In the early 1990s, thousands of healthy Italian babies in a study of whooping cough vaccines got two different amounts of the preservative thimerosal from all their routine shots.
Ten years later, 1,403 of those children took a battery of brain function tests. Researchers found small differences in only two of 24 measurements and those "might be attributable to chance," they wrote in the February issue of the journal Pediatrics, which was released Monday.
Only one case of autism was found, and that was in the group that got the lower level of thimerosal.
Autism is a complex disorder featuring repetitive behaviors and poor social interaction and communication skills. Scientists generally believe genetics plays a role in causing the disorder; a theory that thimerosal is to blame has been repeatedly discounted in scientific studies.
"Put together with the evidence of all the other studies, this tells us there is no reason to worry about the effect of thimerosal in vaccines," said the new study's lead author, Dr. Alberto Tozzi of Bambino Gesu Hospital in Rome.
The debate over thimerosal and autism has been much stronger in the United States than in Italy, Tozzi said. But the researchers recognized a chance to examine the issue by going back to the children who had taken part in the 1990s whooping cough research.
Randomization sets the new study apart. The random assignment of children rules out the chance that factors other than thimerosal, such as education or poverty, caused the results.
Thimerosal, used in some vaccines to prevent the growth of bacteria and fungus, hasn't been in U.S. childhood vaccines since 2001, except for certain flu shots. Italy and other European nations began removing it in 1999. U.S. health officials recommended the removal of thimerosal as a precaution and to reduce the overall exposure of children to mercury.
Safety regulations still require multi-dose vials of vaccines to contain some type of preservative to prevent the spread of infection from contaminated vials.
The study, funded by the U.S. Centers for Disease Control and Prevention, drew praise from outside experts.
"It's yet another well done, peer-reviewed research study that has demonstrated there is no risk of any neurodevelopmental outcomes associated with thimerosal in vaccines," said epidemiologist Jennifer Pinto-Martin of the University of Pennsylvania.
"This becomes the fourth study to look for subtle signs of mercury toxicity and show the answer was 'no,"' said Dr. Paul Offit, chief of infectious diseases at the Children's Hospital of Philadelphia, the author of a book on autism research and the co-inventor of a rotavirus vaccine.
Tozzi said comparing children with no exposure to thimerosal could have improved the study. "However, if thimerosal were a cause of harm, it is likely that this effect would increase with the administered dose," he said.
The children received either 62.5 micrograms or 137.5 micrograms of ethyl mercury from all their shots during their first year of life. Thimerosal breaks down as ethyl mercury in the body. Before the reduction of thimerosal in the United States, the maximum exposure for infants was 187.5 micrograms of ethyl mercury.
The researchers found the children in both groups scored, on average, in the normal range on 11 tests of memory, attention, motor skills and other brain functions.
Those 11 tests included 24 measured outcomes. Small, but statistical differences were found for only two of those areas, and only for girls. The girls with higher exposure scored worse on a finger-tapping test with their dominant hands, and on a vocabulary test in which they were asked to name common objects.
There was no difference in boys on those outcomes or others. Researchers also found no difference in tic disorders. And the one autism case found in the lower-intake group was likely a chance finding, Tozzi said.