It is estimated that autism spectrum disorders (ASD) affect about 1% to 2% of children. Research carried out in twins and families indicate that ASD is highly heritable; however, it is generally believed that while genetic factors play an important role, there is an interplay between genetic and environmental factors in the etiology of this disorder. Various environmental exposures have been implicated, including vaccinations, mercury, air pollution, insecticides, and infection.
Because research studies have demonstrated that about one-third of children with autism have elevated platelet serotonin levels, researchers have speculated that dysfunctional serotonin signaling may play a role in the pathogenesis of this disorder. This hypothesis has also raised concerns regarding prenatal exposure to agents, including certain antidepressants, which may modulate levels of serotonin in the developing fetal brain.
In 2011, we reported on an article suggesting a greater prevalence of autism spectrum disorders (SSRIs) among children exposed to selective serotonin reuptake inhibitors (SSRIs) during pregnancy. This study did not prove a causal relationship between prenatal exposure to SSRIs and autism, nor could it rule out the possibility that exposure to maternal depression (as opposed to exposure to SSRI antidepressants) may increase the risk of giving birth to a child with an ASD. A new study from Sweden investigates the use of antidepressants during pregnancy and the risk of autism spectrum disorders in the offspring.
The Swedish study is a case-control study where 4429 cases of autism spectrum disorder (1828 with and 2601 without intellectual disability) and 43,277 age- and sex-matched controls (without ASD) were identified. The children included in this study were between the ages of 0 and 17 and were part of the Stockholm Youth Cohort. For this cohort, prospectively recorded data on the participants and their first degree relatives has been collected by record linkage with national and regional healthcare, social, and administrative registries.
A history of maternal depression (as identified before the birth of the child) was associated with an increased risk of ASD in the offspring (adjusted odds ratio 1.49, 95% confidence interval 1.08 to 2.08). In contrast, a history of paternal depression was not associated with an increased risk of ASD.
In order to assess the impact of exposure to antidepressants, the investigators performed an analysis of a subsample of the subjects where there was data available regarding the use of antidepressants during pregnancy (reflecting 1679 cases of ASD and 16,845 controls with data on maternal antidepressant use). Women who took antidepressants (either SSRIs or tricyclic antidepressants) were more likely to give birth to a child with autism spectrum disorder. Among women with histories of depression who did not take antidepressants during pregnancy, there was no increase in the risk of ASD.
The authors also attempted to determine if there were any differences in risk of ASD with and without intellectual impairment. (As the name implies, autism spectrum disorders occur along a continuum where some children have significant intellectual disability while others have milder signs of illness but no impairment of intellectual functioning, as in Asperger Syndrome.) Antidepressant exposure was only associated with an increased risk of ASD without intellectual impairment. These associations too were largely observed due to associations with autism without intellectual disability (OR = 2.54, CI 1.37 to 4.68)
What Does This Mean?
First of all, we cannot conclusively determine from this sort of study that using antidepressants during pregnancy increases the risk of having a child with autism. The study observed that there is an association between prenatal exposure to antidepressants and risk of autism, but this sort of study design cannot prove causation.
That said, we cannot rule out that exposure to antidepressants may affect the risk of having a child with ASD; however, there are several different ways to interpret the findings and some other things we must take into consideration:
1. This was not a randomized, controlled trial, and we cannot assume that the women who took antidepressants are identical to the women who did not take antidepressants. In fact, it would be fair to assume that the women who decided to remain on antidepressants had more severe or chronic illness. Therefore, are we selecting for women who may have a genetic vulnerability to psychiatric illness which may also increase the risk of autism spectrum disorders in their children. (Whole-genome studies of affected individuals have shown considerable genetic overlap between autism, schizophrenia and bipolar disorder.)
2. From this study, we cannot determine the impact of depressive symptoms during pregnancy on risk for autism. We cannot assume that all women taking antidepressants were well; the study did not provide any information regarding the presence or absence of depressive symptoms (or anxiety) during pregnancy. In fact, it has been shown in other studies that many women with depression (pregnant or not) continue to have residual depressive symptoms despite treatment with an antidepressant. Thus, it is not entirely clear whether it is the impact of the illness itself or exposure to the medication that is being measured in this study.
3. The authors also suggest that there are may be other exposures that may take place more commonly in women treated with antidepressants than in women who are not depressed. It is possible that antidepressant treatment co-occurs with and may be a marker for these other exposures which affect risk for autism.
- Women with depression are more likely to have a history of childhood abuse. Next week we will review an article which indicates that women with histories of childhood abuse are more likely to have a child with autism.
- Studies have observed that women with depression are less likely to take prenatal vitamins. Could this play a role given that we now have data indicating that supplementation with folate reduces the risk of having a child with autism?
- Women with depression are more likely to use tobacco, alcohol and illicit drugs. Do these exposures have an effect on the risk of autism?
So What Do we Tell Our Patients?
In a thorough and thoughtful discussion of the findings from this study, the authors highlight the clinical dilemma women will now face when considering whether or not to use an antidepressant during pregnancy:
If antidepressants increase the risk of autism spectrum disorder, it would be reasonable to warn women about this possibility. However, if the association actually reflects the risk of autism spectrum disorder related to the non-genetic effects of severe depression during pregnancy, treatment may reduce the risk.
While this study does not prove that taking antidepressants during pregnancy increases the risk of having a child with an autism spectrum disorder, it is bound to raise concerns about the long-term effects of prenatal exposure to antidepressants and is certain to create considerable anxiety for women deciding whether or not they should take antidepressants while pregnant.
It is also important to note that while case-control studies can help researchers identify an association between exposure to a particular drug and an adverse event, they often overestimate the magnitude of the risk. Future studies will help to clarify the magnitude of the risk and will also help to identify other contributing factors; however, without this information we are left to contend with the uncertainty that this sort of study generates.
Ruta Nonacs, MD PhD
Rai D, Lee BK, Dalman C, Golding J, et al. Parental depression, maternal antidepressant use during pregnancy, and risk of autism spectrum disorders: population based case-control study. BMJ 2013; 346: f2059.