The authors believe a mother's immune response to a genital herpes infection could be disrupting the development of a fetus' central nervous system. In other words, the increased risk of autism is not due to direct infection of the fetus.
"It's not herpes, per se; it's the inflammation associated with the infection that's involved in the pathogenesis or pathobiology with this particular syndrome," said study author Dr. W. Ian Lipkin, a professor and director of the Center for Infection and Immunity at Mailman School of Public Health at Columbia University.
However, three experts who were not involved in the new research say it does not provide enough scientific evidence for worry.
"Unfortunately, the analysis conducted in this study has significant flaws, and in fact, the data does not support the claims made by the authors," said Mathew Pletcher, vice president and head of genomic discovery at Autism Speaks.
According to Dr. David Winston Kimberlin, a professor of pediatric infectious diseases at the University of Alabama at Birmingham, "pregnant women should not be worried about HSV-2 (genital herpes) as a cause of autism based upon the findings of this single exploratory research study."
Amalia S. Magaret, a research professor in the department of laboratory medicine at University of Washington, also said the conclusions "are subject to concern."
About one in five women in the United States carries herpes simplex virus 2 (HSV-2), also known as genital herpes
. Symptoms include pain or itching, small red bumps or white blisters, and ulcers that ooze or bleed and then form scabs. This contagious infection, which is usually spread through sex, remains in the body for life. After an initial outbreak, HSV-2 virus buries itself in nerve cells, where it grows increasingly less active over time. Flare-ups can always occur, but they grow less frequent as the body builds up immunity to the virus.
For the new study, Lipkin and his colleagues investigated a possible relationship between the risk for autism and maternal infections during pregnancy, which scientists have suggested may cause neurodevelopmental disorders.
Lipkin and his co-researchers focused specifically on five pathogens known as the TORCH infectious agents: Toxoplasma gondii (a parasite found worldwide), rubella virus (the German measles virus), cytomegalovirus (a common virus related to chicken pox and mononucleosis), and herpes simplex viruses types 1 and 2 (HSV-1 causes oral herpes or cold sores). Infection with any of these agents during pregnancy and transmission to a baby
could lead to a spontaneous miscarriage or birth defects including brain damage and vision problems. In some cases, an infection may threaten the life of the baby
even after birth.
For the study, the research team examined blood samples from mothers enrolled in the Autism Birth Cohort Study overseen by the Norwegian Institute of Public Health, including 412 mothers of children diagnosed with autism and 463 mothers of children without autism.
The researchers took blood samples around week 18 of pregnancy and at birth and then analyzed the samples for levels of antibodies, an indication of the mother's immune system response, to each of the TORCH agents.
High levels of antibodies to HSV-2, but not any of the other infections, were associated with risk for autism spectrum disorder.
"HSV-2 replicates in the vicinity of the placenta, so a little bit goes a long way," Lipkin said. "That's different than what you have with HSV-1, with cytomegalovirus, with the rubella virus, where you would have to have a very large immune and inflammatory responses in order to have high levels in the placenta with an impact on the developing central nervous system."
In all, 13% of mothers in the study tested positive for HSV-2 antibodies at mid-pregnancy. Of these, only 12% reported having HSV lesions before pregnancy or during the first trimester, a likely indication that most infections were asymptomatic.
"We believe the risk is really due to inflammation and immune activation in the mother with trafficking some of these inflammatory molecules across the placenta," Lipkin said.
Evidence of a link between antibodies and autism was found only when exposure occurred during early pregnancy but not at birth. Early pregnancy is when the fetal nervous system undergoes rapid development.
"The brain doesn't develop immediately at some point. It's a process, right?" Lipkin asked. "The risk is associated with early pregnancy. It's not present, at least in this study, with later pregnancy. This is in line with what's been observed in animal models of disease in rodents. So it's biologically plausible."
The effect was seen only in boys, not girls. Lipkin and his colleagues say the number of females with autism in the Norwegian Study is small, so there is not enough evidence to draw conclusions, and further research is needed.
Results not 'compelling'
Lipkin's disclosure of possible research imperfections does not lessen the criticism.
"I do not find the results reported in the article to be compelling," Kimberlin said. For one thing, he wonders, "Why only in boys?"
"Before attention should be paid to this in the general population (as compared with the research community), these results need to be reproduced in other studies," Kimberlin added.
Pletcher joins both Lipkin and Kimberlin in calling for more research.
"I do worry that a paper like this will create concern for some women when there currently is not a scientific basis for that concern," he said.
Although the new study is "reasonable given our current understanding of environmental factors that predispose for autism," he noted that there is a "growing body of evidence that significant inflammation in a pregnant mothers, regardless of the cause, could impact the developing brain of the child."
So it may turn out that HSV-2 and other viral infections during pregnancy increase the chances for autism, "but this work does not provide an answer in either direction," Pletcher said.
Meanwhile, a number of factors "have been shown to impact fetal neurological development and increase the potential for the child to develop autism," he said. "One of these factors are specific types of infections during pregnancy such as rubella. ... Maternal diabetes and the age of both the mother and father can be factors as well. We are also aware of protective factors such as folic acid supplementation."
Magaret found the study's methodology disconcerting.
She noted that the researchers first ran one set of tests to look at whether a maternal infection (measured by a blood test) was associated with autism and found "no association for any of the five viruses." Next, the team ran another set of tests examining antibodies at mid-pregnancy and after birth and here found an association between genital herpes and autism at mid-pregnancy only. Finally, the team ran a final set of tests and discovered that only the highest threshold of genital herpes antibodies in the mother could be linked to "a statistically significant" increase in the rate of autism.
"The more ways you look at data, the more chances you have to find something that is unlikely to be reproduced," Magaret said.
Lipkin stands his ground, maintaining that more research is needed, including new studies of anti-inflammatory drugs, since these could be helpful in reducing the risk of autism.
"I won't say that you should take anti-herpes drugs, because I don't know that those are safe," Lipkin said, adding that if his results can be replicated in a larger group of women, that would provide "more impetus to develop vaccines against herpes viruses as well as other infectious agents that are linked to disease."