QUEBEC CITY— An apparent association exists between assisted reproductive technology (ART) and an increased risk of autism, researchers reported at the 36th Annual Meeting of the Child Neurology Society.

According to the CDC, 1.2% of US births in 2004, or approximately 50,000 children, resulted from ART. The rate was similar for births in Orange County, California, where Pauline A. Filipek, MD, practices and is Associate Professor of Clinical Pediatrics and Clinical Neurology at the University of California School of Medicine in Irvine. Of 48,000 births in Orange County during 2004, “at least 557 were conceived using ART,” Dr. Filipek and colleagues noted.

During her routine neurologic history evaluation for every new patient, Dr. Filipek began to ask questions about the mode of conception, such as “Was this a natural conception, or did you need medication or other help to become pregnant?” and “Do you know what type of ART was used?”

Dr. Filipek and colleagues conducted a study based on a retrospective chart review of all new neurology consultations from a five-year period. Data included date of birth, age at diagnosis of autism (if applicable), single or multiple birth, maternal and paternal education, whether ART was used, and type of insurance, as fewer financial constraints have led to increased use of ART during the past three decades.

Of 1,250 charts reviewed, 715 children had a diagnosis of autism spectrum disorders; among these children, 88 had been conceived using ART. Data analyses compared ART use versus non-ART use. The two non-ART groups included the full sample and those with commercial insurance; the effects of parental age and gender were also analyzed.

“The rate of ART in the entire clinical autism spectrum disorders sample was 10%, which is an eightfold increase over the CDC-reported rate of 1.2% in both the US and California,” the investigators stated. When patients without commercial insurance were excluded, the rate of ART increased 12-fold, to 14%.

Of children with autism conceived with ART, 45% were single births. ART methodologies included 21% who received ovarian stimulation by drugs alone, 30% who had in vitro fertilization with or without intracytoplasmic sperm injection, and 38% by a combination of methods. “Therefore, in vitro manipulation alone is not the causative factor,” Dr. Filipek and colleagues pointed out.

Among all children with autism in her practice, Dr. Filipek found that the male-to-female ratio was 3.5 to 1, and in the ART alone sample, the ratio was 2.7 to 1, which was not significantly different. Parents of children with autism conceived with ART “were significantly older and had more years of education than parents of non-ART children,” the researchers said. “All parents in these cohorts are more than 10 years older than the national average. What is unexpected is the discordance between maternal and paternal years of education for the similar years of paternal education for the non-ART commercial insurance group.”

The ART children who presented for evaluation were 10 months younger than non-ART children with commercial insurance. Dr. Filipek and colleagues noted that this was surprising, given her success at initiating early evaluations of infant siblings of children with autism even before age 6 months, which would have resulted in a decreased age at presentation for the non-ART cohort.           

—Debra Hughes

Return to table of contents