ASSAULT ON AUTISM

Beth Crowell remembers the day in 1989 when her triplets, Casey, Andrew, and Erin, were about 15 months old. Crowell put Erin down on the floor to crawl. "But she just sat there, fixated on the red shag carpeting," says the Housatonic, Mass., mother of four. The toddlers were often sick, and "none of them made eye contact," Crowell recalls. A medical evaluation was devastating: All three babies had autism. Children with autism typically have trouble communicating, interacting socially, and controlling their behavior. Those most severely affected seem to live in a world of their own. Various treatments sometimes reduce symptoms, especially if children are diagnosed early. But there is no cure for autism, which has baffled the medical community since the disorder was first described in 1943.

The Centers for Disease Control and Prevention (CDC) recently estimated that 1 of every 1,000 children may have autism, or 1 in 500 if those with autism-related disorders such as Asperger syndrome are included.

For years, Crowell combed the medical literature trying to figure out what might have gone wrong in her triplets. She doubted that a genetic mutation was solely responsible. Crowell came to suspect that terbutaline, a drug she had taken during pregnancy to prevent premature labor, might have played a role.

A team of researchers in Baltimore found her assertion plausible. They knew of experiments showing that rats exposed to terbutaline before birth had brain abnormalities. More recently, they completed a yet-unpublished clinical study that found a higher-than-expected incidence of autism in both children in sets of fraternal twins whose mothers also took terbutaline during pregnancy. The investigators are Andrew Zimmerman of the Kennedy Krieger Institute, independent researcher Susan Connors, and researchers at Johns Hopkins Medical Institutions.

This research reflects a shift in scientific thinking about what causes autism, and a push to look harder at potential environmental influences.

"For years, the emphasis [in autism research] has been on neurobiology and genetics," says Michael Cuccaro of Duke University in Durham, N.C., a psychiatrist specializing in autism. "It was thought we could identify the causes if we could understand those connections, but we're still left searching for causes. There was a missing piece of the puzzle, which was the environment."

Some scientists are convinced that environmental factors must be at play because autism appears to be increasing rapidly. They argue that genetic factors alone can't account for such rapid growth. For example, California Department of Developmental Services data show that autism cases in the state more than doubled between 1987 and 1998. Scientists from the CDC found a 10-fold increase in autism in Atlanta from 1986 to 1996.

However, perinatal epidemiologist Lisa Croen of Kaiser Permanente's Division of Research in Oakland, Calif., says that some of the apparent increase probably reflects inconsistencies in data-reporting methods and changes in diagnostic criteria over the past 15 years.

A highly controversial piece of the autism picture in the past decade has been the issue of whether childhood vaccines can trigger the disorder. Suspicions arose because autism is often diagnosed around the time when children receive a series of routine vaccinations. A mercury-based vaccine preservative called thimerosal seemed a likely culprit.

In the past 5 years, thimerosal has been phased out of most pediatric vaccines, and a committee of the National Academies' Institute of Medicine in Washington, D.C., has consistently found no conclusive evidence for an autism-vaccine connection. But two new studies are reviving the argument that thimerosal can act as an environmental factor promoting autism in certain children.

Geneticist Thomas Wassink of the University of Iowa in Iowa City says that most researchers studying the genetics of autism now assume that the disorder is caused by interplay between genes and factors from outside the body. He speculates that environmental factors trigger the disorder in children in whom 5 to 15 genes have created an underlying susceptibility. Gene hunters are homing in on several autism-related genes, he says.

Much of the current research on autism is in early stages. Payoffs in treatments or preventive measures are likely to be years to decades away.