder leaves affected individuals dependent on their families for life.
Autism, autistic disorder and autistic spectrum disorder need definitions. They are best understood from a historical perspective.
In 1943 Leo Kanner, a child psychiatrist, described 11 children, most from prosperous and welleducated parents (several were physicians) with "autistic disturbances of affected contact."1 He borrowed the word autistic from the schizophrenia literature, where it denoted the selfcentered thinking, detached behaviors and resulting isolation observed in schizophrenia. Following his report, the term infantile autism emerged.
Kanner's report received little attention in part because the behavioral phenotype was already subsumed under the diagnosis of childhood schizophrenia. However, in the 1950s the psychodynamic theorists resurrected his work and hypothesized that children became autistic because they were not given appropriate emotional nurturing by their "refrigerator" mothers.2,3 This approach to autism was the antithesis of Kanner's speculation that infantile autism had a biologic cause. Infantile autism was formalized as a disorder in 1980 when it was included in the Diagnostic Statistic Manual, third edition (DMSIII).4 It was categorized as a pervasive developmental disorder (PPD), a group of behaviorally defined syndromes that share the characteristic of severe and pervasive disruption of development in "..social interaction skills, communication skills or the presence of stereotyped behavior, interests and activities."4 The other PDDs are Rett's disorder, childhood disintegrative disorder, Asperger's disorder and PDD not otherwise specified (PDD NOS). FDD NOS was included to allow clinicians to indicate that a child had a behavioral phenotype that was similar to one of the defined PDDs but did not meet the full criteria for that disorder. In the next revision of the DSM III infantile autism was changed to autistic disorder (AD), the term used in subsequent editions, including the DSM IV.
With the explosion of research in the past 25 years, the concept has emerged that the behavioral characteristics observed in children with AD are not unique to AD but are distributed in lesser forms and in different combinations throughout a population; AD behaviors represent a portion of this distribution. In the 1990s the descriptive term, autistic spectrum disorder (ASD), was coined in the experimental literature. The term migrated into clinical literature where it is often used as a diagnostic term even though a common clinical definition has not been agreed upon.
PERVASIVE DEVELOPMENTAL DISORDERS
AD and the other PDDs are behaviorally defined syndromes that medicine and the sciences are struggling to define with quantitative, objective diagnostic criteria and to associate the phenotype with a biological marker. At present neither is available although the Autistic Diagnostic Observation Schedule (ADOS) offers a meaningful advancement. Diagnostic criteria are based on a qualitative determination of the seventy of behavioral impairments, and demonstrating that a child has the DSM IV prescribed number of impairments. This qualitative determination is subject to the confounding of observer bias and marked rater-to-rater variance. This confounding variance has been a recurring source of considerable disagreement among physicians, a disagreement that no doubt has perplexed the PCP who is caring for the child and supporting the family. The ADOS offers an observational instrument to reduce the influence of these confounding variables.
AD is a behaviorally defined syndrome characterized by severe impairments in social interaction, the social components of language and inflexible and restricted behaviors or interests. A diagnosis is made based on a behavior phenotype defined in DSM IV.5 Since the diagnosis is based on behaviors, the AD phenotype can overlap with other disorders. For example, a large percent of children with tuberous sclerosis have the AD phenotype. Similarly, approximately 1% of children with AD have tuberous sclerosis. This overlap of phenotype is also seen with congenital brain malformations and untreated phenylketonuria. However, most cases of AD are idiopathic.
AUTISTIC SPECTRUM DISORDER
The basic behavioral traits that comprise autism have not changed since Kanner reported them. However, the interpretation of these traits has changed and so has the type of child receiving an autism diagnosis. The broadest interpretation is represented by the term autistic spectrum disorder (ASD). It assumes that autism is not a discreet disorder but should be viewed as a spectrum of conditions or behaviors unified by impairments in social interactions, communication and restricted behaviors or interests. Since ASD is defined by behaviors and not by pathogenesis or causative agent, a spectrum can include the Kanner-type AD in which there is severe and profound impairments as well as mental retardation, to milder cases such a PDD NOS, to atypical cases such as high functioning AD (AD with IQ > 70), to cases with no impairment of language development and only mild impairments in social interaction and inflexibility such as in Asperger disorder, to children whose impairments in the past were considered below diagnostic threshold for PDD. The dilemma present by this approach is how to distinguish between ASD behavior and behavior that is eccentric, idiosyncratic or odd which are in general considered within the range of normal.
Support for interpreting autism as ASD comes from studies that find AD behavioral traits, albeit with varying severity, distributed through families of children with AD and in the general population.6,7,8 Further support comes from functional brain imaging studies (fMRI) that found similar activation patterns in adults with high functioning autism and Asperger syndrome.9 On the other hand, the term spectrum is most often used in the context of disorders that are clinically distinct and share a common etiology while varying in the severity of symptoms. The validity of grouping behaviorally defined phenotypes of potentially different etiologies has been questioned. AD and the disorders subsumed under ASD are postulated to have heterogeneous etiologies which open the possibility that the autistic behavioral phenotype may represent a final common pathway for a variety of disorders. Determining whether an autistic phenotype represents the final common expression of distinct disorders or represents a spectrum of disorders with shared neuropathological parameters is a question awaiting clarifying studies.
Although the utility of ASD is debated, it has become a popular clinical term, used to define a behavioral phenotype without reference to pathogenesis or etiology. An understanding of this is important because the use of ASD must be accompanied by a vigorous diagnostic evaluation for treatable disorders.
The epidemiology of AD has not changed substantially from the findings reported in the 1980s and early 199Os. It is most often diagnosed in males (male to female ratio of 4 to 1). cases do not segregate by race, level of education, socioeconomic status or geography. Behavioral signs typically appear before the age of three.10 Delays in language (either absence or slow progression) are the most common presenting complaint." A small portion have regression after typical language development. Sixty to seventy-five percent of the children with AD have mental retardation.12,13 Cognitive deficits may be more severe in girls than boys although approximately 40% of boys will have severe to profound mental retardation. Organic, behavioral and emotional comorbidities are common. (Table 1)
AD has implications for the entire family. Mild language, social or psychiatric problems can be present in parents.14,15 Siblings have a greater, albeit small, risk of AD, language disorder, learning disabilities, social problems and psychiatric disorders.
A rise in the prevalence of children with an autistic diagnosis has been established although a specific prevalence rate has not.12,16,17 Population studies in the 1980s and early 1990s found a prevalence ranging from 2 to 10 per 10,000 children.13,17 After 1994 prevalence increased annually with recent estimates reaching as high as 67 per 10,000 for ASD and 4 per 10,000 for AD.12 Since 1994 Rhode Island has mirrored this trend in its special education population.18 In 2003, 605 students with an autistic diagnosis were educated in Rhode Island schools. Understanding the factors behind this rise in prevalence is critical to our understanding of AD and ASD.
Population-based incidence studies, using contemporary diagnostic criteria, are believed to more accurately reflect potential changes in the occurrence of disorders across time periods. One such study in Olmsted County Minnesota (OC) found increases in the incidence of AD from 0.55 per 10,000 children in 1980-1983 to 4.49 per 10,000 children in 1995-1997.'7 A rising incidence among young children was responsible for most of the increased. During the same period the incidence among children older than 10 years of age remained stable. Several putative environmental agents, such as mercury and vaccines, have been intensely studied, but to date no environmental agent has been identified to explain the rise in incidence. Genetic and genomic studies have also failed to explain the rise.
Prevalence studies assessing time trends are subject to inaccuracies that result from changes in diagnostic criteria, inability to validate diagnoses and increased awareness of the disorder across time, all of which have occurred in autism.17,19,20 The broadening of the diagnostic criteria can be seen in the revisions of the DSM since infantile autism was first included in 1980. The use of ASD has broadened the criteria still further. Public and physician awareness has grown substantially. Validating the AD or ASD is short coming in many prevalence studies. A second study of Minnesota children (MS) illustrates the problems. This study found a greater increase in prevalence, to 52 per 10,000, than the OC study which measured incidence.21 This difference can be explained in part by the method of ascertaining subjects and validation differences between the two studies. In the OC study, investigators reviewed medical and school records to determine the diagnosis of each subject using DSM criteria. The MS study used student data reported to the department of education. Such data do not allow diagnostic validation.
The possibility that school data may be biased towards categorizing students as autistic is suggested by the timing of the initial rise in prevalence (between 1991 and 1994 in most studies). Coincident with the initial rise was the inclusion of autism in the list of disabilities eligible for federally mandated special education services in 1991. If this biased diagnosis, studies ascertaining cases using school data would be expected to identify a rise in prevalence. Support for this possibility would be evidence of diagnosis swapping. Croen20 found evidence of diagnosis swapping in California, a state that has reported a dramatic rise in autism. A period of rising prevalence in AD saw a corresponding drop in the prevalence of mental retardation without autism.
The reasons for the disturbing rise in the prevalence of AD and ASD do not suggest a meaningful rise in new cases but a rise due to diagnosis swapping, an increased identification of cases and a broadening of diagnostic criteria to include cases that were considered below the diagnostic threshold in the past. However, the possibility that a portion of the rising prevalence results from a true increased incidence in autism has not been excluded.
AD is categorized as either idiopathic or secondary when the cause is known. Approximately 90% of the cases are idiopathic. A cause of the AS can be identified in approximately 10%. (Table 2) Chromosomal abnormalities account for 5% of the secondary AD. Duplication in'the Prader Willi/ Angelman region (15qll-13) is the most frequent gene mutation. Untreated phenylketonuria was responsible a considerable portion of secondary AD in the past but is now rare. Other causes of AD include prenatal agents such as infection and hypoxia. Although several environmental agents have been implicated as causative agents (e.g., mercury), studies have not provided supporting evidence.
Current conceptualization of the role of genes in AD is that of imparting susceptibility but not causing AD. Genetic heterogeneity is the operating hypothesis in studies with estimates of the number of involved genes ranging from 10-15. Four candidate susceptibility genes are currently under study.22 (Table 3) Linkage studies have identified chromosomes 2, 7, 17, 22 and X, but confirming studies have given inconsistent results. Numerous candidate genes have been identified including most neurotransmitter receptors and proteins for essential brain development (e.g., neuroligin, reelin), but no candidate gene has been consistently found in children with AD. This lack of clarity is not inconsistent with overlapping gene effects. Genes can be epistatic (several genes influencing a behavior) or pleiotrophic (one gene influencing several behaviors).
The inheritance of idiopathic AD is complex and non-mendelian. No single inheritance pattern has been recognized. Inheritance is thought to be multifactoral involving the interaction of genes and epigenetic factors. The evidence for a genetic component in the neuropathogenesis of AD comes from twin and family studies. Twin studies have demonstrated a higher concordance in monozygotic twins than dizygotic. 23 Furthermore, some family pedigrees show an increased recurrence risk within families with one child with AD.24
The recurrence risk to siblings of a child with secondary AD is the risk of the disorder causing AD (e.g., tuberous sclerosis). Determining the empiric risk of idiopathic AD depends on the incidence of AD which is debated (see Prevalence). Barbares! reported an incidence of 4.5/10,000.17 Using this incidence, the empiric risk is less than 0.1%. Some have recommended using prevalence which would raise the risk to 0.5-0.7%. The risk in ASD is debated but is presumed to be greater than AD. The risk increases considerably in families with one child with AD (4%) and still more when there are two or more children with AD (35%).24 Because of the association between AD and language, social and psychiatric problems, families with one child with AD are given an additional risk of 4-6% for one of these problems.
Kanner was the first to speculate that AD had a biological cause. He supported his speculation by observing that several children had large heads. His observation of macrocephaly has been confirmed by anthropomorphic and neuroimaging studies. Macrocephaly is present in 20% of the children with AD, appears late in the first year and resolves by five years old in most children. The high incidence of seizures and mental retardation provides further evidence for a neuropathogenesis.
Studies to elucidate abnormal brain regions in AD have produced inconsistent results while implicating numerous cerebral, cerebellar and brainstem regions. Failure to replicate neuroanatomical studies, in part because of the unavailability of postmortem brain tissue, has hampered an understanding of how brain development is disrupted. The etiologic heterogeneity of AD may also explain inconsistencies between studies. Functional magnetic resonance imaging (flVLRI) offered investigators one method of circumventing the paucity of brain tissue although it is still limited by etiologic heterogeneity. Despite this, it has provided a means of correlating brain dysfunction with brain structures.
Since a hallmark of AD is impaired social components of language and cognition (discussed by Dr. Sheinkopf in this issue), the "social brain" region has received particular attention. 25 Social cognition can be thought of as the ability to recognize, manipulate and behaviorally respond to social information whether in the form of language, another's behaviors or expressions. An important pathway mediating social cognition involves the amygdala, superior temporal sulcus and fusiform gyrus of the orbitofrontal cortex. fMRI studies have demonstrated the abnormal activation of these structures in high functioning autistic individuals.
Of particular interest is the serotonin system: 30% of children with AD have elevated platelet serotonin. Whether this contributes to the neuropathogenesis of AD or is an incidental finding will require further study.
Studies of the autistic brain have produced heterogeneous collections of findings most likely the result of its heterogeneous etiologies. The picture that appears to be emerging is disruption of early brain development with subsequent abnormalities in neuronal morphology and number, synaptic abnormalities and dysfunctional pathways as a consequence.
1. KannerL. Nervous Child 1943;2:217-50.
2. Bettelheim B. The Empty Fortress: Infantile Autism and the Birth of Self. New York: Free Press; 1967.
3. DeMyer MK, Pontius W, et al. J Autism Childhood Schizof h 1972;2:49-66.
4. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. Washington DC: American Psychiatric Association; 1980.
5. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Washington DC: American Psychiatric Association; 1994.
6. Constantino JN, Todd RD. Arch Gen Psychiatry 2003;60:524-30.
7. Piven J, Palmer P, et al. Am } Psychiatry 1997;154:185-90.
8. Spiker D, Lotspeich LJ, et al.. Am]Med Gent 2002;llil4:129-36.
9. Schultz R, Gauthier I, et al. Arch Gen Psychiatry 2000;57:332-43.
10. Rivito ER, Freeman BJ, et al.. Am J Psychiatry \ 989; 146:194-9.
11. De Giacomo A, Fombonne E.. Eur Child Adolesc Psychiatry 1998;7:131-6.
12. Bertrand J, Mars A, et al. Pediatrics 2001;108:1155-61.
13. Bryson SE, Clark BS, Smith IM. Child PscyholPsychiatry 1988;29:433-445.
14. Bailey A, Palferman S, et al. / Autism Dev Disord 1998;28:369-92.
15. Piven J. Palmer P. Am J Psychiatry 1999;156:557-63.
16. Yeargin-Allsopp M, et al. JAMA 2003;289:49-55.
17. Barbaresi W, Katusic S, et al. Arch Pediatr Adolesc Med 2005; 159:37-44.
18. Issue Brief. Rhode Island Kids Count. December 2003.
19. Wing L, Potter D. Ment Retard Dev Disabil ResRev2002;8:\5l-6\.
20. Croen LA, Grether JK, et al. / Autis Dev DwW2002;32:207-15.
21. Gurney JG, Fritz MS, et al. Arch Pediatr Adolesc Med 2003; 157:622-7.
22. Online Mendelian Inheritance in Man. http://www.ncbi.nlm.nih.gov/entrez/query. fcgi?db=OMIM (enter autism)
23 Bailey A, Le Gouteur A, et al. Psychol Med 1995;25:63-78.
24. Wassink TH, Brzustowicz L, et al. MRDD Res Rev 2004; 10:272-83.
25. Brothers L. Concepts Neurosci 1990;1:27-51.
JOSEPH J. HAUUETT, MD, AND VIREN D'SA, MD
Joseph H. Hallett, MD, is Physician-in-Chief, Department of Pediatrics, Memorial Hospital of Rhode Island, and Associate Professor (Clinical) of Pediatrics, Brown Medical School.
VirenD'Sa, MD, is a Developmental Behavioral Pediatrie Fellow, Neurodevelopmental Center, Department of Pediatrics, Memorial Hospital
Joseph J. Hallett, MD
Copyright Rhode Island Medical Society May 2005
Provided by ProQuest Information and Learning Company. All rights Reserved