Journal of Psychosocial Nursing and Mental Health Services

Psychopharmacology 

Antidepressant Medication and Pregnancy: Time for Randomized Controlled Trials

Robert H. Howland, MD

Abstract

Despite the high prevalence of depression and the fairly common use of anti-depressant drugs during pregnancy, no randomized controlled treatment trials (RCTs) prospectively assess the efficacy and safety of antidepressant drugs in depressed pregnant women and then the subsequent effect of drug exposure (versus placebo exposure) on the neonate and later on the child’s development. Observational studies are used to determine whether a drug exposure during pregnancy is associated with an adverse outcome, but they include various confounding factors and other sources of bias. Systematic reviews and meta-analyses, as well as published guidelines from various organizations, all analyze, interpret, and emphasize the same data somewhat differently, and they do not provide consistent or unequivocal guidance for patients or prescribers. Whether antidepressant drugs are efficacious in pregnant women; whether any or certain antidepressant drugs are harmful to the fetus, neonate, or developing child; and whether untreated depression itself is harmful are uncertainties that should now be addressed in rigorous, well-designed, prospective RCTs.

Abstract

Despite the high prevalence of depression and the fairly common use of anti-depressant drugs during pregnancy, no randomized controlled treatment trials (RCTs) prospectively assess the efficacy and safety of antidepressant drugs in depressed pregnant women and then the subsequent effect of drug exposure (versus placebo exposure) on the neonate and later on the child’s development. Observational studies are used to determine whether a drug exposure during pregnancy is associated with an adverse outcome, but they include various confounding factors and other sources of bias. Systematic reviews and meta-analyses, as well as published guidelines from various organizations, all analyze, interpret, and emphasize the same data somewhat differently, and they do not provide consistent or unequivocal guidance for patients or prescribers. Whether antidepressant drugs are efficacious in pregnant women; whether any or certain antidepressant drugs are harmful to the fetus, neonate, or developing child; and whether untreated depression itself is harmful are uncertainties that should now be addressed in rigorous, well-designed, prospective RCTs.

Depression is twice as common in women compared to men, and approximately 10% to 15% of women experience depression during pregnancy and the postpartum period (Gavin et al., 2005). Although actual treatment rates likely vary in clinical practice, studies have found that approximately 8% of pregnant women are prescribed antidepressant drugs (Andrade et al., 2008; Cooper, Willy, Pont, & Ray, 2007). Despite the high prevalence of depression and the fairly common use of antidepressant drugs during pregnancy, there are no randomized placebo-controlled or active-controlled treatment trials (RCTs) that prospectively assess the efficacy and safety of antidepressant drugs in depressed pregnant women and then the subsequent effect of drug exposure (versus placebo exposure) on the neonate and later on the child’s development (Coverdale, McCullough, & Chervenak, 2008; Domínguez, Ramos, Torrents, García, & Carné, 2012). A relatively recent review of the medical literature stated that information on more than 20,000 women exposed to antidepressant agents during pregnancy is available, but none of these exposures included findings from RCTs (Lorenzo, Byers, & Einarson, 2011).

That an antidepressant drug should be expected to be effective in depressed pregnant women is extrapolated from studies of non-pregnant depressed participants. However, the physiological effects that occur in the body and brain during pregnancy—influencing the pharmaco-kinetics and pharmacodynamics of drugs (Thomas & Yates, 2012)—as well as the psychological context of pregnancy, are likely to make the validity of such an extrapolation questionable. Children and older adults provide examples of patient populations in which treatment studies have not necessarily demonstrated outcomes replicating what is seen in adult patients. Efficacy, safety, and tolerability often differ across these groups.

A major concern for patients and clinicians—including nurses—who prescribe and/or monitor medications is whether the prescribed drug(s) have any adverse effects on the fetus, neonate, or developing child. Drug exposure data on pregnancy outcomes can be derived from animal studies, but these studies provide information that is not necessarily applicable to human beings. Prospective and retrospective observational (epidemiological) studies, such as cohort studies, pregnancy exposure registries, birth defect registries, and case-control studies, are used to determine whether a drug exposure during pregnancy is associated with an adverse outcome. Compared to RCTs, observational studies are generally considered more likely to harbor various confounding factors and other sources of bias that make interpretation of the findings difficult. Many studies of antidepressant medications (especially selective serotonin reuptake inhibitor [SSRI] drugs) used administrative databases to link the prescription of the medications in pregnant women with pregnancy outcomes, but these studies were not designed to study drug safety during pregnancy, and most did not necessarily interview mothers or independently examine the babies (Koren & Nordeng, 2012). Moreover, establishing causality (as opposed to simply revealing an association), based on observational studies, also is difficult.

Among experts, there is debate, disagreement, and controversy about interpreting the findings from various studies. Coverdale et al. (2008) conducted a systematic review to determine whether placebo-controlled RCTs of antidepressant medications with pregnant participants are ethically justified. One question they tried to answer was, “Are there documented cases of serious, far-reaching, and irreversible harm to the fetal or neonatal patient?” They carefully assessed whether the evidence supports a judgment of a causal relationship for antidepressant drug exposure and potential teratogencity, spontaneous abortions, pulmonary hypertension in the newborn, preterm delivery or growth restriction of the fetus, and other neonatal problems. Based on their review, they concluded that current evidence does not support a judgment that any of these outcomes is caused by the use of antidepressant drugs during pregnancy.

Healy, Mangin, and Mintzes (2010) severely criticized the analysis and conclusions of Coverdale et al. (2008) in several ways, but with a particular focus on an analysis of the findings from observational studies of the SSRI drug paroxetine (Paxil®) and the risk of cardiac and other defects. According to Healy et al. (2010), 15 studies contain data on paroxetine exposure and the risks of overall defects. Ten studies found that the odds ratio (OR) of a defect was increased with paroxetine (an OR greater than 1.0); three studies identified a decreased risk with paroxetine (an OR less than 1.0); and two studies found no effect of paroxetine (an OR equal to 1.0). They also argued that the statistical significance of the 95% confidence interval surrounding the OR (a standard reporting approach for studies using OR data) should not be used to determine whether the OR is indicative of a clinically significant risk. Most of the studies they cited had an OR that was not statistically significant. Healy et al. (2010) argued that observational studies can indeed be used to establish causality with respect to adverse effects, and they disagreed with Coverdale et al. (2008) that RCTs are necessary or ethical to conduct in pregnant women. It is notable, however, that in another more recent paper (concerning birth defects and the antihistamine drug doxylamine), Healy refers to the “unreliability and potential invalidity of epidemiological evidence” (p. 33) and that “epidemiological evidence in this domain will always be provisional and rarely decisive” (p. 34) (Persaud & Healy, 2012). It is interesting to me that in a conflict of interest declaration, Healy has been an expert witness in paroxetine-induced birth defect cases (Healy et al., 2010).

Scialli (2010) provided a detailed critique of a meta-analysis of studies of paroxetine exposure during pregnancy in relation to cardiovascular malformations. In disagreement with Healy et al. (2010), Scialli stated that the scientific evidence does not support the conclusion that paroxetine causes cardiovascular defects. To my reading, Scialli’s scientific reasoning seems sound, but he reveals in a disclosure that he is a consultant for the manufacturer of Paxil and has testified as an expert witness in Paxil litigation.

The multitude of published studies investigating the effects of antidepressant drug exposure during pregnancy has resulted in little clarity and ongoing uncertainty. Systematic reviews and meta-analyses, as well as published guidelines from various organizations, all analyze, interpret, and emphasize the same data somewhat differently and they do not provide consistent or unequivocal guidance for patients or prescribers. Reading through the papers by Coverdale et al. (2008), Healy et al. (2010), and Scialli (2010) illustrates this problem. In a recent editorial, Steiner (2012) stated, “The question of whether maternal antidepressant treatment during pregnancy is better or worse for the offspring than untreated maternal depression is still mostly unanswered” (p. 1,130). In their recent review, Domar, Moragianni, Ryley, and Urato (2013) stated, “There is no published study which shows an improved pregnancy outcome in women who took SSRIs compared with unexposed women” (p. 161) and “It is unclear from the available evidence whether there is an association between pregnancy complications and depression” (p. 166). In their review, Yonkers et al. (2009) concluded that there is a paucity of information about depression and spontaneous pregnancy loss; that the current state of information does not support or refute an association between major depression and low birthweight or small-for-gestational-age delivery; that available data neither support nor refute a link between major depression and preterm delivery; that there are no studies that link maternal depression to congenital anomalies in their infants; and that little research has looked at the specific effects of depression during pregnancy on late neonatal or post-neonatal development in children.

Conducting RCTs comparing one or more drugs and placebo in patients who are pregnant is typically considered unethical because of the potential risk exposure to the fetus or neonate (Allesee & Gallagher, 2011). Once a product is approved and marketed (based on a small number of RCTs that do not include pregnant women), a much larger number of women of child-bearing age are potentially eligible for enrollment in naturalistic exposure studies. In these naturalistic exposure studies, women and their fetuses are “enrolled” if they knowingly or unknowingly take medication and are pregnant or become pregnant. The outcomes from these naturalistic exposures then form the basis for the various observational investigations that are used to create the current database of knowledge about the safety of antidepressant drugs during pregnancy.

Given three uncertainties—whether antidepressant drugs are efficacious in pregnant women; whether any or certain antidepressant drugs are harmful to the fetus, neonate, or developing child; and whether untreated depression itself is harmful—it seems to me more unethical to enroll a larger number of women in these naturalistic exposure studies than to enroll a smaller number of pregnant women in rigorous well-designed prospective RCTs that will provide better data to address these three uncertainties (Lyerly, Little, & Faden, 2008). Somewhat cynically, Healy et al. (2010) believe that:

given company involvement in the construction of the evidence that increasingly underpins clinical practice, and the bias involved, it is highly unlikely that calls for RCTs of antidepressants in pregnancy will lead to the design and publication of trials that illuminate the real evidence base for the risks and benefits of the clinical practice with pharmaceuticals.

However, there is no reason to believe that large federally funded multicenter collaborative studies, along the lines of the Women’s Health Initiative ( http://www.nhlbi.nih.gov/whi) or the Sequenced Treatment Alternatives to Relieve Depression study ( http://www.nimh.nih.gov/trials/practical/stard/index.shtml), could not be designed and conducted to provide valid data.

Conclusion

Currently, patients and clinicians, including nurses, have to make “informed” decisions based on suboptimal evidence that is difficult to interpret and not uncommonly contradictory, even among experts. Despite likely concerns from many about the care received in clinical research, the quality of informed consent and the clinical monitoring that would occur in an investigational RCT is likely to be superior to what is typical in psychiatric, primary care, or obstetrical/gyneco-logical outpatient settings, where most antidepressant drugs are prescribed for women.

References

  • Allesee, L. & Gallagher, C.M. (2011). Pregnancy and protection: The ethics of limiting a pregnant woman’s participation in clinical trials. Journal of Clinical Research and Bioethics, 2(108). Retrieved from http://www.omicson-line.org/2155-9627/2155-9627-2-108.digi-tal/2155-9627-2-108.html. doi:
  • Andrade, S.E., Raebel, M.A., Brown, J., Lane, K., Livingston, J., Boudreau, D. & Platt, R. (2008). Use of antidepressant medications during pregnancy: A multisite study. American Journal of Obstetrics and Gynecology, 198, 194.e1–194.e5. doi:10.1016/j.ajog.2007.07.036 [CrossRef]
  • Cooper, W.O., Willy, M.E., Pont, S.J. & Ray, W.A. (2007). Increasing use of antidepressants in pregnancy. American Journal of Obstetrics and Gynecology, 196, 544.e1–544.e5. doi:10.1016/j.ajog.2007.01.033 [CrossRef]
  • Coverdale, J.H., McCullough, L.B. & Chervenak, F.A. (2008). The ethics of randomized placebo-controlled trials of antidepressants with pregnant women: A systematic review. Obstetrics and Gynecology, 112, 1361–1368. doi:10.1097/AOG.0b013e31818c2a27 [CrossRef]
  • Domar, A.D., Moragianni, V.A., Ryley, D.A. & Urato, A.C. (2013). The risks of selective serotonin reuptake inhibitor use in infertile women: A review of the impact on fertility, pregnancy, neonatal health and beyond. Human Reproduction, 28, 160–171. doi:10.1093/humrep/des383 [CrossRef]
  • Domínguez, V., Ramos, N., Torrents, A., García, D. & Carné, X. (2012). Clinical trials during pregnancy: What has been done. European Journal of Clinical Pharmacology, 68, 455–458. doi:10.1007/s00228-011-1145-x [CrossRef]
  • Gavin, N.I., Gaynes, B.N., Lohr, K.N., Meltzer-Brody, S., Gartlehner, G. & Swinson, T. (2005). Perinatal depression: A systematic review of prevalence and incidence. Obstetrics and Gynecology, 106, 1071–1083. doi:10.1097/01.AOG.0000183597.31630.db [CrossRef]
  • Healy, D., Mangin, D. & Mintzes, B. (2010). The ethics of randomized placebo controlled trials of antidepressants with pregnant women. International Journal of Risk and Safety in Medicine, 22, 7–16. doi:
  • Koren, G. & Nordeng, H. (2012). Antidepressant use during pregnancy: The benefit-risk ratio. American Journal of Obstetrics and Gynecology, 207, 157–163. doi:10.1016/j.ajog.2012.02.009 [CrossRef]
  • Lorenzo, L., Byers, B. & Einarson, A. (2011). Antidepressant use in pregnancy. Expert Opinion on Drug Safety, 10, 883–889. doi:10.1517/14740338.2011.583917 [CrossRef]
  • Lyerly, A.D., Little, M.O. & Faden, R. (2008). The second wave: Toward responsible inclusion of pregnant women in research. International Journal of Feminist Approaches to Bioethics, 1(2), 5–22. doi:10.2979/FAB.2008.1.2.5 [CrossRef]
  • Persaud, N. & Healy, D. (2012). Epidemiological evidence in forensic pharmacovigilance. International Journal of Risk and Safety in Medicine, 24, 31–35. doi:
  • Scialli, A.R. (2010). Paroxetine exposure during pregnancy and cardiac malformations. Birth Defects Research. Part A, Clinical and Molecular Teratology, 88, 175–177. doi:
  • Steiner, M. (2012). Prenatal exposure to antide-pressants: How safe are they?American Journal of Psychiatry, 169, 1130–1132. doi:10.1176/appi.ajp.2012.12081126 [CrossRef]
  • Thomas, S.H.L. & Yates, L.M. (2012). Prescribing without evidence—Pregnancy. British Journal of Clinical Pharmacology, 74, 691–697. doi:10.1111/j.1365-2125.2012.04332.x [CrossRef]
  • Yonkers, K.A., Wisner, K.L., Stewart, D.E., Oberlander, T.F., Dell, D.L., Stotland, N. & Lockwood, C. (2009). The management of depression during pregnancy: A report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. General Hospital Psychiatry, 31, 403–413. doi:10.1016/j.genhosppsych.2009.04.003 [CrossRef]
Authors

Dr. Howland is Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, Pittsburgh, Pennsylvania.

The author has disclosed no potential conflicts of interest, financial or otherwise.

Address correspondence to Robert H. Howland, MD, Associate Professor of Psychiatry, University of Pittsburgh School of Medicine, Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Pittsburgh PA 15213; e-mail: HowlandRH@upmc.edu.

Posted Online: January 18, 2013

10.3928/02793695-20130109-01

Sign up to receive

Journal E-contents