Antidepressant Safety for Pregnancy & Breastfeeding: Study Notes (Comprehensive)

Learning Objectives

  • Identify challenges related to interpreting studies of medication use in pregnant and breastfeeding women.
  • Detail safety data related to gestational use of antidepressant medications.
  • Discuss the safety of antidepressant use in breastfeeding women.
  • Review prescribing guidelines for preconception, pregnant, and breastfeeding women.

Data: Let’s Talk About Data

  • Agency for Healthcare Research & Quality (AHRQ) 2014 report: “Antidepressant Treatment of Depression During Pregnancy and the Postpartum Period.”
    • Included 6 RCTs and 124 observational studies after evaluating a vast evidence base.
    • McDonagh et al., 2014 (AHRQ): examined benefits/harms of antidepressant use in pregnancy, comparative benefits between specific antidepressants, and comparative benefits between psychotherapy and medication.
  • Conclusion (McDonagh et al., 2014):
    • “Evidence about the comparative benefits and harms of pharmacological treatment of depression in pregnant women was largely inadequate to allow well-informed decisions about treatment.”
    • Reasons: comparison groups were not exclusively depressed women in pregnancy; scarcity of studies in the postpartum period; depression is linked to serious adverse outcomes; thus new research is essential.
  • Data deficits (Andrade 2014; Malm 2011):
    • Most studies have poor design: fail to control for maternal illness, confirm ingestion of meds, or assess treatment response.
    • Many are not prospective with matched controls; blinding in pregnancy is often not done for ethical reasons.
    • Positive findings about antidepressants and adverse outcomes receive press; negative follow-up studies are less publicized.
  • Data quantity (Boyce et al., 2014):
    • Abundant data on antidepressant use in pregnant and breastfeeding women, with focus largely on SSRIs.
    • Much data of questionable quality; no consistent, conclusive evidence for high absolute risks with antidepressant use.
  • Free data sources (summaries, literature, statistics):
    • mothertobaby.org
    • womensmentalhealth.org
    • motherisk.org
    • toxnet.nlm.nih.gov
    • toxnet.nlm.nih.gov/newtoxnet/lactmed.htm

Clinical Challenges & Patient Dilemmas

  • Predatory attorney advertising about antidepressant use in pregnancy and fetal malformations.
  • Frightening pharmacy labeling and counseling material.
  • Conflicting advice from obstetric, primary care, and psychiatric providers.
  • Key to helpful treatment planning: caring, unbiased counseling within a supportive therapeutic alliance.
    • Friedman & Hall, 2013; Payne & Meltzer-Brody, 2009

The Informed Consent Process

  • Domains to cover in informed consent:
    • Risks of untreated and under-treated maternal mental illness.
    • Miscarriage & malformation risk.
    • Effects of antidepressant use on pregnancy outcomes (e.g., preterm birth, mode of delivery, birth weight).
    • Effects of maternal antidepressant use on neonatal adaptation.
    • Long-term effects of antidepressant use on infant and childhood development.
  • Robinson, 2015

Miscarriage & Malformation Risk

  • Miscarriage risk (background): 15\% \text{ to } 20\%; most occur before 12 weeks.
  • Miscarriage risk with antidepressants (Einarson et al., 2009): one prospective study (n=937) across multiple antidepressant classes.
  • Miscarriage risk results:
    • Loss rates with antidepressants less than 15\%, consistent with other studies (e.g., 13\% experimental vs 8\% controls).
    • Unclear whether sub-therapeutic treatment of depression contributes to risk in the experimental group.
    • Two large studies (n ≈ 1,000,000) showed similar miscarriage rates: 12\%\text{ for SSRIs} vs 11\%\text{ for depression without SSRIs} (Kjaersgaard et al., 2013; Andersen et al., 2014).
  • Stillbirth & early infant death:
    • Review of two large registry studies (≈1 million participants each) with adjustments: no increased risk of stillbirth, neonatal mortality, or post-neonatal mortality associated with SSRI use in any trimester (Andrade, 2014).
  • Malformation risk (background): 3\%-5\% in any pregnancy; most data focus on SSRIs; results are conflicting, but pooled data do not show a significant increased risk of malformations with SSRI exposure during pregnancy (March of Dimes, 2001).
  • Malformation detection notes:
    • Antepartum detection (especially cardiac) has increased with high-resolution ultrasound.
    • Many detected cardiac malformations resolve spontaneously; detection sensitivity has increased over time (Remnick, Cohen, & Einarson, 2013; Einarson et al., 2008).

Cardiac Malformations & SSRI Use

  • Medicaid database analysis (Huybrechts et al., 2014): n=949,504; 64,389 on SSRIs (6.8%).
    • Adjusted for depression severity and maternal illness.
    • Conclusion: no substantial increase in risk of cardiac malformations attributable to antidepressant use during the first trimester (Nonacs, 2014).

Specific Antidepressant Medications & Risk

  • SSRIs have the most data; observed risks are generally small and often not statistically significant in pooled analyses.
  • Fluoxetine (Prozac):
    • May be associated with poorer newborn transition.
    • 2015: Large studies with conflicting results on fetal malformation risk; overall, no increased malformation risk in pooled data (Payne & Meltzer-Brody, 2009; Furu et al., 2015; Nonacs, 2015).
  • Sertraline (Zoloft):
    • Often chosen by obstetric providers; data abundant for use in pregnancy and breastfeeding.
    • Does not cross placenta easily; breast milk levels are very low.
    • Omphalocele risk: an “possible association” in one study but no significant increased risk in pooled data (Louik et al., 2007).
  • Citalopram (Celexa):
    • Marginal association with neural tube defects in some datasets.
    • Pooled data reassuring for use in pregnancy and breastfeeding (Grigoriadis et al., 2013; Reefhuis et al., 2015; Sivojelezova et al., 2005).
  • Escitalopram (Lexapro):
    • Newer drug; less data but largely reassuring; one small study (213) largely reassuring (Kleiger-Grossmann et al., 2012).
  • Fluvoxamine (Luvox):
    • Associated with more drug interactions; overall reassuring data in SSRI class (Grigoriadis et al., 2013).
  • Paroxetine (Paxil):
    • Associated with neonatal adaptation syndrome (NAS) risk in some reports; controversy remains.
    • Early unpublished data suggested heart defects risk; FDA changed pregnancy labeling to category D in 2005 based on preliminary results (Abel, 2012).
    • Later prospective data (Einarson et al., 2008; Furu et al., 2015; Huybrechts et al., 2014; Reefhuis et al., 2015) have been negative or mixed; some studies report associations with congenital heart defects.
    • Bottom line: paroxetine may not be contraindicated if it is the only effective agent; consider high-quality fetal echocardiography if concerned (Abel, 2013; Remnick et al., 2013).
  • Other antidepressants:
    • SNRI antidepressants (e.g., venlafaxine): some data; reassuring for malformations but greater NAS risk with use in pregnancy (Ewing et al., 2015; Furu et al., 2015; MotherToBaby, 2013).
    • NDRI: Bupropion (Wellbutrin): reassuring for malformations; some data suggest miscarriage risk and potential ADHD risk in offspring; possible seizure threshold reduction (eclampsia risk concerns) (Einarson et al., 2009; MotherToBaby, 2011).
    • TCAs: less data but generally reassuring for malformations; compliance may be poor due to side effects (Boyce et al., 2014).
    • MAOIs: generally avoided in pregnancy due to sparse safety data and potential labor/delivery interactions (Nonacs, 2009).
    • Other agents: Trazodone, Mirtazapine, Buspirone – limited data; generally reassuring when data exist (Smit, Dolman, & Honig, 2016; MotherToBaby, 2013).
  • Newer antidepressants (Duloxetine, Desvenlafaxine, Vilazodone):
    • Duloxetine (Cymbalta): one 2012 observational study (n=624) reassuring; included in broader reassuring studies (Einarson et al., 2012).
    • Desvenlafaxine (Pristiq): new; no robust data yet; similar to venlafaxine in mechanism; potential NAS risk may be similar (MotherToBaby, 2013).
    • Vilazodone (Viibryd): new in 2011; no human data; does not necessarily rule out use if it’s the only medication that works (MotherToBaby, 2013).

Antidepressants & Pregnancy Outcomes

  • SSRI effects on pregnancy outcomes:
    • May reduce gestation by about 1\text{ week} compared with symptomatic, untreated pregnancies.
    • Generally, birth weight, Apgar scores, and cesarean rates are not clinically different between adequately treated and untreated women with psychiatric symptoms.
    • Li, Liu, & Odouli (2009); Remnick, Cohen, & Einarson (2013).
  • Postpartum hemorrhage (PPH):
    • Baseline risk ~4.1\% in 2009 (up from 1.5\% in 1999).
    • About 8% of maternal deaths are attributable to PPH in the U.S.
    • Five major studies since 2004 show mixed results: three show increased PPH risk, two show no increased risk (Meijer et al., 2004; Salkeld et al., 2008).
    • SSRIs/SNRIs, such as venlafaxine, may be associated with higher PPH risk in some datasets due to effects on platelet aggregation (De Abajo, 2011; Hanley et al., 2016; Palmsten et al., 2013).

Gestational Antidepressant Exposure & the Newborn

  • SSRIs and bleeding risk in newborns: no consistent changes in platelet function tests in newborns exposed to SSRIs during pregnancy (Miller, 2006).
  • Neonatal Adaptation Syndrome (NAS):
    • Also referred to as Post-Neonatal Adaptation Syndrome (PNAS), or SSRI Discontinuation Syndrome.
    • Seen in 10\%\text{ to }15\% of neonates born to mothers on SSRIs & SNRIs; same syndrome occurs in about 6\% of depressed women who discontinue antidepressants in early pregnancy (Ewing et al., 2015; Remnick, Cohen, & Einarson, 2013).
    • Symptoms: jitteriness, hypertonia, feeding issues, irritability, mild respiratory distress.
    • Usually lasts several days; rarely requires intensive interventions; most babies can stay with mothers.
    • 3\% have severe symptoms.
    • Possible mechanisms: toxicity, withdrawal, pharmacologic fetal/gestational factors, and maternal depression (Ewing et al., 2015; Hanley & Oberlander, 2012; Remnick et al., 2013).
  • NAS duration and withdrawal patterns vary by drug half-life;
    • Medications with shorter or longer half-lives, or pronounced withdrawal symptoms in adults, tend to show NAS; notable offenders: fluoxetine\ (long half-life),\ paroxetine,\ venlafaxine (per slide content) (Ewing et al., 2015).
  • Persistent Pulmonary Hypertension of the Newborn (PPHN):
    • Baseline risk ≈ rac{1}{1000}.
    • PPHN risk with SSRI exposure: studies show mixed results (3 negative, 3 positive); two positive studies rely on the same database; overall small absolute risk increase is possible but not definitively causal (Nonacs, 2014; Payne & Meltzer-Brody, 2009; Remnick, Cohen, & Einarson, 2013; Robinson, 2015).
    • FDA warning in 2006; 2011 reversal suggested uncertainty; later data emphasize that if any risk exists, it is small and higher with cesarean delivery, which is more common in depressed/anxious pregnancies.

Fetal/Infant Antidepressant Exposure & Development

  • Short-term findings:
    • Physical: slower fetal head growth; minor, often clinically insignificant lower birth weight (roughly 75\text{ g} less than untreated, symptomatic controls); typical birth weight ~3500\,\text{g}.
    • Respiratory distress; NAS/seizures (rare); disruption in sleep & motor activity (Oberlander et al.; Robinson et al.; Hayes et al.; Suri et al.).
  • Long-term findings:
    • Neurobehavioral: gestational SSRI exposure weakly associated with delays in fine and gross motor skills at age 3; language development may be affected during sensitive language periods; the number/severity of maternal depression episodes impacts language outcomes.
    • ADHD risk: some association with maternal bupropion use in one study; child IQ correlates more with maternal IQ than with depression or antidepressant exposure.
    • Emotional outcomes: increased anxiety at age 3.
    • Autism: baseline prevalence ~1\% in the U.S.; autism diagnoses are more common in boys (about 4.6 times higher). Etiology is multifactorial (genetics, epigenetics, environmental exposures, nutrition).
    • Take-home: long-term effects are not well established; current data are mixed; observational studies cannot prove causality; many studies show no differences between exposed and unexposed groups when controlling for maternal illness and other factors.
  • Gestational SSRI exposure and obstetric/neonatal morbidity: Malm et al., 2015 showed that treating a psychiatric condition with SSRIs during pregnancy was associated with some improved outcomes compared with untreated psychiatric illness.
    • Example outcomes: 16\% lower risk of late preterm birth, 48\% lower risk of very late preterm birth, 30\% lower risk of cesarean section; but lower Apgar scores and higher NICU admissions were also noted in the SSRI group (Yager, 2015).
  • Additional findings: antidepressants may mitigate adverse effects of maternal anxiety on infant auditory processing and sensory gating when mothers are treated appropriately (Hunter et al., 2012).
  • Summary on autism and development: while retrospective observational studies show associations between prenatal SSRI exposure and autism, causality cannot be established; illness-related factors may contribute; many studies report no significant differences when properly controlled (Andrade, 2014; Boukhris, 2015; Clements et al., 2014; Harrington et al., 2014; Hviid et al., 2013).

Antidepressants & Breastfeeding

  • Evidence base strongest for fluoxetine, paroxetine, sertraline, and TCAs.
  • SSRIs generally achieve low levels in breast milk, resulting in low infant exposure.
  • Data support safety for breastfeeding with these medications when clinically indicated (Orosolii & Bellantuono, 2015; Davanzo et al., 2011).
  • Practical guidance: select an agent with established maternal efficacy and reasonable infant exposure data; balance treatment benefits for mother with breastfeeding advantages; counsel about exposure risks and monitor infant well-being.
  • Check safety during breastfeeding at reliable resources (e.g., LactMed).
  • Recommend patient education and documentation of care with reputable sources (Weisskopf et al., 2015).

Your Role as Clinician

  • Generally, choose the agent most likely to work for the patient with reasonable safety data.
  • Dose adequately to address symptoms and discourage relapse from undertreatment.
  • Discuss breastfeeding plans and risks/benefits of medication exposure to the infant.
  • Consider non-pharmacologic treatments as adjuncts (therapy, lifestyle interventions).
  • Use evidence-based resources for safety profiles during breastfeeding (e.g., lactmed).
  • Document and educate: provide patient with scripts and information, and refer to credible sources.
  • Check medication safety profiles at reliable databases (toxnet.nlm.nih.gov/newtoxnet/lactmed.htm).

Prescribing Guidelines for Preconception, Pregnant, & Postpartum Women

  • Prescribing Antidepressants to Reproductive-Age Women:
    1) Trial a reasonable antidepressant before pregnancy when possible, but be comfortable initiating treatment during pregnancy if indicated.
    2) Minimize medication switches between exposures; avoid changes during the first trimester when possible.
    3) Do not discontinue antidepressants during pregnancy or delivery if the patient is moderately to severely symptomatic or at high relapse risk.
    4) Dose adequately to sustain improvement and to prevent relapse due to anxiety/depression or medication gaps.
    5) Discuss breastfeeding plans and concerns; generally there is rarely a need to discontinue antidepressants to prevent infant exposure via breastfeeding.
    6) Emphasize adjunctive/non-medication treatments for any woman with psychiatric symptoms or relapse risk.
  • References: Payne & Meltzer-Brody, 2009; Remnick, Cohen, & Einarson, 2013; Abel, 2013; Hanley & Oberlander, 2012.

Patient Education & Documentation of Care

  • When prescribing psychotropic medications during pregnancy and lactation, develop a counseling script that covers informed consent domains and the known literature.
  • Documentation should indicate that informed consent information was reviewed and questions answered.
  • Refer patients to evidence-based websites: postpartum.net; womensmentalhealth.org; mothertobaby.org.
  • Example documentation phrase: a statement that informed consent information regarding risks of psychiatric illness during pregnancy, breastfeeding, and parenting, as well as those related to medication and other treatment options, was reviewed; questions answered; patient referred to womensmentalhealth.org and provided Mother-to-Baby fact sheets (sertraline in pregnancy).

Evidence-Based Resources for Patients

  • Postpartum Support International: postpartum.net
  • MGH Center for Women’s Mental Health: womensmentalhealth.org
  • MotherToBaby (Teratology Information Service): mothertobaby.org

In Conclusion

  • Clinician role: digest data and clinical experience to help patients make well-informed treatment decisions.
  • Stay current with literature and be flexible as new data emerge.
  • The field is evolving; anticipate further updates to safety data and guidelines.

Final Thoughts

  • The overall picture: gestational antidepressant exposure is associated with small potential risks, but untreated maternal depression also carries substantial risk to both mother and child.
  • Individualized decisions, close monitoring, and shared decision-making with patients are essential.
  • Always balance maternal mental health benefits with potential fetal/neonatal risks, and use the safest effective approach.

Thank You for Your Attention