whom had evidence of herpes simplex infection. The two children in whom the etiology of the episode was not discovered made complete recoveries. Additional reports described an autistic syndrome following herpes simplex encephalitis in a 14-year-old girl (Gillberg, 1986) and an 11-year-old boy (Ghaziuddin et al., 2002). The preceding cases were atypical in that the age of onset of autism was between 5 and 14 years; two children with perinatal herpes simplex encephalitis experienced the onset of autism in early childhood (Ghaziuddin et al., 1992). Another series of 14 children with autism included three whose onset of symptoms closely followed episodes of malaria. However, given that malaria is common in Tanzania, where the series originated, this should not necessarily be regarded as evidence of a mechanistic relationship (Mankoski et al., 2006). A single report described a 9-year-old boy who exhibited changes of late-onset autism associated with anti-N-methyl-D-aspartate receptor antibody positive encephalitis; he recovered with monoclonal antibody therapy (Creten et al., 2011).

The foregoing literature suggests that infectious or inflammatory etiologies may underly some cases of autism, although most of the cases described do not meet current diagnostic criteria for autistic disorder, owing to their late onset. Other studies have implicated dysfunction of the innate immune system in the genesis of some cases of autism. Vargas et al. (2005) described a unique pattern of inflammatory changes in brain tissue obtained at autopsy and in cerebrospinal fluid from living patients (Zimmerman et al., 2005) with established diagnoses of autism, using suitable controls and DSM-IV criteria. Herbert (2005) has suggested that the large brains often reported in children with autism in early life could be explained by inflammatory expansion of the white matter that could also contribute to abnormal central nervous system connectivity. The evidence supporting the concept of autism and a neuroimmune disorder has been reviewed recently (Theoharides et al., 2009).

At a minimum, prior to ascribing autism to vaccination, it would be important to rule out known associations with this phenotype. These include both macroscopic and microscopic structural abnormalities of the brain (Casanova, 2007), particularly minocolumnar architecture (Casanova and Trippe, 2009) as well as specific chromosomal and single-gene defects, including a variety of metabolic disorders and inflammatory or infectious antecedants.


Laboratory animals have been studied for decades as a means to understand both normal physiology and pathogenesis of diseases. Throughout this time, it has become apparent that animal models can be very useful, or alternatively noninformative, depending on the question being addressed.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement