By Suz Redfearn
Special to The Washington Post
Tuesday, July 22, 2003; Page HE01

Scientists writing in the Journal of the American Medical Association have identified the first physical warning sign of autism: small head circumference at birth, followed by rapid and excessive increase in head size during the first year of life.

The researchers' findings, published last week, could lead to earlier identification of autistic children, who now are typically diagnosed at age 3 when teachers and parents begin to notice behavioral problems. Experts say earlier intervention is more beneficial, and the study may help alert doctors to a possible, though not certain, diagnosis of autism in the first months of life.

The research also provides further evidence contradicting the theory that autism could be triggered by vaccinations given just before age 3.

"It's a significant finding that gives us hope as clinicians that we will have, if not a genetic marker, then at least a physiological marker, for identifying a child with autism," said David Holmes, an adviser to the Autism Society of America (ASA) and an adjunct professor of psychology at Princeton University who has studied autism intervention for 30 years. "Currently, all we can do is watch for certain behaviors, most of which don't emerge until a child is older," said Holmes, who is also president of the Eden Family of Services, a New Jersey-based nonprofit organization that provides services to children and adults with autism.

The head-circumference study -- funded by the National Institute of Neurological Disorders and Stroke (NINDS) and underway since 1992 at the University of California, San Diego (UCSD) -- is now expected to become a focus of genetic research into the causes of autism.

The diagnosed incidence of autism is 10 times what it was a decade ago, now striking between two and six children per 1,000, according to the Centers for Disease Control and Prevention (CDC).

Experts say the increased incidence may be attributable to better detection and the recent classification of several autism-like disorders under the umbrella term "autism spectrum disorder." So far, scientists have discovered no cause for autism -- or a reason why it may have spiked in recent years. The disorder is commonly attributed to abnormalities in brain structure or function. Researchers are investigating possible genetic, infectious, metabolic, immunologic and environmental links.

To reach their conclusions, the San Diego researchers compared previously recorded measurements of 48 children who at ages 2 to 5 were diagnosed with autism with those of kids without autism.

Head circumference of the autistic kids at birth was about 5 percent (or two centimeters) smaller than that of the others.

Beginning at two months and ending around 12 months, the autistic children's head growth spiked rapidly. While only 6 percent of infants in the comparison group showed accelerated growth in head circumference between 6 months and 14 months of age, 59 percent of autistic kids did. In the span of six months, the children later diagnosed as autistic underwent the amount of brain growth that normally takes two years. More severely autistic children saw even more overgrowth: those in the study had head sizes smaller than 71 percent of all healthy children at birth; by the end of their first year of life, their heads had grown larger than 95 percent of all normal healthy 1-year-olds.

"Our study shows that something had to be taking place prenatally that was abnormal, which means the beginnings of autism are prenatal," said Eric Courchesne, the study's senior author, a professor in UCSD's department of neurosciences and director of the Center for Autism Research at Children's Hospital and Health Center in San Diego.

Thus, added Courchesne, events like childhood vaccinations, childhood exposure to environmental toxins or pathogens, or unusual gastrointestinal or allergic reactions to food that occur after the undergrowth and subsequent overgrowth do not appear to be plausible causes of the disorder.

But what could it be causing the undergrowth followed by sudden overgrowth?

Researchers don't yet know, but Courchesne said it could reflect either aberrant prenatal conditions or genetic factors that are first expressed in the child's early weeks of life. Courchesne said a follow-up study may provide some answers. "We are just now starting to get that second study underway," he said. "I want to get at least quadruple the size [of the autistic group] and additionally get several hundred normal healthy infant males and females in order check the percent of normals who may show a extreme sudden jump."

Even Courchesne's initial findings have made people in the autism community excited.

"The ability to diagnose early has been one of the holy grails in autism research, and this is certainly a significant step forward toward that goal," said Andy Shih, director of research and programs at the National Association for Autism Research (NAAR), a Princeton, N.J.-based parent group. "If it holds up in a larger study, it may help us to identify potential causes for the disease mechanisms."

There is no quick screening procedure for autism; specialists must spend time observing the child. Autism is expressed differently in each person, but most have problems with communication, both verbal and nonverbal. Many cannot hug a parent or conduct a conversation, and will react badly to changes in routine and engage in repetitive behaviors like spinning or flapping hands. Some with autism can learn and hold jobs, while others require lifelong institutional care. Boys are four times as likely to have the disorder as girls. There is no cure. Much of the research on autism is being done on genetic markers for the disorder and on early diagnosis and intervention. It has been shown that early intervention can serve to rewire bad circuitry in the autistic brain before it gets firmly established.

Courchesne's research dovetails with that of Rebecca Landa, director of the Center for Autism and Related Disorders at Baltimore's Kennedy Krieger Institute and associate professor of psychiatry at the Johns Hopkins School of Medicine. Landa is wrapping up the first study funded by the National Institutes of Health (NIH) to detect autism in children aged 14 months and younger.

Landa plans to publish data showing that many children who will later be diagnosed with autism are showing subtle behavioral signs of the disorder as early as six months after birth. By 14 months, her work shows, a constellation of signs has emerged in most autistic children, making a definitive diagnosis possible. Combined with the new insights into skull growth, the two methods may make early diagnosis more available and dependable.

"Eric Courchesne's work makes my work all the more salient," Landa said.

Courchesne's study does have its limits, Holmes said: The sample size was small, and the measurements were old, taken by health care workers when the children were babies -- not gathered by the study's researchers as the children grew from infancy. Still, say Holmes and others, the new research gives neuroscientists a much better place to start in their quest for the causes of autism.•

Suz Redfearn, a regular contributor to the Health section, wrote about Landa's autism research in our April 15 issue.