Health • Utah researcher says link is slight and not yet understood; it's 'premature' to blame the flu.
This is an archived article that was published on sltrib.com in 2012, and information in the article may be outdated. It is provided only for personal research purposes and may not be reprinted.
(Reuters) Kids whose mothers had the flu while pregnant were slightly more likely to be diagnosed with "infantile autism" before age three in a new Danish study. But the children's overall risk for the developmental disorder was not higher than that of other kids.
Researchers said it's possible that activation of a mother's immune system such as by infection with the influenza virus could affect a fetus's developing brain. But they urged caution with the new findings, especially because of statistical limitations in their number-crunching.
"I really want to emphasize that this is not something you should worry about," said lead author Dr. Hjordis Osk Atladottir, from the University of Aarhus.
"Ninety-nine percent of women with influenza do not have a child with autism," she told Reuters Health. "If it were me that was pregnant, I wouldn't do anything different from before, because our research is so early and exploratory."
Her team's data came from a study that originally recruited more than 100,000 pregnant women in Denmark between 1996 and 2002. The women were called multiple times during their pregnancies, and once afterward, to ask about any new infections they had or medications they had taken.
The study includes 96,736 kids born from that initial cohort who were between 8 and 14 years old at the time of the analysis.
Physician William McMahon, chair of the Department of Psychiatry at the University of Utah, agreed the study should not make "pregnant women feel like they need to do anything, anything different than they've already done."
McMahon, who has been involved in autism research for decades, said, "Probably the bottom line here is influenza in a mom is a potential risk factor for autism, but it's probably only one of many, and its net affect we don't understand yet."
Self-reporting of flu symptoms is not as precise as blood studies to confirm a viral infection, he noted. "To blame it on influenza, it's a little premature."
But there are many good reasons to get vaccinated, he said. "Unless someone has a special condition, [that prohibits getting vaccinated] everybody should have a flu shot," he said.
Using a country-wide register of psychiatric diagnoses, Atladottir and her colleagues found that 1 percent of all kids were diagnosed with autism, including 0.4 percent with infantile autism in which the main symptoms all show up before age three.
There was no link between a range of infections in pregnancy including herpes, coughs and colds and cystitis- and the chance a baby would develop autism or infantile autism, according to the report published Monday in Pediatrics.
And among 808 women who reported having the flu while pregnant, there was no increased risk of autism in their children. However, seven of those babies, or 0.87 percent, were diagnosed with infantile autism, compared to the rate of 0.4 percent among kids in general.
There was also an increased albeit sometimes borderline risk of both autism and infantile autism in babies of women who had fevers for a week or more during pregnancy, as well as mothers who took some types of antibiotics.
Atladottir said there is some research in rodents suggesting women's activated immune cells can cross the placenta and affect chemicals in a fetus's brain. But how those findings apply to humans is still a question mark.
"It's all very unsure now we don't really know anything," she said.
In the United States, about one in 88 children is now diagnosed with autism or a related disorder.
One limitation of the new study, the researchers noted, is that they did 106 statistical tests comparing the risk of autism or infantile autism with various infections and drugs.
In medical research, a significant finding is typically considered one where there is less than a five percent likelihood the result would have occurred by chance.
But when so many calculations are done, scientists would expect that at least some would pass this test of significance, even if there is no real link between the pregnancy variables and autism.
In addition, women's flu reports weren't confirmed by doctors - and the frequency of mistaking the flu for another infection, or vice versa, is "likely to be considerable," the researchers noted.
Because of those limitations, Atladottir said the findings could encourage future research, but shouldn't be at the front of pregnant women's minds.
"We don't want to create panic," she said.
Still, one expert who wasn't involved in the new study thought the researchers were "soft peddling" their conclusions.
"It is highly recommended that women avoid infection during pregnancy, and there are a variety of very practical ways to decrease the likelihood of this," Paul Patterson, who studies the immune system and brain development at the California Institute of Technology in Pasadena, told Reuters Health by email.
The U.S. Centers for Disease Control and Prevention recommend all women get a flu vaccine during pregnancy - in part because serious flu complications are more common in pregnant women.
But, said Patterson, "It is also worth emphasizing that even though the risk (of infantile autism) is significantly increased, the risk is still quite low."