Autism on the Rise

The rate of autism is rising. The number of reported cases has increased 10-fold in the last few decades, from 1 in 2,500 in the 1970s to 1 in 250 in the 1990s. Researchers are looking everywhere for the reason--from drinking water to the womb--with no clear-cut answer to date. In part, the increased incidence can be attributed to a broader definition of autism, which now includes milder forms of the disorder,1 as well as to better diagnostics and greater public awareness.2 But scientists don't know if these reasons explain the entire increase. At a late-April conference entitled "Autism: Deciphering the Puzzle," developmental biologists, geneticists, and neurobiologists gathered to talk about this complex disease. While scientists attending the conference at the California Institute of Technology could not explain the huge jump in incidence, they did voice some hope: there is progress in understanding the condition's biological basis, along with the development of experimental models, and it could lead to better treatments. A Neurologist's Nightmare Described by one participant as "a neurologist's nightmare," autism affects a cohort of complex behaviors, involving impaired language development, the inability to interact socially, and repetitive and restrictive behaviors. Generally, autism is diagnosed in children aged 2 and older because the behaviors can't be observed at an earlier age. But researchers are working on improving the tools for diagnosing autism, particularly in young children. As with many neurological diseases, including apraxia and fetal alcohol syndrome, early intervention improves the prognosis. Patricia Rodier, professor of obstetrics and gynecology at the University of Rochester, described an early intervention test at the conference that is used with infants. Devised by Susan Bryson of York University in Toronto, this test measures a child's ability to shift focus from one stimulus to another. In the first part of the test, one light is turned on, and then as a second light is turned on, the first is shut off. All children will shift their focus from the first to the second light. In the second part of the test, the first light is left on. Here, normal children will disengage from the first to the second light, but autistic children cannot make that shift. Rodier showed dramatic video footage of a 5-year-old autistic child attempting this task. A look of panic came across the child's face, as he realized that he couldn't take his eyes off the first light. In contrast, a severely retarded 6-month-old could refocus her gaze with no problem. One research tool that has eluded workers in this field is an experimental animal model for autism, because the diagnosis relies on human terms such as eye contact, facial expressions, and speech. Bryson's test may be just the ticket; it provides a glimpse into the nervous system but doesn't require learning or intelligence. In the Mind's Eye Many parents of autistic children suffer a heart-breaking burden: often, their youngsters are not emotionally connected to them. A recent University of Washington study, presented to the Society for Research in Child Development, showed that autistic children don't respond to faces, which could explain the emotional distance from their parents. Measuring brain activity with a net of external electrodes, Geraldine Dawson, director of the University of Washington Autism Center, found that the brains of autistic children were electrophysiologically silent when shown pictures of their mothers, while they did respond to other pictures, such as their favorite toys. At the Caltech conference, Eric Courchesne, professor of neurosciences at the University of California, San Diego, presented live, deep-brain scans that back Dawson's work. Using imaging techniques, Courchesne and co-workers showed that the fusiform gyrus, the part of the brain involved with face recognition, is not active when autistic children are shown pictures of faces. Instead, in autistic children, each child displays a different electrophysiological pattern. Why there is decreased activity in the fusiform gyrus is unknown, but Karen Pierce, the principal investigator, offers several explanations. "One possibility is that limited exposure to faces in patients with autism (perhaps due to innate preference, biases of processing style, or learning) results in an underdevelopment or maldevelopment of face-processing systems. Another reason is that the neural substrates involved in face processing (e.g., fusiform gyrus or amygdala) is abnormal in autism." She made her comments after the conference. Thalidomide Revisited Most everyone is familiar with the haunting pictures from the 1960s of so-called thalidomide babies, who were born with deformed limbs after their mothers took this sedative while pregnant. Overlooked in the early studies is that many thalidomide children are autistic--missed, no doubt, because their parents and doctors were dealing with the more obvious and dramatic limb deformities. But in 1994, Swedish researchers reported the surprising finding that thalidomide children had a high incidence of autism,3 and for developmental biologist Rodier, this was helpful news, because the Swedish researchers identified when, during their pregnancies, the women took the drug. Armed with these results, Rodier is developing an animal model for the kinds of brain abnormalities observed with autism, since she knows exactly when in the developmental program she needs to intervene. And she has the environmental agents to do it. Though thalidomide doesn't affect rodents in the same way as humans, Rodier has found that valproic acid, a common anti-seizure drug known to induce autism, causes brain damage in rodents, and precisely in the places expected, based on what's known about this disease. Meeting organizer Paul Patterson describes a promising experimental system being developed in his lab at Caltech that is based on studies linking prenatal infections and immune dysfunction with mental illness. In developing a system that assesses how interactions between the immune system and nervous system affect brain development, Patterson has observed autistic-like behaviors in mice born to mothers exposed to influenza during pregnancy using a battery of behavioral tests. In one experiment, mice are dropped into a box. While normal mice move around the box, frequently stretching to sniff the environment, mice born to infected mothers stay in the corner, clinging to the wall and sniffing only occasionally. Using this test and others, Patterson intends to pick apart the immune response to see what proteins or factors might be involved in explaining the offspring's odd behavior. The evidence for a genetic component to autism is overwhelming and indisputable.4 Consider, for example, that the parents of an autistic child are more likely to have a second autistic child, as opposed to those who have unaffected children. In a normal family, the likelihood is 0.4 percent; if there already is an autistic child, the odds grow to 2 to 3 percent. With identical twins, if one is autistic, the likelihood that the second will have some form of autism is a staggering 90 percent; with fraternal twins, the odds shrink to 2 to 3 percent. Strong sentiment exists, particularly among the parents of autistic children, that environmental factors also are involved here. One popular theory links immunizations, particularly measles, mumps, rubella (MMR), with the onset of autism. Researchers at the Caltech meeting summarily dismissed this notion, because scientific evidence, they say, does not exist. Several recent studies, including an Institute of Medicine report issued April 23, did not show a correlation between MMR immunization and autism.5 In a study published last February in the British Medical Journal,6 there was no abrupt increase in the incidence of autism after the MMR vaccine was introduced. What about other environmental factors? Eric Hollander, professor of psychiatry at the Mt. Sinai School of Medicine and Clinical Director of the Seaver Autism Research Center in New York City, is looking at several factors. Noting that an unusually large number of women at his clinic had pitocin-induced labor, Hollander is currently conducting a survey of some 58,000 births recorded in a national perinatal database to look for a connection between that drug and autism. Hollander also is investigating the possibility that an infectious agent is involved. He has found that in autistic children, a high expression level exists of a particular B-cell marker, D8/17, which is associated with altered sensitivity to streptococcus A. Though much research has been carried out, there is still no complete answer, or answers, as to why more autistic children exist today. "Cautious folks will say that it is really impossible to say for sure what the reason is at this point," Patterson says. "Given the broadening of the diagnostic criteria, the heightened recognition of the disorder by doctors, and the fact that parents only get state funds to help with special education ... if the child has a diagnosis of a severe disorder such as autism, we'll only be able to tell if this is a true rise in incidence after the dust has settled." Laura DeFrancesco can be contacted at defrancesco1@earthlink.net. References 1. E. Fombonne, "The epidemiology of autism: a review," Psychological Medicine, 29: 769-86, 2000. 2. C. Lord et al., "Autism spectrum disorders," Neuron 28(2): 355-63, 2001. 3. K. Strömland, K. et al., "Autism in thalidomide embryopathy: A population study," Developmental Medicine and Child Neurology, 36: 351-6, 1994. 4. A. Bailey, "Autism as a strongly genetic disorder: evidence from a British twin study," Psychological Medicine, 25:63-77, 1995. 5. "Immunization Safety Review: Measles-Mumps-Rubella Vaccine and Autism," Institute of Medicine, April 23, 2001. 6. J.A. Kaye et al., "Mumps, measles, and rubella vaccine and the incidence of autism recorded by general practitioners; a time trend analysis," British Medical Journal, 322:460-3, 2001.

Autism on the Rise

Interested in reading more?

Become a Member of