Autism and the pervasive developmental disorders
Fred R. Volkmar
Ami Klin
Introduction
The pervasive developmental disorders (PDDs) are characterized by patterns of deviance and delay in social-communicative development in the first years of life, which are associated with restricted patterns of interest or behaviour. The prototypic PDD is childhood autism; other conditions included in the PDD class in ICD-10(1) include Rett’s syndrome, childhood disintegrative disorder, Asperger’s syndrome, and atypical autism. Except for one additional category in ICD-10 (hyperkinetic stereotyped movement disorder), the disorders included in ICD-10 and DSM-IV(2) are essentially identical. In this chapter each of these conditions will be reviewed in terms of their clinical features, definition, epidemiology, course, and aetiology; final sections of the chapter address aspects of treatment and prevention for the group of disorders as a whole (Box 9.2.3.1).
Childhood autism
Autism was first recognized by Kanner(3) in his report of 11 children with ‘autistic disturbances of affective contact’. He emphasized two essential diagnostic features: autism and difficulties with change; but he also noted atypical language (when language was present at all). He used Bleuler’s term ‘autistic’ to convey the children’s social isolation. Although children with autism had undoubtedly been previously observed,(4) it was Kanner’s particular genius to so precisely describe the condition. False leads for research arose since the term autism introduced an unintended confusion with schizophrenia. Also, Kanner assumed that the children had normal intellectual potential. Subsequently, it became clear that autism and schizophrenia were distinct and that autism was often associated with intellectual disability.(5) Although describing autism as inborn, Kanner mentioned that parents were very well educated and successful. In turn this led to the idea, common during the 1950s, that autism might somehow result from deviant patterns of care by unusually successful parents. A large body of evidence shows that this is most certainly not the case.(6) It is clear that families of children with autism come from all social classes and circumstances(7) and that the original impression had been the result of referral bias.
Clinical features
Social deficits of a particular type remain a hallmark of autism. The nature of this deficit varies, somewhat, over time but remains a source of great disability to the affected individual throughout life.(8) In younger and more impaired individuals there may be little interest in social interaction; less impaired individuals may come to a passive acceptance of social interaction while older and more cognitively able individuals are more likely to seem highly eccentric and one-sided.(9) Difficulties are observed in the use of eye contact or other non-verbal social cues, in social emotional reciprocity and empathy, in activities involving shared interests with others, and with peer relationships (see Table 9.2.3.1). As Rutter(10) suggested, these problems do not simply reflect cognitive impairment, which is present in about 60 per cent of individuals affected. Abnormalities in communication (not only in language) are also observed. In a substantial minority (perhaps 30 per cent) of cases, the child never acquires the capacity for communicative speech; among individuals who do talk, various unusual features of language are observed such as echolalia, idiosyncratic language, deficits in prosody, and pronoun reversal.(11) Deficits in pragmatic language are particularly striking. As with the social disturbance, the deficits observed are not solely due to intellectual disabilities. Various unusual behaviours are subsumed under the term ‘resistance to change’. These behaviours include literal resistance to change (‘insistence on sameness’), stereotyped and repetitive motor mannerisms, strict adherence to non-functional routines, and interest in non-functional parts of objects. Various other features may be observed, such as unusual sensitivities to aspects of the environment and attachments to unusual objects.
Box 9.2.3.1 The pervasive developmental disorders
Childhood autism/autistic disorder
Rett’s disorder
Childhood disintegrative disorder
Asperger’s disorder
PDD-NOS/atypical autism
Definition
In the 1950s and 1960s there was disagreement about autism, e.g. was it a form of schizophrenia or psychosis? Gradually evidence began to accumulate that suggested the role of central nervous system dysfunction in pathogenesis—for example, high risk for developing seizures. Differences in clinical features, onset and course, and family history also supported the distinctiveness of autism apart from childhood schizophrenia.(12,13) By 1978, there was a substantial body of work on the validity of autism. Rutter synthesized this in his influential definition of autism,(10) which required the presence of patterns of delay and deviance in the areas of social and language development that were not simply the result of developmental delay along with the group of unusual behaviours subsumed under Kanner’s term ‘insistence on sameness’. Onset before 30 months of age was required. In ICD-9(14) infantile autism was still termed a psychotic condition but by 1980 the highly influential DSM-III(15) appeared and recognized autism as a condition apart from schizophrenia, including it in a new class of disorders—the pervasive developmental disorders. The latter term has been the topic of some debate, although a better term has yet to be proposed and, in any case, the term PDD has now come into general usage in both DSM and ICD.(16,17)
The name chosen in DSM-III (‘Infantile autism’) was consistent with Kanner’s original report but reflected a lack of developmental orientation; these concerns were addressed in DSM-III-R,(18) which provided a detailed, and more developmentally oriented, set of diagnostic guidelines. Unfortunately, this definition proved overly inclusive and it became apparent that additional work would be needed. Further impetus was given to this effort by pending changes in ICD-10(1) and DSM-IV. Accordingly, an international field trial was undertaken.(19) Based on the results of this field trial
autism is defined (see Table 9.2.3.1) on the basis of characteristic problems in three areas: social interaction, communication and play, and restricted patterns of interest. By definition, autism must be present by the age of 3 years. ICD-10 provides for various ways in which a diagnosis of atypical autism can be made—for example, because of failure to meet age of onset or behavioural criteria. Data on this system suggest that it has good agreement with the diagnoses of experienced clinicians, avoids the problem of the overdiagnosis of autism in the most mentally handicapped persons, and has reasonably good reliability.
autism is defined (see Table 9.2.3.1) on the basis of characteristic problems in three areas: social interaction, communication and play, and restricted patterns of interest. By definition, autism must be present by the age of 3 years. ICD-10 provides for various ways in which a diagnosis of atypical autism can be made—for example, because of failure to meet age of onset or behavioural criteria. Data on this system suggest that it has good agreement with the diagnoses of experienced clinicians, avoids the problem of the overdiagnosis of autism in the most mentally handicapped persons, and has reasonably good reliability.
Table 9.2.3.1 ICD-10 criteria for childhood autism (F84.0) | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Epidemiology and demographics
Over 40 epidemiological studies of autism and related conditions have been conducted with prevalence estimates ranging from 0.7 per 10 000 to 72.6 per 10 000.(20) In his recent 2005 review, Fombonne notes that prevalence rates are negatively correlated with sample size and there is an apparent trend for increased rates over time. Various considerations (including changes in definition) complicate the interpretation of increased rates. In this review, Fombonne suggests that a reasonable estimate of prevalence is 13 per 10 000 (Box 9.2.3.2). Although higher rates of 1 in 150 are reported if broad definitions are used.
Box 9.2.3.2 Epidemiology of autism and related conditions
Autism | 13 per 10 000 |
Rett’s disorder | 1 per 10 000 girls |
Childhood disintegrative disorder | 1.9 per 100 000 |
Asperger’s syndrome | 4 per 10 000 |
PDD-NOS/atypical autism | 1 per 150 |
Source: Cohen, D.J. and Volkmar, F.R. (eds.) (2005) Handbook of autism and pervasive developmental disorders (3rd edn.) Wiley, New York. | |
A number of studies, including both epidemiological and clinically referred samples, report higher rates of autism in boys than in girls (typically 3.5:1 or 4:1). This ratio varies with IQ level, i.e. females with autism who have IQs in the normal range are probably
20 times less common than males.(21) The explanation for this sex difference remains unclear. It is possible that the degree of insult required to produce autism in females must be greater than for males, but other hypothesis have been raised. Ethnic and cultural differences have been little studied.(20,22)
20 times less common than males.(21) The explanation for this sex difference remains unclear. It is possible that the degree of insult required to produce autism in females must be greater than for males, but other hypothesis have been raised. Ethnic and cultural differences have been little studied.(20,22)
Course and prognosis
Although childhood autism is a chronic disability, there is some suggestion that with early intervention and remediation outcome is improving.(23) For example, the number of individuals with either a ‘good’ or ‘fair’ outcome is now about 50 per cent—a noteworthy increase since 1980.(24) However, even in the highest functioning individuals marked social problems persist.
Changes in the degree of social relatedness, communication, and self-help skills are observed with increases in developmental level. Seizure disorders are observed in up to 25 per cent of individuals and may have their onset at any point, but adolescence and early childhood are particularly common times.(25) Factors that predict long-term outcome include the presence of some communicative speech by the age of 5 or 6 years, and non-verbal intellectual level.
Aetiology
Early interest centred on the possibility that experiential factors might somehow cause autism, but a host of research findings suggests that this is not the case. Rather, a fundamental disturbance in the central nervous system is implicated.
(a) Psychological factors
Disabilities affecting attentional mechanisms, arousal, sensory deficits, memory management, and complex information processing, among others, have been proposed as ‘primary’ deficits underlying the social impairment in autism. Although each of these helps us understand some aspects of the condition, none has, as yet, provided a more comprehensive account of the condition as a whole.(26) Among the most influential recent theories attempting to do that is the hypothesis that posits a lack of a central drive for coherence in children with autism, with the consequent focus on dissociated fragments of their environment rather than integrated ‘wholes’, leading to a fragmentary and overly concrete experience of the world.(27) Another cognitive account of autism posits that the commonly found difficulties in abstracting rules, inhibiting irrelevant responses, shifting attention, and profiting from feedback as well as in maintaining relevant information ‘on-line’—the so-called ‘executive functions’—underlie the social, communicative, and behavioural disabilities in autism.(28) Although both these theories —‘weak central coherence’ and ‘executive dysfunction’—provide insightful new views of well-known clinical features, neither phenomena can be seen as specific to autism relative to other developmental disorders.
Probably the most influential current cognitive hypothesis focuses on mechanisms directly impacting on social understanding. This view, the ‘theory of mind’ hypothesis, posits that autism is caused by the child’s inability to attribute mental states such as beliefs and intentions to others. Devoid of this ability, individuals with autism are thought to be unable to infer the thoughts and motivations of others, thus failing to predict their behaviour and adjust their own actions accordingly, which results in a lack of reciprocity in communication and social contact.(29) Although more than 50 studies have documented such deficits in autism, there are still many limitations to this hypothesis. For example, more able individuals with autism do exhibit ‘theory of mind’ skills—and yet may be totally unable to utilize this capacity in their spontaneous social adjustment. Such phenomena suggest that factors other than a cognitive understanding of mental phenomena are required for a person to meet the demands of everyday social life. For example, the ‘enactive mind’ hypothesis focuses on early emerging and highly conserved mechanisms of socialization that precede the advent of mentalizing abilities, and which culminate in the development of joint attention and perspective taking skills.(30) Of great interest in the past few years has been the confluence of experimental psychological paradigms and functional neuroimaging studies focusing on the same constructs. This new trend is leading to new insights into brain systems subserving basic social mechanisms such as gaze behaviour, face processing, social-affective responses, and thinking about other people’s intentions and beliefs,(31) all of which are greatly compromised in autism.
(b) Biological factors
The importance of biological factors in the pathogenesis of autism is suggested by several lines of evidence. Autism has been associated with a host of medical conditions; but the absence of population data and rigorous diagnostic assessment makes such associations difficult to interpret. For example, early reports suggested an association with congenital rubella, but this now seems questionable given the diagnostic dilemmas and the observation that ‘autisticlike’ features diminish over time.(32) Gillberg(33) argues that medical conditions may be associated with autism in as many as one-third of the cases, but Rutter and colleagues(34) suggest that a more reasonable figure would be roughly 10 per cent of all cases. The strongest associations are with fragile X syndrome and tuberous sclerosis— both conditions having a strong genetic component (Box 9.2.3.3).
Box 9.2.3.3 Medical conditions associated with autism
Seizure disorder (epilepsy)
Fragile X syndrome
Tuberous sclerosis
Fragile X syndrome is an X-linked intellectual disability syndrome involving a mutation characterized by a triplet repeat of cytosine-guanine-guanine (CGG) that may amplify with succeeding generations. It is associated with a characteristic facial appearance, enlarged testicles, intellectual disability, and some autistic features. Early reports suggesting high rates of fragile X in autism have now been modified; fragile X affects perhaps 1 to 2 per cent of all individuals with autism.(20)
The autosomal dominant disorder tuberous sclerosis is characterized by abnormal tissue growth, or benign tumours (hamartomas), in the brain and in other organs. The condition, which may affect 1 in 10 000 individuals, is variably expressed; the phenotype ranges from minor skin problems or seizures to severe intellectual disability with intractable seizures. The rate of this condition in autism (0.4-2.8 per cent) is significantly increased.(20)
Autism is a strongly genetic condition. Studies of monozygotic and dizygotic twins revealed much higher levels of concordance for monozygotic relative to dizygotic twin pairs, but also elevated
rates of concordance in dizygotic twins relative to population rates.(35) General studies suggest that the recurrence risk of autism in siblings is in the order of between 2 and 10 per cent—which is a substantial increase in risk over population rates. There also appear to be higher rates of social and language problems and rigid patterns of behaviour in siblings and close relatives, raising the possibility that what is inherited is a broader phenotype reminiscent of autism but which also may reflect a more general predisposition to developmental difficulties. Recent work also suggests elevated rates of anxiety and mood disorders in family members. Specific modes of inheritance remain unclear. It now appears that several interacting genes are probably involved in the pathogenesis of autism. Efforts are underway to identify susceptibility genes and trace their impact on brain development. Although several studies have shown increased rates of pre-, peri-, and neonatal complications in children with autism, it is possible that some of these difficulties may reflect a genetic vulnerability in the child or that there may be an interaction of genetic and perinatal factors.(35) A recent report has noted the presence of placental abnormalities in pregnancies of children with autism.
rates of concordance in dizygotic twins relative to population rates.(35) General studies suggest that the recurrence risk of autism in siblings is in the order of between 2 and 10 per cent—which is a substantial increase in risk over population rates. There also appear to be higher rates of social and language problems and rigid patterns of behaviour in siblings and close relatives, raising the possibility that what is inherited is a broader phenotype reminiscent of autism but which also may reflect a more general predisposition to developmental difficulties. Recent work also suggests elevated rates of anxiety and mood disorders in family members. Specific modes of inheritance remain unclear. It now appears that several interacting genes are probably involved in the pathogenesis of autism. Efforts are underway to identify susceptibility genes and trace their impact on brain development. Although several studies have shown increased rates of pre-, peri-, and neonatal complications in children with autism, it is possible that some of these difficulties may reflect a genetic vulnerability in the child or that there may be an interaction of genetic and perinatal factors.(35) A recent report has noted the presence of placental abnormalities in pregnancies of children with autism.
Attempts have been made to identify neuropathological and neuroanatomical correlates of autism. Areas of interest have included the cortical areas responsible for language and social interaction (frontal and temporal lobes) as well as the neostriatum and cerebellum.(36) The report of reduced cerebellar size in the neocerebellar vermal lobules VI and VII has not proven readily replicable. Some individuals with autism have enlarged brains and head sizes, with some evidence that abnormal growth occurs in the first 2 years of life.(37) Neuropathological studies have suggested possible cellular changes in areas of the brain such as the hippocampus and amygdala and changes in the cytoarchitecture of the brain, e.g. in the arrangement of cortical ‘minicolumns’.
Various neurotransmitter systems have been studied. Probably the most robust finding has been the observation that about one-third of the children with autism have increased peripheral levels of serotonin. This finding is not specific to autism and its significance remains unclear.(38) Studies of other neurotransmitters such as dopamine produced inconsistent findings. The possible involvement of dopamine is suggested by the high levels of stereotyped behaviours in autism—behaviours, which can be induced in animals by the administration of agents (stimulants) that affect levels of dopamine in the brain. Agents such as the neuroleptics, which block dopamine receptors, are effective in reducing the stereotyped and hyperactive behaviours of many autistic children. Another hypothesis has centred on the possible role of endogenous opioids, in that overproduction of such compounds might lead to social withdrawal and unusual sensitivities and behaviours. This has led to the administration of opioid antagonists such as naltrexone in autism; unfortunately results have been disappointing. Studies of the immune system in autism have been relatively uncommon and findings inconsistent.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
