Comment on "The Effectiveness of Various Postpartum Depression Treatments and the Impact of Antidepressant Drugs on Nursing Infants"

D.M. Campagne, PhD


March 19, 2004

Editor's note:
This letter is in response to: "The Effectiveness of Various Postpartum Depression Treatments and the Impact of Antidepressant Drugs on Nursing Infants," by Dwenda Gjerdingen, originally published in
J Am Board Fam Pract. 2003;16:372-382 and reposted in Medscape on January 8, 2004.

Rightfully, postpartum depression is getting more attention, and any documented article helps to focus that attention on the facts. However, an important part of Gjerdingen's "Results and Conclusions" is stated in a categorical way that is not supported by the facts. It claims that, "Of the more frequently studied antidepressant drugs in breastfeeding women, paroxetine, sertraline, and nortriptyline have not been found to have adverse effects on infants."[1]

There is reasonable doubt as to possible side effects of selective serotonin reuptake inhibitors (SSRIs) in the long term, based upon several studies that indicate they do exist, as I recently quoted.[2] Although Gjerdingen rightfully criticizes the methodologic weaknesses of the studies she cites, such as small sample size, out-of-date methods, and short-term follow up, she omits studies that do document the possible effects of SSRIs on children's growth rates.[3]

Gjerdingen cites older studies by Hendrick and colleagues,[4] but does not include a very relevant recent study by the same author[5] that concludes that "infants exposed to maternal depression lasting 2 months or more appear to experience significantly lower weight gain," thus referring directly to postpartum depression where antidepressant treatment usually extends to well over 2 months. More important, she omits a recent much larger, methodologically sound study that concludes, "For those infants exposed to SRIs (including fluoxetine, fluvoxamine, paroxetine, and sertraline) gestational age, birth weight, and Apgar scores were all statistically significantly lower compared with those of nonexposed infants." The same study found that infants exposed to SSRIs were twice as likely to be born at or before 36 weeks, with its inherent risks, and thus equals that particular risk of SSRI use with the risk of smoking during pregnancy, that is, twice as high as normal.[6]

Gjerdingen cites Stowe and colleagues from 1997[7] as a proof that sertraline has no adverse infant-related clinical outcomes. However, the authors gave their conclusions with 2 caveats: first, that sertraline and its metabolite were present in the breast milk of all the subjects, although below detection limits of most commercial laboratories at that time, and second, that metabolite monitoring is (more) important in determining infant exposure. The same author published a study in 2003[8] that confirms sertraline to be detectable in 18% and its metabolite in 50% of the infants in the sample. The fact that no adverse events were reported or documented in any infant of the study is completely irrelevant for determining the safety of sertraline in the long run, and should not be used as an argument when considering possible risks.

Any drug that affects infant growth, in or ex utero, should be regarded with suspicion and should not be written up as a recommendable solution for postpartum depression, comparable with innocuous and effective methods such as psychotherapy, nurse home visits, and group therapy. The available evidence indicates that even when infant serum levels are low or undetectable, side effects may occur. There is an ample body of evidence that SSRIs taken during pregnancy cause neonates to suffer withdrawal symptoms, which all by itself is a matter for concern.

Postpartum depression is an issue that should not be taken lightly, but, as history shows, the short-term benefits of drugs that have not been sufficiently studied do not weigh up against the tragic long-term disasters they might provoke. Think of thalidomide. Gjerdingen admits that fluoxetine should be avoided in breastfeeding women. As recent as 2001, a review in a reputable journal still "confirmed" the suitability of long-term fluoxetine treatment in pregnancy.[9] US Food and Drug Administration reticence to approve SSRIs for use during lactation should be the guideline.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.