Ascertainment Bias an Issue in SSRI Birth Defect Debate

March 1, 2010, 10:30AM. By Gordon Gibb

Toronto, ONRecent studies have linked paroxetine, a selective serotonin reuptake inhibitor (SSRI) frequently prescribed to women for depression and anxiety, to birth defects like PPHN (primary pulmonary hypertension of the newborn). Patients have filed lawsuits against the manufacturers of the antidepressant for failing disclose the potential risks of SSRI use by pregnant women.

The debate over SSRI use by expectant mothers continues to reverberate in the medical community. On the one hand, SSRIs have been known to cause birth defects; on the other hand, untreated depression can cause harm to both mother and baby. Which is the lesser of the two evils?

In a column published on 2/1/10 in Family Practice News, Dr. Gideon Koren acknowledges that SSRI use may pose a risk to the fetus, "there is consensus on one point: if there is a risk, it is very small."

Dr. Koren, a professor of pediatrics, pharmacy, pharmacology, medicine and medical genetics at the University of Toronto and head of the Research Leadership in Better Pharmacotherapy During Pregnancy and Lactation at Toronto's Hospital for Sick Children, notes that studies found Paxil (paroxetine) to be safe during pregnancy for at least a decade following approval in the US in 1992.

That all changed in 2005 when registry data emerged "that appeared to suggest an association between prenatal exposure to paroxetine and a higher-than-expected rate of congenital cardiac malformations," writes Dr. Koren.

Subsequent studies, mostly based on results from administrative databases, produced contradictory results. "Some studies found an association between paroxetine exposure and an increased risk of cardiac malformations, in particular ventricular septal defects (VSD)," says Dr. Koren. "But others did not find this association, and in fact suggested an increased risk for cardiac malformations with other SSRIs, such as sertraline or citalopram. There have also been several meta-analyses, again with mixed results."

For Dr. Koren, the most convincing evidence that an SSRI such as paroxetine does not increase the risk of cardiovascular malformation stems from an international study of infants exposed to paroxetine in the first trimester. According to the study (Am. J. Psychiatry 2008;165:749–52), cardiovascular malformation rate among 1,174 infants exposed to paroxetine in utero when compared with an unexposed group was the same: 0.7 percent v. 1 percent for the general population.

Koren notes that he and his associates performed a meta-analysis of literature between 1985 and 2006, which revealed that first-trimester use of paroxetine was associated with a slight increase in the cardiac malformation birth defect. However, "the use of ultrasound during pregnancy…was 30 percent higher among the women who were on antidepressants during pregnancy, and the babies of women who were on SSRIs had about twice as many echocardiograms during their first year of life than the babies of women who were not on an SSRI during pregnancy. In addition, about four times as many women on paroxetine were using it to treat anxiety than were women on other SSRIs."

The study concludes: "Until we settle this issue of ascertainment bias in this situation, we cannot be certain that in utero exposure to paroxetine is associated with an increased risk of cardiac malformations."

Dr. Koren is director of the Motherisk Program at the Hospital for Sick Children in Toronto and holds the Ivey Chair in Molecular Toxicology at the Department of Medicine, University of Western Ontario, London.

On October 15 of last year, a Philadelphia jury awarded $2.5 million to the family of Lyam Kilker, who was born with serious heart defects. Lyam's mother was prescribed paroxetine while she was pregnant.