The last 15-20 years have brought enormous attention to the relevant clinical issues regarding prescribing antidepressants during pregnancy. Concern about the effects of fetal exposure to selective serotonin reuptake inhibitors (SSRIs) is appropriate given the consistent data that approximately 7% of women use antidepressants during pregnancy, and that risk for relapse of depression during pregnancy in women who have stopped antidepressants during pregnancy is very high.
We have learned so much from studies of relevant questions regarding SSRI exposure. Concerns about increased risk for organ malformation have been set aside. An extraordinary number of studies across a broad range of patients around the globe looked at the issue of risk for organ malformation following in utero SSRI exposure – even looking specifically at risk for cardiac malformations, which had been an earlier concern in the literature – with the evidence supporting absence of increased risk. Also clarified has been, first, the absence of risk of complications such as persistent pulmonary hypertension of the newborn (PPHN) and, second, a delineation of the prevalence and clinical implications of transient neonatal symptoms such as jitteriness and tachypnea in offspring of women who used antidepressants during pregnancy – so-called “poor neonatal adaptation syndrome.”
However, for so many clinicians and for patients, the missing piece in the risk-benefit equation has been the issue of long-term neurodevelopmental sequelae in children whose mothers used antidepressants during pregnancy. While the accumulated data have shown sparse evidence linking SSRI exposure with autism or attention-deficit/hyperactivity disorder (ADHD), the evidence has been mixed regarding neurobehavioral sequelae associated with fetal exposure using developmental outcomes such as language ability, cognition, academic performance, language, math, and other cognitive outcomes. As far back as the 1990s, colleagues in Canada failed to show a difference in neurobehavioral outcomes in 5- to 7-year-old children whose mothers used SSRIs or older tricyclic antidepressants during pregnancy compared to nonexposed women (N Engl J Med. 1997 Jan 23;336:). Even early on, it was noted that one of the strongest predictors of neurodevelopmental outcome was untreated maternal psychiatric illness.
Since those early studies and over the last decade, there have been numerous small studies with conflicting data regarding a whole host of neurodevelopmental outcomes with inconsistent methodologies, different assessments, and failure to control for the presence or absence of maternal psychiatric illness during pregnancy – one of the most critical predictors of neurodevelopmental outcome and one we are beginning to appreciate plays a very significant role.
Most recently, the authors of a very large population-based retrospective cohort study in Denmark linked population-based registries with obstetrical data and examined language and math performance among 575,369 public schoolchildren whose mothers used or didn’t use antidepressants during pregnancy (JAMA. 2021 Nov 2;326:). These investigators found a decrease in mean test scores for language (53.4 vs. 56.6) and math (52.1 vs. 57.4) in children whose mothers received antidepressant prescriptions during pregnancy compared with children who did not have that exposure. However, when they adjusted for maternal psychiatric illness and other relevant confounders, the finding went to null for language (adjusted difference, –0.1; 95% confidence interval, –0.6 to 0.3), but did not for math (adjusted difference, −2.2; 95% CI, −2.7 to −1.6). The results ultimately showed a modest finding for exposure and a small decrement in mathematical performance. The takeaway is that antidepressant use may be a proxy for neurodevelopmental deficit but is unlikely to be the etiology or direct cause of that deficit.
With that said, patients and their doctors can be reassured with respect to how much we have learned about SSRIs during pregnancy across the last decade. Yet there are appropriate concerns about long-term neurodevelopmental sequelae in this patient population. I think that what we can say in 2022 is that there is a growing appreciation for the effect of maternal psychiatric illness on long-term outcomes in children and the effect of maternal psychiatric illness on risk for postpartum depression, which we know influences long-term neurodevelopmental outcomes in children. Perhaps more than in years past, there is now also an appreciation of the effect of a dysregulated stress axis on the intrauterine fetal neuronal programming, which is perhaps the newest frontier, and which may hold the answers with respect to how to weigh the effect of maternal psychiatric illness on decisions about psychotropic use during pregnancy. But for today, there is an appreciation that exposure to maternal psychopathology is not a benign exposure.
Although some of the data remain incomplete, in 2022, patients will continue to make individual decisions based on the available data, factoring in the effect of maternal adversity in a more deliberate way and with a refined lens through with which to see their options with respect to using or not using SSRIs during pregnancy.
is the director of the Ammon-Pinizzotto Center for Women’s Mental Health at Massachusetts General Hospital in Boston, which provides information resources and conducts clinical care and research in reproductive mental health. He has been a consultant to manufacturers of psychiatric medications. Email Dr. Cohen at .