A Review of the records at the Autism Research Unit


Paul Shattock
Autism Research Unit, University of Sunderland, UK


Paper presented at the Durham Conference 1998





During the course of the research undertaken at the Autism Research Unit, parents are requested to complete a specially designed questionnaire about their child, including general demographic information, and specific information relating to diagnosis, pre and post - partum circumstances and current medical health conditions. Several groupings emerged from the data suggesting common links in the medical profiles of sub-groups of children. Completed parental reports (n=392) were therefore analysed for data pertaining to four main areas: pre- and post partum conditions, incidence of ‘diagnosed’ gastrointestinal disease, incidence of ‘diagnosed’ viral infection and genetic disease. Results indicated that although only a proportion of children were diagnosed with problems in these areas, the incidence and possible role of these factors in subgroups of children with autism does require more thorough investigation. The results are discussed with reference to a metabolic rationale for the aetiology of autism.





The metabolic hypothesis of autism (Shattock & Savery, 1996; Reichelt et al, 1996; Reichelt et al, 1997) aims to provide a comprehensive account of the aetiology and presentation of the condition, providing both data on the biochemical processes involved (Reichelt et al, 1981; Shattock et al, 1990; Waring & Ngong, 1993; Gardner, 1994; D’Eufemia et al, 1996; Wakefield et al, 1998; Mills et al, 1998) and the application of intervention (Knivsberg et al, 1990; Knivsberg et al, 1995; Whiteley et al, 1999). Not only does the metabolic theory account for differences in levels of functioning, but it also provides the flexibility to accommodate research from many different disciplines, including the possibility that external as well as intrinsic factors may be involved in the development of the syndrome. For many years, parents as well as professionals have speculated as to the nature of these external factors and their possible role. A particularly useful tool for scientific research into the nature of these external components is the epidemiological study of the autistic proband for instances of other known factors or disease states accompanying the autistic syndrome. Subsequent comparison of the aetiology of these diseases for any common thread (causal mechanism and/or clinical presentation) may yield important clues for determining the genetic / biological processes involved in autism, and guidelines for the use of appropriate therapeutic measures.







The files of 391 people were analysed (327 males and 65 female; age range = 37 years 3 months - 1 year 5 months). 209 were diagnosed with autism. 48 were diagnosed with Asperger syndrome. 6 were diagnosed with pervasive developmental disorder and 4 with atypical autism. 49 were diagnosed with autistic spectrum disorder, and 47 with autistic tendencies. 23 were diagnosed with semantic - pragmatic disorder, 5 with dyspraxia and 1 child with global developmental delay. All children had received their diagnosis from registered clinicians, although resources did not allow for a confirmation of diagnosis.


Figure 1: Diagnoses of participants (figure not included in electronic internet version)





The ARU Autism Checklist is a specially designed questionnaire to be completed by parents of people with autism and associated spectrum disorders who are participating in the urinary investigation offered at the Unit. The questionnaire is divided into five main sections to obtain: general Information (personal details of individual and parents), diagnosis, medical notes, family information, early developmental history and food and eating/drinking patterns.





Results revealed that multiple factors emerged from the analysis of completed parental questionnaires, specifically in four distinct areas: early developmental history (peri - and post partum conditions), diagnosed gastrointestinal problems, diagnosed viral infection and genetic disease.



Adverse peri- and post partum conditions.


109 parents (27.9%) indicated the presence of at least one adverse factor taking place either pre- or post partum. The most frequently cited factors to occur during birth were: evidence of the umbilical cord around the child’s neck (n=33; 8.4%), indications of foetal distress (n=25; 6.4%), and the child turning blue (n=10; 2.6%). 10 subjects were born pre-term (2.6%) (5 children = 3 weeks pre-term; 2 children = 4 weeks; 1 child = 5 weeks; 2 children = 6 weeks) and 10 subjects (2.6%) were described as having an aided delivery (e.g. forceps delivery, ventouse / suction method). 8 subjects (2.0%) were delivered by emergency caesarean section and 3 subjects (<0.1%) were breech birth. 12 people (3.1%) were diagnosed with heart complaints including: 9 (2.3%) were found to have problems with irregular heart beat or heart murmurs, 2 (<0.1%) were born with a hole in the heart, and 1 (<0.1%) was diagnosed with a non-specified heart complaint. Other problems also highlighted include: infant pneumonia (n=2; <0.1%), hypoglycaemia (n=2; <0.1%), under-active thyroid gland (n=2, <0.1%), pulmonary embolism (n=1, <0.1%), kidney failure (n=1, <0.1%), and liver infection (n=1, <0.1%). Analysis of pre-natal factors revealed that 8 mothers (2.0%) cited adverse factors during pregnancy. These included: maternal viral infection (n=5; 1.3%), maternal cytomegalovirus (CMV) (n=1; <0.1%), maternal gastrointestinal problems (n=2; <0.1%) (including food poisoning and non-specified gastrointestinal infection), and maternal diabetes (n=1; <0.1%). 3 mothers (<0.1%) suffered with pre-eclampsia during pregnancy, and 3 mothers (<0.1%) reported problems relating to a threatened miscarriage during pregnancy.



Gastrointestinal problems


5 people (1.3%) had been diagnosed as suffering from gastroenteritis at some point in their lives. 3 (<0.1%) had been diagnosed with problems relating to pyloric / duodenal stenosis (narrowing of the outlet from the stomach to the small intestine). Other individual accounts of gastrointestinal problems included: gastro-oesphogal reflux (n=2; <0.1%), dysfunctional gut syndrome / intestinal dysbiosis (n=2;<0.1%), gastric infection (n=2; <0.1%), bowel blockage (n=1; <0.1%), ulcerative colitis (n=1; <0.1%), megacolon and megarectum (n=1; <0.1%), salmonella infection (n=1; <0.1%) and E.coli infection (n=1; <0.1%). Coeliac disease or Crohn’s disease were not diagnosed in any of the participants but were diagnosed as present in other family members (coeliac disease - n=4, Crohn’s disease - n=1). There was also an individual report of ulcerative colitis in one father.


Figure 2: Adverse factors at parturition (figure not included in electronic internet version)



Viral infection


Of all the known diagnosed viral infections experienced by participants during childhood, chicken pox was the most prevalent (n=16; 4.1%). Of these 16 cases, 8 (2.5%) were reported to have contracted chickenpox before symptom onset and diagnosis, and in 2 of these 8 cases this was directly related by the parent/carer to symptom onset. A specific time of chicken pox infection was not obtained from 6 of the 16 records. This was not however, the most common infection diagnosed. 19 people (4.8%) were reported to have contracted mystery viral infections, which were never explained or identified following medical consultation, but 11 cases (2.8%) preceded the onset of symptoms (3 of the 19 cases did not specify the time of infection). 2 of these 11 reports directly related episodes of infection with symptom onset. Insufficient information was given about the nature and symptoms of this infection to form any substantial explanation of it’s possible relevance, although individual reports suggested that the symptoms were similar to influenza. 4 children (1.0%) were diagnosed with meningitis (viral), with another child diagnosed with encephalitis (viral) (<0.1%); all occurring before symptom onset. Other viral infections reported include: measles (n=4; 1.0%) (with a further 2 children contracting the measles virus after vaccination), pertussis (n=3; <0.1%), rubella (n=2; <0.1%) (with an additional report of infection after vaccination), Herpes simplex (n=2; <0.1%), Scarlet fever (n=1; <0.1%), Hepatitis B (n=1; <0.1%) and pneumonia (viral) (n=2; <0.1%). Analysis of parental episodes of viral infection revealed 2 cases of viral meningitis (a sibling and a non-specified other family member).


Figure 3: Viral and chromosomal factors (figure not included in electronic internet version)



Genetic disease.


Four subjects (1.0%) were identified as having chromosomal problems (deletion, rearrangement, inversion). Other factors reported include: infantile Batten disease (a genetic disorder characterised by mental impairment, seizures and gradual loss of motor skills) (n=1; <0.1%), and one subject (<0.1%) being investigated for Saethre-Chotzen syndrome (malformation of head, face and/or skeletal system). Two participants were also diagnosed with anaemia (<0.1%) (one case of hemolytic anemia).





The current results found that over a quarter of parents recorded the presence of at least one adverse factor during parturition (the most common being the umbilical cord wrapped around the child’s neck). Over 15% of participants had a history of diagnosed diseases occurring concurrently with the diagnosis of autism spectral disorder, the majority categorised as viral, gastrointestinal or genetic in origin and nature. Conditions at parturition have been the subject of extensive clinical study (e.g. Folstein, 1985). Previous research has pointed to an increased incidence of adverse perinatal conditions appearing with greater regularity at the birth of children with autism (Coleman & Gillberg, 1985). These have included: perinatal anoxia (Sloan, 1978), induced labours (Mason-Brothers et al, 1987) and breech birth (Finegan & Quarrington, 1980). Reports from the current study describe matching factors. Similarly, findings from previous research investigating the prevalence of diseases co-existing with the autistic syndrome (Ritvo et al, 1990; Rossi et al, 1996) have found that a multitude of diseases (viral, bacterial and genetic) were present and related to the autistic syndrome (through either a genetic link or damage to the CNS), although no causal process was specified. The present study cannot offer any direct support for the relationship between certain adverse factors or disease states detected during the course of the research and the autistic syndrome, although the results are compatible with the hypothesis stating autism to be the result of a metabolic disorder (Shattock & Savery, 1996) (see Figure 4). The diagram shown in Figure 4 suggests that there are two main points of access for the passage of peptides and other compounds through to the Central Nervous System (CNS): the gastrointestinal membrane and the blood-brain-barrier (BBB); with the possibility that a malfunction in one or both of these barriers would lead to a subsequent rise in peptides and other compounds entering the CNS. Physical and infectious insults, including perinatal factors such as anoxia and viral infection such as meningitis, have been suggested to be one such factor causing malfunction in the gastrointestinal membrane (Taylor, Dzus & Colgan, 1998 - hypoxic damage of the intestinal mucosa; de Maar et al, 1996 - increased intestinal permeability in patients with asymptomatic cytomegalovirus) and the BBB (Shattock & Lowdon, 1991).



Figure 4: Autism as a metabolic disorder (Shattock & Savery, 1996).



The possible impact of gastrointestinal (GI) diseases upon intestinal permeability has been catalogued elsewhere (Gupta, 1996; Wakefield et al, 1998). The finding that gastrointestinal disease such as coeliac disease and Crohn’s disease, although not diagnosed in any of the people reviewed, was found in the families of people with autism (and do have a genetic component (Balas et al, 1997)), together with the similarity in mechanisms between for example coeliac disease and the model shown in Figure 4, does suggest that a genetic predisposition for gastrointestinal disease or disease mechanism could exist in autism. Ritvo et al (1990) concluded that there was a relationship between autism and diseases affecting the CNS, albeit inferential, because of the lack of direct neuropathological evidence. The current advances in research examining the mechanism by which these diseases may be involved, superimposed on the proposed metabolic rationale for autism, represents a further step towards providing data for that relationship. Caution must be taken in the interpretation of these results. The findings were derived from a questionnaire completed by parents of people with autism. The study authors did not have access to subject’s complete medical history, and as such cannot be considered to have accounted for all facets of illness or disease states outside the diagnosis. With this in mind however, the number and variation of factors emerging from the current study, suggests that the metabolic hypothesis of autism does provide a platform to explain why the aetiology of autism spectrum disorders may have eluded medical science at present. Individuality in factors such as parturition and the incidence of disease states, combined with the variations in developmental profiles which epitomise the autistic syndrome, would imply that the aetiological factors of autism are going to be different for different individuals. There may however, be a common thread related to how the factors of aetiology are manifested, which could be summarised using the metabolic rationale for autism.





Related Articles:


Home Testing & Sanitizer: