CHICAGO – The risk of autism is mildly increased in women who experience an intra-amniotic infection during labor, according to a study reported at the annual meeting of the Infectious Diseases Society for Obstetrics and Gynecology. A separate study, also reported at the meeting, found no such link, however.
In the first study, a retrospective birth-cohort analysis, the rate of intra-amniotic infection (IAI) was 7.27% among mothers whose children developed autism and 6.51% among those with typically developing children. After adjustment for maternal age, parity, and insurance payer, the risk of having a child with autism was increased 6% in mothers with an intra-amniotic infection (odds ratio, 1.06).
Dr. Cheryl Walker
"There’s probably something small going on there," Dr. Cheryl Walker of the University of California, Davis, MIND (Medical Investigation of Neurodevelopmental Disorders) Institute reported.
"We’re putting it out there because there is very little, if any, connection between these specific perinatal birth-related events and autism, but a lot of people are very concerned about them," she added.
The analysis was based on administrative data from 20,206 children with autism and 6,028,353 controls born from January 1991 through December 2001 in California and recorded in the state’s massive Office of Statewide Health Planning and Development (OSHPD) database. IAI, which complicates up to 4% of term deliveries, was identified using the ICD-9 diagnostic code 658.4.
Dr. Walker pointed out that IAI is often underdiagnosed and poorly documented in medical records, leading to under-reporting to OSHPD. In addition, subtle subclinical states cannot be assessed from administrative data.
"Our results likely underestimate a mild association between intra-amniotic infection and autism," she acknowledged.
In contrast, analysis of data from a case-control study called CHARGE found no association between autism and IAI. CHARGE (Childhood Autism Risks from Genetics and the Environment) was launched in 2003, and includes nearly 1,500 patients from northern California, some of whom have autism spectrum disorders and others who are typically developing or have developmental delays but not ASD.
This substudy of CHARGE included only those for whom medical records were available from their births, and analyzed 509 mothers of children with ASD and 325 controls. IAI was evaluated as overt IAI documented in the maternal medical record; covert IAI defined as the presence of at least one IAI symptom of uterine tenderness, foul vaginal discharge, purulent amniotic fluid, and maternal or fetal tachycardia; IAI risk defined on a four-point scale ranging from no risk to overt IAI; and finally, IAI risk with or without the use of antibiotics in labor.
No significant differences were observed between 396 evaluable mothers of children with autism/ASD and 247 mothers of typically developing children in rates of overt IAI (3.9% vs. 4.6%), presence of at least one IAI symptom (9.8% vs. 9.9%), or any IAI risk (20.2% vs. 18.6%), Dr. Walker reported.
Rates also were similar between the autism and typically developing groups for rupture of membranes greater than 18 hours (6.5% vs. 4.1%), antibiotic use in labor (26.3% vs. 22.4%), and any IAI risk without antibiotics (12% vs. 10%).
After adjustment for maternal age, payer, and birth order, IAI risk was similar among mothers of children with autism and those without (adjusted OR, 1.08).
"The neurobiologists tell us that we should be focusing more on the first and second trimesters, so maybe exposure to infection at the time of birth is too late to have much of an effect," Dr. Walker said in an interview. "Most of the neuronal migration has probably occurred by the time of the infection."
In addition, there was the potential for a type II statistical error in the CHARGE study and suboptimal classification of predictors and outcomes in the OSHPD study.
"Case-control designs are great for rare outcomes, but don’t help with rare exposures; and huge administrative data sets help with power at the cost of covariate misclassification," she said.
The studies were prompted by a growing body of evidence that IAI is associated with adverse neurodevelopment and cerebral palsy, and that inflammation plays a role in autism. Many children with autism have altered immune systems; lower levels of circulating immunoglobulin G antibodies; impaired responses to a variety of stimulation assays in vitro, including immunizations; and very unusual natural killer cell function on a variety of levels, Dr. Walker explained.
Dr. Walker reported financial relationships with Merck and Graceway. Her coauthors reported no conflicts.