06/26/2007 09:58 am ET | Updated Nov 17, 2011

Epidemiology On The Dock

After two weeks of examining arcane biological theories for Michelle Cedillo's autism, the federal "vaccine court" on Monday heard a final day of testimony that centered on two dozen population studies of the possible link between vaccines and autism--evidence that has convinced the world's major health agencies there is no such connection.

The case was put forward by Eric Fombonne, a leading autism expert who described evidence from Japanese, European and North American studies that refute the link. Thimerosal was removed from the vaccine schedule in Canada and Denmark in the 1990s, while Japanese children have not routinely received MMR vaccination since 1992. These situations created experimental conditions for scientists to examine the thesis that removing thimerosal--or the triple MMR shot--would cause autism rates to decline. In no case did this occur. To the contrary, autism prevalence rates have increased in every country where awareness of autism has increased.

Similarly, Fombonne testified that studies have found no link between vaccination and regressive autism or autism-associated irritated bowel syndromes, as proponents of a vaccine link have suggested.

The parents of Michelle Cedillo, a severely autistic 12-year-old, have put forward the theory that her immune system was weakened by thimerosal in vaccines she received as an infant and that measles virus in her MMR vaccine at 16 months delivered a crushing blow to her system. Cedillo's is the first of 4,800 cases that will be examined by the court, a branch of the federal Court of Claims.

There have been no epidemiological studies specifically designed to test this double-shot hypothesis. But Fombonne, on cross-examination by lead attorney Tom Powers, was able to point to data in studies that showed autism rates unaffected in children who had gotten different doses of thimerosal before receiving the MMR shot.

In his cross-examination, Powers attempted to show that since there might be many forms of autism, including one specifically linked to vaccine damage, epidemiological studies wouldn't be sensitive enough to turn up the link. But Fombonne turned back this line of questioning when he pointed out that there are few if any known links between autism symptoms and clearly defined etiology.

Interestingly, when Powers was trying to dispute the significance of studies showing the increase in autism prevalence rates following removal of thimerosal, he borrowed a concept that Fombonne and others have been using to explain the autism "epidemic"--namely that it is changes in diagnostic and educational criteria, better awareness and assessment that have led to the increase in prevalence rates. Without specifically saying so, Powers seemed to be suggesting that increases in autism assessment might have masked declines in vaccine-related autism.

Later, however, Powers turned this logic on its head by accusing Fombonne of attributing the entire increase in autism to diagnostic substitution, better case ascertainment and increased awareness. Fombonne said that one could demonstrate, in many countries, that much of this increase was due to these factors. "Can we say all the increase is entirely accountable by that?

No, we can't. The hypothesis that there might be something in the environment contributing to a small extent to prevalence figures needs still to be entertained. But there is no lead in terms of a good candidate for environmental factor, putting aside immunizations, which in my view has been disproven."

Powers and Fombonne had a lively exchange over the Centers for Disease Control's Verstraeten study (Pediatrics, 2004), which has been the focus of anti-vaccine activists' efforts to demonize the CDC. The study showed a rate of only about 20 autistic children per 10,000, much lower than expected rates. But Fombonne argued there was no indication this under-assessment of autism had affected the study's capacity to determine the relationship of thimerosal to autism.

Friday featured the testimony, by telephone, of British scientist Nicholas Chadwick, who was working as a graduate student with Andrew Wakefield in the mid-1990s when Wakefield began investigating the thesis that the live measles virus in the measles-mumps-rubella shot had caused chronic infections that gave children gut problems and autism. Chadwick tested the gut samples for evidence of measles, he said, but found that nine initially positive tests were, in fact, false positives. He reported the results to Wakefield, but the latter did no mention them in his 1998 Lancet paper that sparked the worldwide MMR controversy. Chadwick's whose name was not included on the paper, said he withdrew in consternation over Wakefield's action.

Jeffrey Brent, a pediatrician and medical toxicologist at the University of Colorado, testified against the thimerosal contribution to autism. He pointed out there was not a single study indicating that any form of mercury could cause serious neurological symptoms in the dosages that were used in vaccines.

Brent countered testimony the previous week by H.V. Aposhian, the plaintiff's witness, who had argued that the dose of thimerosal wasn't relevant to its potential to cause harm. To the contrary, Brent said, the dose very much determines the harm that a substance of this kind can harm. He countered in vitro studies that showed damage to immune cells by thimerosal, explaining that the exposure to a cell in a petri dish was far more likely to cause damage than an equivalent dosage in a living system. Even animal studies often are not replicated in people, he noted.

The bottom line, Brent testified, is "there is no such thing as a poisonous substance. There are only poisonous dosages." In the mercury poisoning cases discussed in the literature and in this case, Brent noted, the blood-mercury levels ranged from 22,000 to 500,000 micrograms per liter of blood. When vaccinated with 12.5 micrograms of ethyl mercury contained in the hepatitis B shot, a baby typically has a blood mercury level of 1.5 micrograms/liter. No mercury at all is detectable within a week.

The Cedillos are claiming that thimerosal damaged Michelle's immune system to the extent that when she was vaccinated with MMR nine months later, it caused brain damage. "That couldn't possibly be the case," Brent testified.

Brent also took on the thimerosal critics' claim that thimerosal levels in children exceeded EPA guidelines. In fact, they exceeded the reference dose for methyl mercury, which is not the same. First, methyl mercury stays in the body longer than ethyl mercury. Second, and more importantly, Brent testified, the reference dose is not a threshold amount above which toxicity occurs. It is a lifetime level of mercury in the blood that is ten times less than the levels that seemed to cause harm to the children of pilot whale-eating moms on an island of Denmark.

"These children, by the way, are clinically normal, but you can find subtle deficits in learning, memory and language," Brent noted. "We're not talking about autism or immune suppression here."

So, while children who received all the scheduled vaccines in the 1990s exceeded the EPA's reference dose on days they were vaccinated, their bodies would no longer exceed the reference dose over a period of time.

Brent concluded by noting that not a single reputable medical or public health organization in the world supports the hypothesis of a link between vaccines and autism.

Final statements were scheduled for Tuesday.