* Explain that SSRIs are the standard first-line treatment for depression, along with psychotherapy.

    * Explain that recent evidence-based treatment recommendations for depression during pregnancy indicated that SSRIs may be prescribed, though in women with less severe depression or less risk of relapse, tapering the drugs may be considered.

    * Explain that this study was a retrospective analysis of medical records, a relatively weak form of evidence.

Babies born to women taking selective serotonin reuptake inhibitors for depression during pregnancy were at increased risk for septal heart defects, especially when the mothers were on more than one of these drugs, researchers said.

In a cohort study of nearly half a million Danish children, the odds ratio for septal heart defects was 1.99 (95% CI 1.13 to 3.53) for those whose mothers received an SSRI while pregnant compared with those not receiving such drugs, according to Lars Henning Pedersen, MD, PhD, of Aarhus University in Aarhus, Denmark, and colleagues online in the British Medical Journal.

They also found that, among women taking multiple SSRIs during pregnancies, babies had a nearly five-fold increase in risk (OR 4.70, 95% CI 1.74 to 12.7).

Significant increases in risk were found for most individual SSRI drugs, the researchers said.

"Our results suggest a class effect of the SSRI on heart defects," they concluded.

On the other hand, the risks remained relatively small, as the prevalence of septal heart defects was 0.9% among children born to mothers on one SSRI and 2.1% when the mothers had been prescribed multiple kinds.

Pedersen and colleagues also could not exclude the possibility that something about depression itself, rather than the treatments, accounted for the increase in heart defects.

In an accompanying comment, Christina Chambers, PhD, MPH, of the School of Medicine, University of California San Diego, said the results, while provocative, do not yet warrant a recommendation that depressed pregnant women should avoid SSRIs.

"This and other studies suggest that the absolute risk for the individual pregnant woman is very low," she wrote, noting that suboptimal treatment for depression carries its own risks.

Pedersen and colleagues analyzed national registry data on more than 493,000 births in Denmark from 1996 to 2003. The data included prescriptions filled by mothers-to-be as well as the medical status of their babies at birth.

Among this large pool of expectant mothers, 1,370 had filled prescriptions for SSRIs during pregnancies.

The researchers looked for associations between SSRI use and a host of birth defects involving every major body system. Although some important-seeming odds ratios were found -- for example, 2.65 for cleft palate -- only septal heart defects reached statistical significance.

For individual SSRI drugs, Pedersen and colleagues calculated the following odds ratios for septal heart defects, relative to no exposure to these drugs:

    * Fluoxetine (Prozac): 1.34 (95% CI 0.33 to 5.41)
    * Citalopram (Celexa): 2.52 (95% CI 1.04 to 6.10)
    * Paroxetine (Paxil): 0.76 (95% CI 0.11 to 5.43)
    * Sertraline (Zoloft): 3.25 (95% CI 1.21 to 8.75)

But the researchers, as well as Chambers, cautioned against interpreting these findings as suggesting one SSRI was safer or riskier than others.

Pedersen and colleagues noted that variations in compliance or errors in records could have affected the findings. They also acknowledged that previous studies of individual drugs have not consistently identified birth-defect risks associated with SSRIs.

Oddly, paroxetine i