Antidepressants and pregnancy

(P. Gueritot, Getty Images)

Upon learning they are pregnant, most women dutifully nix the alcohol, sushi and caffeine.

But what about antidepressants?

Headlines about the potential risks of antidepressants on a developing fetus, including miscarriage, premature birth and newborn breathing problems, have produced angst for many moms on medication. But of greater concern, some experts say, are the harmful effects of untreated depression on the baby, including low birth weight, irritability and developmental delays.

"There's no path that's really risk-free," said Pec Indman, a San Jose, Calif.-based therapist who specializes in perinatal mood and anxiety disorders. "Illness has risk, and treatment has risk."

Which is riskier, of course, is the tough question.

Limited knowledge

Dozens of studies exploring the safety of antidepressants during pregnancy, especially the popular class of antidepressants known as selective serotonin reuptake inhibitors, or SSRIs, have yielded mixed results and suffered from design limitations.

A chief limitation is that it's not clear if adverse outcomes are the fault of the medication or the depression itself, said Dr. Kimberly Yonkers, professor of psychiatry and of obstetrics, gynecology and reproductive services at Yale School of Public Health.

So far the most consistent data have linked antidepressants with preterm delivery, though studies haven't shown any long-term effects on the child, Yonkers said. A study last year that linked antidepressant use in the first trimester with a fourfold increase in the chance of a child having autism was met with a flurry of warnings from experts to not read into the findings, as it was the first and only study to make the association and there were methodological weaknesses.

Putting risk in perspective

One of the more common consequences of antidepressant exposure to babies is post-delivery withdrawal-like symptoms, including tremors, respiratory problems, feeding difficulties and jitteriness. In a 2006 Israeli study published in the Archives of Pediatric and Adolescent Medicine, a third of the antidepressant-exposed babies experienced such symptoms, compared with none in the control group, though most symptoms resolved themselves within a few days.

Other potential dangers pose a very small risk.

Writing last month in the British Medical Journal, European researchers found that taking SSRIs during the second half of pregnancy can more than double the risk that newborns will develop persistent pulmonary hypertension, a condition that prevents babies from getting enough oxygen into their bloodstream and, if severe, can result in multiple organ damage and death.

But that means the risk jumps from 1.2 in 1,000 babies to 3 in 1,000 babies, which is still very rare.

To compare, the risk of any woman developing a blood clot during pregnancy is 2 in 1,000 and of having a stillbirth is 6 in 1,000, said Dr. Avi Patil, who specializes in maternal-fetal medicine at Duke University Medical Center.

"Many more women will have depression in pregnancy that needs to be treated than will have this adverse event," said Patil, who in December published a review article in the journal Obstetrical and Gynecological Survey examining data on common antidepressants taken during pregnancy.

Left untreated, depression itself could harm gestating babies. Babies born to depressed moms have shown elevated cortisol levels and reduced serotonin levels, plus depressed moms are less likely to seek regular prenatal care or take prenatal vitamins and more likely to smoke or drink alcohol, Indman said. They also are more likely to experience postpartum depression, hampering their ability to interact with their new baby, which can interfere with development.

Studies have associated depressed moms with preterm delivery, low birth weight and low Apgar scores (a test to measure the health of a newborn). Untreated anxiety has been linked to babies with inconsolable crying, sleep problems and developmental delays, Indman said.

A mother's options

Women with mild depression might consider gradually tapering off their medication dose before they conceive (if they have the luxury of planning) or in the third trimester (if they don't) to decrease the risk of withdrawal symptoms in the newborn, Patil said. But if it's likely the mom will relapse, it's best to stay on. He recommends patients seek guidance from their OB-GYN and mental health providers to determine the best course of action, including alternative therapies.

It's never a good idea to discontinue medication abruptly. In a 2000 study of women who stopped taking antidepressants and anti-anxiety pills upon learning they were pregnant, 70 percent reported adverse physical or psychological effects, 30 percent had suicidal thoughts and 10 percent were admitted to hospitals.

"It was pretty sad," said study co-author Adrienne Einarson, a nurse who has published extensively on the topic. "It can be more harmful for women not to be treated, but no one seems to look at it that way."

Einarson, who is semiretired as assistant director of Motherisk, a counseling service for pregnant women based at The Hospital for Sick Children in Toronto, worries the stigma and misinformation about the risks might dissuade women from taking medication they need. Injury lawyers haven't helped; last year, a family won a $2.5 million settlement from GlaxoSmithKline after a jury found it negligent for not warning the mom's physician of the risks of taking Paxil during her pregnancy. Her son was born with a heart defect.

Several studies have linked paroxetine, the generic form of Paxil, to heart malformations, but others have found no greater risk, Einarson said.

"Now they're going to think for the rest of their lives that it was her fault because she took Paxil," she said.


15%: Contrary to decades of conventional wisdom that held that pregnancy warded off mental illness with happy-making hormones and the bliss of impending motherhood, depression occurs in up to 15 percent of pregnant women, according to a 2004 study that reviewed different prevalence estimates.