Selective serotonin reuptake inhibitors and congenital malformationsBMJ 2009; 339 doi: http://dx.doi.org/10.1136/bmj.b3525 (Published 23 September 2009) Cite this as: BMJ 2009;339:b3525
- Christina Chambers, associate professor
- 1Division of Dysmorphology and Teratology, Departments of Pediatrics and Family and Preventive Medicine, School of Medicine, University of California San Diego, 9500 Gilman Drive, MC 0828, La Jolla, CA 92093-0828, USA
Major depressive disorder in women is most common during their childbearing years, and about 13% of women in the United States have taken an antidepressant drug during pregnancy.1 2 In the past 20 years, selective serotonin reuptake inhibitors (SSRIs) have become a mainstay of treatment in women with major depressive disorder; however, concerns persist about safety for the developing fetus. This is counterbalanced by equally compelling concerns about the consequences of undertreatment for mother and child.3
In the linked population based cohort study from Denmark (doi:10.1136/bmj.b3569), Pedersen and colleagues confirm a previously reported doubling of risk for septal heart defects after early exposure in pregnancy to SSRIs (odds ratio 1.99, 95% confidence interval 1.12 to 3.53).4 However, in contrast to previous studies, redemptions of prescriptions for citalopram and sertraline, but not paroxetine or fluoxetine, were significantly associated with this group of heart defects.5 6 7 8 Furthermore, unlike two previous large case-control studies conducted in the US, no association was noted with anencephaly, omphalocele, craniosynostosis, or right ventricular outflow tract defects.7 8
Lack of consistency across these studies with respect to specific malformations and specific drugs makes it difficult to translate the findings into clinical practice. One of the fundamental principles of teratology is that teratogenic exposures induce specific patterns of malformation, and not an increase in the incidence of every defect. In other words, if some or all SSRIs are teratogenic, we would expect to see similar findings for specific drug exposures and specific defects in all studies.
One explanation for this inconsistency, assuming that SSRIs do cause specific birth defects, is differences in study designs. For example, although Pedersen and colleagues linked records for 496 881 singleton live born infants, they identified only 1370 mothers who redeemed multiple prescriptions for an SSRI in the perinatal period. Therefore, the study may have been insufficiently powered to detect the previously suggested twofold to threefold increased risk for anencephaly, omphalocele, craniosynostosis, or right ventricular outflow tract defects, all of which occur at least an order of magnitude less frequently than septal defects.
Alternatively, these findings could be spurious and attributable in observational studies to unmeasured or inadequately controlled confounding factors, such as maternal obesity, alcohol, tobacco, or periconceptional use of folic acid supplements; confounding by the mother’s underlying condition; or detection bias, in which mothers being treated for major depressive disorders are more likely to seek out or receive more comprehensive prenatal and postnatal testing of their children.
How does Pedersen and colleagues’ study contribute to clinicians’ and patients’ decisions about the use of SSRIs in pregnancy, and how should this be weighed against the risks of non-treatment? The answer remains as before—if an increased risk for major congenital malformations does exist, this study and others suggest that the absolute risk for the individual pregnant woman is very low. Furthermore, each of the more commonly used drugs in this class has been implicated in at least one study, so it is difficult to conclude that one SSRI is “safer” than another.
We need information from larger studies of specific SSRIs, with study designs that control for maternal disease type and severity, comorbidities, and other exposures. In addition, studies of basic science might elucidate the mechanisms involved in inducing specific birth defects to support the biological plausibility of a causal association.
In August 2009, the American College of Obstetrics and Gynecology released a joint statement with the American Psychiatric Association on treatment recommendations for depression during pregnancy.9 Briefly, the recommendations state that women with major depressive disorder who are contemplating pregnancy or who are currently pregnant can start or continue taking their drugs. Women who prefer to avoid or discontinue drugs may benefit from psychotherapy, although this will depend on their psychiatric history. Women should be informed about the possible risks and benefits of their treatment choices, and ongoing consultation between the patient’s obstetrician and psychiatrist is needed during pregnancy, to determine and carry out the most appropriate and acceptable treatment plan.
Most drugs taken by pregnant women have not been well studied, or studied at all with respect to safety of the fetus.10 Although research about SSRIs and pregnancy outcomes is plentiful, it does not necessarily provide definitive answers for clinical practice. Clinicians and patients need to balance the small risks associated with SSRIs against those associated with undertreatment or no treatment.
Cite this as: BMJ 2009;339:b3525
CC has received grant funding from pharmaceutical companies including Amgen, Abbott, Bristol Myers Squibb, Sanofi-Pasteur, Teva, Sandoz, Kali, Barr, and Apotex, some of which manufacture or distribute selective serotonin reuptake inhibitors.
Provenance and peer review: Commissioned; not externally peer reviewed.