• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

One Hundred Males With Asperger Syndrome: a Clinical Study of Background and Associated Factors

Posted on: Saturday, 30 October 2004, 03:00 CDT

The objective of this study was to investigate the background and associated factors in a representative group of young males with Asperger syndrome (AS) presenting at a specialized autism clinic. One hundred males aged 5 years 6 months to 24 years 6 months, with a mean age of 11 years 4 months (SD 3y 10mo), who had a clinical diagnosis of AS were included in the study. An in-depth review of their medical records and neuropsychological test data was performed. There was a high rate (51%) of non-verbal learning disability (defined as Verbal IQ more than 15 points higher than Performance IQ), but otherwise there was little or no support for the notion of right-hemisphere brain dysfunction being at the core of the syndrome. There was a very high rate of close relatives with autism spectrum problems, but also high rates of prenatal and perinatal problems, including prematurity and postmaturity. In comparison with general population data, those with AS very often had a combination of genetic and prenatal and perinatal risk factors. Non-verbal learning disability test results applied in about half the group. There was a subgroup of individuals with AS who had macrocephalus. However, there was no support for an association of AS with low body mass index.

Asperger syndrome (AS; Asperger 1944, Wing 1981, Gillberg and Gillberg 1989) is an autism spectrum disorder believed to be closely genetically and clinically linked to the International Classification of Diseases' (ICD-10) 'childhood autism', and the Diagnostic and Statistical Manual of Mental Disorders', 4th edition (DSM-IV) 'autistic disorder' (World Health Organization 1993/ American Psychiatric Association 1994). By many it is seen to represent autism in individuals with high IQ or, specifically, autism associated with good or excellent verbal abilities. Whether it differs significantly from so-called 'high-functioning' autism is a contentious issue (Mesibov et al. 1998, Gillberg 1999, Wing and Potter 2002). The concept of 'high-functioning autism' may be a misnomer in that, literally, it suggests that it is the autism that is mild. If anything, it is the cognitive impairment that is mild; the 'autism' may be just as disabling as in so-called 'low- functioning' cases.

There have been no studies of large groups of individuals diagnosed as having AS. To our knowledge, no study comprising 100 patients or more has ever been published. Few, if any, studies - except one controlled small clinical case series (Gillberg 1989) - have looked at the associated clinical and background factors in AS. The present exploratory study was undertaken with a view to filling in some of the gaps in our present knowledge in this respect. In particular, we wanted to examine whether there is (1) anything typical about the social demographic background in AS, (2) a particular pattern of neuropsychiatrie problems in relatives, (3) a high rate of prenatal and perinatal problems, (4) a typical set of comorbid conditions, and/or (5) a neuropsychological profile that will help refine the diagnostic process and suggest a specific type of underlying brain dysfunction.

The diagnostic criteria for AS published by the World Health Organization and the American Psychiatric Association are not clinically relevant, in that virtually no individual with an autism spectrum disorder meets them (Leekam et al. 2000). However, those diagnosed under the Gillberg and Gillberg (1989) criteria (operatioiialized in more detail by Gillberg; 1991) comprise a large subgroup (slightly under half) of all those in the autism spectrum. The Gillberg criteria were based on the original descriptions of cases provided by Hans Asperger, whose cases do not meet World Health Organization or American Psychiatric Association criteria for the disorder carrying his name (Miller and Ozonoff 1997). In the present study, therefore, the Gillberg (1991) operationalized criteria for AS will be used.

Methods

The study is based on medical records and psychological test results that were collected in a systematic fashion in a clinical setting.

PARTICIPANTS

The study group consisted of 100 patients diagnosed as having AS, as described by Gillberg (1991). All 100 individuals also met the ICD-10 criteria for AS and DSM-IV criteria for Asperger 's disorder, except the one related to completely normal development in the first 3 years of life. They were all included in a long-term follow-up study of AS that is currently under way in our centre. All patients included had consulted the psychiatric services at the Child Neuropsychiatrie Clinic (CNC) at Gteborg. The CNC is a statewide diagnostic service for autism spectrum disorders. It is also the autism spectrum disorders centre for diagnosis and habilitation of those living in the city of Gteborg and the surrounding region of West Sweden (population about 1.6 million). All patients included had been born between 1967 and 1988. They had been selected so as to include all patients with AS who would be 16 years or older on 30 June 2004 (the census date for inclusion in our continuing follow- up study), with a Full-scale IQ (FSIQ) of more than 70 on the Wechsler Intelligence Scale for Children - Revised (WISC-R), the Wechsler Intelligence Scale for Children, 3rd edition (WISC-III), or the Wechsler Adult Intelligence Scale - Revised (WAIS-R), and who had had a documented follow-up period from original diagnosis at the clinic of at least 5 years at that census date. The patients had been assessed and diagnosed over a period of 15 years between 1985 and 1999 (nine patients from 1985 to 1989, 62 patients from 1990 to 1994, and 29 patients from 1995 to 1999).

During this period the senior author (CG), an expert on AS, had been in charge of clinical services at the CNC. Although spcifie diagnostic interviews and specific laboratory testing have changed (i.e. expanded) during these years, basic clinical and medical work- up (particularly medical histoiy taking) has undergone little change. Thus, for instance, family history relating to neurodevelopmental and neuropsychiatrie disorders has been taken in a fairly uniform fashion over the past 20 years. Here we report on the background and associated factors in this group of patients. All patients were male (because we could not locate more than seven female patients meeting the inclusion criteria described above). They had been living in the city of Gteborg (n=57), the surrounding region (West Sweden plus the county of Hailand; n=31) or another part of Sweden (n= 12) at the time of diagnosis at the CNC.

PROCEDURE

In-depth medical-psychiatric records are kept at the CNC for all patients. All patients included in the study had been examined by at least one specialist child neuropsychiatrist/neurologist with abundant experience in the field of autism spectrum disorders and at least one specialist child neuropsychologist with expertise in the field. Diagnostic interviews had been in-depth in all cases, usually employing either the Childhood Autism Rating Scale (Mesibov et al. 1989), the Handicaps, Behaviours and Skills Schedule (Wing 1980), and/or a check-list version of the AS Diagnostic Interview (Gillberg 1991, Gillberg et al. 2001), and all children had been tested on one of the Wechsler scales: the WISC-R (Wechsler 1974), the WISC-III (Wechsler 1992), or the WAIS-R (Wechsler 1981) in 40, 52, and eight patients respectively. For seven of the individuals tested on the WISC-III, verbal subtest scaled scores were not available because, at the time of testing, Swedish norms for these subtests were not available and it was not considered reasonable to use US norms for these language-loaded tests. These individuals had their Verbal IQ (VIQ) estimated from the psychologists' detailed descriptions of verbal ability, in accordance with a special scale constructed for internal use at our clinic so as to 'standardize' the descriptions of verbal ability with US norms for IQ (details are available upon request from the authors). For six of those tested with the WISC-R, only stanine results noted in the medical records (and no raw scores) were available and IQ was estimated by approximation of those results compared with an IQ-level scale. For two additional patients only stanine results for Performance IQ (PIQ) were available, and PIQ was estimated on the basis of these.

Most of the patients (n=79) also had had an AS Screening Questionnaire (ASSQ; Ehlers and Gillberg 1993) completed by one of their parents (usually the mother). In 33 cases the ASSQ was completed by a teacher. Parents (or teachers) of patients examined before 1992 did not receive the ASSQ, which was constructed in 1991. A few questionnaires were also missing from the medical records of the patients diagnosed after 1992.

Diagnoses were always based on the comprehensive information from clinical psychiatric examination made by experts on AS (Christopher Gillberg, Carina Gillberg, Stephan Ehlers, Maria Rastam, and Peder Rasmussen). This included an interview with the parent, educational information, questionnaire data, and child neuropsychiatrie examination including a neuropsychological evaluation. Diagnoses have been made in accordance with the DSM system since 1985. AS ('autistic psychopathy according to Asperger' before 1982), because of the expertise of the senior author, has b\een diagnosed since the mid-1970s. The clinical criteria used for the Asperger diagnosis were those later published by Gillberg (1991; see Table I). All individuals in the present study would also fit the symptom (but not necessarily the normal early development) criteria for Asperger's disorder/AS according to DSM-IV/ICD-10.

After locating possible cases through medical and research registers held at the clinic, all patients were reviewed by medical record analysis to confirm that the clinical diagnosis of AS had been correct at first evaluation. This assessment was performed by a junior colleague (the first author), who now has about five years' in-depth clinical experience working with individuals with autism spectrum disorders. From the medical records, the following information was systematically extracted for 79 to 100% of all 100 individuals: age at diagnosis; social demographic factors; familial/ hereditary factors (autistic and AS symptoms were always recorded in detail whenever clinical suspicion arose that a first-degree or second-degree relative might be suffering from an autism spectrum disorder) ; maternal age at birth of child; pregnancy complications; gestational duration; perinatal and neonatal events or factors; weight, length, and head circumference at birth; early development (age at walking, age at talking); macrocephalus; body mass index (BMI); comorbidity; physical findings and medical disorder/ laboratory findings at the time of original diagnosis; ASSQ results; education; and neuropsychological test results.

ETHICS

The study was approved by the Ethics Committee at the Sahlgrenska Academy, Gothenburg University, Sweden.

Results

Results will be presented as means and numbers (often equivalent to percentages, given that the number of patients with AS included was 100). In some instances where data were not available for all cases, both numbers (n) and percentages are given.

AGE AT DIAGNOSIS

Mean age at diagnosis was 11 years 4 months (SD 3y 10mo; range 5y 6mo to 24y 6mo). Mean diagnosis age for those born between 1967 and 1979 inclusive was 15 years 5 months (SD 3y 10mo; n = 27) and for those born between 1980 and 1988 inclusive it was 9 years 11 months (SD 2y6mo; n=73;.p<0.01); 89% were 15 years or younger at the time of diagnosis.

SOCIAL DEMOGRAPHlC FACTORS

Occupation of the fathers was registered for 84 patients and that of the mothers for 98 patients. Fifteen of the fathers (18%) responded 'engineer'; there were also four responses of 'skilled technician' and four of 'nurse'. Among the mothers, 'teacher' was the most common response (n=9), followed by 'nurse' (n=7), and 'secretary' (n=6). Two mothers responded 'engineer'. Altogether, 'engineer' was the most common parental profession (17 children had a father or a mother who was an 'engineer'), followed by 'nurse' (n=11) and 'teacher' (n= 10). There were also five responses of 'physician' and three of 'dentist' among the parents.

Two of the patients were in foster care. None of those in the study had been adopted. Two had been born in another country (Chile and Great Britain) : in the first case to parents born in Chile; in the second the father was from Great Britain and the mother from Sweden. In 17 cases one or both parents had been born outside Sweden (11 mothers and 11 fathers). A total of eight mothers or fathers came from Finland, three from Poland, two each from France, Great Britain, and Chile, and finally one each from the Czech Republic, Yugoslavia, Hungary, the USA, and India. None of the mothers had migrated to Sweden during pregnancy.

Table I: Neonatal problems

FAMILIAL/HEREDITARY FACTORS

Autism spectrum problems

The study group included two pairs of brothers with AS. In 70 cases there was one or more relatives (parents, siblings, grandparents, cousins, grandparents' siblings or their children) who had (or had had) documented major problems in the fields of one or more of (autistic-like) social interaction, reciprocal communication, and/or behaviour/interest patterns. Four of these 70 families contained an individual (one brother, two sisters, and one sibling of a grandparent) who had been diagnosed as suffering from childhood autism. In 28 families there was at least one relative with a full symptom description consistent with a diagnosis of AS, according to the examining doctor and based on the criteria of Gillberg and Gillberg (1989) : five brothers, one sister, 15 fathers, one paternal uncle, three maternal grandfathers, one maternal grandmother, one maternal aunt, and one maternal cousin. The remainder (n=38) had at least one relative with one to five (usually two to four) symptoms of autism as listed in the DSM-IV

Learning problems

Reading/writing disorders and/or dyscalculia had been reported in the relatives of 32 patients. Twelve individuals had a relative with learning disability* (three of whom had Down syndrome). Late language development in relatives was reported to have been present in 11 patients.

Psychiatric problems

Depression requiring medication was noted in the families of 15 individuals, but schizophrenia in only two. Thirteen individuals with AS had a relative with severe alcoholism, and three had relatives with drug addiction. Other psychiatric disorders were found in the relatives of 16 patients (panic attacks, paranoia, psychosis, and unspecified psychiatric problems). Many of those with unspecified psychiatric problems had been hospitalized because of their psychiatric illness and were believed to have had a severe psychiatric illness. However, because the descriptions in the medical records were limited it was not possible to arrive at a more specific diagnosis. In five cases a relative had committed suicide. Anorexia nervosa had been diagnosed in the relatives of two individuals. Attentiondeficit-hyperactivity disorder (ADHD) had been diagnosed in only one relative, but unspecified hyperactivity was noted in the relatives of nine patients. Hypomania had been diagnosed in relatives of three patients and mania in one. Tics occurred in the relatives of four patients, but none was described as having Tourette syndrome. Obsessive-compulsive disorder in a relative was noted in one medical record.

Epilepsy

Seven patients with AS had a relative with epilepsy documented in (non-psychiatric, usually neurological) medical records.

MATERNAL AGE

Mean maternal age at the time of birth of the index child was 28 years 6 months (SD 4y 2mo; n=93; range 18y 6mo to 41y 6mo). Mean maternal age in the general population during this period was 27 years 7 months (SD 10 months; Swedish

Medical Birth Registry 2004, Statistics Sweden 2004).

PREGNANCY COMPLICATIONS

Infections

Ten mothers had experienced a major viral or bacterial infection during pregnancy. These included cases of suspected mumps and herpes infection, and documented influenza A infection, streptococcal tonsillitis, pyelonephritis, and other urinary tract infections.

Other major problems in pregnancy

Second-trimester or third-trimester bleeding had occurred in 10 cases, which was about double the incidence in normal pregnancies (Berglund and Lindmark 2000). Ten mothers had had preeclampsia, which occurs in about 5 to 6% of the normal population (Berglund and Lindmark 2000). Four mothers had had isolated hypertension (1 to 4% in the general population; Berglund and Lindmark 2000). Three had had generalized edema, and two of these had been treated with diuretics. One mother had experienced isolated proteinuria.

Four mothers had had major infertility problems for several years before giving birth to the index child. Nephrolithiasis had occurred in three mothers during pregnancy, and sciatic pain in another two. One mother had suffered from diabetes mellitus, one from Crohn's disease (one of the mothers who had two sons with AS), and one from alcoholism. One mother had fainted twice during the 16th week of pregnancy. There was one twin pregnancy. Alpha-fetoprotein had been elevated in one mother, who had suffered episodes of bleeding in the first and second trimesters and trauma to the abdomen in the sixth month of pregnancy. The child was later found to have ataxic diplegia.

GESTATIONAL DURATION

Duration of pregnancy showed an uneven distribution in comparison with the normal population: 19% (17 of 89) had been born either preterm (36 weeks or less; n=8) or postmaturely (42 weeks or more ; n=9). Mean gestational age at birth was 39.3 weeks (SD 2.6) for the whole group (=89), and 40 weeks (SD 1.3) after the preterm births had been excluded (n = 81). In the general population, 10.7% of children were born either preterm (4.1%) or postmaturely (6.6%) in the years studied (Swedish Medical Birth Registry 2001).

PERINATAL EVENTS OR FACTORS

Data on perinatal factors were available for 96 of the 100 children. 'Normal delivery' (without induction) had occurred in 62 cases (65%). In the general population normal delivery occurs for about 80% of all children (Berglund and Lindmark 2000). In seven cases, induction was the only abnormal aspect of delivery, and in a further two induction occurred in conjunction with other events. The rate of induction in the general population is about 5 to 6% (Berglund and Lindmark 2000).

Caesarean section had been performed in 17 (18%) cases. In the general population, Caesarean section had a mean occurrence of 10.4% of deliveries (range 5.3 to 12%) over the period studied (Swedish Medical Birth Registry 2001). In Sweden, Caesarean section occurs more often in university clinics (about 15%) and the vast majority of the patients in our study were born at a university clinic. Of the Caesarean sections, seven were emergencies due to ablatio placentae (two cases), threatening asphyxia (four cases), and suspected maternal HSVII infection (one case). The 10 remaining section cases were performed electively because of small pelvis in the mother (four cases), cervical fistula (one case), placenta praevia \(one case), breech presentation (three cases), and 'maternal indication' (in one case the mother requested a Caesarean section because she was physically exhausted).

Vacuum extraction had been performed in 10 cases, but one of these ended in Caesarean section because vacuum extraction failed after three attempts. Forceps delivery had been performed in one case (the child born in Chile). In the general population, about 6% of deliveries end with instrumental delivery (Swedish Medical Birth Registry 2001). In this study an additional four vaginal deliveries were noted as abnormal: two because the infant was in breech presentation and another two because of abnormal infant presentation of another kind. Breech presentation has an occurrence rate of about 2 to 3% in the general population (Berglund and Lindmark 2000). In our group five patients were in breech position by the time of delivery: two were born the normal way and three were delivered by Caesarean section, which was in line with rates in the general population (Swedish Medical Birth Registry 2001).

APGAR SCORES

Mean Apgar scores were 8.6 (SD 1.6), 9.6 (SD 1.0), and 9.8 (SD 0.6) after 1, 5, and 10 minutes respectively (n=84). Five children (6%) had an Apgar score of less than 7 at 1 minute, and three of these scored less than 3; that is, they had had a severe postnatal asphyxia at 1 minute. About 3% of newborns have an Apgar score of less than 7 at birth, which is consistent with postnatal asphyxia (Berglund and Lindmark 2000). Five children (6%) had an Apgar score of less than 7 at 5 minutes of age, compared with just above 1% in the normal population (Swedish Medical Birth Registiy 2001).

WHIGHT, LENGTH, AND HEAD CIRCUMFERENCE AT BIRTH

Mean birthweight for the whole group was 3443g (SD 703; n=99) compared with 3544g for the general population of boys born during that period (Swedish Medical Birth Registry 2001). For those born after 37 weeks'gestation it was 3572g (SD 530; n=91), which was almost exactly the same as for the general population at 40 weeks' gestation, about 3590g (Swedish growth charts: Karlberg and Taranger 1976). Four of the participants weighed more than 4500g and eight less than 2500g. This is in accordance with recent general population figures for Scandinavia (Eaton et al. 2001). Birth length was 49.7cm (SD 3.3; n=93) for the whole group, and 50.4cm (SD 2.1; n = 85) when preterm births were excluded (in the general population this figure was 50.7cm). Head circumference, in contrast, was 35cm (SD 2.3; n=85; general population 35cm), or 35.4cm (SD 1.8; n=78) with preterm births excluded (Karlberg and Taranger 1976). Three patients had a hcacl circumference above the 97.5th cellule (Knutltzon et al. 1988).

NEONATAL FACTORS

For 58 of 99 children, some kind of abnormality was noted in their neonatal record (Table I). Twenty-four had been admitted to a paediatric ward because of various problems, but only 11 of these (including 7 preterm births) had been treated for more than 3 days. The duration of paediatric treatment varied between 1 day and 8 weeks. Twenty-two had had hyperbilirubinemia (plasma bilirubin more than 200mol/l), but only 10 had a medical record of being treated with phototherapy. Hyperbilirubinemia occurs in about 10% of newborn infants (Clarkson et al. 1984).

EARLY DEVELOPMENT

Mean age at walking unsupported was 13.8 months (SD 3.2; n=85). In the general population the mean age for independent ambulation is about 12 months (Stanitski et al. 2000, Carruth and Skinner 2002). The children born preterm (1 to 3 months preterm) had started walking at a mean age of 16.9 months (SD 3.8; n=7).

Forty-five of 92 children (49%) for whom fairly detailed data about early language development were available, clearly did not have normal language development at 2 years of age. It cannot be concluded that the remainder had normal language development.

MACROCEPHALUS AFTER THE NEWBORN PERIOD

Twelve of 78 patients (15%) who had their head circumference recorded at the time of diagnosis of AS had a head size corresponding to the 97.5th centile or higher (Knudtzon et al. 1988). Only one of these patients had a head circumference above the 97.5th centile at birth (but data for head circumference at birth for four of the 12 patients were missing). Two further patients had macrocephalus at birth but not at follow-up. (Preeclampsia section was performed in five cases: in three because of a small pelvis in the mother, in one on 'maternal indication', and in one because of breech presentation.) Mean head circumference for our study group was 55.1cm (SD 2.4), which at 11 years 3 months is slightly above the 60th centile compared with Scandinavian growth charts (Knudtzon et al. 1988).

Table II: Comorbid conditions registered in medical records of 100 patients with Asperger syndrome at a mean age of 11 years 3 months

BMI

For 94 patients, height and weight were registered at the time of diagnosis. Twenty-three (24%) had a BMI for age above the 90th centile; 18 of these had a BMI above the 97th centile. Three were below the 10th centile, one of whom was below the 3rd centile.

COMORBIDlTY AS REFLECTED IN MEDICAL RECORD DIAGNOSES

At the time of the diagnosis of AS at the CNC, comorbid conditions were registered in cases in which it was felt that the second (or third) diagnosis contributed equally to the child's impairment, as did the condition of AS itself (Table II). The policy has changed in the past 5 years: now any significant impairing comorbidity is registered as a separate diagnosis. This means that the 33 patients with a registered additional diagnosis represent a very conservative estimate of comorbidity in AS. Six of these 33 had more than one additional diagnosis. Tic disorders were the most commonly registered additional diagnosis (Tourette syndrome in 10 and simple tics in six patients). Only four cases of ADHD were recorded. However, impairing attention deficits were extremely common, even when no additional diagnosis was assigned. The Wechsler profiles (Verbal, Perceptual, Distractibility) showed poor results on the Distractibility factor (Arithmetic, Digit Span, Digit Symbol) in a majority of all patients, suggesting a strong link between AS and ADHD symptoms.

MEDICAL WORK-UP

Neuroimaging

Of the 100 patients, 48 had been examined with each or all of single-photon emission computed tomography (SPECT), magnetic resonance imaging (MRI), or computerized tomography (CT). SPECT had been performed on 16 patients, 15 of whom showed 'abnormal' results (without there being any indication of abnormality on MRI [n=1] or CT [n=6] in those seven who were exposed to these more 'neuroanatomical' investigations). The most common SPECT finding was pronounced hypoperfusion in the temporal region (n= 14), with left- sided predominance in nine patients and right-sided predominance in five. There was additional hypoperfusion in the frontal lobe in six patients, with equal numbers showing left-sided and right-sided predominance. Parietal hypoperfusion occurred in four patients, with predominance for the right and left side in two patients each. The occipital lobe was hypoperfused in seven patients, with a right- sided predominance in five. MRI had been performed on a further eight patients, of whom two showed pathology. One showed small perivascular spaces in the putamen bilaterally and the other had a malformation of the cerebellum. CT had been performed on a further 24 patients, none of whom showed any clear abnormality.

Neurophysiology

Electroencephalograms (EEGs) had been performed on at least 56 patients, 20 of whom (36%) showed abnormal results. However, clear epileptogenic discharges were seen in only 13% of those examined. Of the four patients with clinical epilepsy, only two had an EEG in their medical records at the CNC. Of these two, only one had clearly epileptic EEG discharges.

Auditory brainstem responses had been recorded in 44 patients. Of these, eight (18%) were clearly pathological (seven with prolongation of the I-V interval and three with abnormality of individual waves; two had both types of finding). No more than 2% of the general population of children would be expected to show a similar pathology (Rosenhall et al. 2003).

Chromosomal examinations

All patients had been clinically screened for physical symptoms of fragile X syndrome. All families of children with autism spectrum disorders were offered karyotyping of the child, but if parents were hesitant and there was no major clinical suspicion of a chromosomal disorder the issue was not pursued. Analysis for fragile X had been performed in 52 patients; all were found negative for fragile X and there was no clinical suspicion of this syndrome in the remaining 48 patients. Karyotyping had been performed in 53 patients, five of whom had some kind of abnormality. Three of these were translocations: one with (?;15)(p32.1;q24.1) - mother had the same abnormality and recognized that she had some similarities to her son, but so did the father and he had no chromosomal abnormality; one with (13;17)(q1;p1) -de novo; this patient had a sister with autism, a mother with anorexia, and an aunt (father's half-sister) with learning disability; and one with (5;11) (q13;q13.3)-his father with AS symptoms had a similar karyotype (q14;q14.4). One patient had fragile Y (his father had the same, but there was no report of AS problems of the father in the medical records), and one had 21p+ and was described as 'a copy of his father', as judged by the mother, but the father had no chromosomal abnormality.

ASPERGER SYNDROME SCREENING QUESTIONNAIRE RESULTS

Parents had completed the ASSQ (possible scores range from O to 54) for 79 of the patients. Nine of these had completed a 19-item version of the ASSQ; scores for these werez-converted. Of the 79 ASSQs, 66% scored 19 or above (shown to be strongly associated with a diagnosis of AS or another autism spectrum disorder); and 8\1% had scores of 15 or above. In the general population, ASSQ scores are very much lower (Ehlers and Gillberg 1993). The range of scores in the present group was 5 to 43, and the mean was 23.3 (SD 8.7).

ASSQ scores were divided into groups based on VIQ (n = 77). Patients with a VIQ of 70 to 84 had the highest scores on ASSQ: 26.1 (SD9.5; n=9).Those with a VIQ of 85 to 114 had a mean score of 22.3 (SD 8.9; n=42) and those with a VIQ of 115 and above had a mean score of 24 (SD 8.5 ; n=26).

ASSQ results were also subdivided according to age. The highest scores were found in the group 10 to 15 years old, namely 23.9 (SD 9.2; n=39). The scores were lower in the group 5 to 9 years old, namely 20.2 (SD 10.2; n=31), and in the group 16years and above, namely 20.5 (SD 9.1; n=9).

In 33 patients the ASSQ had been completed by a teacher; the mean result was 28.3 (SD 9.3; a score above 22 has been shown to be strongly associated with AS or another autism spectrum disorder). Twenty-seven of these also had had an ASSQ completed by one or both parents; the mean result of parents' scores was 21.5 (SD 9.5) and that of teachers' scores was 27.0 (SD 9.6) in this group.

EDUCATION OR SPECIAL EDUCATION

The vast majority of patients were of compulsory or comprehensive school age at the time of first diagnosis (n=82). One patient was still in kindergarten and 11 were at preparatory school. One young man studied at university and another had left his university studies unfinished. Two had no organized activity at all and for two individuals data were absent. Because all the schoolboys were of normal or near-normal IQ (15 had an IQ in the range 70 to 85; 85 had an IQ of more than 85) they would have been expected to attend mainstream classrooms without special education support. This applied to only 42 of the 82 individuals who were at school by the time of diagnosis. Thirteen had had a personal assistant (half-time or full-time), and nine had had special education provided by a special education teacher in the mainstream classroom. Another nine were in special needs classrooms or schools. Three patients had been in classrooms for individuals with autism spectrum disorders. Three had had their education in schools for children with learning disability. One had had individual education with the help of an assistant. Two had had some kind of special education or service provision specifically because of their social and communication problems.

NEUROPSYCHOLOGICAL FINDINGS AT ORIGINAL DIAGNOSIS

IQ scores did not differ significantly across the three Wechsler scales (Table III). There were significant differences between VIQ and PIQ on all three scales (p<0.001 for the child scales, p<0.05 for the adult scale). In 50 of the 98 patients (51%) for whom sufficient data were available for the estimation of differences between VIQ and PIQ, VIQ was 15 or more points greater than PIQ; in 30 of these the difference was of 20 points or more. For a further nine, VIQ was 11 to 14 points greater than PIQ. Five individuals had a PlQ that was 20 or more points greater than VIQ; for one the difference was 15 to 19 points, and for eight it was 11 to 14 points. These directional differences in VIQ and PIQ were highly significant (p<0.0001). There were also significant differences across the three Kaufman factors: Verbal Comprehension (11.6, SD 3.2), Perceptual Organization (9.6, SD 2.7), and Distractibility (8.5, SD2.9;p<0.01).

Given the mean IQ of 101 in the study group, we considered all subscale scores of 7 or less as being indicative of an unexpectedly poor performance. (Scaled scores range from 1 to 19, 10 being the mean; therefore, a score of 7 or less represents a rough measure that IQ in that domain is about 70 or less.) Digit Symbol (Coding) was the subtest that most consistently showed a trough (54% of all patients; Table IV). Other subscales on which the patients with AS showed particularly high rates of poor performance were Digit Span, Object Assembly, Arithmetic, and Picture Arrangement (40%, 38%, 35%, and 34% respectively). Low rates of poor performance were seen on Information, Similarities, and Vocabulary (14%, 14%, and 17% respectively). We tested all subscale scores against all other subscale scores (55 comparisons) and found 22 (40%) significant differences (p<0.01 to allow for the effect of multiple comparisons).

Discussion

This study includes the largest clinical series of young male patients with AS published so far. The sample is not necessarily representative of all patients with AS in the general population. However, given that participants were selected from registers held at the only specialist autism spectrum disorder clinic in the region so as to include all patients from a particular age and follow-up period cohort, it is likely that they are representative of clinically referred patients with the diagnosis. The diagnostic criteria used were those of Gillberg and Gillberg (1989), which were based on the clinical descriptions of Hans Asperger's patients. All patients included met the Asperger symptom criteria of the ICD-10/ DSM-IV but few, if any, had perfectly normal development in the first 3 years of life.

A major drawback is the lack of a similarly examined comparison group. However, it is not immediately obvious what would have constituted a reasonable contrast group. A high-functioning group with the diagnosis of autistic disorder would have to be matched for IQ, and such a group was not available on the register, except a handful of patients with an IQ in the range 70 to 90. The clinical diagnosis of autistic disorder is almost never made in individuals with an FSIQ of 100 or more (Gillberg 2002). To make up for the lack of comparison groups, our findings were compared with general population data in instances where such data were available for appropriate comparisons.

Mean age at diagnosis was 11 years 4 months, which is consistent with earlier findings (Howlin and Asgharian 1999). Children with AS, in spite of having problems with social interaction, communication, and behaviour from a very early age in most cases, are often not recognized as having an autism spectrum disorder until well into their school years. This tendency for diagnosis to be 'delayed' is possibly even more pronounced in females with the disorder (Kopp and Gillberg 2003).

Table III: Results on Wechsler scales in 98 patients with Asperger syndrome at time of original diagnosis

Table IV: Results on Wechsler scales subtests at time of original diagnosis

Age at diagnosis was significantly lower in children born in the 1980s than in those born in the 1970s and earlier, which probably reflects the fact that as knowledge about AS in the wider medical community spreads, the age for clinical concern and referral for specialist evaluation goes down. Many of the patients (about 50%) were in special education classrooms - in spite of having a normal or high IQ - or had their own personal assistant or special education teacher at school, which is testimony to the degree of overall clinical impairment.

There were a large number of engineers among the fathers. This is in accord with findings from the UK (Baron-Cohen et al. 2001). Among these were many of the fathers who, according to the medical records, had the full clinical picture of AS. By and large, however, the patients in this study came from a wide variety of socioeconomic circumstances.

More than two thirds of the patients had first-degree, second- degree, or third-degree relatives with major social interaction, communication, or behavioural/interest pattern problems/ peculiarities suggestive of autism, AS, or the 'broader autism phenotype' (Le Couteur et al. 1996). Three of these were first- degree relatives with diagnosed autism, and a further 21 had first- degree relatives who had been described in enough detail in the medical records to suggest the full syndrome of AS. There were also two pairs of brothers with AS in the study. Even though no control group was included, it seems clear that this degree of familial clustering of autistic type symptoms or problems is far beyond what can be expected in the general population. It is also possible that the rate of various kinds of learning disorder (including Down syndrome) might have been increased in the extended families of these patients (more than one in three had a first-degree, second- degree, or third-degree relative with such problems) but, again, given the lack of a control group and the relatively high rate of such disorders in the general population (Gillberg et al. 1992), no definitive conclusions can be drawn in this respect.

Pregnancy complications (including severe infections, second trimester bleeding, and preeclampsia) had occurred for 31% of the patients. It is not possible to draw conclusions about the possible effects that these might have had on later child development. The same holds for perinatal complications, which were extremely common (about 60%). However, overall, severe perinatal complications were rare. In contrast, prematurity and postmaturity seem to have been much overrepresented in comparison with the general population.

Macrocephalus has been reported to be much overrepresented in autism spectrum disorders (Woodhouse et al. 1996, Deutsch and Joseph 2003). The rate of macrocephalus, (defined as head circumference above the 97.5th centile for age and sex) in the present sample (15%), was very similar to that reported recently in a controlled study of 50 patients with AS (Gillberg and de Souza 2002). Macrocephalus at birth did not seem to be more prevalent in our AS group than in the normal population. Thus, the findings may lend support to the notion that in the subgroup of people with autism spectrum disorders, head growth in the first several years of life may be accelerated. Nevertheless, the vast majority of all children with AS do not have macrocephalus (at birth or during late childh\ood).

BMI was not lower than in the general population; in fact, there was a trend in the opposite direction. This finding contrasts with results obtained by some previous authors (Hebebrand et al. 1997, Sobanski et al. 1999, Boite et al. 2002), who all reported higher rates of low BMI in boys with AS. One study (Boite et al. 2002) suggested that low body weight in AS might rather be associated with hyperactivity or other variables not yet known.

Only a minority of the patients had undergone neuroimaging. However, in those on whom such examinations had been performed, the striking discrepancy between the almost universal pronounced degree of dysfunction demonstrated by SPECT and the low rate of abnormality on MRI/CT is worth mentioning. It would seem likely, on the basis of these limited data, that AS is associated with brain dysfunction but not with gross structural changes. There was also a high rate of auditory brainstem response abnormality, indicating brainstem dysfunction in a subgroup of patients (Klin and Volkmar 2003, Rosenhall et al. 2003).

About 1 out of every 10 of those who had had a chromosomal culture had some kind of minor or major abnormality. The relevance of this finding is difficult to judge. For a few patients, one of the parents had the same or similar type of translocation, which could be taken to support the notion of 'balanced translocations'. However, the parents themselves had features of AS in these cases, making conclusions in this respect no more than tentative.

The rate of comorbid diagnoses (including tic disorders and ADHD) registered in the medical records was relatively low in comparison with the high rate of coexistence with other neuropsychiatric disorders/problems in AS found even in total population samples (Ehlers and Gillberg 1993). However, at the time at which the patients in the present study were diagnosed, comorbid conditions were registered in the medical records only when it was felt that the second (or third) diagnosis contributed equally to the child's impairment as did the condition of AS itself. This policy has changed over the past several years, when any significant impairing comorbidity has been registered as a separate diagnosis. This means that the figure of 33% is almost definitely a gross underestimate of the real rate of coexisting impairing problems, such as attention deficits, tics, and depression.

The ASSQ results were very much in line with those reported in previous studies (Ehlers and Gillberg 1993, Clark et al. 1999, Ehlers et al. 1999, Kadesj et al. 1999, Fraser 2002), supporting the view that the ASSQ might be a useful parent questionnaire tool for identifying young patients with a high risk of having an autism spectrum disorder.

The neuropsychological findings were among the most clear-cut of the present study. More than half the group had a considerably higher VlQ than PIQ, whereas only 6% had the converse discrepancy. Children with a much higher VIQ than PIQ are often described as having 'non-verbal learning disability' (Rourke 1988); this in turn has been suggested to be associated with right-hemisphere dysfunction, which has also been proposed to be linked to a particular Wechsler scale subtest profile with poor results on Arithmetic, Digit Symbol (Coding), and Digit Span. This was exactly what we found in our sample. However, some would also count poor results on the Information subtest among the indicators of righthemisphere dysfunction. The study group had their peak performance on this subtest. Interestingly, the SPECT results did not suggest right-hemisphere dysfunction to be more common than left- hemisphere or bilateral dysfunction; if anything, our findings tended to run counter to this notion. The study group also had low scores on Object Assembly (perhaps indicating poor central coherence) and Picture Arrangement (perhaps indicating poor 'theory of mind'). The single most striking finding was the rate of 54% very poor results on the Digit Symbol (Coding) subtest. This subtest is very sensitive to reduced processing speed. A clinical impression of reduced Overall speed' had been noted in many of the medical records reviewed for the present study.

In summary, this study has shown familiarity for autism spectrum disorders to be very common indeed in AS. It has also clearly demonstrated that many individuals with AS qualify for a diagnosis of, or are closely clinically akin to, so-called 'non-verbal learning disability' with markedly higher VIQ than PIQ in many cases. There was some support for temporal lobe dysfunction according to SPECT, but no convincing evidence suggesting right- hemisphere dysfunction (except for the neuropsychological results) as being the most common underlying feature in AS. However, these results must be interpreted with great caution in view of the very small number of patients who had undergone any kind of anatomic or functional neuroimaging in this study.

List of abbreviations

DOI: 10.1017/S0012162204001100

Accepted for publication 3rd March 2004.

* US usage: mental retardation.

References

American Psychiatric Association. (1994) Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Association.

Asperger H. (1944) Die autistischen Psychopathen im Kindesalter. Arch Psychiatr Nervenkrankh 117: 76-136. (In German)

Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E. (2001) The autism-spectrum quotient (AQ): evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians J Autism Dev Disord 31: 5-17.

Berglund A, Lindmark G. (2000) Risk assessment at the end of pregnancy is a poor predictor for complications at delivery. Acta Obstet Gynecol Scand 79:854-860.

Bolte S, Ozkara N, Poustka F. (2002) Autism spectrum disorders and low body weight: is there really a systematic association? Int J Eat Disord 31: 349-351.

Carruth BR, Skinner JD. (2002) Feeding behaviors and other motor development in healthy children (2-24 months). J. Am Coll Nutr 21: 88-96.

Clark T, Feehan C, Tinline C, Vostanis P (1999) Autistic symptoms in children with attention deficit-hyperactivity disorder. Eur Child Adolesc Psychiatry 8: 50-55.

Clarkson JE, Cowan JO, Herbison GP (1984) Jaundice in full term healthy neonates - a population study. Aust Paediatr J 20:303-308.

Deutsch CK, Joseph RM. (2003) Brief report: cognitive correlates of enlarged head circumference in children with autism. J Autism Dev Disord 33: 209-215.

Eaton WW; Mortensen PB, Thomsen PH, Frydenberg M. (2001) Obstetric complications and risk for severe psychopathology in childhood. J Autism Dev Disord 31: 279-285.

Ehlers S, Gillberg C. (1993) The epidemiology of Asperger syndrome. A total population study. J Child Psychol Psychiatry 34: 1327-1350.

Ehlers S, Gillberg C, Wing L. (1999) A screening questionnaire for Asperger syndrome and other high-functioning autism spectrum disorders in school age children. J Autism Dev Disord 29:129-141.

Fraser WI. (2002) Forensic learning disabilities: the evidence base. Executive summary.J Intellect Disabil Res 46 (Suppl 1): 1-5.

Gillberg C. (1989) Asperger syndrome in 23 Swedish children. Dev Med Child Neurol 31: 520-531.

Gillberg C. (1991) Clinical and neurobiological aspects of Asperger syndrome in six family studies. In: Frith U, editor. Autism and Asperger Syndrome. Cambridge: Cambridge University Press. p 122- 146.

Gillberg C. (1999) Neurodevelopmental processes and psychological functioning in autism. Dev Psychopathol 11: 567-587.

Gillberg C. (2002) A Guide to Asperger Syndrome. Cambridge: Cambridge University Press.

Gillberg C, de Souza L. (2002) Head circumference in autism, Asperger syndrome, and ADHD: a comparative study. Dev Med Child Neurol 44: 296-300.

Gillberg IC, Gillberg C. (1989) Asperger syndrome - some epidemiological considerations: a research note. J Child Psychol Psychiatr 30: 631-638.

Gillberg C, Gillberg IC, Rastam M, Wentz E. (2001) The Asperger Syndrome (and high functioning autism) Diagnostic Interview (ASDI) : a preliminary study of a new structured clinical interview. Autism 5: 57-66.

Gillberg C, Gillberg IC, Steffenburg S. (1992) Siblings and parents of children with autism: a controlled population-based study. Dev Med Child Neurol 34: 389-398.

Hebebrand J, Henninghausen K, Nau S, Himmelmann GW, Schulz E, Schafer H, Remschmidt H. (1997) Low body weight in male children and adolescents with schizoid personality disorder or Asperger's disorder. Acta Psychiatr Scand 96: 64-67.

Howlin R Asgharian A. (1999) The diagnosis of autism and Asperger syndrome: findings from a survey of 770 families. Dev Med Child Neurol 41: 834-839.

Kadesj B, Gillberg C, Hagberg B. (1999) Brief report: autism and Asperger syndrome in seven-year-old children. A total population study. J Autism Dev Disord 29: 327-331.

Karlberg P, Taranger J. (1976) Physical growth from birth to 16 years and longitudinal outcome of the study during the same age period. Acta Paediatr Scand 258 (Suppig: 7-76.

Klin A, Volkmar FR. (2003) Asperger syndrome: diagnosis and external validity. Child Adolesc Psychiatr Clin N Am 12: 1-13.

Knudtzon J, Waaler PE, Skjaerven R, Solberg LK, Steen J. (1988) New Norwegian percentage charts for height, weight and head circumference forage groups 0-17years. Tidsskr Nor Laegeforen 108:2125-2135. (In Norwegian)

Kopp S, Gillberg C. (2003) Swedish child and adolescent psychiatric out-patients: a five year cohort. Eur Child Adolesc Psychiatry 12: 30-35.

Le Couteur A, Bailey A, Goode S, Pickles A, Robertson S, Gottesnian I, RutterM. (1996) A broader phenotype of autism: the clinical spectrum in twins. J Child Psychol Psychiatry 37: 785-801.

Leekam S, Libby S, Wing L, Gould J, Gillberg C. (2000) Comparison of ICD-10 and Gillberg's criteria for Asperger syndrome. Autism 4: 11-28.

Mesibov GB, Kunce L, Schopler E. (1998) Asperger Syndrome or High Functioning Autism? Current Issues in Autism, New York: PlenumPress.

Mesibov GB, Schopler E, Schaffer B, Michal N. (1989) Use of the childhood autism rating scale with autistic adolescents and adults. J Am Acad Child Adolesc Psychiatry 28: 538-541.

Miller JN, OzonoffS. (1997) Did Asperger's cases have Asperger disorder? A research note. J Child Psychol Psychiatry 38: 247-251.

Rosenhall V, Nordin Y Brantberg K, Gillberg C. (2003) Autism and auditory brain stem responses. Ear Hear 24: 206-214.

Rourke BP (1988) Socioemotional disturbances of learning disabled children J Consult Clin Psychol 56:801-810.

Sobanski E, Marcus A, Hennighausen K, Hebebrand J, Schmidt MH. (1999) Further evidence for a low body weight in male children and adolescents with Asperger's disorder. Eur Child Adolesc Psychiatry 8: 312-314.

Stanitski DF, Nietert PJ, Stanitski CL, Nadjarian RK, Barfield W (2000) Relationship of factors affecting age of onset of independent ambulation.J Pediatr Orthop 20: 686-688.

Statistics Sweden. (2004) Data on Maternal Age 1971-1977. Orebro: Stastiss Sweden.

Swedish Medical Birth Registry. (2001) Data on Delivery and Birth Statistics in Sweden 1973-1988. Linkoping: The National Board of Health and Welfare.

Swedish Medical Birth Registry. (2004) Data on Maternal Age 1978- 1988. Stockholm: The National Board of Health and Welfare.

Wechsler D. (197'4) Manual of the Wechsler Intelligence Scale for Children-Revised. New York: The Psychological Corporation Limited.

Wechsler D. (1981) WAIS-R Manual. Wechsler Adult Intelligence Scale -Revised. San Antonio, TX: The Psychological Corporation Limited.

WechslerD. (1992) Wechsler Intelligence Scale for Children. Third Edition-Revised. New York: The Psychological Corporation Limited. World Health Organization. (1993) The International Classification of Diseases (Classification of Mental and Behavioural Disorders). Geneva: World Health Organization.

Wing L. (1980) Handicap, Behaviour and Skills Schedule. London: MRC Social Psychiatry Unit.

Wing L. (1981) Asperger's syndrome: a clinical account. Psychol Med 11: 115-129.

Wing L, Potter D. (2002) The epidemiology of autistic spectrum disorders: is the prevalence rising? Ment Retard Dev Disabil Res Rev 8:151-161.

Woodhouse W Bailey A, Rutter M, Bolton P, Baird G, Le Couteur A. (1996) Head circumference in autism and other pervasive developmental disorders. J Child Psychol Psychiatry 37: 665-671.

Mats Cederlund* MD, Department of Child and Adolescent Psychiatry, Queen Silvia's Hospital for Children and Adolescents, University of Gteborg, Sweden.

Christopher Gillberg MD PhD, Department of Child and Adolescent Psychiatry, St George's Hospital Medical School, University of London, London, UK.

* Correspondence to first author at CNC, Queen Silvia's Hospital for Children and Adolescents, Otterhallegatan 12 A, S-411 18, Gteborg, Sweden.

E-mail: mats.cederlund@vgregion.se

Copyright Mac Keith Press Oct 2004

Source: Developmental Medicine and Child Neurology

More News in this Category