Publications

Commentary on 'The New Autism': One Family's Perspective

A.J. Wakefield

Introduction
Christopher (born 30 May 1993), is a child who has been seen and investigated in the Centre for Paediatric Gastroenterology at the Royal Free Hospital. His presentation, although atypical of classical autism, is consistent with that of the great majority of the many children with an autistic spectrum disorder (ASD) that we have investigated for gastrointestinal symptoms. From the outset it is important to accord the initiative for the study of these children to a small but indefatigable cohort of parents, including Christopher's mother. His case is rendered all the more significant by the fact that his grandfather, a respected paediatrician with special training in childhood development and infectious diseases, has reached his own conclusions as to the origins of his grandson's problems.

Clinical features
Broadly, Christopher's life has evolved through four phases: normal early development prior to 18-20 months of age; developmental and physical deterioration that appears to have stabilized somewhat (at least in terms of language) up to 4 years of age; an acute dramatic deterioration in the summer of 1997, followed some time later by a period of recovery. I will review these phases in turn.

Christopher was a normal, healthy, happy child. His mother's story is corroborated by the experienced eye of his grandfather, and the medical and photographic record of his first 18 months. He received all of his immunizations according to schedule.

Associated with Christopher's developmental arrest and subsequent regression (particularly in terms of loss of play, language and social interaction) was a physical deterioration that seemed to be accelerated by an unspecified infectious event at around 20 months. Symptoms that are a recurring theme for these children include coarsening of hair, excessive thirst, sleep disturbances, loss of temperature control, altered pain threshold, dietary restriction and changes in complexion with dark circles around the eyes. Gastrointestinal symptoms are frequently reported and are described later. As in Christopher's case, problems with coordination not infrequently accompany behavioural symptoms, particularly during the initial stages of the encephalopathy, and are variously referred to in the medical records as clumsiness, gait disturbance or ataxia. The child who was walking starts bumping into things or reverts to crawling; the child who could feed himself with a utensil can no longer do so. A history of frequent antibiotic use is typical although this has not been systematically compared using a control population of developmentally normal children. Review of Christopher's records shows that he received one course of antibiotics in the first year of life and 14 further courses before age 3 years, prescribed for ear and upper respiratory tract infections.

Prior to the summer of 1997, Christopher was making some improvements in language although his behaviour was still very challenging. Then he crashed. He underwent a secondary developmental regression that was as rapid as it was dramatic. Prominent features at this stage were aggression and loss of speech and socialization.

Latterly, Christopher has entered a phase of recovery. Rather than accepting the professional advice that she should come to terms with her son's end-stage condition, Christopher's mother steadfastly refused to endorse conventional psychological and exclusively genetic models for her son's autism; in addition to applied behavioural analysis therapy(ABA) she has strived to find treatments that have their rationale in an understanding of the biological basis for Christopher's disease. I have no expertise in ABA, although it is well regarded by many parents. I will, however, discuss dietary intervention in the context of the emerging evidence for associated gastrointestinal pathology.

A gut connection
Evidently, there exists a subset of children with autism who have an immune-mediated gastrointestinal pathology (Furlano et al., 2001; Wakefield et al., 1998, 2000). Clinically, the presence of this pathology may be difficult to infer from what can be somewhat abstruse symptomatology, including sudden, unprovoked temper tantrums, aggression and self-injurious behaviour. The observation that these symptoms may peak prior to bowel evacuation, and are relieved by the latter (observed in the clinical setting of bowel preparation prior to colonoscopy) are a clue that they may reflect visceral pain. Alternatively children may suffer frank diarrhoea, constipation or alternating bowel habit. Abdominal bloating, dietary restriction and change in growth trajectory are also seen. Upper and lower gastrointestinal endoscopy and biopsy are indicated in the presence of such symptoms.

Briefly, the mucosal lesion in affected children represents a novel lymphocytic entero-colitis, characterized by a relatively consistent pattern of lymphocyte and eosinophil infiltrate with a variable degree of acute inflammation (Furlano et al., 2001; Wakefield et al., 1998, 2000). In particular, both CD8+cells and ?dT cells are present at high density. Increase in lamina propria HLA-DR expression with absence of this marker on colonic epithelium, suggests a TH2-skewed response in autistic children. Serum IgG of affected children co-localizes with complement C1q at the epithelial basolateral membrane, suggesting autoimmunity (Torrente et al., 2000). This is not seen in either histologically normal and disease controls, or in children with cerebral palsy. In view of the increasing evidence for gut epithelial dysfunction in autism, these changes are indicative of a specific and possibly important lesion.

Functionally, the pathology is variously associated with dysmotility, with faecal impaction and oesophageal reflux. Increased paracellular intestinal permeability (Horvath et al., 2000), reduced brush border enzyme activity (Horvath, Papdimitriou, Rabsztyn, Drachenberg, & Tildon, 1999), defective sulphation of ingested phenolic amines such as paracetamol (Alberti, Pirrone, Elia, Waring, & Romano, 1999), and cognitive and behavioural responses to exclusion diets (Cade et al., 2000; Knivsberg, Reichelt, Nodland, & Hoien, 1995) have been reported. In an open-label study, the antibiotic vancomycin - given to reduce putative intestinal bacterial dysbiosis - induced striking cognitive responses in children with regressive autism, maintained only during the period of administration (Sandler et al., 2000). These features suggest an enterocolonic encephalopathy that may be analogous, in certain respects, to hepatic encephalopathy (Wakefield et al., 2002).

Systemic immunity
Particularly during the disintegrative phase of his encephalopathy, Christopher suffered recurrent infections and underwent surgical removal of tonsils and adenoids at three and a half years of age. This is consistent with the observation that many affected children suffer from recurrent prolonged infections, particularly of the upper respiratory tract, with a high rate of dietary allergy, eczema and adenotonsillar hypertrophy. Routine immunological investigation frequently reveals lymphopenia, affecting both CD4+and CD8+populations. Consistent with allergic predisposition, IgA is usually in the lower quartile of the normal range (Ludviksson, Elriksson, Ardal, Sigfusson, & Valdimarsson, 1992) and a raised IgG1 and low IgG2 and -4 are relatively common. Functionally, the majority of children assessed show unresponsiveness for all common recall antigens on cutaneous delayed-hypersensitivity testing, in significant contrast with age-matched controls (Murch et al., 1999). Lymphopenia and systemic immune dysregulation have been reported by other groups (Gupta, Aggarwal, Rashanravan, & Lee, 1998; Warren et al., 1996). We have found a history of organ-specific autoimmunity in first-degree family members in 30-50% of cases, as described elsewhere (Comi, Zimmerman, Frye, Law, & Peeden, 1999).

In summary, within the autistic spectrum there is a substantial group of children with what may be a primary autoimmune intestinal pathology. Affected children exhibit dysregulated systemic and mucosal immune responses. Understanding the neurochemical and immune disturbances associated with this intestinal pathology and any associated gut-brain interaction in these children may help to develop rational therapeutic approaches.

The problem in the kitchen: Autism and opioid excess
Christopher's mother reports a clear cognitive and behavioural response to dietary manipulation. Those working in this field will be familiar with the proposition that some forms of autism may arise from the toxic effects of intestinal products on the developing brain, in particular through the inappropriate central activity of dietary-derived opioid peptides (exorphins) from the gut (Reichelt, Ekrem, & Scott, 1990; Reichelt, Knivsberg, Nodland, & Lind, 1994; Sun & Cade, 1999). These include gliadmorphine and β-caesomorphine from the respective substrates, cereal gliadin and bovine casein. Such a mechanism is plausible and supported by recent data, including our own (Reichelt et al., 1990; Wakefield et al., 1998, 2000). Under normal circumstances abundant dietary peptides with potential opioid activity, are digested by brush border peptidases. Recent data indicate that brush border enzyme activity is impaired in autistic children with associated gastrointestinal pathology (Horvath et al., 1999). Increased intestinal permeability and impaired digestive enzyme activity leading to an increased luminal opioid load, may contribute to any systemic excess. This excess may lead to direct effects within the central nervous system (CNS) and indirectly, by influencing the activity or breakdown of endogenous opioids (endorphins) by inhibition of tissue endopeptidases. An opioid-mediated intestinal dysmotility would provide one rational explanation for the clinical paradox of an enterocolitis associated with chronic constipation and reflux oesophagitis.

Aetiology
As with the great majority of cases of autism, the cause of the developmental regression in these children is not known. However, in many children referred to us, parental reports suggest a temporal relationship with exposure to the measles mumps rubella (MMR) vaccine (Wakefield et al., 1998), an association reported by others (Jyonouchi, Sun, & Le, 2001). This clearly represents a highly sensitive medical, legal and political issue, which has made objective assessment difficult.

One problem is the timing of regression - all be this unusual in typical autism - which often overlaps with that of routine vaccine administration. The frequently delayed and stuttering onset of symptoms makes exact timing imprecise, and retrospective epidemiological analysis particularly difficult (Spitzer, Aitken, Dell'Aniello, & Davis, 2001). In contrast, late regression beyond the age of 3 years, and further episodes of sustained regression in the same child, are very unusual. This particular feature of Christopher's illness, a bimodal pattern of regression, provides a clue to possible aetiology. His secondary regression in the summer of 1997 followed a booster MMR vaccine. Of our first 60 cases studied in depth (Wakefield et al., 2000), 11 children received a booster measles-containing vaccine (MMR or MR). Nine suffered developmental regression - noted contemporaneously - after both doses, while one suffered late regression after the second dose only. Such re-challenge cases, if they are substantiated, would constitute strong evidence of a causal association (Stratton, Gable, Shetty, & McCormick, 2001). More recently, measles virus RNA and protein have been detected in reactive ileal lymphoid tissue in affected children (including Christopher) at significantly higher prevalence than paediatric controls (Murch et al., 1999; Uhlmann, Martin, Shiels, Wakefield, & O'Leary, 2002), and Singh, Lin, and Yang (1998) have linked atypical serological responses to MV and myelin basic protein (MBP) to autism.

Although these observations cannot confirm a causal link, it is interesting to note that there are precedents for a role for measles, mumps and rubella viruses, in their natural form, in childhood developmental disorders, including autism (Deykin & MacMahon, 1979), disintegrative disorder (Rutter, Taylor, & Hersor, 1994) and developmental regression (Weibel, Caserta, & Benor, 1998).

The temporal association of regression with combined MMR vaccine rather than the monovalent (MV) measles vaccine raises the possibility of a compound effect of the concurrent exposure. Certainly MV is capable of exhibiting interference phenomena with other viruses, including as part of MMR vaccine (Bunyak, Weibel, Whitman, Stokes, & Hilleman, 1969; Minekawa et al., 1974). Such interference may modify the immune response and increase the risk of persistent infection. The possibility that polyvalent MMR might represent a compound risk for mucosal immunopathology is consistent with the data of Montgomery, Morris, Pounder, and Wakefield (1999), who found that close temporal exposure to measles and mumps was a risk factor for later inflammatory bowel disease.

It is important to consider, for example, in the design of any epidemiological study, why some children might react aberrantly to a vaccine when the great majority do not. Possible associated risk factors, that are frequently reported include: vaccination during an intercurrent infection or following recent antibiotic use, a history of atopy including food allergy, exposure to multiple vaccines concurrently, and a strong family history of autoimmune disease, as reported elsewhere (Comi et al., 1999; Gupta et al., 1998).

Maternal immunity: A role in pathology?
Transfer of a mother's immune memory, via placenta and breast milk, protects her offspring from infection and modifies the latter's immune response, allowing him/her to encounter infection and develop natural immunity (Zinkernagal, 2001). What are the implications for an offspring exposed to an infection when the mother's immune system has been inadequately instructed or has failed to heed the lessons of a specific infectious exposure? It is likely that the component viruses of MMR may be transmitted to foetus or infant; for example, rubella virus may cause in uteroinfection and is excreted in breast milk. In the absence of passively acquired specific antibody the naïve infant immune system will be unable to prevent such viruses from establishing infection; subsequent subversion of viral clearance by a profoundly immunomodulatory virus (such as measles) and an immature and inadequately primed cellular immune system, may create the necessary conditions for chronic immunopathology in the infant.

Christopher's mother received two doses of MMR as an adult, prior to Christopher's conception. Despite this, she failed to seroconvert to rubella. The role of either these exposures, or the genetic profile that defines a non-seroconverter status, to subsequent immunopathology in the offspring, are not known, although it would be naïve to dismiss them from a biological standpoint. Dr F.E. Yazbak, Christopher's grandfather, has reported a high rate of developmental disorders - principally autism - in the offspring of rubella-susceptible mothers who were revaccinated in the postpartum period with either the MMR or the monovalent rubella vaccine (Yazbak & Lang-Radosh, 2001). Of 60 such mothers, 45 have children diagnosed with ASD; another 10 women have children with autistic symptoms, attention deficit disorder/attention deficit hyperactivity disorder (ADD/ADHD) or other developmental delays. By the authors' own admission, there are biases in this study that limit the extent of any conclusions that can be drawn. As such, this should be seen as an hypothesis-generating observation, certainly demanding of a follow-up case-control study using, for example, the Center for Disease Control and Prevention's (CDC) registry of women receiving rubella vaccine in pregnancy.

Finally, beyond the clinical pathology there are, as for so many of these children, parents who are devoted, articulate and dedicated, while at the same time, angry, frustrated and distrustful of a medical system that they feel has failed their children. For my colleagues and I it has been, and remains, immensely rewarding and instructive - gratification that is offset by the knowledge that, for children like Christopher, so much precious time has been lost to territorial and political imperative. For medicine this episodemight stand as a lesson in humility - in deference to the potent evolutionary forces of parental instinct and recidivist microorganisms. To the child, 'these things' do not, as Christopher's mother was told 'just happen'. As a first step, there is no substitute for listening.

References
Alberti, A., Pirrone, P., Elia, M., Waring, R.H., & Romano, C. (1999). Sulphation deficit in 'low-functioning' autistic children: A pilot study. Biological Psychiatry,46, 420-424. Buynak, E.B., Weibel, R.E., Whitman, J.E., Stokes, J., & Hilleman, M.R. (1969). Combined live measles mumps rubella virus vaccines. Journal of the American Medical Association, 207, 2259-2262.

Cade, R., Privette, M., Fregly, M., Rowland, N., Sun, Z., Zele, V., Wagemaker, H., & Edelstein, C. (2000). Autism and schizophrenia: Intestinal disorders. Nutritional Neuroscience, 3, 57-72.

Comi, A.M., Zimmerman, A.W., Frye, V.H., Law, P.A., & Peeden, J.H. (1999). Familial clustering of autoimmune disorders and evaluation of medical risks in autism. Journal of Child Neurology, 14, 388-394.

Deykin, E.Y., & MacMahon, B. (1979). Viral exposure and autism. American Journal of Epidemiology, 109, 628-638.

Furlano, R., Anthony, A., Day, R., et al. (2001). Quantitative immunohistochemistry shows colonic epithelial pathology and T cell infiltration in autistic enterocolitis. Journal of Pediatrics, 138, 366-372.

Gupta, S., Aggarwal, S., Rashanravan, B., & Lee, T. (1998). Th1- and Th2-like cytokines in CD4+ and CD8+ T cells in autism. Journal of Neuroimmunology, 85, 106-109.

Horvath, K., Papdimitriou, J.C., Rabsztyn, A., Drachenberg, C., & Tildon, J.T. (1999). Gastrointestinal abnormalities in children with autistic disorder. Journal of Paediatrics, 135, 559-563.

Horvath, K., Zielke, R.H., Collins, R.M., Rabsztyn, A., Mederios, L.A., & Perman, J. (2000). Secretin improves intestinal permeability in autistic children. Journal of Paediatric Gastroenterology and Nutrition, 31(Suppl. 2), A112.

Jyonouchi, H., Sun, S., & Le, H. (2001). Proinflammatory and regulatory cytokines associated with innate and adaptive immune responses in children with autistic spectrum disorders and developmental regression. Journal of Neuroimmunology, 120, 170-179.

Knivsberg, A.-M., Reichelt, K.-L., Nodland, M., & Hoien, T. (1995). Autistic syndromes and diet: A follow-up study. Scandinavian Journal of Educational Research, 39, 223-236.

Ludviksson, B.R., Elriksson, T.H., Ardal, B., Sigfusson, A., & Valdimarsson, H. (1992). Correlation between serum immunoglobulin A concentrations and allergic manifestations in infants. Journal of Pediatrics, 121, 23-27.

Minekawa, Y., Ueda, S., Yamanishi, K., Ogino, T., Takahashi, M., & Okuno, Y. (1974). Studies on live rubella vaccine V. Quantitative aspects of interference between rubella, measles and mumps viruses in their trivalent vaccine. Biken Journal, 17, 161-167.

Montgomery, S.M., Morris, D.L., Pounder, R.E., & Wakefield, A.J. (1999). Paramyxovirus infections in childhood and subsequent inflammatory bowel disease. Gastroenterology, 116, 796-803.

Murch, S.H., Anthony, A., Thompson, M., et al. (1999). Ileo-colonic lymphoid nodular hyperplasia is associated with immunodeficiency in children with developmental disorders. Gut, 44, A127.

Reichelt, K.-L., Ekrem, J., & Scott, H. (1990). Gluten, milk proteins and autism: Dietary intervention effects on behavior and peptide secretion. Journal of Applied Nutrition,42, 1-11.

Reichelt, K.-L., Knivsberg, A.-M., Nodland, M., & Lind, G. (1994). Nature and consequences of hyperpeptiduria and bovine casomorphins found in autistic syndromes. Developmental Brain Dysfunction, 7, 71-85.

Rutter, M., Taylor, E., & Hersor, L. (1994). Child and adolescent psychiatry. London: Blackwell Scientific.

Sandler, R.H., Finegold, S.M., Bolte, E.R., et al. (2000). Short-term benefit from oral vancomycin treatment of regressive-onset autism. Journal of Child Neurology, 15, 429-435.

Singh, V.K., Lin, S.X., & Yang, V.C. (1998). Serological association of measles virus and human herpesvirus-6 with brain autoantibodies in autism. Clinical Immunology and Immunopathology, 89, 105-108.

Spitzer, W.O., Aitken, K.J., Dell'Aniello, S., & Davis, M.W.L. (2001). The natural history of autistic syndrome in British children unexposed to MMR. Adverse Drug Reactions and Toxocological Reviews,20, 1-4.

Stratton, K., Gable, A., Shetty, P., & McCormick, M. (2001). Immunization safety review: Measles-mumps-rubella vaccine and autism. Washington, DC: National Academy Press. Available: www.iom.edu/imsafety

Sun, Z., & Cade, J.R. (1999). A peptide found in schizophrenia and autism causes behavioral changes in rats. Autism, 3, 85-95.

Torrente, F., Machado, N., Perez-Machado, M., et al. (2000). Enteropathy with T cell infiltration and epithelial IgG deposition in autism. Journal of Paediatric Gastroenterology and Nutrition,31 (Suppl. 2), A546.

Uhlmann, V., Martin, C., Shiels, O., Wakefield, A.J., & O'Leary, J.J. (2002). Possible viral pathogenesis of a novel paediatric inflammatory bowel disease. Molecular Pathology(in press).

Wakefield, A.J., Anthony, A., Murch, S.H., et al. (2000). Enterocolitis in children with developmental disorders. American Journal of Gastroenterology, 95, 2285-2295.

Wakefield, A.J., Murch, S.H., Anthony, A., et al. (1998). Ileal-lymphoid nodular hyperplasian non-specific colitis, and pervasive developmental disorder in children. Lancet,351, 637-641.

Wakefield, A.J., Puleston, J., Montgomery, S.M., Anthony, A., O'Leary, J.J., & Murch, S.H. (2002) Entero-colonic encephalopathy, autism and opioid receptor ligands. Alimentary Pharmacology and Therapy, (in press).

Warren, R.P., Singh, V.K., Averett, R.E., et al. (1996). Immunogenetic studies in autism and related disorders. Molecular and Clinical Neuropathology, 28, 77-81.

Weibel, R.E., Caserta, V., & Benor, D.E. (1998). Acute encephalopathy followed by permanent brain injury or death associated with further attenuated measles vaccines: A review of claims submitted to the national vaccine injury compensation program. Paediatrics, 101, 383-387.

Yazbak, F.E., & Lang-Radosh, K. (2001). Adverse outcomes associated with postpartum rubella or MMR vaccine. Medical Sentinel,6, 95-99.

Zinkernagel, R.M. (2001). Maternal antibodies, childhood infections and autoimmune disease. New England Journal of Medicine, 345, 1331-1335.

Return to Top

Join our email list

Dr. Andrew Wakefield | Dr. Arthur Krigsman | Dr. Bryan Jepson | Dr. Doreen Granpeesheh | Kelly Barnhill, MBA, CN, CNN