Correspondence
Birth Outcomes in Pregnant Women Taking Fluoxetine
N Engl J Med 1997; 336:872-873March 20, 1997
- Article
To the Editor:
Chambers et al. (Oct. 3 issue)1 report birth outcomes after prenatal exposure to fluoxetine but fail to provide results for infants of mothers with mood disorders who were not treated or who received other antidepressant drugs. Although their findings of similar rates of major malformations in fluoxetine-exposed infants and unexposed infants are consistent with previous reports,2 the findings of higher rates of perinatal complications after third-trimester exposure to fluoxetine are not.3
The authors propose that the occurrence of three minor malformations in a single infant is a marker for embryonic disruption and, specifically, central nervous system disruption. These criteria are at least controversial. Furthermore, the reliability of the diagnosis of minor malformations even among well-trained examiners is poor, and nearly half the infants in the study were not examined for the presence of minor malformations, suggesting selection bias.
Another problem with this study is that neither the reasons for and duration of admission to a special-care nursery nor the clinical importance of “poor neonatal adaptation” is discussed. Did any infant not leave the hospital with the mother or require follow-up after discharge? Could the assessment of neonatal status have been influenced by maternal treatment status?
The critical problem, however, is the absence of a relevant control group of women. Higher rates of perinatal complications, including lower birth weight, neonatal distress, and prematurity, have been described in the offspring of mothers with mood and anxiety disorders who did not take any psychotropic drug during pregnancy. Given the precautions about the use of psychotropic drugs during pregnancy, the women who continued to take fluoxetine into the third trimester probably had more severe psychiatric illness.
It is good clinical practice to avoid unnecessary exposure to drugs, including fluoxetine, during pregnancy. However, given the risk of relapse of major depression after the discontinuation of antidepressant-drug therapy4 and the risk of puerperal worsening of mood in women with histories of depression or depression during pregnancy,5 antidepressant-drug therapy should not be discontinued late in pregnancy without careful consideration.
5 ReferencesLee S. Cohen, M.D.
Jerrold F. Rosenbaum, M.D.
Massachusetts General Hospital, Boston, MA 021141
Chambers CD ,Johnson KA ,Dick LM ,Felix RJ ,Jones KL . Birth outcomes in pregnant women taking fluoxetine. N Engl J Med 1996;335:1010-1015
Full Text | Web of Science | Medline2
Pastuszak A ,Schick-Boschetto B ,Zuber C , et al. Pregnancy outcome following first-trimester exposure to fluoxetine (Prozac). JAMA 1993;269:2246-2248
CrossRef | Web of Science | Medline3
Goldstein DJ . Effects of third trimester fluoxetine exposure on the newborn. J Clin Psychopharmacol 1995;15:417-420
CrossRef | Web of Science | Medline4
Kupfer DJ ,Frank E ,Perel JM , et al. Five-year outcome for maintenance therapies in recurrent depression. Arch Gen Psychiatry 1992;49:769-773
Web of Science | Medline5
O'Hara MW ,Zekoski EM ,Philipps LH ,Wright EJ . A controlled prospective study of postpartum mood disorders: comparison of childbearing and nonchildbearing women. J Abnorm Psychol 1990;99:3-15
CrossRef | Web of Science | Medline
To the Editor:
Any assessment of the effect of the treatment of maternal psychiatric disorders during pregnancy on prematurity and neonatal well-being requires similar diagnostic groups for comparison, given the effects of the psychiatric disorder on these outcomes.1,2 Chambers et al. attempted to create a control group on the basis of the trimester of exposure: the exposed-early group stopped taking fluoxetine in the first or second trimester, and the exposed-late group took the drug into the third trimester. However, the groups were mismatched with regard to smoking status, alcohol ingestion, and maternal weight gain. These factors may be indicators of underlying psychiatric disorders and have been associated with decreased birth weight and an increased risk of prematurity and neonatal complications.3 These limitations call the validity of the control group in this study into question.
There is no evidence that the use of recommended doses of fluoxetine results in sustained weight loss during pregnancy. When the authors included maternal weight gain (a predictor of birth weight) in their model, the relation between the birth weight of term infants and fluoxe-tine exposure was no longer significant. The smaller weight gain in the fluoxetine-treated mothers may have been due to the individual (or combined) effects of depression, continued smoking or alcohol ingestion, or concomitant psychotropic-drug therapy.
We examined all prospectively identified fluoxetine-exposed infants (all trimesters) in our registry as of April 1996 for prematurity and neonatal well-being (Table 1Table 1
Outcomes of 123 Prospectively Identified Fluoxetine-Exposed Infants (All Trimesters) as of April 1996.). Our group, demographically comparable to the exposed-late group of Chambers et al., had rates of prematurity, stillbirth, and admission to the intensive care nursery similar to those of the authors' control group and other large cohort norms, a finding that is inconsistent with an association between fluoxetine exposure and neonatal complications.Overinterpretation of the results reported by Chambers et al. could result in the treatment of fewer pregnant women with depression, with consequences one desires to avoid.
3 ReferencesDavid J. Goldstein, M.D., Ph.D.
Karen L. Sundell, B.S.
Lois A. Corbin, R.N.
Lilly Research Laboratories, Indianapolis, IN 462851
Orr ST ,Miller CA . Maternal depressive symptoms and the risk of poor pregnancy outcome: review of the literature and preliminary findings. Epidemiol Rev 1995;17:165-171
Web of Science | Medline2
Steer RA ,Scholl TO ,Hediger ML ,Fischer RL . Self-reported depression and negative pregnancy outcomes. J Clin Epidemiol 1992;45:1093-1099
CrossRef | Web of Science | Medline3
Calabrese JR ,Gulledge AD . Psychotropics during pregnancy and lactation: a review. Psychosomatics 1985;26:413-6, 424
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We agree that an ideal control group would consist of pregnant women with severe depression who were not taking any medications. However, we do not believe that such a group, which is similar to ours with respect to other variables, is likely to exist.
With respect to possible bias in the assessment of minor anomalies in our study, a single examiner, who was unaware of the infants' drug exposure, performed all the evaluations. Furthermore, in that minor anomalies are by definition subtle defects of little or no cosmetic or functional importance, the women in our sample would have had no knowledge of their existence before deciding whether to participate in the follow-up phase of our study.
The studies that found an increased rate of perinatal complications in the offspring of mothers with mood disorders who were not treated concerned mothers who differed markedly from those we studied. Both Steer et al.1 and Orr and Miller2 studied inner-city women of low socioeconomic status who were predominantly nonwhite, whereas we studied mostly white women with above-average socioeconomic status.
Goldstein et al. suggest that comparing the pregnancy outcome in women who took fluoxetine until delivery with that in women who stopped earlier in pregnancy is inappropriate owing to differing rates of smoking, alcohol ingestion, and maternal weight gain. However, each of these factors was taken into account in the regression models, and we reported significantly higher relative risks after adjustment for these factors.
With respect to perinatal complications, the data presented by Goldstein et al. suffer from the same limitations as those from any registry: many pregnancies or outcomes may be unreported, and the documentation of outcomes is inconsistent.
Finally, we concur that fluoxetine therapy should not be discontinued without careful consideration. However, we also believe that the results of our study can be used to make a more informed decision regarding this issue.
2 ReferencesKenneth Lyons Jones, M.D.
Kathleen A. Johnson, B.A.
Christina D. Chambers, B.A.
University of California, San Diego, School of Medicine, La Jolla, CA 920931
Steer RA ,Scholl TO ,Hediger ML ,Fischer RL . Self-reported depression and negative pregnancy outcomes. J Clin Epidemiol 1992;45:1093-1099
CrossRef | Web of Science | Medline2
Orr ST ,Miller CA . Maternal depressive symptoms and the risk of poor pregnancy outcome: review of the literature and preliminary findings. Epidemiol Rev 1995;17:165-171
Web of Science | Medline
- Citing Articles (5)
Citing Articles
1
Mary Macdougall, Meir Steiner. (2003) Treatment of premenstrual dysphoria with selective serotonin re-uptake inhibitors: focus on safety. Expert Opinion on Drug Safety 2:2, 161-166
CrossRef2
D. Jeffrey Newport, Molly M. Wilcox, Zachary N. Stowe. (2001) Antidepressants during pregnancy and lactation: Defining exposure and treatment issues. Seminars in Perinatology 25:3, 177-190
CrossRef3
LORI L. ALTSHULER, LEE S. COHEN, MARGARET L. MOLINE, DAVID A. KAHN, DANIEL CARPENTER, JOHN P. DOCHERTY, RUTH W. ROSS. (2001) Treatment of Depression in Women: A Summary of the Expert Consensus Guidelines. Journal of Psychiatric Practice 7:3, 185-208
CrossRef4
Yuri Vedernikov, Sandra Bolanos, Egle Bytautiene, Eva Fulep, George R. Saade, Robert E. Garfield. (2000) Effect of fluoxetine on contractile activity of pregnant rat uterine rings. American Journal of Obstetrics and Gynecology 182:2, 296-299
CrossRef5
David J. Goldstein, Karen Sundell. (1999) A review of the safety of selective serotonin reuptake inhibitors during pregnancy. Human Psychopharmacology: Clinical and Experimental 14:5, 319-324
CrossRef
