Antidepressant Use in Pregnancy: a Critical Review Focused on Risks and Controversies

Review Antidepressant use in pregnancy: a critical review focused on risks and controversies

Byatt N, Deligiannidis KM, Freeman MP. Antidepressant use in N. Byatt1, K. M. Deligiannidis2, pregnancy: a critical review focused on risks and controversies. M. P. Freeman3 1Psychiatry and Obstetrics & Gynecology, Psychosomatic Objective: Conflicting data have led to controversy regarding Medicine, Women’s Mental Health, University of antidepressant use during pregnancy. The objectives of this study are to Massachusetts Medical School/UMass Memorial i) review the risks of untreated depression and anxiety, ii) review the Medical Center, Worcester, MA, USA, 2Psychiatry and literature on risks of exposure to antidepressants during pregnancy, iii) Obstetrics & Gynecology, Depression Specialty Clinic, discuss the strengths and weaknesses of the different study designs used Women’s Mental Health Specialty Clinic, Center for to evaluate those risks, and iv) provide clinical recommendations. Psychopharmacologic Research and Treatment, Method: MEDLINE/PubMed was searched for reports and studies on University of Massachusetts Medical School/UMass the risk of first-trimester teratogenicity, postnatal adaptation syndrome Memorial Medical Center, Worcester, MA, USA and 3Psychiatry, Perinatal and Reproductive Psychiatry (PNAS), and persistent pulmonary hypertension (PPHN) with in utero Program, Harvard Medical School/Massachusetts antidepressant exposure. General Hospital, Boston, MA, USA Results: While some individual studies suggest associations between some specific major malformations, the findings are inconsistent. Key words: antidepressant; in utero; teratogenicity; Therefore, the absolute risks appear small. PNAS occurs in up to 30% postnatal adaptation syndrome; persistent pulmonary of neonates exposed to antidepressants. In some studies, PPHN has hypertension been weakly associated with in utero antidepressant exposure, while in Nancy Byatt, Psychiatry and Obstetrics & Gynecology, other studies, there has been no association. Psychosomatic Medicine, Women’s Mental Health, Conclusion: Exposures of concern include that of untreated maternal University of Massachusetts Medical School/UMass illness as well as medication exposure. It is vital to have a careful Memorial Medical Center, 55 Lake Avenue North, discussion, tailored to each patient, which incorporates the evidence to Worcester, MA 01655, USA. date and considers methodological and statistical limitations. Past E-mail: [email protected] medication trials, previous success with symptom remission, and women’s preference should guide treatment decisions. Accepted for publication October 10, 2012

Summations • No single type of malformation has been consistently observed across studies with any commonly used antidepressant. Some individual studies suggest associations between particular selective serotonin reuptake inhibitors (SSRIs) and speciﬁc birth defects. • Postnatal adaptation syndrome (PNAS) occurs in up to 30% of neonates exposed to antidepressants in late pregnancy. • Some studies ﬁnd a small association between persistent pulmonary hypertension of the newborn (PPHN) and SSRI use, although other studies do not.

Considerations • SSRIs remain the most studied antidepressants in pregnancy. Less data are available for serotonin– norepinephrine reuptake inhibitors (SNRIs), mirtazapine, nefazodone, trazodone and vilazodone. • The current evidence base is limited by data that do not i) systematically assess infants, ii) use appropriate control groups, iii) use blind raters of the neonates and iv) take into account maternal diagnosis or symptoms or other confounding variables. • The evidence regarding the risk of PPHN because of in utero antidepressant exposure is inconclusive.

94 Antidepressant use in pregnancy

adverse effects, women are likely to stop antide- Introduction pressant treatment during attempts to conceive or During the reproductive years, a significant pro- pregnancy (28, 29). portion of women experienced depressive and anx- Due to conflicting reports in the literature on the iety disorders (1). Approximately 18.4% of women risk of first-trimester teratogenicity, postnatal suffer from antenatal depression, and as many as adaptation syndrome (PNAS), and persistent pul- 19.2% of mothers develop a depressive disorder monary hypertension (PPHN) with in utero anti- within several weeks of delivery (2). Anxiety disor- depressant exposure, a review of these topics is of ders are also common, with a prevalence rate of great clinical importance. In this review, selected 21.7% during the 3rd trimester of pregnancy, and topics of controversy regarding antidepressant use 11.1% during the first 3 postpartum months (3, 4). in pregnancy are presented, to provide clarification Untreated depression and anxiety during preg- during clinical decision-making and risk/benefit nancy negatively impacts mother and fetus/child. assessment and discussion. A review of all possible Women are more likely to experience inadequate outcomes is beyond the scope of this study. maternal weight gain (5) and abuse substances (6). Although not reviewed in this article, there are Depression in pregnancy is also associated with associations between antidepressant in utero expo- preeclampsia, preterm birth (7–10), increased risk sure and measures of birth outcome, including an for delivery of a low birth weight infant (11), elec- increased rate of spontaneous abortion (30–35) tive termination of the pregnancy (12), postpartum (36–39), low birth weight (35, 40, 41), and reduced depression (13), and anxiety. Depression is associ- gestational age or preterm birth (35, 40, 42–44) in ated with fetal distress (14) and an increased risk depressed women exposed to antidepressants. of neonatal care unit admission and caesarian Recent studies have raised questions about possi- delivery (15). ble associations with antidepressant use in Postpartum depression can negatively impact pregnancy, including autism (45) and effects on child development and has been associated with long-term neurocognitive development (46). While difficult infant and childhood temperament (16, all associations with in utero exposure are impor- 17) and attachment insecurity (16). Maternal tant, we limited our review to topics that have depression may also lead to emotional and func- become the main controversies in antidepressant tional disability in children including cognitive use during pregnancy in recent years. delays (18), behavioural problems (16), and difficulties with social interaction (19). Children of Aims of the study depressed mothers are at increased risk of developmental delay, impaired language development, and The purposes of this review are to i) review the lower IQ scores (20, 21). The impact of maternal risks of untreated depression and anxiety, ii) depression has effects beyond infancy, as one-third review the literature on risks of exposure to com- of school-aged children of depressed mothers suf- monly used antidepressants during pregnancy, iii) fer from depressive, anxiety, or disruptive disor- discuss the strengths and weaknesses of the differ- ders (22). Effective treatment of maternal ent study designs used to evaluate those risks, and depression mitigates this negative impact (23). iv) provide clinical recommendations. Perinatal depression can also be fatal; maternal suicide accounts for up to 20% of postpartum Material and methods deaths in depressed women (24). Stress and anxiety during pregnancy influence A search was performed for the English language maternal behaviour and birth outcomes: maternal literature indexed on MEDLINE/PubMed for the tobacco smoking, caffeine consumption, poor period between 1966 and 2012 using the following nutrition and exercise, preterm labor, preterm key terms: antidepressant, selective serotonin reup- birth, and low birth weight are associated with pre- take inhibitor (SSRI), serotonin-norepinephrine natal anxiety (25). Antenatal anxiety may increase reuptake inhibitor (SNRI), noradrenergic and spe- the risk of childhood developmental and psychiat- cific serotonin antidepressant (NaSSA), norepi- ric disorders. It may adversely affect infant emo- nephrine reuptake inhibitor (NRI), fluoxetine, tional development (26) and has been associated paroxetine, sertraline, citalopram, escitalopram, with reductions in gray matter density in young fluvoxamine, venlafaxine, mirtazapine, reboxetine, children (27). Treatment of perinatal depressive duloxetine, bupropion, trazodone, nefazodone, and anxiety disorders is of paramount importance and vilazodone in association with antenatal to mitigate these risks. Despite the risk of relapse depression, maternal, pregnancy, prenatal expo- to depressive episodes and anxiety and associated sure, malformation, in utero exposure, neonatal

95 Byatt et al. complications, gestational, neonatal health, neona- prospective interview data on drug use. For pre- tal outcome, birth outcome, PPHN, birth defects, scription data, they recommended avoiding the use and congenital heart defect. To determine whether of prescriptions given earlier than 1 month before other relevant articles were not identified in the ini- the last menstrual period because of the increased tial search, all articles were cross-referenced. Origi- percentage of women who did not use the medica- nal observational studies, case reports, and case tion in pregnancy (50). series were included. Across all study types, the risks of the underlying untreated disorder and other associated confounding factors are rarely taken into account. Results The strategy of using depressed women without drug treatment as a comparison may not be ade- First-trimester exposure: is there evidence of teratogenicity? quate given the likelihood that illness severity cov- In the USA, approximately 1 in every 33 infants aries with medication use. Comparisons of (3%) is born with a major birth defect (47). Major different SSRI effects may also be complicated by birth defects or malformations typically require the fact that this class of medications is used for medical or surgical intervention for a structural or many conditions other than depression, including functional abnormality. Antidepressants that affect anxiety disorders (51). SSRI preference may also serotonergic tone could putatively increase the differ according to indication, and prescribing pat- incidence of congenital malformations because terns may differ between populations. To control serotonin is important in aspects of early embry- for maternal illness severity, Oberlander et al. used onic development that impact development of the a propensity score method that allowed them to neural tube, branchial arch, and heart. control for maternal illness severity and other Despite the high prevalence of birth defects characteristics that impact neonatal outcomes. overall in the population, each specific type of Unfortunately, even with their large population major birth defect is generally rare, and studies level data set, propensity score matching led to a that are inadequately powered or controlled may reduced sample size for comparison groups (52). overestimate risk and association. For example, of The use of different study designs and limita- the major birth defects studied in association with tions of each may explain why findings of malfor- antidepressant use, omphalocele occurs in 1 per mations are inconsistent. The data presented in 5386 births, gastroschisis occurs in 1 per 2229, Table 1 are difficult to interpret because it is diffi- anencephaly 1 in every 4859, and craniosynostosis cult to differentiate whether adverse outcome is in 4–10 per 10 000 births in the general population associated with underlying illness, the medication (47). While congenital heart disease is relatively itself, or other unknown factors associated with common (48, 49), studies examining the effects of either or both. drug exposure should ideally use the rate of unexposed infants with identical ascertainment methods Selective serotonin reuptake inhibitors. Prior to as a comparison rate for the malformation(s) 2005, studies had not suggested an increased risk under study, rather than the 1% incidence of con- of major congenital malformations with in utero genital heart disease (48, 49) in the general popula- exposure to SSRIs (35, 36, 43, 53–59). These stud- tion, which makes interpretation of findings ies, as well as those since, were generally limited by difficult. insufficient power, confounding variables, con- Different methodologies have been used to study cerns with the method of birth outcome classifica- the risks of teratogenicity with antidepressant use tion, and limited exposure information (60). Over in pregnancy. Some use retrospective case–control the past several years, a number of studies have studies, which carry the risk of recall bias and large found associations between specific SSRIs and non-response rates. Other prospective controlled malformations, although the findings are inconsis- studies, often performed in teratology information tent across studies, making their interpretation dif- centers, use detailed drug use information in a ficult. A teratogen would be expected to cause a small number of women. A final type of study uses similar type of malformation consistently across data from drug registries and administrative data- studies. While this has not been the case with bases, which carry the risk of exposure misclassifi- SSRIs, sporadic reports have generated concerns. cation given that it is not clear whether women Since 2005, the data have been conflicting and who purchased the drug took it as prescribed. inconsistent (61) with regard to whether individual When they recently compared interview data with SSRIs are associated with an increased risk of con- data from a prescription registry, Kallen et al. genital malformations. Discrepant findings have found the most valid results are obtained through fueled confusion, and it is unclear whether un-rep-

96 Table 1. Maternal use of antidepressants and the risk of congenital defects

Study Design Antidepressant studied N Findings

Pedersen, et al. (2009) (79) Population-based, Fluoxetine, citalopram, 493 113 unexposed infants and Increased prevalence of septal defects with prescription cohort study paroxetine, sertraline 1370 infants exposed to SSRI of >1 SSRI (2.1%) vs. a single SSRI (0.9%) vs. not prescribed an SSRI (0.5%) Increased prevalence of septal heart defects after first-trimester exposure to sertraline (1.5%) or citalopram (1.1%) yet not fluoxetine (0.6%) or paroxetine (0.3%) Overall no association with major malformations or non-cardiac malformations Wogelius, et al. (2006b) (64) Population-based, Fluoxetine, citalopram, 150 780 women with no SSRI Increased risk of congenital malformations with cohort paroxetine, sertraline prescription vs. 1,051 women prescribed SSRI early in pregnancy compared with who filled SRRI prescription unexposed infants (3.4% vs. 4.9%) 30 days before conception to (aRR = 1.34; 95% 1.00–1.79) and late in pregnancy end of first trimester vs. 453 (6.8%) (aRR = 1.85; 95% CI 1.25–2.71) filled prescription during SSRI use not associated with any specific malformation second or third trimester Wen SW, et al. (2006) (80) Population-based, Fluoxetine, citalopram, 972 women exposed to an No increased risk of major (OR = 0.98; 95% CI 0.59–1.64) retrospective cohort paroxetine, sertraline, SSRI; 3878 not exposed or minor malformations (OR = 1.02; 95% 0.69–1.51) fluvoxamine with prenatal SSRI exposure Reis and Kallen (2010) (41) Population-based, SSRIs, bupropion, 14 821 women with 15 017 Increased risk of general teratogenicity after exposure prospective cohort trazodone, SNRIs, infants with early exposure, to fluoxetine (OR = 1.31; 95% CI 0.85–2.02) and of TCAs, MAOIs, later exposure or both and cardiovascular defects after paroxetine exposure unspecified compared with 1 236 053 (OR = 1.66; 95% CI 1.09–2.53) antidepressant infants in the general population Risk of hypospadias with SSRI exposure (OR = 1.30; 95% CI 0.94–1.80) and higher with paroxetine (OR = 2.45; 95% CI 1.12–4.64) TCAs (primarily clomipramine) associated with a higher risk of teratogenicity overall Risk of cystic kidney disease was elevated for SSRI (n = 9) (OR = 2.39; 95% CI 1.09–4.54) Kallen and Otterbla Population-based, Citalopram, sertraline, 6555 infants exposed to Increase in the risk of heart defects and atrial and Olausson (2007) (51) prospective cohort fluoxetine, paroxetine first-trimester SSRI use ventricular defects with paroxetine exposure No overall increase in congenital malformations with pregnancy in use Antidepressant SSRI exposure (AOR = 0.89; 95% CI 0.79–1.07) Paroxetine was associated with an increased risk of ventricular and atrial septum defects (OR = 1.81; 95% CI = 0.96–3.09) No association between craniosynostosis or omphalocele and maternal SSRI use Einarson A, et al. (2008) (74) Prospective, cohort Paroxetine 3379 infants with first-trimester Paroxetine is not associated with an increased risk paroxetine exposure vs. 1174 of cardiovascular birth defects unpublished cases vs. 2061 infants from published databases vs. an unexposed cohort 97 98 Table 1. (Continued) al. et Byatt

Study Design Antidepressant studied N Findings

Einarson A, et al. (2009) (81) Prospective, cohort Bupropion, citalopram, 928 women exposed to Prevalence of cardiac malformations was below the escitalopram, fluvoxamine, antidepressants in pregnancy prevalence rate at 0.6% nefazodone, paroxetine, vs. 928 controls mirtazapine, fluoxetine, trazodone, venlafaxine, sertraline Einarson A, et al. (2011) (82) Prospective, cohort Bupropion, citalopram, 1243 with first-trimester No association between antidepressant exposure fluvoxamine, nefazodone, antidepressant exposure vs. 89 and malformations paroxetine, mirtazapine, women exposed to >1 antidepressant fluoxetine, trazodone, vs., 89 exposed to 1 antidepressant venlafaxine, sertraline vs. 89 not exposed Klieger-Grossman, Prospective, cohort Escitalopram 212 women exposed to citalopram vs. Escitalopram not associated with increased risk of et al. (2011) (85) 212 exposed to other antidepressants major malformation vs. 212 exposed to non-teratogens Malm H, et al. (2011) (73) Population-based, Fluoxetine, citalopram, 6881 mother and child pairs with Overall major congenital abnormalities were not retrospective cohort paroxetine, sertraline, first-trimester SSRI exposure vs. 618 associated with SSRI exposure fluvoxamine, escitalopram 727 mother and child pairs with no (AOR = 1.08; 95% CI 0.96–1.22) exposure Fluoxetine associated with increased risk of ventricular septal defects (AOR = 2.03; 95% CI 1.28–3.21) Paroxetine associated with increased risk of ventricular outflow defects (AOR 4.68; 95% CI 1.48–14.74) Citalopram associated with neural tube defects (AOR = 2.46; 95% CI 1.20–5.07) Cole JA, et al. (2007) (66) Case control Paroxetine as compared 815 infants with exposure to Increased overall rate of malformations in infants after with all other antidepressants paroxetine monotherapy vs. 1020 first-trimester paroxetine monotherapy exposure including SSRIs, SNRIs, infants with mono- or polytherapy (AOR = 1.89; 95% CI 1.20–2.98), mono- or polytherapy TCAs, serotonin-2 antagonist exposure vs. 4936 infants with (AOR = 1.76; 95% CI 1.18–2.64) reuptake inhibitors, MAOIs other antidepressant mono- or No increase in cardiovascular malformations with polytherapy exposure vs. and a paroxetine monotherapy (AOR = 1.46; 95% CI 0.74–2.88) subset of 4198 infants with other or for mono- or polytherapy antidepressant monotherapy (AOR = 1.68; 95% CI 0.95–2.97) exposure Cole et al. (2007) (93) Case control Bupropion compared with all 1213 infants with first-trimester No increase in malformations with first-trimester other antidepressants bupropion exposure vs. 4743 infants bupropion exposure including SSRIs, SNRIs, with first-trimester other TCAs, serotonin-2 antagonist antidepressant exposure vs. 1049 reuptake inhibitors, MAOIs infants with bupropion exposure outside first trimester Table 1. (Continued)

Study Design Antidepressant studied N Findings

Louik, et al. (2007) (62) Case control Fluoxetine, citalopram, 9849 infants with malformations vs. SSRI use not associated with increased risk of paroxetine, sertraline, 5860 control infants craniosynostosis (OR = 0.8; 95% CI 0.2–3.5) omphalocele fluvoxamine, venlafaxine, (OR = 1.4; 95% CI 0.4–4.5) or heart defects overall escitalopram, bupropion (OR = 1.2; 95% 0.9–1.6) Sertraline associated with omphalocele (OR = 5.7; 95% CI 1.6–20.7) and septal defects (OR = 2.0; 95% CI 1.2–4.0) Paroxetine associated with right ventricular outflow tract obstruction defects (OR = 3.3; 95% CI 1.3–8.6) Berard, et al. (2007) (69) Case control Paroxetine compared with all 1403 women with antidepressant No association between first-trimester exposure to other antidepressants including exposure vs. infants without paroxetine (OR = 1.38; 95% CI 0.49–3.92) or other SSRIs, bupropion, trazodone, malformations SSRIs (OR = 0.89; 95% CI 0.28–2.84) and major congenital SNRIs, TCAs malformations serotonin-2 antagonist reuptake Infants exposed to >25 mg of paroxetine had an increased inhibitors, MAOIs risk of major congenital anomalies (AOR = 2.23; 95% CI 1.19–4.17) or major cardiac malformations (AOR = 3.07; 95% CI 1.00–9.42) Alwan S, et al. (2007) (60) Population-based, Citalopram, sertraline, 9622 infants with major birth SSRI use not associated with most congenital heart defects case control fluoxetine, paroxetine defects vs. 4092 control infants or other birth defects SSRI use was associated with anencephaly (AOR = 2.4; 95% CI 1.1–5.1), cranosynostosis (AOR = 2.5; 95% CI 1.5–4.0), and omphalocele (AOR = 2.8, 95% CI 1.3–5.7) Bakker, et al. (2010) (72) Population-based, Paroxetine 678 cases with heart defects vs. Increased OR for atrial septal defects case control 615 controls (AOR = 5.7; 95% CI 1.4–23.7) but not for heart defects overall (AOR = 1.5; 95% 0.5–4.0), after preconception and/or first-trimester paroxetine exposure Bakker, et al. (2010) (76) Population-based, Fluoxetine 4255 infants, 178 of which were Association between fluoxetine and infantile hypertrophic case control exposed to fluoxetine in the first stenosis (1.7% infants exposed to fluoxetine vs. 0.2% trimester non-exposed vs. infants with other malformations (AOR = 9.8; 95% CI 1.5–62.0). Ramos, et al. (2008) (83) Retrospective case SSRIs, TCAs, SNRIs, 2329 women, of which 189 infants No association between antidepressant use in the first control bupropion, mirtazapine, had major congenital malformation, trimester and major congenital malformations pregnancy in use Antidepressant moclobemide, trazodone, vs. 2140 without malformation (AOR = 1.10; 95% 0.75–1.62) nefazadone No association between duration of antidepressant exposure and the prevalence of major congenital malformations Antidepressant class exposure was not associated with major congenital malformations Wisner, et al. (2009) (84) Prospective, controlled SSRIs 238 women; 131 without SSRI No association between continuous or first trimester and observational exposure or MDD vs. 71 with minor or major malformations partial or continuous SSRI exposure vs. partial or continuous MDD (n = 36) 99 Byatt et al.

licated results represent true associations. One 2.5) 7.5) – – example of conﬂicting ﬁndings is the retrospective 1.6) – case–control studies by Alwan et al. and Louik et al. that linked the use of SSRI drugs with rare malformations. Both studies carry a risk for recall 1.7; 95% CI 1.1 3.3; 95% CI 1.5 = 1.3; 95% CI 1.1 = bias and have a high rate of non-responders (60, = 62). Other studies have other methodological limi- 6) (OR 4.39)

– tations, for example, Wogelius et al. (63) identiﬁed = malformations from discharge diagnoses in women with and without intrauterine SSRI exposure. Examining physicians were aware of SSRI expo-

2.17; 95% CI 1.07 sure, which may have increased vigilance during = examinations, and led to surveillance bias and septal defects, 1.7% (n and overall malformations (OR risk of cardiac malformations (OR risk of non-syndromic congenital heartvs. malformations 1.6% of non-exposed newborns (RR overestimation of risks because of the identiﬁca- SSRI exposure was associated with an overall increased SSRI exposure associated with two-fold increased Sertraline was associated with an increased risk of tion of less serious malformations, for example, mild cardiac defects that would not otherwise have been identiﬁed. Other studies are so small that they lack statistical power to detect anything, but extreme risk increases that are unlikely to occur with SSRI exposure. A small increased absolute risk of rare defects such as omphalocele, anencephaly, craniosynostosis (60), cystic kidney, and congenital heart defects –

tio; CI, confidence interval; MAOI, monoamine oxidase inhibitor; SNRI, serotonin norepinephrine reuptake (51, 64 66) has also been reported. Importantly, speciﬁc patterns of congenital malformations have prescriptions during early pregnancy in 8 of 235 (3.4%)heart vs. congenital defects in 1083 of(1.6%) 67 non-exposed 636 newborns not been demonstrated with SSRIs across studies, 2062 of 216 042 women had SSRI Congenital heart defects identified and teratogenicity is usually determined by a consistent risk and pattern of malformation. Among the SSRIs, paroxetine has become the most complicated in terms of reported risks and recommendations for use in pregnancy.

Paroxetine. In 2005, a small study performed by GlaxoSmithKline suggested an increase in cardiac malformations in infants exposed to paroxetine in fluoxetine, paroxetine, citalopram, escitalopram fluoxetine, paroxetine, citalopram, escitalopram utero compared with controls. As a result, Glaxo- Sertraline, fluvoxamine, SmithKline modified the prescribing information to include a warning regarding the risk of cardiac malformations with antenatal paroxetine exposure. This study was not peer-reviewed or published, but was presented to the US FDA and cited elsewhere (67). As this recommendation was based on non- peer-reviewed, unpublished data, and a relatively small sample size, its findings are difficult to interpret (68). Since then, multiple studies (41, 51, 60, 66, 69–73), although not all (65, 74), have found observational an association between prenatal paroxetine exposure and an increased risk of congenital malformations, yet the causality and magnitude of that risk are unclear. A recent meta-analysis based on research prior to 2006 found that paroxetine was associated with 1.7-fold risk increase of cardiac malformation (71). A later review, however, concluded that it is not practical to use a meta-analysis to examine the TCA, tricyclic; AOR, adjusted odds ratio;inhibitor. SSRI, selective serotonin reuptake inhibitor; RR, relative risk; aRR, adjusted relative risk; OR, odds ra Table 1. (Continued) Study Design Antidepressant studied N Findings Kornum JB, et al. (2010) (78) Population-based prevalence study Sertraline, fluvoxamine, Merlob, et al. (2009) (61) Prospective, controlled, safety of paroxetine in pregnancy given the limita-

100 Antidepressant use in pregnancy tions in the methodology of the published studies antidepressants. Compared with the SSRIs, there (75). Another meta-analysis that examined 37 are fewer reports on the reproductive safety pro- studies from January of 1992 through September files of other antidepressants. While the available 2008 provides further evidence of an increased risk evidence is extremely limited, studies examining of major congenital malformations with paroxetine venlafaxine, duloxetine, nefazodone, and mirtaza- exposure. The authors concluded that first-trimes- pine do not suggest an increased risk of congenital ter paroxetine exposure is associated with an malformations (32, 39, 81, 89, 90). An increased increased prevalence of combined cardiac defects risk of left outflow tract heart defects has been [prevalence odds ratio (POR) = 1.46%; 95% CI inconsistently demonstrated in association with 1.17–1.82] and aggregated defects (POR = 1.24; bupropion. Overall, the limited studies have shown 95% CI 1.08–1.43) (70). reassuring, but not definitive data regarding the reproductive safety of venlafaxine, trazodone, Other SSRIs. Specific individual studies have mirtazapine, and bupropion. found malformations associated with specific SSRIs as isolated reports. These include hypertro- Bupropion. Bupropion has efficacy for smoking phic stenosis (76), congenital heart defects, and cessation, and tobacco use is associated with birth other major abnormalities (41, 65, 73, 77) with flu- complications. Understanding the safety of bupro- oxetine, omphalocele (62, 77–79) and cardiac sep- pion in pregnancy is important as it may be a tal defects (62, 77, 79) with sertraline, and potential treatment option for women with depres- omphalocele (62, 77–79), congenital heart defects sion and nicotine dependence. In 1997, Glaxo- (62, 77, 79), and neural tube defects (73) with cita- SmithKline established a Bupropion Pregnancy lopram. Other studies, however, have not sug- Registry and by March 2008, 3.6% and 1.3% of gested an association between fluoxetine (62, 64, infants exposed to bupropion were reported to 69, 77–84), sertraline (41, 51, 69, 73, 80–84), or cit- have congenital abnormalities and congenital heart alopram (41, 80–82, 84) and major congenital defects, respectively. Congenital heart defects were abnormalities. While the data are very limited, found in both retrospective and prospective escitalopram has not been associated with risk of reports (91), raising concern that first-trimester bu- major malformation (62, 73, 81, 85). propion exposure might increase the risk of congenital heart defects. The first prospective study on Meta-analyses. When interpreting this data, it is bupropion use in pregnancy did not demonstrate imperative to consider meta-analyses, because the an increased risk of major malformation with bu- power of individual studies may not be adequate propion exposure; however, the sample was small and large numbers of subjects are needed to dem- with only enough power to detect a five-fold onstrate association with congenital malforma- increased risk (37). A retrospective case–control tions. Five meta-analyses have investigated the risk study that examined the risk of bupropion expo- for major malformations in association with anti- sure 1 month prior to conception until 3 months depressant use during pregnancy. Four of these after conception found that exposed infants were studies found no statistically significant increased more likely to have left outflow tract heart defects, risk of major malformations in the first trimester but not other defects (OR = 2.6; 95% CI 1.2–5.7) of pregnancy (31, 86–88). The fifth meta-analysis (92). Another case–control study did not find an found an increased risk of cardiac malformations increased risk of congenital malformations when in infants exposed to paroxetine in the first they compared first-trimester bupropion exposure trimester (71). to (1) first-trimester exposure to other antidepres- In summary, paroxetine and other SSRIs have sants and (2) bupropion exposure outside the first not consistently been demonstrated to be associ- trimester (93). Other studies (37) (92) are reassur- ated with particular birth defects. Paroxetine use ing as they do not demonstrate an increased risk of during early pregnancy has been the most contro- congenital malformations following exposure to versial, as it has been associated with an increased bupropion during pregnancy. Even with the possi- risk of overall major malformations, particularly ble increased risk of congenital heart defects (91, atrial and ventricular septal defects in several 92), the absolute risk of a congenital heart defect studies (51, 60, 66, 69, 72), but assessments of remains low at 2.1/1000 births in exposed infants large databases have not supported this finding (92) when compared with the estimated prevalence (65, 74). of 0.82/1000 births in the general population (49).

Serotonin–norepinephrine reuptake inhibitors (SNRIs), SNRIs and other non-SSRI antidepressants. There norepinephrine reuptake inhibitors (NRI), and other are a very limited number of studies examining

101 Byatt et al.

SNRIs (such as venlafaxine and duloxetine). rological development, which all involve serotonin. A large prospective cohort study that included the Premature neonates may be more susceptible to SNRI venlafaxine and other non-SSRI antidepres- PNAS given their immature lung and central ner- sants – mirtazapine, nefazadone, bupropion, and vous systems. trazodone – found that the prevalence of cardiac Since PNAS was first noted in 1973 (94), expo- malformations was well below the prevalence rate sure to antidepressants during late pregnancy has at 0.6% in the antidepressants as a group (81). been associated with infant irritability, abnormal Data obtained from the Swedish Medical Birth crying, tremor, lethargy, hypoactivity, decreased Registry also do not suggest an increased risk of feeding, tachypnea, and respiratory distress (95– congenital malformations after exposure to SNRI/ 99). As data emerge, this cluster of symptoms is NRIs. However, because of small sample size, it increasingly referred to as PNAS. It has also been only would have detected a marked teratogenic referred to as ‘neonatal behavioural syndrome’ or effect (89). ‘poor neonatal adaptation syndrome’ with studies focusing on a collection of symptoms including Mirtazapine. A prospective, comparative study irritability, tachypnea, hypothermia, and hypogly- that examined whether exposure to mirtazapine cemia (96, 100–102). Clinical signs and symptoms during organogenesis increased the rate of major usually develop from birth to days after delivery malformations found that mirtazapine was and are time limited. While they usually resolve not associated with an increased risk of major mal- within days or weeks of delivery (95), symptoms formation (39). have been reported to last as long as 6 weeks (103). Severity and length are impacted by multiple Trazodone/nefazodone/vilazodone. Data on trazo- factors including dose, timing and duration of done and nefazodone are limited with the excep- exposure, and SSRI pharmacology including half- tion of a multicenter prospective controlled study, life, presence of active metabolites, and maternal which found that trazodone and nefazadone did and infant hepatic cytochrome P450 isoenzyme not increase the rates of major malformation genotype, among others (104). above the baseline (90). A literature review did not As shown in Table 2, Different methodologies reveal any data on vilazodone. have been used to examine the relationship between PNAS and antidepressants. Databases of Summary. Recently, the safety of SSRIs in preg- adverse drug event reports have the advantage of nancy has been challenged by data from large pop- being able to detect effects too rare to be detected ulation-based studies. Reports are difficult to in clinical trials. Similar to case reports, however, interpret because of the lack of consistent findings databases of adverse drug reaction reports are lim- and the inability to assess cause and effect from ited by lack of incidence rate determination, und- association studies. Most reports do not take into erreporting and reporter bias (95). The account the underlying psychiatric condition and identification of PNAS symptoms related to anti- variables that may not be controlled for that would depressant exposure is also complicated by the differ between groups being compared. While some challenge of determining whether the symptoms individual studies suggest associations between are due to maternal illness or medication exposure SSRIs and some specific major malformations, the (105). Important considerations in evaluating the findings are inconsistently observed; therefore, the data include the following: i) whether systematic absolute risks appear small. While the limited avail- assessments of infant were conducted, ii) whether able data suggest a possible association between bu- appropriate control groups were included in the propion and congenital heart defects (91, 92), the study, iii) whether raters of the infants were absolute risk appears low. Although the very lim- blinded to antidepressant exposure, iv) whether ited studies examining SNRIs have been reassuring, maternal diagnosis or symptoms were taken into further investigation is needed before the risks consideration, and v) whether other confounding associated with their use may be fully understood. variables may contribute to neonatal symptoms. Numerous mechanisms have been proposed for PNAS including serotonin toxicity (104), oversti- Later pregnancy controversies mulation of serotonin (101), and infant genotype Postnatal adaptation syndrome (PNAS). All SSRIs (106). Many PNAS symptoms overlap those found cross the placenta, carrying the potential to in adult SSRI discontinuation syndrome, choliner- increase serotonin concentrations in the developing gic overdrive, and serotonin syndrome. Several fetus. Increased serotonin concentrations may case reports note infant toxicity after in utero impact fetal cardiovascular, respiratory, and neu- exposure to paroxetine (106) and fluoxetine (97,

102 Antidepressant use in pregnancy

Table 2. Maternal use of antidepressants and the risk of postnatal adaptation syndrome (PNAS)

Study Design Antidepressant studied N Findings

Chambers et al. (1996) (35) Prospective, controlled Fluoxetine 73 infants with PNAS associated with late third-trimester exposure only compared antidepressant with early exposure exposure vs. 101 (aRR = 8.7) infants with first- and Late pregnancy exposure second-trimester associated with NICU antidepressant admission (RR = 2.6) exposure Oberlander, et al. (2004) (104) Prospective, controlled Paroxetine, fluoxetine, sertraline 28 infants with late PNAS symptoms in 39% follow-up or citalopram alone or in pregnancy SSRI of infants with SSRI and combination with clonazepam exposure vs. 18 clonazepam exposure vs. infants with SSRI 25% exposed to SSRI and and clonazepam 9% of controls exposure vs. 23 (unadjusted RR = 3.5) non-exposed infants Costei, et al. (2002) (102) Prospective, controlled Paroxetine 55 with late exposure Third-trimester paroxetine to paroxetine, 27 exposure was associated infants with early with PNAs symptoms in exposure, and 27 22% of exposed compared with no exposure with 11% controls (unadjusted RR = 4.0) Zeskind and Stephens (2004) (156) Prospective, controlled Fluoxetine, citalopram, paroxetine, 17 SSRI-exposed vs. SSRI-exposed infants sertraline, fluvoxamine 17 non-exposed exhibited a wide range infants of neurobehavioural outcomes (p>0.05) Laine, et al. (2003) (101) Prospective, controlled Citalopram, fluoxetine 20 mothers and infants Exposed infants had lower with late pregnancy Apgar scores and a four-fold antidepressant increase in serotonergic exposure vs. 20 symptom score in first 4 mothers and infants days of life (P = 0.008) without exposure As serotonergic symptoms increased, cord blood 5-H1AA levels decreased (P = 0.007) Galbally, et al. (2009) (118) Prospective, controlled Fluoxetine, citalopram, paroxetine, 27 women exposed Third-trimester antidepressant sertraline, fluvoxamine, mianserin, to medication vs exposure associated with mirtazapine, escitalopram matched controls increased risk of PNAS, increased rate of admission to the special care nursery and higher rates of jaundice no difference between specific antidepressants Maschi, et al. (2008) (120) Prospective, controlled Fluoxetine, citalopram, paroxetine, 200 neonates with No difference between sertraline, fluvoxamine antidepressant antidepressant exposure exposure vs. 1200 and PNAS controls Levinson-Castiel, et al. (2006) (116) Matched control SSRIs, venlafaxine 120 infants; 60 with PNAS occurred in 30% of prolonged infants with SSRI exposure antidepressant vs. none in non-exposed exposure vs. 60 group controls 13% met criteria for severe neonatal dose–response effect. Cohen, et al. (115) Retrospective, cohort study Fluoxetine 64 mother–infant Increased risk of special pairs care nursery admissions after exposure to fluoxetine in late trimester when compared with early trimester exposure (18.9% vs. 9.1%)

103 Byatt et al.

Table 2. (Continued)

Study Design Antidepressant studied N Findings

Rampono, et al. (2009) (157) Prospective, observational Fluoxetine, citalopram, paroxetine, 56 mother–infant Neonatal abstinence scores sertraline, escitalopram, pairs; 27 exposed were higher (p < 0.05) on day fluvoxamine, venlafaxine to SSRI vs. 11 to 1 in exposed infants compared venlafaxine vs. with controls 18 controls Exposed infants had higher serotonin scores and Brazelton Neonatal Behavioral Assessment Scale (BNBAS) scores for motor and autonomic clusters, habituation, and social interactive (P<0.05) Suri, et al. (2007) (44) Prospective, naturalistic Sertraline, fluoxetine, paroxetine, 90 women comprised Infants of women with MDD study bupropion, nortriptyline of 49 with MDD and exposed to taking antidepressant antidepressants had vs. 22 with MDD greater rates of admission either not taking to the special care nursery antidepressant or then women with MDD who limited exposure were not treated with vs. 19 healthy controls antidepressants and healthy women (21%, 9%, and 0%, respectively) Suri et al. (2011) (158) Prospective, naturalistic Sertraline, fluoxetine, paroxetine, 64 women; 33 with MDD No differences between study bupropion, nortriptyline and antidepressants vs. groups regarding Apgar 16 with a history of scores, special nursery MDD who were either admissions or on the not treated with BNBAS antidepressant or had limited exposure vs. 15 healthy controls Hendrick et al. (2003) (55) Prospective case series Fluoxetine, citalopram, paroxetine, 131 infants exposed 10.5% of infants sertraline, fluvoxamine, venlafaxine to SSRIs exhibited PNAS Sit et al. (2011) (159) Prospective case series Fluoxetine, citalopram, paroxetine, 21 mother–infant pairs No association between sertraline, fluvoxamine, nortriptyline with antidepressant cord-to-maternal exposure concentration ratios and perinatal events Ferreira et al. (2007) (117) Retrospective chart review Fluoxetine, citalopram, paroxetine, 66 neonates with SSRI Behavioural signs in 77.6% sertraline exposure vs. 90 without of SSRI-exposed neonates exposure compared with 41% non-exposed Tremors, agitation, spasms, hypotonia, irritability, and sleep disturbances were reported in 63.2% of exposed infants and respiratory effects in 40.8% Kallen (2004) (40) Prospective Swedish Birth Paroxetine, fluoxetine, sertraline, 555 infants with late Increased risk for respiratory Registry citalopram SSRI exposure vs. 728 distress (OR = 2.21), controls hypoglycemia (OR = 1.62), convulsions (OR = 1.9) low Apgar score (OR = 2.33) with maternal use of antidepressants Effects were not specific to any SSRI Warburton et al. (2010) (52) Retrospective register study Fluoxetine, citalopram, paroxetine, Infants exposed to No difference in neonatal sertraline, fluvoxamine antidepressants in symptoms among women the last 14 days of exposed to antidepressant pregnancy vs. infants in the last 14 days of exposed earlier in pregnancy when compared pregnancy with those who were not Sanz et al. (2005) (100) WHO database case series Paroxetine, fluoxetine, sertraline, 93 infants with late 69% of cases with neonatal review citalopram SSRI exposure behavioural symptoms were associated with paroxetine (n = 64), 14 with fluoxetine, 9 with sertraline, and 7 with citalopram

104 Antidepressant use in pregnancy

107–109). Several of those reports note elevated SSRIs and SNRIs may both cause PNAS symp- infant serum levels, supporting toxicity as the toms. A prospective observational study compar- cause (97, 107, 108, 110). Other reports document ing placental transfer and neonatal effects of SSRI discontinuation symptoms following in utero anti- and SNRI (venlafaxine) exposure in pregnancy to depressant exposure (110), specifically with sertra- non-exposed matched controls found that both line (111), paroxetine (98–100, 112, 113), and SSRIs and venlafaxine transferred across the pla- venlafaxine (114). centa and were associated with PNAS symptoms In 1996, Chambers et al. (35) published the first (121). An analysis of the World Health Organiza- cohort study examining PNAS and found that late tion adverse events database reported 94 cases of pregnancy exposure to fluoxetine was associated PNAS after in utero exposure to SSRIs and venla- with an increased special care nursery admission faxine. PNAS symptoms in infants were noted rate when compared with exposure earlier in preg- after maternal fluoxetine, citalopram, or paroxe- nancy. A similar study that retrospectively com- tine exposure (100). However, the total number of pared exposure with fluoxetine early and late in women using these medications was not reported, pregnancy found an increased risk of special care so the incidence of neonatal symptoms was not nursery admissions after late pregnancy exposure possible to ascertain. to fluoxetine (115). In 2004, the FDA suggested (122) that providers Another study that prospectively compared neo- consider tapering antidepressants in the third tri- natal complications in infants found that third-trimester. While discontinuation is intuitive if antide- mester paroxetine exposure had a high rate of pressants are associated with neonatal symptoms, neonatal complications compared with controls this recommendation did not receive formal clini- (22% vs. 5.5%, respectively) (102). Using the same cal study prior to its release. Since this recommen- definition as Chambers et al., Oberlander et al. dation, Warburton (2010) and colleagues assessed found that thirty per cent of infants exposed to babies of mothers who were not exposed to antide- SSRI alone or in combination with clonazepam pressants in the last 14 days of pregnancy com- showed symptoms of poor neonatal adaptation, pared with those who were not exposed and 25% and 39%, respectively (104). Another hospi- factored in maternal psychiatric symptoms and tal-based cohort study found PNAS in 30% of other possible cofounding variables into their anal- infants exposed to SSRIs (116). This rate of 30% is ysis. After accounting for confounding variables, higher than found by Costei et al. (22%) (102) and there was no difference in PNAS among women Hendrick et al. (10.5%) (55), but very similar to exposed to antidepressant in the last 14 days of that found by Chambers et al. (31.5%) (35) and pregnancy when compared with those who were Oberlander et al. (30%) (104). The data collected not. However, the total number of women using by Costei may not represent the true rate as data these medications was not reported, so the inci- were collected later after delivery and cases may dence of neonatal symptoms was not possible to have been missed. A retrospective chart review ascertain (52). This study was also limited by the study reported a higher rate of behavioural signs inclusion of subjects on fluoxetine, which may have (77.6%) in SSRI-exposed infants compared with been in the subjects’ system given its long half-life 41% of the non-exposed (117). of 2 weeks. In a prospective comparison study, using cord PNAS appears to be multifactorial in nature, blood levels of 5-hydroxyindoleacetic acid with late pregnancy antidepressant use accounting, (5-H1AA) and a modified Serotonin Syndrome in part, for neonatal symptoms. It is important to Scale, infants exposed to citalopram or fluoxetine consider that maternal anxiety has also been asso- in late pregnancy had lower Apgar scores and ciated with changes in infant behaviour and self- more serotonergic symptoms than infants not regulation (123). Exposure to antidepressants in exposed (101). Other prospective studies have also pregnancy, regardless of timing, has been associ- found that third-trimester exposure to antidepres- ated with PNAS. While the available evidence is sants is associated with PNAS symptoms (118) and conflicting, the overall data suggest that PNAS can greater special care nursery admission rates (119). occur in neonates exposed to SSRIs and SNRIs, While it lacked statistical significance, a prospec- yet have most often been reported after exposure to tive controlled cohort noted a trend for increased paroxetine, fluoxetine, and venlafaxine (84, 120). risk of PNAS symptoms in infants exposed to antidepressants in the second and third trimester. Persistent pulmonary hypertension of the newborn Mild symptoms may have been underreported (PPHN). Pulmonary hypertension is a normal because of recall bias during data collection inter- and required state for the fetus in utero because views with mothers (120). the placenta, as opposed to the lung, is responsible

105 Byatt et al. for gas exchange. At birth, the lung replaces the cially when compared with the prevalence found in placenta as the primary site of gas exchange, and the general population (0.07–0.10%). A subse- there is a rapid drop in pulmonary vascular resis- quent case–control study found that SSRI use after tance and a resultant increase in pulmonary blood the 20th week of pregnancy was significantly asso- flow. Multiple chemical pathways are responsible ciated with PPHN (adjusted odds ratio for this cardiopulmonary transition (124). (AOR) = 6.1), but use of SSRIs or other antide- PPHN can result whenever cardiopulmonary pressants prior to 20-week gestation was not (138). transition does not occur. PPHN is a rare disorder A retrospective analysis of data from the Swedish that occurs in approximately 1–2 per 1000 births Medical Birth Register also found an association (125). Reduced length of gestation and premature between PPHN and maternal SSRI use during birth has been associated with increased risk of early pregnancy [relative risk (RR) = 2.4] and late PPHN (121). Infants with PPHN present within pregnancy (RR = 3.6); though, the absolute risk twelve hours of birth with cyanosis and mild respi- was small (139). This group’s most recent analysis ratory distress and can develop severe respiratory that also included tricyclics, monoamine oxidase failure requiring intubation and mechanical venti- inhibitors, SNRIs, and other antidepressants dem- lation (126). Even with therapy, PPHN can be fatal onstrated an increased RR of PPHN for antide- in approximately 10–20% of cases (127), depend- pressant exposure in early pregnancy (RR = 2.30), ing upon the etiology (128). Approximately 25% for later exposure (RR = 2.56), and for both early of infants with moderate-to-severe PPHN will and later exposure (RR = 3.44) (41). demonstrate significant neurodevelopmental Most of the aforementioned studies (41, 140, impairment at 12–24 months (129). 141) restricted their analyses to infants delivered Respiratory insufficiency is one symptom of after 34 completed weeks of gestation because PNAS and may represent the presence of a mild shorter gestation is associated with an increased form of PPHN and not PNAS, per se (130, 131). It risk of PPHN. A recent population-based cohort is not certain that SSRIs are associated with the study used national health register data demon- development of PPHN, but there is some evidence strated an increased risk of PPHN associated with that longer periods of in utero SSRI exposure may SSRI prescription after 20-week gestation and be associated with increasing risk, and severity, of before 8-week gestation (142). The risk of PPHN neonatal respiratory complications (132). How associated with exposure to individual SSRIs was SSRIs may affect neonatal respiratory complica- of a similar magnitude, suggesting a class effect. tions is under investigation. The accumulation of This study restricted their analyses to infants deliv- SSRIs in the lungs may result in high circulating ered after 33-week gestation and was the largest levels of serotonin which, through its vasoconstric- completed study of the relationship between in ute- tive effects, increases pulmonary vascular ro SSRI exposure and PPHN. resistance and may cause proliferation of smooth As shown in Table 3, other studies have sug- muscle cells in the fetal lung (133), but not all stud- gested no association between antidepressant use ies have found elevated serotonin levels (134). during pregnancy and PPHN. Two studies (143, Another possible mechanism may involve the 144) compared the prevalence of PPHN among inhibitory effect of SSRIs on nitric oxide synthesis, infants whose mothers were exposed to antidepres- an essential vasodilator that regulates vascular sants in the third trimester of pregnancy compared tone (135), but other studies do not support that with infants not exposed and did not find any dif- hypothesis (134). Importantly, through unidenti- ference. The Wichman et al. study (144) is limited fied mechanisms, depression and the use of SSRIS by a lack of delineation between women treated in during pregnancy have been associated with a late pregnancy and during the first trimester. reduced length of gestation and increased risk of Because late trimester use has been identified as a premature birth (84) that itself is associated with possible risk factor for PPHN, the inclusion of an increased risk of PPHN. Additionally, genetic women who were treated in the first trimester may factors contribute to the risk of developing PPHN have contributed to the negative finding. Addition- (128). Functional polymorphisms in the serotonin ally, the small size of both of these studies makes it transporter promoter region may modulate the likely that a possible association would have been risk of PPHN in both adults (136) and infants with missed. Most recently, a case–controlled study of in utero exposure to SSRIs (128, 137). 11 923 births, including 20 cases of PPHN, demon- In 1996, Chambers et al. (35) first reported that strated an increased risk of PPHN with cesarean late in utero exposure to fluoxetine was associated delivery prior to the onset of labor (OR = 4.9) but with an increased risk of PPHN when compared not with SSRI use during the second half of preg- with first-trimester exposure (2.7% vs. 0%), espe- nancy (145).

106 Antidepressant use in pregnancy

Table 3. Maternal use of antidepressants and the risk of persistent pulmonary hypertension of the newborn (PPHN)

Study Design Antidepressant studied N Findings

Chambers CD et al. (1996) (35) Prospective observational Fluoxetine 228 exposed women vs. Late in utero exposure cohort 254 controls associated with increased risk of PPHN as compared with first-trimester exposure (2.7% vs. 0%) Chambers et al. (2006) (138) Multicenter case control Citalopram, fluoxetine, 377 women whose infants After the 20th week of paroxetine, sertraline, had PPHN vs. 836 matched pregnancy, antidepressant TCA, bupropion, venlafaxine, controls use was associated with trazodone PPHN (AOR = 6.1), but use of antidepressants prior to 20-week gestation was not Kallen and Olausson (2008) (139) Population-based retrospective Citalopram, sertraline, 504 infants with PPHN of Early pregnancy SSRI use cohort fluoxetine, mirtazapine, 831 324 births; 11 of which (RR = 2.4) and late paroxetine had antidepressant exposure pregnancy use (after 34 weeks) (RR = 3.6) associated with PPHN Andrade et al. (2009) (143) Retrospective review SSRI, TCA, miscellaneous 1104 exposed infants vs. Prevalence of PPHN among 1104 matched controls infants with any third-trimester antidepressant exposure = 1.81 per 1000 infants No association among those infants exposed to SSRIs in third trimester (2.14 per 1000) vs. unexposed infants (2.72 per 1000) Wichman et al. (2009) (144) Retrospective review Citalopram, venlafaxine, 24 406 women with no SSRI 16 newborns diagnosed with escitalopram, paroxetine, use vs. 808 women with SSRI PPHN, but none of whom fluoxetine, sertraline, exposure had exposure to SSRI more than 1 SSRI (0.07% vs. 0.0% p>0.99) Wilson, et al. (2010) (160) Prospective database case SSRI 11 923 births Use of SSRIs in the second control half of pregnancy was identified in 5% of the controls but none of the cases (OR = 0) Reis and Kallen (2010) (41) Population-based, prospective TCA, SSRIs, MAOIs, 12 914 women with early Increased relative risk of cohort SNRIs (other antidepressant) exposure vs. 5987 with later PPHN for antidepressant exposure vs. 4080 with both exposure in early pregnancy early and late exposure vs. (RR = 2.30), for later exposure 1 062 190 women without (RR = 2.56) and for both early exposure and later exposure (RR = 3.44) Kieler et al. (2011) (142) Population-based, prospective SSRIs, other antidepressant 28 067 women filled a Increased risk of PPHN for SSRI cohort; national health prescription for SSRI vs. 1 588 prescription in late pregnancy registers 140 women without (AOR = 2.1; 3 per 1000 liveborn infants compared with the background incidence of 1.2 per 1000) and before 8-week gestation (AOR = 1.4) Women without SSRI use and with prior hospitalization for a psychiatric disorder were at increased risk of having infant with PPHN (AOR = 1.3) Women with a prior hospitalization for a psychiatric disorder and late gestation SSRI prescription were at greater risk (AOR = 3.1)

TCA, tricyclic; AOR, adjusted odds ratio; SSRI, selective serotonin reuptake inhibitor; RR, relative risk; OR, odds ratio; MAOI, monoamine oxidase inhibitor; SNRI, serotonin– norepinephrine reuptake inhibitor.

In 2006, based on data from the Chambers of pregnancy. This advisory resulted in changes to et al. study, the FDA published a Public Health drug labeling to include a risk of PPHN with the Advisory regarding an increased risk of PPHN use of antidepressants during late pregnancy. In associated with the use of SSRIs after the 20th week December 2011, the FDA released a Drug Safety

107 Byatt et al.

Communication (146) which stated that there is lines the inherent and varying difficulties conduct- insufficient evidence that antidepressant exposure ing epidemiological studies of this nature. In such during pregnancy causes PPHN. This recent state- a complicated area of study, no individual study is ment is an accurate reflection of the current litera- definitive. Statistical significance does not necessar- ture that has reported either a small association ily translate into a valuable clinical or epidemio- between PPHN and maternal antidepressant use logical finding. For example, an OR of 1.2 during pregnancy or no association. compared with 1.7 may be statistically significant, Although studies have also reported an associa- yet not important from an epidemiological or clini- tion between mode of delivery and PPHN (140, cal perspective. 141, 147–149), several were small and lacked con- The current evidence for malformations is lim- trol groups. Studies that have investigated the role ited because of inconsistent findings and limited of cesarean delivery and the risk of PPHN are lim- methodology of the published studies (150). Few ited because it is not clear whether the risk of studies have controlled for maternal illness and PPHN is increased secondary to the mode of deliv- therefore do not take into account whether repro- ery or intrauterine fetal distress. ductive outcomes are due to maternal illness or antidepressant exposure. Many studies also group malformation types together to detect a difference Clinical implications and discussion in risk. For example, some studies have grouped The treatment of women during pregnancy is com- together all septal defects or congenital heart dis- plex, and clinical decisions should be based on the ease, encompassing a variety of birth defects and risks, benefits, and alternatives to psychopharma- a range of severity. As a result, it is not known cological treatment. All risks need to be consid- whether or not the reported increased risk for car- ered, including those of untreated maternal diac defects is for minor, moderate, or severe psychiatric illness and the known and unknown forms, which each carries varying medical risks to potential risks of psychotropic medication. To pro- the infant. Larger sample sizes are required to vide optimal clinical care to women and their improve the identification of particular terato- developing child, it is imperative to consider risks genic patterns, so that specific birth defects associ- of treatment in the context of illness severity, con- ated with antidepressants can be isolated sequences of no treatment and under-treatment consistently and reproduced across studies. While and individual treatment preferences. the large studies that do not show evidence of teratogenicity are reassuring, the literature also con- tains a number of studies that suggest concerns. Interpreting a conflicting evidence base The current evidence base for PNAS also has Clinicians face the challenge of interpreting an limitations in that it does not: i) systematically expanding and sometimes controversial evidence assess infants, ii) use appropriate control groups, base that is summarized in Table 4. Comparing the iii) use blind raters of the neonates, and iv) take available studies is difficult because of methodo- into account maternal diagnosis or symptoms or logical weaknesses and differences in study design, other confounding variables. Despite these limita- outcome measures, and exposure (72). This under- tions, these study findings vary less than those of the teratogenicity or PPHN studies and therefore more strongly suggest an association between Table 4. Results summary antidepressant use in pregnancy and PNAS. Teratogenicity Similarly, the PPHN literature is limited by While individual studies suggest associations between selective serotonin reuptake inhibitors (SSRIs) and some specific major malformations, the findings are small and/or uncontrolled studies. There are other inconsistent; therefore, the absolute risks appear small reported risk factors, including race, method of SSRIs remain one of the best-studied classes of medications used in pregnancy delivery, obesity, asthma, and diabetes (141) that Less data are available for serotonin–norepinephrine reuptake inhibitors, many studies do not take into account. While the mirtazapine, nefazadone, trazodone, and vilazodone While an increased risk of left outflow tract heart defects has been inconsistently PPHN literature is inconclusive, the available evi- demonstrated in association with bupropion, the absolute risk of a congenital dence reports either a small association between heart defect remains low PPHN and maternal antidepressant use during Postnatal Adaptation Syndrome (PNAS) pregnancy or no association. PNAS has been reported to occur in up to 30% of neonates exposed to antidepressants in late pregnancy and has most often been reported after exposure to paroxetine, fluoxetine, and venlafaxine Persistent pulmonary hypertension (PPHN) Clinical importance of results The available evidence reports either a small association between PPHN and maternal antidepressant use during pregnancy or no association Given the inconclusive evidence, it is vital to have a careful discussion tailored to each patient that

108 Antidepressant use in pregnancy incorporates the evidence to date on the risks and pressants during pregnancy while carefully benefits of antidepressant use in pregnancy. We considering the risk of untreated illness. What recommend educating women about the risks of poses a greater risk: exposure to untreated illness exposure to antidepressants throughout preg- or the antidepressant? If the benefits of treatment nancy, which should include the US Food and outweigh the risks, then the medication should be Drug Administration categorization. Unfortu- prescribed based on an individualized risk/benefit nately, the US FDA antidepressant risk categori- assessment and discussion. Past medication trials zation is limited and does not adequately inform and previous success with symptom remission and the decision-making process. The limitations of women’s preference should guide treatment deci- this system include lack of a requirement for sys- sions. To avoid exposure to more than one antide- tematic human data, lack of clear differentiation pressant or under-treated illness, one should between pregnancy categories, lack of incorpora- choose a medication with known efficacy for indi- tion of risks of the maternal illness, and potential vidual women. The goal of treatment should be to benefits of the medication. Newer drugs are usually maximize non-medication evidence-based treat- much more poorly studied in human pregnancy, as ments, such as specific forms of psychotherapy, little human data are required before a drug comes and remission of the maternal symptoms, with to market. Therefore, healthcare providers and judicious use of pharmacotherapy when indicated. patients who rely on the US FDA categories may Unless there is a reason to use another class of receive an oversimplified and often inaccurate antidepressant, SSRIs are generally considered assessment of the knowledge base regarding a spe- first-line in pregnancy. SSRIs are well character- cific medication in human pregnancy. Despite ized, and even though risks have been reported, numerous studies, information regarding possible the preponderance of data is reassuring. It makes teratogenicity of antidepressants has not been less sense to preferentially use a medication with updated since an FDA public health advisory in less available human data. When selecting an anti- 2005 was released concerning paroxetine and the depressant for a pregnant woman who has not had possible associated risk of increased cardiac mal- past medication trials, many providers prefer fluo- formations. The FDA is currently working on risk xetine because of the amount of data available and categorization that will include more meaningful lack of long-term developmental sequela in chil- and useful information to clinicians (151). How- dren (152). A current response or history of a posi- ever, with the current limitations of the FDA cate- tive response to medication should help determine gories, it is crucial to incorporate the current which medication to continue or initiate. The bene- research literature into the decision-making and fits of discontinuing an effective medication often informed consent process. do not outweigh the risks of relapse or of exposing mother and fetus to a second antidepressant medication during pregnancy. Translating results into clinical practice To maximize the treatment of depression and As summarized in Table 5, we recommend weigh- minimize risks of maternal and fetal exposure to ing the risks and benefits of treatment with antide- antidepressants and untreated depression, women should receive the minimal effective dose of an antidepressant. The increased dose requirements Table 5. Clinical points of emphasis across gestation (153, 154) should be considered Unless there is a reason to use another class of antidepressant, selective serotonin and weighed against the risks when determining reuptake inhibitors are generally considered first-line in pregnancy the minimal effective dose. It is also important to A current response or history of a positive response to medication should help determine which medication to continue or initiate avoid under-treatment as residual depression and Strongly consider using an antidepressant that the woman has responded to in the medication exposure represent dual exposures for past, to avoid unnecessary exposures during pregnancy the fetus. Maximize non-medication evidence-based treatments Polypharmacy with multiple psychotropic medi- To avoid exposure to both illness and medication, use lowest possible dose while also avoiding under-treatment cations should also be avoided, unless there is a Avoid polypharmacy with multiple psychotropic medications, unless there is a clear indication for the use of multiple pharmaco- clear indication for the use of multiple pharmacotherapies therapies. Some studies suggest that the use of Tapering antidepressants in the third trimester has not been shown the decrease SSRIs in combination with benzodiazepines may in the incidence of postnatal adaptation syndrome or improve infant outcomes, and it carries the risk of precipitating relapse of depression increase the risk of congenital heart malformations Exposure to untreated illness can be mitigated by preconception counseling, (77) and PNAS (104). Another study did not dem- psychoeducation regarding risks and benefits of treatment and no treatment, onstrate an increased risk of congenital malforma- close clinical monitoring, and a careful treatment plan tailored to the each woman tions after exposure to multiple antidepressants (82). It is important to consider that approximately

109 Byatt et al.

50% of pregnancies are unplanned and many 5. Bodnar LM, Wisner KL, Moses-Kolko E, Sit DK, women enter the first trimester on several medica- Hanusa BH. Prepregnancy body mass index, gestational tions. For example, a positive response to taking weight gain, and the likelihood of major depressive disorder during pregnancy. J Clin Psychiatry two antidepressants concurrently may indicate that 2009;70:1290–1296. the risk of relapse with discontinuation of either 6. Flynn HA, Chermack ST. Prenatal alcohol use: the role outweighs the risk of polypharmacy. of lifetime problems with alcohol, drugs, depression, and Despite the FDA’s suggestion in 2004 (122) that violence. J Stud Alcohol Drugs 2008;69:500–509. 7. Cripe SM, Frederick IO, Qiu C, Williams MA. Risk of providers consider tapering antidepressants in the preterm delivery and hypertensive disorders of preg- third trimester, there is no evidence suggesting this nancy in relation to maternal co-morbid mood and approach reduces the incidence of PNAS or migraine disorders during pregnancy. Paediatr Perinat improves infant outcomes. It also carries the risk Epidemiol 2011;25:116–123. of precipitating relapse or postpartum depression, 8. Wisner KL, Zarin DA, Holmboe ES et al. Risk-benefit particularly in high-risk individuals (52, 95). decision making for treatment of depression during pregnancy. Am J Psychiatry 2000;157:1933–1940. Individualized treatment recommendations or 9. Wadhwa PD, Sandman CA, Porto M, Dunkel-Schetter C, plans should aim to diminish both the risk of Garite TJ. The association between prenatal stress and exposure to untreated illness and the antidepres- infant birth weight and gestational age at birth: a pro- sant. Exposure to untreated illness can be spective investigation. Am J Obstet Gynecol – mitigated by preconception counseling, psychoed- 1993;169:858 865. 10. Istvan J. Stress, anxiety, and birth outcomes: a critical ucation regarding risks and benefits of treatment review of the evidence. Psychol Bull 1986;100:331–348. and no treatment, close clinical monitoring, and a 11. Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar careful treatment plan tailored to the each woman S, Katon WJ. A meta-analysis of depression during preg- (155). The risks associated with antidepressant use nancy and the risk of preterm birth, low birth weight, and intrauterine growth restriction. Arch Gen Psychia- can be mitigated using antidepressants with reas- try 2010;67:1012–1024. suring human data, using the minimal effective 12. Suri R, Altshuler LA, Mintz J. Depression and the deci- dose, and avoidance of polypharmacy. Outcomes sion to abort. Am J Psychiatry 2004;161:1502. for mother and child can be optimized by the utili- 13. Gotlib IH, Whiffen VE, Wallace PM, Mount JH. Pro- zation of multidisciplinary approach that takes spective investigation of postpartum depression: factors into account risks of treatment and no treatment involved in onset and recovery. J Abnorm Psychol 1991;100:122–132. and each woman’s preferences for treatment. This 14. Jablensky AV, Morgan V, Zubrick SR, Bower C, Yella- can be accomplished by a careful discussion, tai- chich LA. Pregnancy, delivery, and neonatal complica- lored to each patient, which incorporates the evi- tions in a population cohort of women with dence to date and considers methodological and schizophrenia and major affective disorders. Am J Psy- – statistical limitations. Past medication trials, previ- chiatry 2005;162:79 91. 15. Chung TK, Lau TK, Yip AS, Chiu HF, Lee DT. Antepar- ous success with symptom remission, and women’s tum depressive symptomatology is associated with preference should guide treatment decisions. adverse obstetric and neonatal outcomes. Psychosom Med 2001;63:830–834. 16. Forman DR, O’hara MW, Stuart S, Gorman LL, Larsen Declaration of interests KE, Coy KC. Effective treatment for postpartum depression is not sufficient to improve the developing mother- The authors declare that they have no conflict of interest. child relationship. Dev Psychopathol 2007;19:585–602. 17. Britton JR. Infant temperament and maternal anxiety References and depressed mood in the early postpartum period. Women Health 2011;51:55–71. 1. Vesga-Lopez O, Blanco C, Keyes K, Olfson M, Grant 18. Sohr-Preston SL, Scaramella LV. Implications of timing BF, Hasin DS. Psychiatric disorders in pregnant and of maternal depressive symptoms for early cognitive and postpartum women in the United States. Arch Gen Psy- language development. Clin Child Fam Psychol Rev chiatry 2008;65:805–815. 2006;9:65–83. 2. Gavin NI, Gaynes BN, Lohr KN, Meltzer-Brody S, 19. Downey G, Coyne JC. Children of depressed parents: an Gartlehner G, Swinson T. Perinatal depression: a system- integrative review. Psychol Bull 1990;108:50–76. atic review of prevalence and incidence. Obstet Gynecol 20. Deave T, Heron J, Evans J, Emond A. The impact of 2005;106:1071–1083. maternal depression in pregnancy on early child devel- 3. Borri C, Mauri M, Oppo A et al. Axis I psychopathology opment. BJOG 2008;115:1043–1051. and functional impairment at the third month of preg- 21. Paulson JF, Keefe HA, Leiferman JA. Early parental nancy: results from the perinatal depression-research depression and child language development. J Child Psy- and screening unit (PND-ReScU) study. J Clin Psychia- chol Psychiatry 2009;50:254–262. try 2008;69:1617–1624. 22. Pilowsky DJ, Wickramaratne PJ, Rush AJ et al. Children 4. Reck C, Struben K, Backenstrass M et al. Prevalence, of currently depressed mothers: a STAR*D ancillary onset and comorbidity of postpartum anxiety and study. J Clin Psychiatry 2006;67:126–136. depressive disorders. Acta Psychiatr Scand 2008;118: 23. Weissman MM, Pilowsky DJ, Wickramaratne PJ et al. 459–468. Remissions in maternal depression and child psychopa-

110 Antidepressant use in pregnancy

thology: a STAR*D-child report. JAMA 2006;295: 41. Reis M, Kallen B. Delivery outcome after maternal use 1389–1398. of antidepressant drugs in pregnancy: an update using 24. Lindahl V, Pearson JL, Colpe L. Prevalence of suicidality Swedish data. Psychol Med 2010;40:1723–1733. during pregnancy and the postpartum. Arch Womens 42. Lund N, Pedersen LH, Henriksen TB. Selective serotonin Ment Health 2005;8:77–87. reuptake inhibitor exposure in utero and pregnancy out- 25. Lobel M, Cannella DL, Graham JE, Devincent C, Schnei- comes. Arch Pediatr Adolesc Med 2009;163:949–954. der J, Meyer BA. Pregnancy-specific stress, prenatal 43. Simon GE, Cunningham ML, Davis RL. Outcomes of pre- health behaviors, and birth outcomes. Health Psychol natal antidepressant exposure. Am J Psychiatry 2008;27:604–615. 2002;159:2055–2061. 26. Beydoun H, Saftlas AF. Physical and mental health out- 44. Suri R, Altshuler L, Hellemann G, Burt VK, Aquino A, comes of prenatal maternal stress in human and animal Mintz J. Effects of antenatal depression and anti- studies: a review of recent evidence. Paediatr Perinat depressant treatment on gestational age at birth and Epidemiol 2008;22:438–466. risk of preterm birth. Am J Psychiatry 2007;164:1206– 27. Buss C, Davis EP, Muftuler LT, Head K, Sandman CA. 1213. High pregnancy anxiety during mid-gestation is associ- 45. Croen LA, Grether JK, Yoshida CK, Odouli R, Hendrick ated with decreased gray matter density in 6-9-year-old V. Antidepressant use during pregnancy and childhood children. Psychoneuroendocrinology 2010;35:141–153. autism spectrum disorders. Arch Gen Psychiatry 28. Cohen LS, Altshuler LL, Harlow BL et al. Relapse of 2011;68:1104–1112. major depression during pregnancy in women who 46. Gentile S, Galbally M. Prenatal exposure to antidepres- maintain or discontinue antidepressant treatment. sant medications and neurodevelopmental outcomes: a JAMA 2006;295:499–507. systematic review. J Affect Disord 2011;128:1–9. 29. Bennett IM, Marcus SC, Palmer SC, Coyne JC. Preg- 47. Rynn L, Cragan J, Correa A. Update on overall preva- nancy-related discontinuation of antidepressants and lence of major birth defects–Atlanta, Georgia, 1978– depression care visits among Medicaid recipients. Psy- 2005. MMWR Morb Mortal Wkly Rep 2008;57:1–5. chiatr Serv 2010;61:386–391. 48. Hoffman JI, Kaplan S. The incidence of congenital heart 30. Hemels ME, Einarson A, Koren G, Lanctot KL, Einarson disease. J Am Coll Cardiol 2002;39:1890–1900. TR. Antidepressant use during pregnancy and the rates 49. Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle of spontaneous abortions: a meta-analysis. Ann Phar- WT, Correa A. Prevalence of congenital heart defects in macother 2005;39:803–809. metropolitan Atlanta, 1998-2005. J Pediatr 2008;153: 31. Rahimi R, Nikfar S, Abdollahi M. Pregnancy outcomes 807–813. following exposure to serotonin reuptake inhibitors: a 50. Kallen B, Nilsson E, Olausson PO. Antidepressant use meta-analysis of clinical trials. Reprod Toxicol during pregnancy: comparison of data obtained from a 2006;22:571–575. prescription register and from antenatal care records. 32. Einarson A, Fatoye B, Sarkar M et al. Pregnancy out- Eur J Clin Pharmacol 2011;67:839–845. come following gestational exposure to venlafaxine: a 51. Kallen BA, Otterblad Olausson P. Maternal use of selec- multicenter prospective controlled study. Am J Psychia- tive serotonin re-uptake inhibitors in early pregnancy try 2001;158:1728–1730. and infant congenital malformations. Birth Defects Res 33. Einarson A, Choi J, Einarson TR, Koren G. Rates of A Clin Mol Teratol 2007;79:301–308. spontaneous and therapeutic abortions following use of 52. Warburton W, Hertzman C, Oberlander TF. A register antidepressants in pregnancy: results from a large pro- study of the impact of stopping third trimester selective spective database. J Obstet Gynaecol Can 2009;31:452– serotonin reuptake inhibitor exposure on neonatal 456. health. Acta Psychiatr Scand 2010;121:471–479. 34. Nakhai-Pour HR, Broy P, Berard A. Use of antidepres- 53. Malm H, Klaukka T, Neuvonen PJ. Risks associated with sants during pregnancy and the risk of spontaneous selective serotonin reuptake inhibitors in pregnancy. abortion. CMAJ 2010;182:1031–1037. Obstet Gynecol 2005;106:1289–1296. 35. Chambers CD, Johnson KA, Dick LM, Felix RJ, Jones 54. Pastuszak A, Schick-Boschetto B, Zuber C et al. Preg- KL. Birth outcomes in pregnant women taking fluoxe- nancy outcome following first-trimester exposure to flu- tine. N Engl J Med 1996;335:1010–1015. oxetine (Prozac). JAMA 1993;269:2246–2248. 36. Kulin NA, Pastuszak A, Sage SR et al. Pregnancy out- 55. Hendrick V, Smith LM, Suri R, Hwang S, Haynes D, come following maternal use of the new selective seroto- Altshuler L. Birth outcomes after prenatal exposure to nin reuptake inhibitors: a prospective controlled antidepressant medication. Am J Obstet Gynecol multicenter study. JAMA 1998;279:609–610. 2003;188:812–815. 37. Chun-Fai-Chan B, Koren G, Fayez I et al. Pregnancy out- 56. Goldstein DJ, Corbin LA, Sundell KL. Effects of first- come of women exposed to bupropion during preg- trimester fluoxetine exposure on the newborn. Obstet nancy: a prospective comparative study. Am J Obstet Gynecol 1997;89:713–718. Gynecol 2005;192:932–936. 57. Ericson A, Kallen B, Wiholm B. Delivery outcome after 38. Yaris F, Kadioglu M, Kesim M et al. Newer antidepres- the use of antidepressants in early pregnancy. Eur J Clin sants in pregnancy: prospective outcome of a case series. Pharmacol 1999;55:503–508. Reprod Toxicol 2004;19:235–238. 58. Sivojelezova A, Shuhaiber S, Sarkissian L, Einarson A, 39. Djulus J, Koren G, Einarson TR et al. Exposure to mirt- Koren G. Citalopram use in pregnancy: prospective azapine during pregnancy: a prospective, comparative comparative evaluation of pregnancy and fetal outcome. study of birth outcomes. J Clin Psychiatry 2006;67: Am J Obstet Gynecol 2005;193:2004–2009. 1280–1284. 59. Hallberg P, Sjoblom V. The use of selective serotonin 40. Kallen B. Neonate characteristics after maternal use of reuptake inhibitors during pregnancy and breast-feed- antidepressants in late pregnancy. Arch Pediatr Adolesc ing: a review and clinical aspects. J Clin Psychopharma- Med 2004;158:312–316. col 2005;25:59–73.

111 Byatt et al.

60. Alwan S, Reefhuis J, Rasmussen SA, Olney RS, Friedman sure surveillance study. Pharmacoepidemiol Drug Saf JM. Use of selective serotonin-reuptake inhibitors in 2010;19:808–813. pregnancy and the risk of birth defects. N Engl J Med 77. Oberlander TF, Warburton W, Misri S, Riggs W, Aghaja- 2007;356:2684–2692. nian J, Hertzman C. Major congenital malformations fol- 61. Merlob P, Birk E, Sirota L et al. Are selective serotonin lowing prenatal exposure to serotonin reuptake reuptake inhibitors cardiac teratogens? Echocardio- inhibitors and benzodiazepines using population-based graphic screening of newborns with persistent heart health data. Birth Defects Res B Dev Reprod Toxicol murmur. Birth Defects Res A Clin Mol Teratol 2008;83:68–76. 2009;85:837–841. 78. Kornum JB, Nielsen RB, Pedersen L, Mortensen PB, 62. Louik C, Lin AE, Werler MM, Hernandez-Diaz S, Mitch- Norgaard M. Use of selective serotonin-reuptake inhibi- ell AA. First-trimester use of selective serotonin-reup- tors during early pregnancy and risk of congenital take inhibitors and the risk of birth defects. N Engl J malformations: updated analysis. Clin Epidemiol Med 2007;356:2675–2683. 2010;2:29–36. 63. Birndorf CA, Madden A, Portera L, Leon AC. Psychiat- 79. Pedersen LH, Henriksen TB, Vestergaard M, Olsen J, ric symptoms, functional impairment, and receptivity Bech BH. Selective serotonin reuptake inhibitors in preg- toward mental health treatment among obstetrical nancy and congenital malformations: population based patients. Int J Psychiatry Med 2001;31:355–365. cohort study. BMJ 2009;339:b3569. 64. Wogelius P, Norgaard M, Gislum M et al. Maternal use 80. Wen SW, Yang Q, Garner P et al. Selective serotonin re- of selective serotonin reuptake inhibitors and risk of uptake inhibitors and adverse pregnancy outcomes. Am congenital malformations. Epidemiology 2006;17:701– J Obstet Gynecol 2006;194:961–966. 704. 81. Einarson A, Choi J, Einarson TR, Koren G. Incidence of 65. Diav-Citrin O, Shechtman S, Weinbaum D et al. Paroxe- major malformations in infants following antidepressant tine and fluoxetine in pregnancy: a prospective, multi- exposure in pregnancy: results of a large prospective centre, controlled, observational study. Br J Clin cohort study. Can J Psychiatry 2009;54:242–246. Pharmacol 2008;66:695–705. 82. Einarson A, Choi J, Koren G, Einarson T. Outcomes of 66. Cole JA, Ephross SA, Cosmatos IS, Walker AM. Paroxe- infants exposed to multiple antidepressants during preg- tine in the first trimester and the prevalence of congenital nancy: results of a cohort study. J Popul Ther Clin Phar- malformations. Pharmacoepidemiol Drug Saf macol 2011;18:e390–e396. 2007;16:1075–1085. 83. Ramos E, St-Andre M, Rey E, Oraichi D, Berard A. 67. Williams M, Wooltorton E. Paroxetine (Paxil) and con- Duration of antidepressant use during pregnancy and genital malformations. CMAJ 2005;173:1320–1321. risk of major congenital malformations. Br J Psychiatry 68. Einarson A, Koren G. Counseling pregnant women trea- 2008;192:344–350. ted with paroxetine. Concern about cardiac malforma- 84. Wisner KL, Sit DK, Hanusa BH et al. Major depres- tions. Can Fam Physician 2006;52:593–594. sion and antidepressant treatment: impact on pregnancy 69. Berard A, Ramos E, Rey E, Blais L, St-Andre M, Oraichi and neonatal outcomes. Am J Psychiatry 2009;166: D. First trimester exposure to paroxetine and risk of car- 557–566. diac malformations in infants: the importance of dosage. 85. Klieger-Grossmann C, Weitzner B, Panchaud A et al. Birth Defects Res B Dev Reprod Toxicol 2007;80:18–27. Pregnancy outcomes following use of escitalopram: a 70. Wurst KE, Poole C, Ephross SA, Olshan AF. First tri- prospective comparative cohort study. J Clin Pharmacol mester paroxetine use and the prevalence of congenital, 2011;52:766–70. specifically cardiac, defects: a meta-analysis of epidemio- 86. Addis A, Koren G. Safety of fluoxetine during the first logical studies. Birth Defects Res A Clin Mol Teratol trimester of pregnancy: a meta-analytical review of epi- 2010;88:159–170. demiological studies. Psychol Med 2000;30:89–94. 71. Bar-Oz B, Einarson T, Einarson A et al. Paroxetine and 87. Einarson TR, Einarson A. Newer antidepressants in preg- congenital malformations: meta-Analysis and consider- nancy and rates of major malformations: a meta- ation of potential confounding factors. Clin Ther analysis of prospective comparative studies. Pharma- 2007;29:918–926. coepidemiol Drug Saf 2005;14:823–827. 72. Bakker MK, Kerstjens-Frederikse WS, Buys CH, De 88. O’brien L, Einarson TR, Sarkar M, Einarson A, Koren Walle HE, De Jong-Van Den Berg LT. First-trimester G. Does paroxetine cause cardiac malformations? use of paroxetine and congenital heart defects: a popula- J Obstet Gynaecol Can 2008;30:696–701. tion-based case-control study. Birth Defects Res A Clin 89. Lennestal R, Kallen B. Delivery outcome in relation to Mol Teratol 2010;88:94–100. maternal use of some recently introduced antidepres- 73. Malm H, Artama M, Gissler M, Ritvanen A. Selective sants. J Clin Psychopharmacol 2007;27:607–613. serotonin reuptake inhibitors and risk for major congen- 90. Einarson A, Bonari L, Voyer-Lavigne S et al. A multicen- ital anomalies. Obstet Gynecol 2011;118:111–120. tre prospective controlled study to determine the safety 74. Einarson A, Pistelli A, Desantis M et al. Evaluation of of trazodone and nefazodone use during pregnancy. Can the risk of congenital cardiovascular defects associated J Psychiatry 2003;48:106–110. with use of paroxetine during pregnancy. Am J Psychia- 91. GlaxoSmithKline. The Bupropion Pregnancy Registry try 2008;165:749–752. Final Report. 2008; Available from: http://pregnancy- 75. Gentile S, Bellantuono C. Selective serotonin reuptake registry.gsk.com/documents/bup_report_final_2008.pdf inhibitor exposure during early pregnancy and the risk 92. Alwan S, Reefhuis J, Botto LD, Rasmussen SA, Correa of fetal major malformations: focus on paroxetine. A, Friedman JM. Maternal use of bupropion and risk for J Clin Psychiatry 2009;70:414–422. congenital heart defects. Am J Obstet Gynecol 76. Bakker MK, De Walle HE, Wilffert B, De Jong-Van 2010;203:52.e1–6. Den Berg LT. Fluoxetine and infantile hypertrophic 93. Cole JA, Modell JG, Haight BR, Cosmatos IS, Stoler pylorus stenosis: a signal from a birth defects-drug expo- JM, Walker AM. Bupropion in pregnancy and the

112 Antidepressant use in pregnancy

prevalence of congenital malformations. Pharmacoepi- 113. Herbst F, Gortner L. [Paroxetine withdrawal syndrome demiol Drug Saf 2007;16:474–484. as differential diagnosis of acute neonatal encephalopa- 94. Webster PA. Withdrawal symptoms in neonates associ- thy?]. Z Geburtshilfe Neonatol 2003;207:232–234. ated with maternal antidepressant therapy. Lancet 114. De Moor RA, Mourad L, Ter Haar J, Egberts AC. [With- 1973;2:318–319. drawal symptoms in a neonate following exposure to 95. Moses-Kolko EL, Bogen D, Perel J et al. Neonatal signs venlafaxine during pregnancy]. Ned Tijdschr Geneeskd after late in utero exposure to serotonin reuptake inhibi- 2003;147:1370–1372. tors: literature review and implications for clinical appli- 115. Cohen LS, Heller VL, Bailey JW, Grush L, Ablon JS, cations. JAMA 2005;293:2372–2383. Bouffard SM. Birth outcomes following prenatal expo- 96. Nordeng H, Ystrom E, Einarson A. Perception of risk sure to fluoxetine. Biol Psychiatry 2000;48:996–1000. regarding the use of medications and other exposures 116. Levinson-Castiel R, Merlob P, Linder N, Sirota L, during pregnancy. Eur J Clin Pharmacol 2010;66:207– Klinger G. Neonatal abstinence syndrome after in utero 214. exposure to selective serotonin reuptake inhibitors in 97. Spencer MJ. Fluoxetine hydrochloride (Prozac) toxicity term infants. Arch Pediatr Adolesc Med 2006;160:173– in a neonate. Pediatrics 1993;92:721–722. 176. 98. Dahl ML, Olhager E, Ahlner J. Paroxetine withdrawal 117. Ferreira E, Carceller AM, Agogue C et al. Effects of syndrome in a neonate. Br J Psychiatry 1997;171:391– selective serotonin reuptake inhibitors and venlafaxine 392. during pregnancy in term and preterm neonates. Pediat- 99. Nijhuis IJ, Kok-Van Rooij GW, Bosschaart AN. With- rics 2007;119:52–59. drawal reactions of a premature neonate after maternal 118. Galbally M, Lewis AJ, Lum J, Buist A. Serotonin discon- use of paroxetine. Arch Dis Child Fetal Neonatal Ed tinuation syndrome following in utero exposure to anti- 2001;84:F77. depressant medication: prospective controlled study. 100. Sanz EJ, De-Las-Cuevas C, Kiuru A, Bate A, Edwards R. Aust N Z J Psychiatry 2009;43:846–854. Selective serotonin reuptake inhibitors in pregnant 119. Suri R, Altshuler L, Hendrick V, Rasgon N, Lee E, Mintz women and neonatal withdrawal syndrome: a database J. The impact of depression and fluoxetine treatment on analysis. Lancet 2005;365:482–487. obstetrical outcome. Arch Womens Ment Health 101. Laine K, Heikkinen T, Ekblad U, Kero P. Effects of expo- 2004;7:193–200. sure to selective serotonin reuptake inhibitors during 120. Maschi S, Clavenna A, Campi R, Schiavetti B, Bernat M, pregnancy on serotonergic symptoms in newborns and Bonati M. Neonatal outcome following pregnancy expo- cord blood monoamine and prolactin concentrations. sure to antidepressants: a prospective controlled cohort Arch Gen Psychiatry 2003;60:720–726. study. BJOG 2008;115:283–289. 102. Costei AM, Kozer E, Ho T, Ito S, Koren G. Perinatal 121. Hibbard JU, Wilkins I, Sun L et al. Respiratory morbid- outcome following third trimester exposure to paroxe- ity in late preterm births. JAMA 2010;304:419–425. tine. Arch Pediatr Adolesc Med 2002;156:1129–1132. 122. Administration US Food and Drug Administration. 103. Alehan F, Saygi S, Tarcan A, Gurakan B. Prolonged FDA Medwatch Drug Alert on Effexor (venlafaxine neonatal complications after in utero exposure to HCL) and Effexor XR (venlafaxine HCL) 2004 [cited fluoxetine. J Matern Fetal Neonatal Med. 2008;21:921– 2011 12/22/11]; Available from: http://www.fda.gov/ 923. Safety/MedWatch/SafetyInformation/SafetyAlertsfor 104. Oberlander TF, Misri S, Fitzgerald CE, Kostaras X, HumanMedicalProducts/ucm154975.htm. Rurak D, Riggs W. Pharmacologic factors associated 123. Hernandez-Martinez C, Arija V, Balaguer A, Cavalle P, with transient neonatal symptoms following prenatal Canals J. Do the emotional states of pregnant women psychotropic medication exposure. J Clin Psychiatry affect neonatal behaviour? Early Hum Dev [Research 2004;65:230–237. Support, Non-U.S. Govt]. 2008;84:745–750. 105. Oberlander TF, Warburton W, Misri S, Aghajanian J, 124. Steinhorn RH. Neonatal pulmonary hypertension. Pedi- Hertzman C. Neonatal outcomes after prenatal exposure atr Crit Care Med 2010;11:S79–S84. to selective serotonin reuptake inhibitor antidepressants 125. The Neonatal Inhaled Nitric Oxide Study Group. and maternal depression using population-based linked Inhaled nitric oxide in term and near-term infants: neu- health data. Arch Gen Psychiatry 2006;63:898–906. rodevelopmental follow-up of the neonatal inhaled nitric 106. Knoppert DC, Nimkar R, Principi T, Yuen D. Paroxetine oxide study group (NINOS). J Pediatr 2000;136:611– toxicity in a newborn after in utero exposure: clinical 617. symptoms correlate with serum levels. Ther Drug Monit 126. Davidson D, Barefield ES, Kattwinkel J et al. Inhaled 2006;28:5–7. nitric oxide for the early treatment of persistent pulmo- 107. Kwon P, Lefkowitz W. Poor neonatal adaptation in term nary hypertension of the term newborn: a randomized, infant. Fluoxetine toxicity. Pediatr Ann 2008;37:131–133. double-masked, placebo-controlled, dose-response, mul- 108. Mhanna MJ, Bennet JB II, Izatt SD. Potential fluoxetine ticenter study. The I-NO/PPHN Study Group. Pediat- chloride (Prozac) toxicity in a newborn. Pediatrics rics 1998;101:325–334. 1997;100:158–159. 127. Greenough A, Khetriwal B. Pulmonary hypertension in 109. Mohan CG, Moore JJ. Fluoxetine toxicity in a preterm the newborn. Paediatr Respir Rev 2005;6:111–116. infant. J Perinatol 2000;20:445–446. 128. Occhiogrosso M, Omran SS, Altemus M. Persistent pul- 110. Anbu AT, Theodore A. Fluoxetine withdrawal syndrome monary hypertension of the newborn and selective sero- in the newborn. Indian Pediatr 2006;43:66–69. tonin reuptake inhibitors: lessons from clinical and 111. Kent LS, Laidlaw JD. Suspected congenital sertraline translational studies. Am J Psychiatry 2012;169: dependence. Br J Psychiatry 1995;167:412–413. 134–40. 112. Boucher N, Bairam A, Beaulac-Baillargeon L. A new 129. Group NINOS. Inhaled nitric oxide in term and look at the neonate’s clinical presentation after in utero near-term infants: neurodevelopmental follow-up of the exposure to antidepressants in late pregnancy. J Clin neonatal inhaled nitric oxide study group (NINOS). Psychopharmacol 2008;28:334–339. J Pediatr 2000;136:611–617.

113 Byatt et al.

130. Koren G, Boucher N. Adverse effects in neonates 146. Administration USFAD. FDA Drug Safety Communi- exposed to SSRIs and SNRI in late gestation–Motherisk cation: Selective serotonin reuptake inhibitor (SSRI) Update 2008. Can J Clin Pharmacol 2009;16:e66–e67. antidepressant use during pregnancy and reports of a 131. Boucher N, Koren G, Beaulac-Baillargeon L. Maternal rare heart and lung condition in newborn babies. 12/4/ use of venlafaxine near term: correlation between neona- 11 ed2011. tal effects and plasma concentrations. Ther Drug Monit 147. Keszler M, Carbone MT, Cox C, Schumacher RE. Severe 2009;31:404–409. respiratory failure after elective repeat cesarean delivery: 132. Oberlander TF, Warburton W, Misri S, Aghajanian J, a potentially preventable condition leading to extracor- Hertzman C. Effects of timing and duration of gesta- poreal membrane oxygenation. Pediatrics 1992;89:670– tional exposure to serotonin reuptake inhibitor antide- 672. pressants: population-based study. Br J Psychiatry 148. Levine EM, Ghai V, Barton JJ, Strom CM. Mode of 2008;192:338–343. delivery and risk of respiratory diseases in newborns. 133. Eddahibi S, Raffestin B, Hamon M, Adnot S. Is the seroto- Obstet Gynecol 2001;97:439–442. nin transporter involved in the pathogenesis of pulmo- 149. Reece EA, Moya F, Yazigi R, Holford T, Duncan C, nary hypertension? J Lab Clin Med 2002;139:194–201. Ehrenkranz RA. Persistent pulmonary hypertension: 134. Fornaro E, Li D, Pan J, Belik J. Prenatal exposure to flu- assessment of perinatal risk factors. Obstet Gynecol oxetine induces fetal pulmonary hypertension in the rat. 1987;70:696–700. Am J Respir Crit Care Med 2007;176:1035–1040. 150. Gentile S. Selective serotonin reuptake inhibitor expo- 135. Dakshinamurti S. Pathophysiologic mechanisms of per- sure during early pregnancy and the risk of birth defects. sistent pulmonary hypertension of the newborn. Pediatr Acta Psychiatr Scand. [Review]. 2011;123:266–275. Pulmonol 2005;39:492–503. 151. US Food and Drug Administration. Pregnancy and 136. Eddahibi S, Humbert M, Fadel E et al. Serotonin trans- Drug Labeling 2008 [cited 2012 11/6/12]; Available porter overexpression is responsible for pulmonary from: http://www.fda.gov/Drugs/DevelopmentAppro- artery smooth muscle hyperplasia in primary pulmonary valProcess/DevelopmentResources/Labeling/ucm093307. hypertension. J Clin Invest 2001;108:1141–1150. htm. 137. Oberlander TF, Bonaguro RJ, Misri S, Papsdorf M, Ross 152. Nulman I, Rovet J, Stewart DE et al. Child development CJ, Simpson EM. Infant serotonin transporter (SLC6A4) following exposure to tricyclic antidepressants or fluoxe- promoter genotype is associated with adverse neonatal tine throughout fetal life: a prospective, controlled outcomes after prenatal exposure to serotonin reuptake study. Am J Psychiatry 2002;159:1889–1895. inhibitor medications. Mol Psychiatry 2008;13:65–73. 153. Sit DK, Perel JM, Helsel JC, Wisner KL. Changes in 138. Chambers CD, Hernandez-Diaz S, Van Marter LJ et al. antidepressant metabolism and dosing across pregnancy Selective serotonin-reuptake inhibitors and risk of per- and early postpartum. J Clin Psychiatry 2008;69:652– sistent pulmonary hypertension of the newborn. N Engl 658. J Med 2006;354:579–587. 154. Hostetter A, Stowe ZN, Strader JRJR, Mclaughlin E, 139. Kallen B, Olausson PO. Maternal use of selective seroto- Llewellyn A. Dose of selective serotonin uptake inhibi- nin re-uptake inhibitors and persistent pulmonary tors across pregnancy: clinical implications. Depress hypertension of the newborn. Pharmacoepidemiol Drug Anxiety 2000;11:51–57. Saf 2008;17:801–806. 155. Yonkers KA, Wisner KL, Stewart DE et al. The man- 140. Heritage CK, Cunningham MD. Association of elective agement of depression during pregnancy: a report from repeat cesarean delivery and persistent pulmonary the American Psychiatric Association and the American hypertension of the newborn. Am J Obstet Gynecol College of Obstetricians and Gynecologists. Gen Hosp 1985;152:627–629. Psychiatry 2009;31:403–413. 141. Hernandez-Diaz S, Van Marter LJ, Werler MM, Louik 156. Zeskind PS, Stephens LE. Maternal selective serotonin C, Mitchell AA. Risk factors for persistent pulmonary reuptake inhibitor use during pregnancy and newborn hypertension of the newborn. Pediatrics 2007;120:e272– neurobehavior. Pediatrics 2004;113:368–375. e282. 157. Rampono J, Simmer K, Ilett KF et al. Placental transfer 142. Kieler H, Artama M, Engeland A et al. Selective seroto- of SSRI and SNRI antidepressants and effects on the nin reuptake inhibitors during pregnancy and risk of neonate. Pharmacopsychiatry 2009;42:95–100. persistent pulmonary hypertension in the newborn: pop- 158. Suri R, Hellemann G, Stowe ZN, Cohen LS, Aquino A, ulation based cohort study from the five Nordic coun- Altshuler LL. A prospective, naturalistic, blinded study tries. BMJ 2011;344:d8012. of early neurobehavioral outcomes for infants following 143. Andrade SE, Mcphillips H, Loren D et al. Antidepressant prenatal antidepressant exposure. J Clin Psychiatry medication use and risk of persistent pulmonary hyper- 2011;72:1002–1007. tension of the newborn. Pharmacoepidemiol Drug Saf 159. Sit D, Perel JM, Wisniewski SR, Helsel JC, Luther JF, 2009;18:246–252. Wisner KL. Mother-infant antidepressant concentra- 144. Wichman CL, Moore KM, Lang TR, St Sauver JL, Heise tions, maternal depression, and perinatal events. J Clin RH Jr, Watson WJ. Congenital heart disease associated Psychiatry 2011;72:994–1001. with selective serotonin reuptake inhibitor use during 160. Wilson KL, Zelig CM, Harvey JP, Cunningham BS, pregnancy. Mayo Clin Proc 2009;84:23–27. Dolinsky BM, Napolitano PG. Persistent pulmonary 145. Wilson KL, Zelig CM, Harvey JP, Cunningham BS, hypertension of the newborn is associated with mode of Dolinsky BM, Napolitano PG. Persistent pulmonary delivery and not with maternal use of selective serotonin hypertension of the newborn is associated with mode of reuptake inhibitors. Am J Perinatol 2011;28:19–24. delivery and not with maternal use of selective serotonin reuptake inhibitors. Am J Perinatol 2010;28:19–24.

114