The concern that vaccines might trigger autism was first sparked a decade ago by a British study—since refuted—showing that the measles part of the measles-mumps-rubella shot caused intestinal inflammation and allowed toxins to enter the bloodstream and attack the central nervous system. Other experts speculated that thimerosal, a preservative containing mercury, was the culprit, and it was removed from children's vaccines in 2001 (though most flu vaccines still contain it). The CDC is currently conducting a study of 1,200 children to see whether a thimerosal-autism link really does exist, while some scientists wonder whether a small percentage of cases are, like Hannah Poling's, triggered by multiple vaccinations. Poling was found to have mitochondrial disease, a nerve disorder causing autismlike symptoms that appeared to be brought on by her immunizations. "Mitochondrial disease often occurs in the later stages of a viral illness, and it's proper reasoning to think that vaccines could do what viruses do," in terms of immune reactions, says neurologist Bruce Cohen, a mitochondrial disease expert at the Cleveland Clinic.
A search for markers. The answer could lie in gene studies. "We'd like to know if there are particular markers that signal undetectable diseases like a subclinical mitochrondrial disorder," says Fauci. A 2007 study already found that certain mutations affect a person's susceptibility to fevers after smallpox vaccination; the researchers say they may also predict other responses, like febrile seizures linked to the MMR vaccine.
It's important to keep the risks in perspective: More than 95 percent of kids sail through their shots with, at most, a little fussiness, according to Renee Jenkins, president of the American Academy of Pediatrics. A small percentage experience an overactive immune response like redness, swelling, or pain at the injection site, high fever, or extreme irritability, but severe complications like anaphylactic shock are extremely rare (see graphic). Still, how to account for the fact that once familiar diseases like measles and mumps have become nearly as rare as the adverse reactions?
"It's one thing to take a risk [with a medication] if you actually have a disease, but taking a risk when the goal is prevention of a very rare disease is less tolerable," says Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania. Menactra, for example, protects against bacterial meningitis, which strikes about 1 in 100,000 people per year and kills about 1 in a million. But it also may cause Guillain-Barré syndrome, a temporary but severe paralysis triggered by an overactive immune system, in 1 to 2 teens per million who are vaccinated, according to Iskander.
New vaccines like Menactra and Gardasil pose unknown safety risks because, like any drug submitted for FDA approval, they only need to be tested in several thousand people. "These trials simply aren't big enough to detect rare events that only come to light after 1 million or more doses are distributed," says Iskander. The original vaccine against rotavirus, which causes severe diarrhea and dehydration in infants, was tested on fewer than 1,300 American infants before it was approved in 1998; a year later, after being given to 1.5 million babies, RotaShield was pulled from the market because 13 reported cases of severe intestinal blockages were attributed to the vaccine. The meningitis vaccine Menactra was studied in just over 7,500 people before it was approved in 2005 for adults and kids over age 11. It wasn't until last February, after 15 million doses had been administered, that the CDC announced a "small increased risk" of Guillain-Barré that needs to be studied further.
Hit or miss. The CDC's current system of detecting rare problems is hit or miss. Perhaps the crudest tool is the Vaccine Adverse Event Reporting system, which relies on doctors and patients to file a report if they suspect symptoms have been caused by a vaccine. Many problems filed with VAERS have nothing to do with vaccinations; real adverse events often go unreported. A better monitoring system, the agency's Vaccine Safety Datalink, regularly scans 5.5 million anonymous health records provided by managed care organizations to see whether new vaccines are associated with a spike in certain conditions. Still, even the Datalink database doesn't hold enough teens to definitively prove a causal link between Guillain-Barré and Menactra, says Harvard Medical School professor and vaccine researcher Richard Platt. He and his colleagues recently established a surveillance system that includes 50 million people and are using it to check for Menactra-related Guillain-Barré cases in more than 9 million young people ages 11 to 21. Platt expects to publish results sometime in 2009.