Prenatal Antidepressant Use: Time for a Pregnant Pause?

This is the nightmare scenario. Your patient is a 32-year-old mother of 4 who has just learned from her gynecologist that she is 2 months pregnant. She has had 7 major depressive episodes over the past 8 years, 3 of which were accompanied by serious suicide attempts. Two episodes occurred postpartum and were accompanied by profound social withdrawal, poor nutritional intake, and thoughts of harming herself or her infant. She is currently taking paroxetine(Drug information on paroxetine) (Paxil), 50 mg/d, with positive results and has been in remission from depression for nearly 2 years. However, the patient has just read a story on the Internet linking maternal antidepressant use during pregnancy with birth defects. She is asking you if she should stop taking the antidepressant at this time. What is your response, and how does it square with the latest research?

Until quite recently, many psychiatrists were confident that SSRIs were relatively safe during gestation, based on studies done in the 1990s.¹ For example, a study by Pastuszak and coworkers² found no evidence of teratogenicity in 128 women taking fluoxetine(Drug information on fluoxetine) (Prozac) during the first trimester, compared with matched controls. While there was a trend toward higher miscarriage rates in the fluoxetine group, the risk was small (relative risk, 1.9) and comparable to that of tricyclics. Furthermore, depression itself may also raise the risk of miscarriage.

A study by Chambers and associates³ found no significant differences between fluoxetine treatment in pregnant women and controls in the number of spontaneous pregnancy losses or major structural anomalies; however, the incidence of 3 or more minor anomalies was significantly higher in the fluoxetine cohort, and women who took fluoxetine during the third trimester were at increased risk for perinatal complications. The Chambers study was criticized on a variety of methodologic grounds, including failure to control for coexisting medical illnesses.⁴ Indeed, summarizing the published reports of SSRI use during gestation (N = 2219)--including use of fluoxetine, citalopram(Drug information on citalopram) (Celexa), paroxetine, sertraline(Drug information on sertraline) (Zoloft), and fluvoxamine(Drug information on fluvoxamine) (Luvox)--Newport and colleagues⁵ concluded that, collectively, "these data provide no evidence that prenatal SSRI exposure is associated with an increased incidence of congenital malformation."

The data from the 1990s also suggested little long-term neurobehavioral risk to children exposed to antidepressants in utero. For example, Nulman and associates⁴ found that in utero exposure to either tricyclics or fluoxetine did not affect global IQ, language development, or behavioral development in preschool children. Finally, from 1990 to 2000, reports of neonatal withdrawal syndromes associated with SSRIs were starting to trickle in, and their clinical significance was underplayed.

In most respects, the overall safety of SSRIs and related antidepressants during pregnancy is still supported by studies from the last 5 years--but some yellow lights are now flashing. In December 2005, for example, the FDA issued a public health advisory concerning first-trimester exposure to paroxetine and the risk of congenital malformations.⁶ The FDA had reviewed 2 recent, unpublished epidemiologic studies: 1 using a Swedish national registry database and the other using a US insurance claims database. The net results suggested that the risk of cardiac malformations (eg, atrial or ventricular septal defects) increased 1.5- to 2-fold in infants exposed to paroxetine, compared with unexposed infants or those exposed to other antidepressants. The absolute number of affected infants was small: about 1.5% to 2% in infants exposed to paroxetine versus 1% among controls. The FDA subsequently requested that the manufacturer of paroxetine change its pregnancy category from C to D and add new warnings to its labeling information.

Shortly after this news broke, Chambers and associates⁷ published a study reporting a significant association between the use of SSRIs after the 20th week of pregnancy and the presence of persistent pulmonary hyper-tension of the newborn (PPHN)--a condition associated with significant morbidity and mortality. For example, infants with PPHN may manifest cognitive delay, neurologic abnormalities, and hearing loss.

Finally, to complete the "triple threat," reports of a neonatal abstinence syndrome associated with SSRIs have increased in the past year. One study found that as many as 30% of neonates with third-trimester SSRI exposure showed some type of withdrawal syndrome.⁸ This appeared to peak around 48 hours after birth and was characterized by lack of crying, increased muscle tone, irritability, abnormal breathing, and disrupted sleep.