New research further debunks any link between measles vaccine and autism, work that comes as the United States is experiencing a surge in measles cases fuelled by children left unvaccinated.

Years of research with the measles, mumps and rubella vaccine, better known as MMR, have concluded that it doesn't cause autism.

Still, some parents' fears persist, in part because of one 1998 British study that linked the vaccine with a subgroup of autistic children who also have serious gastrointestinal problems. That study reported that measles virus was lingering in the children's bowels.

Only now have researchers rigorously retested that finding, taking samples of youngsters' intestines to hunt for signs of virus with the most modern genetic technology. There is no evidence that MMR plays any role, the international team — which included researchers who first raised the issue — reported Wednesday.

"Although in fact there was evidence that this vaccine was safe in the bulk of the population, it had not been previously assessed with respect to kids with autism and GI complaints," said Dr. W. Ian Lipkin of Columbia University College of Physicians and Surgeons, who led the work published in PLoS One, the online journal of the Public Library of Science.

"We are confident there is no link between MMR and autism," Lipkin said.

Added co-author Dr. Larry Pickering of the Centers for Disease Control and Prevention: "I feel very certain that it is a safe vaccine."

Measles, a highly infectious virus best known for its red skin rash, once routinely sickened thousands of children a year and killed hundreds, until childhood vaccinations made it a rarity in this country.

But so far this year, the U.S. has counted 131 measles cases, the most in a decade. Most patients were unvaccinated. Some were infants too young for their first MMR shot, but nearly half involved children whose parents rejected vaccination, the CDC reported last month.

No one knows just how many autism patients also suffer gastrointestinal disorders, pain that they may not be able to communicate. But Lipkin said that by some estimates, up to a quarter may be affected.

The MMR fear was that the vaccine's weakened measles virus somehow lodged in and inflamed intestines, allowing waste products to escape and reach the central nervous system, Lipkin said. So his team had two questions: Does measles virus really persist in children with both disorders and not other youngsters? And did vaccination precede the GI complaints which in turn preceded autism?

Researchers studied 25 children with both autism and GI disorders, and another 13 children with the same GI disorders but no neurologic problems. The youngsters — the average age was five — all were undergoing colonoscopies for their GI conditions anyway, allowing tissue samples to be tested for genetic traces of measles virus. All had been vaccinated at younger ages.

The tests uncovered traces of measles genetic material in the bowels of one boy with autism — and one boy without autism. That doesn't prove virus never temporarily lodged in more children, but it contradicts the earlier study that raised concern.

Nor was there a relationship with vaccine timing: Just five of the 25 autistic children had MMR precede GI complaints that in turn preceded autism symptoms.

Researchers consulted some prominent vaccine critics in designing the study. California advocate Rick Rollens praised the work but said it didn't eliminate other vaccine concerns that deserve similar study. Meanwhile, he said it should draw much-needed attention to the suffering of patients like his son, who has both autism and GI disorders.

"No longer can mainstream medicine ignore the parents' claims of significant GI distress," he said.