Emerging therapy

Chelation therapy

DMSA (succimer), DMPS, EDTA, “heavy metal detox”, provoked urine testing

The idea behind it was tested and failed. There is no trial evidence that it helps an autistic child, there is a documented record of children harmed, and at least one child has died. This is the clearest no on this site.

Where it stands, condition by condition

The same therapy can be well supported for one problem and completely untested for another. This is the single most common place families are misled.

What it is

Chelating agents are drugs that bind metal ions so they can be excreted. Succimer (DMSA), DMPS and the EDTA salts are the ones in use; they are given by mouth, by infusion, and in the private market also as creams, suppositories and unlicensed oral products.

In poisoning medicine these are important drugs with a narrow, well-defined job. The practice described on this page is different: giving them to a child who has no diagnosed poisoning, on the theory that removing metals will improve a developmental condition.

That practice sits almost entirely outside mainstream medicine, is not covered by any paediatric guideline, and is sold privately.

How it is meant to work

The proposal was that autism is caused, or worsened, by mercury — originally from the preservative thimerosal in vaccines, later from heavy metals generally — and that removing it would improve the child.

That hypothesis has been tested and it failed. Thimerosal was removed from childhood vaccines in the early 2000s and autism diagnoses did not fall. Large epidemiological studies found no association. There is no reproducible evidence that autistic children carry a higher body burden of mercury or lead than other children.

So the mechanism is not merely unproven in the way that exosomes or peptides are unproven. The specific claim it rests on was investigated and found to be wrong, and the practice continued anyway.

What has actually been tested

• The Cochrane review of pharmaceutical chelation for autism concluded that there is no clinical trial evidence that it is effective. Twenty years of private practice has not produced a single good trial showing benefit.
• A planned National Institutes of Health trial of succimer in autistic children was cancelled in 2008 on safety grounds, after animal data indicated that the drug could impair cognition in animals that had not been exposed to lead. The agency judged the risk to children unjustifiable in the absence of poisoning.
• Regulators have acted against the products. The United States FDA issued warning letters to companies marketing over-the-counter chelation products for autism and other conditions, stating they were unapproved drugs making unproven claims.
• What exists instead of evidence is testimonial: before-and-after accounts from clinics and parents, collected without controls, without blinding and without comparison to the ordinary course of a developing child.

What we still do not know

• Very little that matters. The question of whether chelation helps autistic children is not an open scientific question being worked on — it is a question that was asked, examined and answered.
• What is genuinely unresolved is narrower: the long-term consequences of repeated mineral depletion in children who were chelated during the years this was popular.

Risks and costs

• Death. In 2005 a five-year-old autistic boy died of cardiac arrest in the United States after being given disodium EDTA; the cause was severe hypocalcaemia. The CDC subsequently reported deaths associated with hypocalcaemia during EDTA chelation. This is not a theoretical risk recited for completeness — it happened to a child receiving this treatment for autism.
• Mineral depletion. Chelators are not selective. They remove zinc, copper, iron and calcium along with anything else, from a child who needs those minerals to grow and to build a nervous system.
• Kidney and liver injury, documented with these agents.
• Repeated intravenous access in a young child: sedation, distress, line infection.
• The non-medical harm. Courses are expensive and open-ended, the child is told something is wrong with their body that is not wrong with it, and the family's attention and money go here instead of to education, communication and therapy that would have helped.

Questions to ask before you agree

1. Has my child a confirmed heavy metal poisoning, diagnosed on an unprovoked blood level, by a toxicologist? If not, there is nothing to chelate.
2. If I am being shown a urine test, was a chelating agent given before the sample — and is the reference range printed beside it for provoked or unprovoked samples?
3. Which randomised controlled trial shows this improves autism? (There is none. The answer to this question is the point of asking it.)
4. Who is monitoring calcium, kidney function and trace minerals, and how often?
5. Would you put in writing that you consider this an evidence-based treatment for my child's condition?

More in this section