Limited Conclusive Data Exist on Perinatal Risk Factors for Autism

From Medscape Education Clinical Briefs

News Author: Deborah Brauser
CME Author: Laurie Barclay, MD

07/18/2011

Clinical Context

The cause of autism is not specifically identified and may be multifactorial.
Although the neuropathologic mechanism is nonspecific, studies have shown macroscopic, microscopic, and functional brain abnormalities.
For the past 4 decades, epidemiologic research on risk factors for autism has focused on perinatal and neonatal exposures.
Obstetric and neonatal complications have possibly been linked to autism risk, but specific complications associated with increased risk have not been identified, and the magnitude of effect has been inconsistent across studies. The goal of the study by Gardener and colleagues was to review studies of perinatal and neonatal risk factors for autism and to conduct a meta-analysis as appropriate.

Study Synopsis and Perspective

There is scant conclusive evidence of specific perinatal or neonatal risk factors for autism, according to a new meta-analysis of 40 studies.
The investigators write that although several of the 60 factors evaluated (including abnormal fetus presentation at birth, umbilical cord complications, and low birth weight) were linked to autism risk, "there is insufficient evidence to implicate any 1 factor in autism etiology."
However, the findings suggest that a combination of multiple neonatal complications may indeed increase the risk for autism development.
"Autism has no known cure and can have devastating effects on families, which really underscores the importance of trying to discover risk factors, in particular those that are modifiable and that we may be able to lessen with improved prenatal care, for example," lead author Hannah Gardener, ScD, epidemiologist at the University of Miami, Florida, told Medscape Medical News.

"We did find several significant factors, but whether those are causal or simply markers associated with an increased risk is not perfectly understood," added Dr. Gardener, who was at the Harvard School of Public Health in Boston, Massachusetts, at the time of the study.
She noted that prenatal monitoring is important not only for decreasing risk for autism but for also ensuring the health of the fetus and neonate in many health respects.
"Overall, I think the message to the general public is that it's important to realize that the vast majority of children who experience these complications will not end up having autism. Although we may have found several risk factors, it doesn't mean that a child with any of them is definitely going to have the disorder,"
"The risk is still very low. Each of the factors only increased the risk incrementally, so women should not be overly concerned."
The study was published online July 11 in Pediatrics.
 
Assessing Specific Complications
"Although many studies support the hypothesis that obstetrical and neonatal complications may increase the risk of autism, the specific complications, magnitude of effect, and overall conclusions of these studies have been inconsistent," write the researchers.
"A lot of these results were inconclusive or contradictory. There were also several limitations to these individual studies, including relatively small sample sizes. Even large studies may not get that many autism cases and so you may miss associations that are there," said Dr. Gardener.
She noted that another problem with examining just 1 study is that an association may be detected "that isn't really there" and is instead due to chance or to "the randomness of nature."
Because of these limitations, the investigators decided to evaluate data from all studies conducted through March 2007 that looked at the association between autism and prenatal, perinatal, and neonatal factors.
The investigators reported their findings on exposures during pregnancy in 2009 (Br J Psychiatry. 2009:195:7-14).

The results from that analysis showed that advanced parental age, maternal medication use, bleeding, gestational diabetes, and having a mother born abroad were the strongest risk factors for the disorder. However, no association was found with previous fetal loss, proteinuria, or maternal hypertension.
For this meta-analysis, they concentrated on 40 studies that examined more than 60 perinatal and neonatal risk factors.
 
More Research Needed
Results showed that the following were all associated with a significantly increased autism risk:

• Fetal distress;
• Birth injury or trauma;
• Multiple births;
• Maternal hemorrhaging during labor and delivery;
• Summer birth;
• Small for gestational age;
• Congenital malformation;
• Low Apgar score;
• Feeding difficulties;
• Meconium aspiration; and
• Neonatal anemia or jaundice.

Factors found not to be significantly associated with autism risk included:

• Anesthesia use;
• Assisted vaginal delivery;
• Post-term birth;
• High birth weight; and
• Head circumference.

"The complications identified in the current analysis may be the result of previous prenatal complications and/or may operate in combination with prior prenatal complications to impact autism risk," write the researchers.
"Additional research that considers the joint and independent effects of adverse conditions during these various time periods is required to address these possibilities."
They add that the risk factor associations may also affect only those who are genetically vulnerable.
"However, the correlated occurrence of many of these complications limits the ability to determine which, if any, are independently associated with the disorder."
Dr. Gardener, who called the meta-analysis "important," hopes that it will motivate further studies.
"However, I also hope it doesn't cause a woman to worry that, for example, if she's due to give birth in July that the child will definitely be born with autism. That is absolutely not true. A lot of these risk factors are very common so I hope there isn't undue concern over them."


Pediatrics. Published online July 11, 2011.