ANNOTATION (Neeta Shenai)

The Finding: The use of any antidepressant during the second and third trimester was associated with an increased risk of autism spectrum disorders (ASDs) (31 exposed infants; adjusted hazard ratio, 1.87; 95% CI, 1.15-3.04). Use of SSRIs during the second and/or third trimester was associated with an increased risk of ASD (22 exposed infants; adjusted hazard ratio, 2.17; 95% CI, 1.20-3.93).

Strength: This was a large, register-based study of an ongoing population-based cohort, the Québec Pregnancy/Children Cohort. A very large number of singleton full-term infants (145,456) were analyzed.
Weaknesses:
• Antidepressant exposure was defined as having at least one prescription filled at any time during pregnancy or a prescription filled before pregnancy that overlapped the first day of gestation, however it could not confirm whether or not women actually took the medication.
• While a very large number of births were analyzed, only 1054 children (0.7%) were diagnosed with ASD. The numbers of children diagnosed with ASD were very small (n=31 with any antidepressant and n=22 with SSRIs) as compared to the 145,456 records reviewed. Children were identified as having ASD if at least one diagnosis of ASD was documented in the medical record, but independent clinical confirmation was not completed.
• The study does not fully account for the fact that women suffering from psychiatric illnesses already have a greater risk of having children with ASD. Although the authors controlled for maternal depression, there are not reliable measures of severity. Women who choose to take antidepressants during pregnancy are inherently a different group of women than those who choose not to take medications during pregnancy, despite having the same diagnosis. Therefore, it makes it very difficult to determine if the increased risk is due to the antidepressant medications or because women were experiencing more severe depression.

Relevance: Given the number of factors that may account for an increased risk for ASDs, it is difficult (if not impossible) to differentiate the impact of in utero exposure to antidepressants from other exposures and genetic factors. The association between in utero antidepressant exposure and ASD may be, all or at least in part, the result of confounding factors, specifically the underlying indication for antidepressant use.

ANNOTATION (Meredith Leigh Spada)

The Finding: This 2015 study utilizes data from the first hospital-based pregnancy registry for second-generation antipsychotics in North America that prospectively evaluates risk of malformations among infants exposed in utero to antipsychotics. Of the three infants in the exposed group that developed major malformations, two of the malformations were cardiac in nature (transposition of the great arteries and ventricular septal defect), with the third being an imperforate hymen. All three infants were notably exposed to multiple agents, including other non-antipsychotic psychotropics. While the unadjusted odds ratio conveyed a 1.25 times greater risk for infants exposed to second-generation antipsychotics compared to those infants not exposed, adjustment for a number of confounding variables decreased this risk closer to 1, indicating that this preliminary finding may be an overestimate.

Strength and Weaknesses: The most notable strengths of this study involve its design, particularly related to its prospective analysis and the use of both medical records and participant interviews to provide in-depth information regarding the outcomes. The comparison group, consisting of women with psychiatric illnesses who are not using medications, helps to limit potential confounding from any pathophysiologic or behavioral contributions for those with psychiatric illness. The study additionally looked at clinically relevant factors such as chronicity of illness, which again facilitates the generalizability and relevance to the psychiatric patient population.
Weaknesses include the current size of the registry, as determining a statistically significant increased risk in rare malformations would take far more subjects than currently enrolled. In addition, determining risks related to specific agents is difficult given that a large number of subjects in both groups (79% in exposed and 44% in unexposed groups) were exposed to multiple psychiatric medications, including multiple antipsychotics. Finally, generalizability is further challenged by the potential selection bias, as the majority of participants were white, married, college-educated women who were motivated to participate in a longitudinal study.
Relevance: The data from this first American pregnancy registry looking at second-generation antipsychotic agents enables a prospective and longitudinal evaluation of the potential effect of these agents on infants exposed in utero. Second-generation antipsychotics are being used for a variety of reasons in clinical practice, ranging from sleep aid to mood stabilization and treatment of agitation or depression. As their clinical use becomes more common, the information obtained will enable better risk-benefit discussions with pregnant women regarding exposure to infants. This is particularly relevant now that the old classification of reproductive safety is being replaced with a system that details all available data of known risk of fetal exposure in order to facilitate such discussions with pregnant women and their families.