December 18, 2013
Pregnancy: Mixed Results SSRI-Autism Link
By John Gever, Deputy Managing Editor, MedPage Today
This study could not detect a significant association between maternal use of SSRIs during pregnancy and autism spectrum disorder in the offspring.
However, there was a significantly increased risk among women who received SSRIs before pregnancy but not during pregnancy.
Children of women using selective serotonin reuptake inhibitor (SSRI) antidepressants during pregnancy were not at increased risk of autism when compared with other children, a large registry analysis in Denmark showed.
The adjusted rate ratio for autism spectrum cases in children born to women who used SSRIs during pregnancy versus other children was 1.20 (95% CI 0.90 to 1.61), according to Anders Hviid, DrMedSci, of the Statens Serum Institut in Copenhagen, and colleagues.
On the other hand, the incidence of autism spectrum disorders in those born to women who used the drugs prior to becoming pregnant and then stopped was 46% higher than in other children (adjusted rate ratio 1.46, 95% CI 1.17-1.81), the researchers reported in the Dec. 19 issue of the New England Journal of Medicine.
Hviid and colleagues concluded that SSRIs themselves probably do not raise the risk of autism spectrum disorders, but rather it is the indications for which the drugs are prescribed.
More than 600,000 births from 1996 to 2005, followed through 2009, were included in the study.
Adjustments in the analysis were made to account for other factors that, in univariate analysis, were associated with increased autism risk. Among them were other psychiatric disorders beside depression in the mother, other classes of drugs that mothers were using, mothers' age at delivery, their parity status, their employment status, smoking history, place of residence within Denmark, and child's age at follow-up.
Hviid and colleagues concluded that, because the increased autism risk was seen with pre-pregnancy use of SSRIs but not with use during pregnancy, it "may be related to the indications for their use rather than an effect of these drugs. This highlights the potential effect of confounding by indication in our study and similar studies of SSRIs and the importance of being able to adequately take this confounding into account in the study design."
Other researchers contacted by MedPage Today cautioned that the findings fell short of a complete exoneration, but were reassuring nonetheless.
"The findings, while not definitive, suggest that SSRIs do not confer a substantially increased risk of autism to offspring," said Diane Treadwell-Deering, MD, of Texas Children's Hospital in Austin, in an email.
"It seems more likely that there are multiple factors that each confer relatively small increased risks for autism. With regard to SSRIs, the study indicates that the indications for the use of the SSRIs, rather than the SSRIs specifically, may be associated with an increased risk for autism in offspring," she said.
Andrew Adesman, MD, of Steven & Alexandra Cohen Children's Medical Center of New York in New Hyde Park, agreed that the study should reduce concern that had been generated by previous smaller studies. He suggested that future studies should examine "to what extent a mother's history of depression or other psychiatric conditions -- not the use of SSRI medications -- poses a risk for an autism spectrum disorder."
Much of the concern arose from a case-control study in 2011 that suggested a potential doubling in autism risk with fetal exposure to SSRIs. A meta-analysis published last year found a similar increase in risk.
For the current study, Hviid and colleagues culled records on some 626,000 births in Denmark over a 10-year period ending in 2005, including data on the mother's medical history and prescriptions as well as outcomes in the children.
The researchers determined that 6,068 of the births involved SSRI use during pregnancy, with 52 of the offspring eventually receiving diagnoses of autism spectrum disorders.
Without adjustments, the incidence rate for autism in these cases was 122.6 per 100,000 person years, compared with 77.0 per 100,000 in the entire sample. The crude rate ratio of 1.62 (95% CI 1.23 to 2.13) was significant, but when other factors also associated with autism spectrum diagnoses were included in the analysis, the association shrank.
Many of those other factors had univariate associations as strong as for SSRI use during pregnancy. They included maternal diagnoses of schizophrenia (RR 3.56), eating disorders (RR 2.10), and substance abuse (1.79), and residence in the Copenhagen area versus the Jutland and Funen regions (RR 2.05).
"Genetic factors carry the most significant risks for autism," Treadwell-Deering told MedPage Today. "However, the identification of environmental factors, especially those that are preventable, is of vital importance."
Limitations to the study included reliance on administrative records and the lack of genetic data. Also, Hviid and colleagues pointed out that the rate of SSRI use in pregnancy in their data was far lower than studies in the U.S. have indicated (0.97% versus about 5.6%).
The study was funded by the Danish Health and Medicines Authority.
Study authors had no other disclosures.