Some parents blame their children’s autism on mercury in vaccines.R. Faris/Corbis

The leading US government funder of autism research is drawing fire over its proposal to run a randomized clinical trial of a treatment widely viewed by experts to be useless and potentially harmful, but that is broadly used for autism.

Chelation therapy, in which agents such as dimercaptosuccinic acid (DMSA) are used to bind metal ions in the blood so that they can be excreted easily, is an approved treatment for heavy-metal poisoning. Parents are using such therapy on children with autism because of their belief — which has been scientifically discredited — that mercury from vaccinations caused their children’s condition.

In May, investigators at the US National Institute of Mental Health (NIMH) in Bethesda, Maryland, won approval from the US Food and Drug Administration to use DMSA in a trial of children with autism who are aged four to ten years and have detectable, but not toxic, levels of mercury or lead in their blood. The trial, ‘Mercury Chelation to Treat Autism’, is now under ethics review and has not enrolled any patients.

Critics say the trial will put children at risk for what is certain to be no medical gain. The American Academy of Pediatrics has concluded that there is no justification for giving children DMSA in the absence of very high levels of heavy-metal exposure, notes epidemiologist Ellen Silbergeld of the Johns Hopkins Bloomberg School of Public Health in Baltimore, Maryland. “I don’t know why we have to do this experiment again on children.”

Louis Cooper, a vaccine expert and former president of the American Academy of Pediatrics, says that despite a growing scientific consensus that mercury in vaccines does not cause autism, many parents still think that it is a problem. He says that a well-designed study of chelation in children with autism would “respond to these parents’ deeply held beliefs in the most careful, ethical way”. Cooper explains that, with informed consent, such a study could ensure that the only families to be enrolled would be those already determined to try chelation.

Others argue that the study doesn’t make scientific sense because autism has not been documented as a symptom of high-dose mercury poisoning and that, even if it were, the damage that mercury does to cells is permanent and not reversible; chelation simply prevents additional harm. They say that the NIMH is bowing to political pressure because a growing number of parents with children who have autism use chelation therapy or want to use it. NIMH director Tom Insel denies pandering to families and says that the idea for the study “came up in the first place because we were getting reports that this was a therapy in broad use and there were very substantial questions about both its efficacy and its safety”.

Because chelators bind indiscriminately to metal ions, they can deplete the body of essential metals such as copper, zinc, selenium and calcium. In Portersville, Pennsylvania, in 2005, a 5-year-old boy with autism died from cardiac arrest after being injected with a chelation agent. (The proposed study uses an oral agent.)

Recruitment for the study, initially proposed in 2006 by Susan Swedo, who heads autism research at the NIMH, was put on hold and additional review undertaken after a report in February 2007 showed that chelating agents could cause cognitive problems in rats (D. Stangle et al. Environ. Health Perspect. 115, 201–209; 2007).

Last week Insel said that because the study involves children, carries more than minimal risk and offers no demonstrable benefit to the participants, regulations dictate that it must be referred to a US Department of Health and Human Services panel for ethics approval.

Insel says that the study’s in-limbo status shouldn’t be read as its demise. “It hasn’t been killed by any means. There’s certainly a possibility that this project will go forward.”