The multiple recent reports regarding the reproductive safety of the selective serotonin reuptake inhibitors have raised concerns about a spectrum of potentially adverse outcomes associated with SSRI use during pregnancy. But these reports have produced conflicting results and have been confusing for many patients and clinicians.
To date, no professional medical association has issued formal guidelines regarding treatment with SSRIs during pregnancy. However, in December, the American College of Obstetricians and Gynecologists Committee on Obstetric Practice published an opinion on this topic, which is noteworthy in its clarity and balanced review of the existing data. The opinion artfully underscores the relevant issues for reproductive-age women who suffer from depression and who are being treated with antidepressants (Obstet. Gynecol. 2006;108:1601–3).
The ACOG committee makes clear there is no specific treatment algorithm for this population, recommending that “treatment with all SSRIs or selective norepinephrine reuptake inhibitors or both during pregnancy be individualized” and that decisions about treating depression “should incorporate the clinical expertise of the mental health clinician and obstetrician, and the process should actively engage the patient's values and perceptions when framing the discussion of the risks and benefits of treatment.”
The committee points out that many studies have not identified an increased risk for major congenital malformations associated with SSRI use during pregnancy. This is noteworthy, since more is known about the teratogenic risk of SSRIs than about most medications prescribed to pregnant women, including, as examples, various anti-infectives, antiemetics, and medicines used to treat gastroesophageal reflux.
The ACOG opinion does refer to the recent unpublished reports from a Swedish registry and a U.S. claims database associating first-trimester exposure to paroxetine with a greater risk of atrial and ventricular septal defects. (This prompted the Food and Drug Administration to change paroxetine's pregnancy category label from C to D in late 2005.) Interestingly, the committee does not suggest that paroxetine absolutely is contraindicated during pregnancy, but wisely states that paroxetine should be avoided when planning pregnancy and during pregnancy.
Also referenced in the opinion are reports of syndromes of perinatal toxicity or poor neonatal adaptation associated with SSRI exposure in late pregnancy—transient neonatal symptoms that include jitteriness, mild respiratory distress, weak cry, and neonatal ICU admissions—findings that I have addressed in previous columns. With respect to persistent pulmonary hypertension of the newborn (PPHN), the more serious outcome, the opinion summarizes the case control study that found a sixfold greater risk in PPHN associated with SSRI use after 20 weeks' gestation.
The PPHN finding was the topic of another FDA public health advisory in 2006, which, as the ACOG opinion points out, also addressed the risk of depression relapse. That advisory cites our study, which found relapse was fivefold higher among women who stopped antidepressants during pregnancy than it was among those who continued with treatment.
The opinion then states that “the potential risk of SSRI use throughout pregnancy must be considered in the context of the risk of relapse of depression if maintenance treatment is discontinued” and refers to the increased risks of low weight gain and alcohol and substance abuse that are associated with untreated depression.
The ACOG opinion advises that treatment decisions need to be carefully tailored, and “optimally, shared decision making among obstetric and mental health clinicians and women should occur before pregnancy.” But, it concludes, 50% of pregnancies are unplanned; hence, preconception planning will not always be possible “and decisions regarding treatment with SSRIs will undoubtedly occur during gestation.”
In our experience, we see directly what prompts women to make decisions about SSRI use during pregnancy. These decisions are not necessarily driven by severity or duration of illness, or number of past episodes, but more from their perception of risk and the willingness to accept the risk of depression relapse vs. fetal exposure to medication. Every patient applies her own calculus with respect to the amount of risk she and her partner may be willing to accept.
When it comes to prescribing SSRIs, we need to collaborate with patients on a decision in circumstances in which we have imperfect estimates of risk on both sides of the risk-benefit equation. As in any other clinical situation, we tailor treatment based on the clinical scenarios and patients' wishes.