Diagnosis and epidemiology

Major depressive disorder (MDD) is defined as a period of at least 2 weeks of low mood, or loss of interest or pleasure, associated with at least five of the following: (1) change in appetite or weight; (2) insomnia or hypersomnia; (3) psychomotor symptoms such as restlessness or retardation (slowed speech, thought, or movements); (4) decreased energy or fatigue; (5) sense of worthlessness or guilt, hopelessness, or helplessness; (6) difficulty concentrating or making decisions; and (7) recurrent thoughts of death, dying, or suicide . The prevalence of depression in women is 10–20% following puberty, approximately twice the prevalence rate in men . Depression is underrecognized and undertreated in prenatal care . The self-report Edinburgh Postnatal Depression Scale (EPDS) , which was developed to identify postpartum depression (PPD), is commonly used to screen for depression during pregnancy. In addition, the Patient Health Questionnaire (PHQ-9) has been validated for the detection of prenatal depression . The diagnosis of depression in pregnant women can be complicated due to the overlap of symptoms of normal pregnancy (e.g., sleep changes, appetite changes, and fatigue) with some of the diagnostic symptoms of MDD . Two-item case-finding questions can also be used to identify patients that may warrant further assessment for a depressive disorder . An EPDS score ≥10 or a PHQ-9 score ≥10 suggests the presence of a possible depression. A positive screening is not diagnostic; therefore, it should lead to further diagnostic assessment and potential initiation of treatment . It is important to assess psychosocial risk factors in pregnant woman, which include unstable housing, low resources, poor social support, intimate partner violence, active psychiatric disorders, and substance abuse . Screening for perinatal depression is most effective when accompanied by the assessment of psychosocial issues and the availability of integrated treatment and support options .

In a systematic review of studies in which depression was evaluated by a structured clinical interview, the point prevalence of depression (MDD and less severe depression) was 11% in the first trimester with a drop to 8.5% in the second and third trimesters . The point prevalence of MDD ranged from 1% to 5.6% through pregnancy . Pregnancy does not appear to be a time of increased prevalence rates of depression compared to other time periods during women’s reproductive years, but the perinatal period may be considered a time of risk of a major depressive episode in some women. Risk factors for prenatal depression include being adolescent, single, financially disadvantaged, African American, Hispanic or Asian, and having previous MDD, current anxiety, medical problems, psychosocial stress, intimate partner violence, poor social support, and lower education . Depression during pregnancy can lead to poor health behaviors such as poor compliance with prenatal care, increased smoking and substance abuse, poor nutrition, lack of exercise, and not taking prenatal vitamins and prescribed medications . Suicidal ideation can be elevated in pregnant women with depression .

Prenatal depression, anxiety, stress, and pregnancy

It is difficult to define and separate the effects of untreated prenatal depression, anxiety, and stress; so many studies examine these effects together. Moreover, stress can be both a stimulus, such as an unexpected stressful life event during pregnancy, and a response to a stressor involving anxiety and distress . In the developing fetus, serotonin has roles as a growth factor for its own and related neural systems, and as a modulatory neurotransmitter . Prenatal stress, anxiety, and low mood are likely to influence fetal development, birth outcome, and infant and child development through alterations to the serotonin system and the hypothalamic–pituitary–adrenal (HPA) axis hormones, particularly corticotropin-releasing hormone (CRH) . Other factors that may be involved include an increase in pro-inflammatory cytokines, a reduced placental 11 beta-hydroxy steroid dehydrogenase type-2 enzyme, increased catecholamines, and potential epigenetic changes to placental genes involved in the HPA and serotonin systems .

With prenatal stress and depression, studies have reported altered fetal response to vibroacoustic stimulation, altered fetal heart rate variability, altered motor activity, and altered behavioral reactivity and development . Preterm birth (PTB), which occurs in approximately 12% of pregnancies, and low birth weight (LBW), which occurs in approximately 8% of pregnancies, can both lead to neonatal morbidity and long-term negative consequences for development. Although a few studies have reported that untreated prenatal depression is not associated with PTB or LBW, recent meta-analyses have reported consistent findings of increased risks of both conditions with maternal depression. A meta-analysis of studies conducted through 2009 reported that untreated prenatal depression was associated with small but significant increased risks of PTB and LBW . A review of studies through 2010 reported that depression during pregnancy was associated with an increased risk of PTB and a decreased intention to breastfeed . Another recent review of studies through 2011 reported that elevated depression, particularly in the first half of pregnancy, increased the risk of PTB and small-for-gestational age, and possibly LBW .

Prenatal depression, anxiety, and stress and child development

The effects of untreated prenatal depression on the infant and child are often difficult to separate from continued maternal depression and anxiety, paternal mood symptoms, postpartum caregiving, and other environmental factors. Studies examining short-term and long-term effects from exposure to prenatal maternal stress, anxiety, and depression are increasing. In the infant, reported effects of prenatal depression have included altered neonatal behavioral scores, decreased vagal tone, altered cortisol reactivity, altered reactivity to pain or stress, altered temperament, increased irritability, altered attention, sleep problems, and delayed neuromotor development . A prospective cohort study reported that prenatal depression in mothers who did not report postpartum depressive symptoms was associated with a developmental delay in an 18-month-old offspring . Young children have been reported to have internalizing behaviors, externalizing behaviors, fearful temperament and anxiety, and delayed motor and cognitive development . In older children, prenatal stress and depression have been associated with altered stress response, attention deficit hyperactivity disorder, depression, and anxiety disorders .

Nonpharmacological treatment of MDD during pregnancy

Nonpharmacological strategies can be utilized as a single or an adjunctive treatment for depression in pregnant women who want to minimize fetal exposure to psychotropic medications. Pregnant women prefer psychotherapy to medication for the prevention of recurrence of depression as well as for the treatment of depression . Meta-analyses of psychotherapy modalities in pregnancy and postpartum women have suggested that psychotherapy is moderately effective when compared to usual care, see review . Interpersonal psychotherapy (IPT) is particularly applicable to pregnancy with its attention to role transition, interpersonal issues, and building social support. A recent RCT reported that IPT and parenting education were both helpful for improving depressive symptoms in pregnant women . Studies have also suggested benefit for depression during pregnancy with cognitive-behavior therapy (CBT) .

Mixed results have been reported with omega-3 fatty acids, folate, St. John’s Wort, light therapy, and acupuncture . Exercise may decrease depression scores in pregnant women . Yoga and massage during pregnancy may be helpful for mood and anxiety symptoms . There are increasing reports of benefit for mood with repetitive transcranial magnetic stimulation . Pregnant women with severe symptoms who are not responsive to medications or to nonpharmacological treatments may require electroconvulsive therapy (ECT). Although ECT has been used safely in pregnancy, there are specific maternal precautions, and there have been adverse events reported such as fetal heart rate reduction, uterine contractions, premature labor, and fetal demise .

Antidepressant treatment of MDD during pregnancy

Selective serotonin reuptake inhibitors (SSRIs) are used by approximately 7% of pregnant women in the United States , with lower prevalence estimates in Europe . Some women discontinue their antidepressant once pregnancy is recognized, either on their own or at the advice of a clinician in order to minimize exposure of the fetus to the antidepressant , while others continue to use antidepressants throughout pregnancy. In a prospective study of women with prior depression who were euthymic on an antidepressant at conception, women who discontinued their antidepressant had a five times increased risk ratio of depression relapse compared to women who maintained their antidepressant through pregnancy . Although a more recent community study did not confirm that antidepressant medication was protective against relapse during pregnancy , these studies suggested that a relapse is more likely in women with more previous episodes, longer length of illness, and an episode in the 6 months prior to conception .

If a pregnant woman takes an antidepressant, it is important to try to achieve remission so that the fetus is not exposed to both the underlying illness and medication. Preferred medications in pregnancy are those with the best safety data, fewer metabolites, higher protein binding to decrease placental passage, and fewer interactions with other medications . Due to pregnancy-related changes in absorption, distribution, metabolism, and elimination, an increase in antidepressant dosage is often indicated to maintain a therapeutic effect . Fetal exposure to antidepressants occurs through the umbilical cord, across the placenta by transporter genes, and through absorption from amniotic fluid. Antidepressant exposure in the fetus is influenced by the cord-to-maternal ratio of the antidepressant level, half-life of antidepressant, peak and trough levels of antidepressant, genetic polymorphisms of drug metabolism enzymes and transporter proteins, and the unbound fraction of antidepressant in the fetus .

Prenatal exposure to antidepressants

Results of studies examining the effects of antidepressant exposure on birth outcomes and child development have been inconsistent. As women with psychiatric symptoms are not randomized to take antidepressants or placebo, it is difficult to separate out the contribution from the underlying mood disorder and the associated behaviors (e.g., poor prenatal health behaviors, concomitant medications, smoking, substance abuse, and obesity) that could be confounding variables . Moreover, women receive prescriptions for antidepressant medications for several indications that are also not controlled for, for example, anxiety disorders . Some domains of behavior and neurodevelopment may be adversely affected by SSRI exposure, while others may be more sensitive to the effects of maternal mental illness . Exposure to SSRIs may influence the development and “fine-tuning” of the serotonin system in the fetus, possibly by decreasing an astroglia-specific calcium-binding protein S100B or by changes in genes encoding the serotonin transporter protein . The HPA axis is also influenced by SSRI exposure . It is difficult to identify specific effects from SSRI exposure as untreated prenatal stress, anxiety, and depression may lead to similar outcomes .

Antidepressants and miscarriage

A large population-based study of pregnancies in Denmark recently reported a 12% rate of spontaneous abortion with exposure to antidepressants, which was a relative risk of 1.14 (95% confidence interval (CI) 1.10–1.18) compared to no antidepressant exposure . However, the relative risk decreased to 1.0 when antidepressant exposure was compared to spontaneous abortion rates in women with depression who did not take an antidepressant during pregnancy. Another recent large study of pregnancies in Denmark reported a 1.27 adjusted hazard ratio (95% CI 1.22–1.33) of having a miscarriage after exposure to an SSRI compared to no exposure . However, this hazard ratio was almost equivalent to the 1.24 hazard ratio (95% CI 1.18–1.30) of miscarriage after discontinuing an SSRI up to a year before pregnancy. As the underlying depression can be a “confound by indication,” both of the studies suggest that the increased hazard ratio may be due to underlying disease or behaviors rather than to the SSRI.

Antidepressants and birth outcome

The recent AHRQ systematic review of studies through July 2013 reported that based on low-strength evidence, neonates of women with depression who took SSRIs during pregnancy did not have a significantly increased risk of PTB compared to neonates of untreated women; the odds ratio (OR) was 1.87 (95% CI 0.89–3.89) . A meta-analysis of studies through 2012 reported that antidepressant use in pregnancy was significantly associated with LBW (1.44 relative risk ratio (95% CI 1.21–1.70)) and PTB (1.69 relative risk ratio (95% CI 1.52–1.88)) compared to no exposure . The significant increased risk was evident even with controlling for exposure to maternal depression. Another recent review of studies through 2012 reported a pooled OR of 1.96 (95% CI 1.62–2.38) for PTB with the third-trimester antidepressant use that remained significant when adjustments were made for a diagnosis of depression . A review of studies through 2010 reported a significant increased risk of PTB (pooled OR of 1.55 (95% CI, 1.38–1.74)) with prenatal exposure to antidepressants . This review commented that the risk of PTB averaged 3 days shorter gestational age, considered a small effect . Exposure to antidepressants was associated with a significantly lower birth weight (mean difference [grams], −74; 95% CI, −117 to −31) compared to no exposure . The 74 g mean difference for birth weight was considered a small effect. This review also reported small, but statistically significant, decreases in gestational age, 1-min, and 5-min Apgar scores with prenatal antidepressant exposure . In summary, multiple studies, several with large registry samples, have reported an increased risk of PTB (OR ranging from 1.5 to 2) with prenatal exposure to antidepressants; however, studies vary in whether or not the increased risk is statistically significant. Overall, the risk is considered modest, with pregnancy perhaps being shortened by 3–5 days .

Antidepressants and congenital malformations

A recent large cohort study reported that there was no substantial increase in the risk of cardiac malformations with first-trimester exposure to antidepressants . A review of studies through 2011 reported that if an increased risk of major congenital malformations does exist, it is a small increase in the absolute risk, and no consistent organ anomaly has been reported with antidepressants as a class . Another recent review of studies reported that there is no increased risk of congenital major malformations with antidepressants, but there may be a statistically significant, but not clinically significant, increased risk of cardiovascular malformations with antidepressants, particularly paroxetine . A recent meta-analysis reported a small, but significant, increased risk of cardiac malformations with paroxetine . Although the recently published cohort study did not find a significant association between the use of paroxetine and the right ventricular outflow tract obstruction , another recent large population-based cohort study in the UK reported an adjusted OR of 1.78 (95% CI 1.09–2.88) of increased heart anomalies with paroxetine use . The recent AHRQ systematic review of studies through July 2013 reported that although there may be a small increased risk of malformations with paroxetine and fluoxetine, direct evidence is insufficient to draw conclusions . An increased risk of left outflow tract heart defects has been reported with bupropion . One small study found that Chiari I malformations at the base of the skull were more common at 1–2 years of age following prenatal SSRI exposure . Further larger studies are needed to determine if this is a genuine observation. The Food and Drug Association (FDA) issued a warning in 2005 that paroxetine exposure during the first trimester may increase the risk of cardiac malformations, and paroxetine was changed to pregnancy category D. Fetal echocardiography with first-trimester paroxetine exposure has been suggested .

Study results continue to be inconsistent, and the small absolute increased risks can be due to the antidepressant, the underlying illness, or other unknown factors . One study reported increased rates of congenital malformations with SSRI exposure, but an increase was also found in women who have “paused” their SSRI throughout pregnancy, suggesting confounding by indication . The mechanism by which antidepressants might increase cardiac malformations is not clear, but altered serotonergic influence on cardiac development is hypothesized. Most studies report increased OR values between 1.2 and 2 with SSRI exposure, considered a small increase in the absolute risk of major congenital malformations, which may not be of clinical significance . Overall, SSRIs are not considered to be major teratogens .

Antidepressants and persistent pulmonary hypertension of the newborn

Persistent pulmonary hypertension of the newborn (PPHN) occurs in 1.9 infants per 1000 live births with a 10–20% mortality risk that can vary with etiology . Risk factors for PPHN include congenital malformations, PTB, meconium aspiration, maternal obesity, and cesarean section . Etiological theories include an increased vasoconstriction of smooth muscle cells and/or a decreased production of nitric oxide, a potent vasodilator. SSRIs may influence serotonin effects on pulmonary vascular smooth muscle endothelium, and they may increase vasoconstriction . PPHN may represent the severe end of a spectrum of respiratory difficulties with SSRI exposure . The increased risk has been noted with several SSRIs, and it is considered a class effect .

A recent meta-analysis of seven studies (five cohort and two case–control studies) reported that exposure to SSRIs in late pregnancy was associated with a 2.5 increased risk ratio (95% CI 1.32–4.73) of PPHN, while exposure to SSRIs in early pregnancy was not associated with PPHN . The absolute risk difference for the development of PPHN after SSRI exposure in late pregnancy was 2.9–3.5 per 1000 live births. Congenital malformations and meconium aspiration were not significant moderator variables, but this meta-analysis was not able to assess for the effects of PTB, cesarean section, obesity, or maternal depression . In July 2006, the US FDA and Health Canada issued a Public Health Advisory advising careful consideration of the potential benefits and risks of SSRIs for each individual pregnant woman based on the reports from one study that PPHN was six times more common in babies whose mothers took an SSRI antidepressant after the 20th week of the pregnancy compared to babies whose mothers did not take an antidepressant. In December 2011, the US FDA issued a safety announcement advising health-care professionals not to alter their current clinical practice of treating depression during pregnancy due to conflicting findings of studies of SSRI use during pregnancy and the development of PPHN.

Antidepressants and poor neonatal adaptation syndrome

Exposure to an antidepressant at the end of the third trimester may lead to a poor neonatal adaptation syndrome (PNAS), also described as a neonatal toxicity or a neonatal behavioral syndrome. Neonatal symptoms including respiratory distress, cyanosis, apnea, seizures, temperature instability, feeding difficulty, vomiting, hypoglycemia, hypotonia, hypertonia, hyperreflexia, tremor, jitteriness, irritability, and persistent crying occur in up to 30% of newborns exposed to SSRIs and serotonin and norepinephrine reuptake inhibitors (SNRIs). The recent AHRQ systematic review of studies through July 2013 reported that based on low-strength evidence, neonates of women with depression who took SSRIs during pregnancy had a higher risk of respiratory distress than neonates of untreated women . Another recent review reported that antidepressant exposure during pregnancy was associated with a five times increased OR (95% CI, 3.25–7.90) of PNAS, respiratory distress, and tremors compared to no exposure . Although an FDA alert in 2004 warned that neonates exposed to SSRIs or SNRIs late in the third trimester may develop complications following delivery requiring prolonged hospitalization, respiratory support, and tube feeding, the symptoms of PNAS are generally mild and usually resolve within days to 2 weeks .

It is unclear what causes PNAS; potential etiologies include serotonergic toxicity due to accumulation of the antidepressant in the neonate and withdrawal signs from the abrupt discontinuation of neonate exposure to the antidepressant . Symptoms of PNAS may differ by antidepressant, maternal factors, or neonatal factors . Some studies have reported that PNAS is more likely with third-trimester exposure to fluoxetine, paroxetine, and venlafaxine . Exposure to prenatal depression without medication also leads to neonatal symptoms , and many studies of PNAS have not controlled for maternal depression or included systematic blinded ratings. To date, there is no evidence that tapering the dose of antidepressant prior to expected delivery lessens PNAS, and this strategy could increase the risk of postnatal depression . Future studies of PNAS need to include the use of a validated neonatal behavioral symptom scale, assessments of infants by evaluators blind to maternal SSRI use, control for maternal psychiatric disorders and other variables, and follow-up to evaluate long-term sequelae .

Tricyclic antidepressants and pregnancy

Although most recent studies have examined the newer antidepressants, SSRIs and SNRIs, tricyclic antidepressants (TCAs) have a role in some perinatal women. Overall, TCAs are not associated with major congenital anomalies or PPHN . TCAs are associated with an increased risk of PTB, LBW, and neonatal symptoms . Due to neonatal symptoms consistent with both withdrawal and toxicity, the least anticholinergic TCAs (nortriptyline and desipramine) are the preferred TCAs in pregnancy . Clomipramine exposure may increase the risk of cardiac defects, and it may be associated with more severe and prolonged neonatal symptoms .

Long-term effects of antidepressants during pregnancy

A recent systematic review of prenatal exposure to antidepressants and developmental outcomes reported inconsistent findings . Some studies suggest delayed motor development and motor control, and possibly delayed language, but no studies have reported a significant association with cognitive problems . When adverse effects are noted, they appear to be within normal limits and resolve with age. It is difficult to separate the effects of prenatal exposure to antidepressants from the exposure to the underlying prenatal illness and the influence of postpartum factors . It is well established that continued exposure to untreated maternal depression during and beyond the first postnatal year continues to be a critical risk factor for developmental problems in childhood, and it contributes to long-term psychopathology .

Recent studies investigating prenatal SSRI exposure effects on infant and child development continue to report mixed results. Prenatal exposure to SSRIs, but not maternal depression, was associated with altered motor movements in 3–4-month-old infants , and lower gross motor, social–emotional, and adaptive behavior in 10-month-old infants . Positive reports with SSRI use during pregnancy compared to no exposure have included an increased infant readiness to interact at 3 months of age , an increased infant attention measured by P50 auditory sensory gating , and an accelerated infant speech perception . A recent study reported that prenatal exposure to untreated depression, not exposure to prenatal antidepressant use, was associated with behavioral or emotional problems in children at 4–5 years of age . Another study reported that children between 3 and 7 years of age exposed in utero to SSRIs or to venlafaxine did not differ from children exposed to untreated maternal depression on cognitive or behavioral measures .

One of the more concerning possible adverse effects from prenatal SSRI exposure is the development of autism or autism spectrum disorder (ASD), but studies report mixed findings. In addition, untreated maternal depression may be a risk factor for ASD. Four studies have reported an increased risk. The first study reported an almost four times greater risk of ASD with exposure to SSRIs in the first trimester , a second study reported a three times greater risk of ASD without intellectual disability , a third study reported a three times greater risk of ASD in boys , and the fourth study reported an increased risk of autistic traits . Some of these studies did not examine the presence or severity of depression during pregnancy, or adjust for some behaviors that might be associated with prenatal depression. Two recent studies have not found an association between SSRI use during pregnancy and increased risk of ASD .

Other potential effects of prenatal antidepressant exposure

Increased infant mortality in the first postdelivery month and year has been reported with prenatal antidepressant use; however, two recent studies reported no association between SSRIs during pregnancy and increased stillbirth, neonatal , or postneonatal mortality . Antidepressants have also been associated with pregnancy-induced hypertension and preeclampsia . The potential confound of underlying depression is important in the interpretation of the results of these studies.

Diagnosis and epidemiology

Major depressive disorder (MDD) is defined as a period of at least 2 weeks of low mood, or loss of interest or pleasure, associated with at least five of the following: (1) change in appetite or weight; (2) insomnia or hypersomnia; (3) psychomotor symptoms such as restlessness or retardation (slowed speech, thought, or movements); (4) decreased energy or fatigue; (5) sense of worthlessness or guilt, hopelessness, or helplessness; (6) difficulty concentrating or making decisions; and (7) recurrent thoughts of death, dying, or suicide . The prevalence of depression in women is 10–20% following puberty, approximately twice the prevalence rate in men . Depression is underrecognized and undertreated in prenatal care . The self-report Edinburgh Postnatal Depression Scale (EPDS) , which was developed to identify postpartum depression (PPD), is commonly used to screen for depression during pregnancy. In addition, the Patient Health Questionnaire (PHQ-9) has been validated for the detection of prenatal depression . The diagnosis of depression in pregnant women can be complicated due to the overlap of symptoms of normal pregnancy (e.g., sleep changes, appetite changes, and fatigue) with some of the diagnostic symptoms of MDD . Two-item case-finding questions can also be used to identify patients that may warrant further assessment for a depressive disorder . An EPDS score ≥10 or a PHQ-9 score ≥10 suggests the presence of a possible depression. A positive screening is not diagnostic; therefore, it should lead to further diagnostic assessment and potential initiation of treatment . It is important to assess psychosocial risk factors in pregnant woman, which include unstable housing, low resources, poor social support, intimate partner violence, active psychiatric disorders, and substance abuse . Screening for perinatal depression is most effective when accompanied by the assessment of psychosocial issues and the availability of integrated treatment and support options .

In a systematic review of studies in which depression was evaluated by a structured clinical interview, the point prevalence of depression (MDD and less severe depression) was 11% in the first trimester with a drop to 8.5% in the second and third trimesters . The point prevalence of MDD ranged from 1% to 5.6% through pregnancy . Pregnancy does not appear to be a time of increased prevalence rates of depression compared to other time periods during women’s reproductive years, but the perinatal period may be considered a time of risk of a major depressive episode in some women. Risk factors for prenatal depression include being adolescent, single, financially disadvantaged, African American, Hispanic or Asian, and having previous MDD, current anxiety, medical problems, psychosocial stress, intimate partner violence, poor social support, and lower education . Depression during pregnancy can lead to poor health behaviors such as poor compliance with prenatal care, increased smoking and substance abuse, poor nutrition, lack of exercise, and not taking prenatal vitamins and prescribed medications . Suicidal ideation can be elevated in pregnant women with depression .

Prenatal depression, anxiety, stress, and pregnancy

It is difficult to define and separate the effects of untreated prenatal depression, anxiety, and stress; so many studies examine these effects together. Moreover, stress can be both a stimulus, such as an unexpected stressful life event during pregnancy, and a response to a stressor involving anxiety and distress . In the developing fetus, serotonin has roles as a growth factor for its own and related neural systems, and as a modulatory neurotransmitter . Prenatal stress, anxiety, and low mood are likely to influence fetal development, birth outcome, and infant and child development through alterations to the serotonin system and the hypothalamic–pituitary–adrenal (HPA) axis hormones, particularly corticotropin-releasing hormone (CRH) . Other factors that may be involved include an increase in pro-inflammatory cytokines, a reduced placental 11 beta-hydroxy steroid dehydrogenase type-2 enzyme, increased catecholamines, and potential epigenetic changes to placental genes involved in the HPA and serotonin systems .

With prenatal stress and depression, studies have reported altered fetal response to vibroacoustic stimulation, altered fetal heart rate variability, altered motor activity, and altered behavioral reactivity and development . Preterm birth (PTB), which occurs in approximately 12% of pregnancies, and low birth weight (LBW), which occurs in approximately 8% of pregnancies, can both lead to neonatal morbidity and long-term negative consequences for development. Although a few studies have reported that untreated prenatal depression is not associated with PTB or LBW, recent meta-analyses have reported consistent findings of increased risks of both conditions with maternal depression. A meta-analysis of studies conducted through 2009 reported that untreated prenatal depression was associated with small but significant increased risks of PTB and LBW . A review of studies through 2010 reported that depression during pregnancy was associated with an increased risk of PTB and a decreased intention to breastfeed . Another recent review of studies through 2011 reported that elevated depression, particularly in the first half of pregnancy, increased the risk of PTB and small-for-gestational age, and possibly LBW .

Prenatal depression, anxiety, and stress and child development

The effects of untreated prenatal depression on the infant and child are often difficult to separate from continued maternal depression and anxiety, paternal mood symptoms, postpartum caregiving, and other environmental factors. Studies examining short-term and long-term effects from exposure to prenatal maternal stress, anxiety, and depression are increasing. In the infant, reported effects of prenatal depression have included altered neonatal behavioral scores, decreased vagal tone, altered cortisol reactivity, altered reactivity to pain or stress, altered temperament, increased irritability, altered attention, sleep problems, and delayed neuromotor development . A prospective cohort study reported that prenatal depression in mothers who did not report postpartum depressive symptoms was associated with a developmental delay in an 18-month-old offspring . Young children have been reported to have internalizing behaviors, externalizing behaviors, fearful temperament and anxiety, and delayed motor and cognitive development . In older children, prenatal stress and depression have been associated with altered stress response, attention deficit hyperactivity disorder, depression, and anxiety disorders .

Nonpharmacological treatment of MDD during pregnancy

Nonpharmacological strategies can be utilized as a single or an adjunctive treatment for depression in pregnant women who want to minimize fetal exposure to psychotropic medications. Pregnant women prefer psychotherapy to medication for the prevention of recurrence of depression as well as for the treatment of depression . Meta-analyses of psychotherapy modalities in pregnancy and postpartum women have suggested that psychotherapy is moderately effective when compared to usual care, see review . Interpersonal psychotherapy (IPT) is particularly applicable to pregnancy with its attention to role transition, interpersonal issues, and building social support. A recent RCT reported that IPT and parenting education were both helpful for improving depressive symptoms in pregnant women . Studies have also suggested benefit for depression during pregnancy with cognitive-behavior therapy (CBT) .

Mixed results have been reported with omega-3 fatty acids, folate, St. John’s Wort, light therapy, and acupuncture . Exercise may decrease depression scores in pregnant women . Yoga and massage during pregnancy may be helpful for mood and anxiety symptoms . There are increasing reports of benefit for mood with repetitive transcranial magnetic stimulation . Pregnant women with severe symptoms who are not responsive to medications or to nonpharmacological treatments may require electroconvulsive therapy (ECT). Although ECT has been used safely in pregnancy, there are specific maternal precautions, and there have been adverse events reported such as fetal heart rate reduction, uterine contractions, premature labor, and fetal demise .

Antidepressant treatment of MDD during pregnancy

Selective serotonin reuptake inhibitors (SSRIs) are used by approximately 7% of pregnant women in the United States , with lower prevalence estimates in Europe . Some women discontinue their antidepressant once pregnancy is recognized, either on their own or at the advice of a clinician in order to minimize exposure of the fetus to the antidepressant , while others continue to use antidepressants throughout pregnancy. In a prospective study of women with prior depression who were euthymic on an antidepressant at conception, women who discontinued their antidepressant had a five times increased risk ratio of depression relapse compared to women who maintained their antidepressant through pregnancy . Although a more recent community study did not confirm that antidepressant medication was protective against relapse during pregnancy , these studies suggested that a relapse is more likely in women with more previous episodes, longer length of illness, and an episode in the 6 months prior to conception .

If a pregnant woman takes an antidepressant, it is important to try to achieve remission so that the fetus is not exposed to both the underlying illness and medication. Preferred medications in pregnancy are those with the best safety data, fewer metabolites, higher protein binding to decrease placental passage, and fewer interactions with other medications . Due to pregnancy-related changes in absorption, distribution, metabolism, and elimination, an increase in antidepressant dosage is often indicated to maintain a therapeutic effect . Fetal exposure to antidepressants occurs through the umbilical cord, across the placenta by transporter genes, and through absorption from amniotic fluid. Antidepressant exposure in the fetus is influenced by the cord-to-maternal ratio of the antidepressant level, half-life of antidepressant, peak and trough levels of antidepressant, genetic polymorphisms of drug metabolism enzymes and transporter proteins, and the unbound fraction of antidepressant in the fetus .

Prenatal exposure to antidepressants

Results of studies examining the effects of antidepressant exposure on birth outcomes and child development have been inconsistent. As women with psychiatric symptoms are not randomized to take antidepressants or placebo, it is difficult to separate out the contribution from the underlying mood disorder and the associated behaviors (e.g., poor prenatal health behaviors, concomitant medications, smoking, substance abuse, and obesity) that could be confounding variables . Moreover, women receive prescriptions for antidepressant medications for several indications that are also not controlled for, for example, anxiety disorders . Some domains of behavior and neurodevelopment may be adversely affected by SSRI exposure, while others may be more sensitive to the effects of maternal mental illness . Exposure to SSRIs may influence the development and “fine-tuning” of the serotonin system in the fetus, possibly by decreasing an astroglia-specific calcium-binding protein S100B or by changes in genes encoding the serotonin transporter protein . The HPA axis is also influenced by SSRI exposure . It is difficult to identify specific effects from SSRI exposure as untreated prenatal stress, anxiety, and depression may lead to similar outcomes .

Antidepressants and miscarriage

A large population-based study of pregnancies in Denmark recently reported a 12% rate of spontaneous abortion with exposure to antidepressants, which was a relative risk of 1.14 (95% confidence interval (CI) 1.10–1.18) compared to no antidepressant exposure . However, the relative risk decreased to 1.0 when antidepressant exposure was compared to spontaneous abortion rates in women with depression who did not take an antidepressant during pregnancy. Another recent large study of pregnancies in Denmark reported a 1.27 adjusted hazard ratio (95% CI 1.22–1.33) of having a miscarriage after exposure to an SSRI compared to no exposure . However, this hazard ratio was almost equivalent to the 1.24 hazard ratio (95% CI 1.18–1.30) of miscarriage after discontinuing an SSRI up to a year before pregnancy. As the underlying depression can be a “confound by indication,” both of the studies suggest that the increased hazard ratio may be due to underlying disease or behaviors rather than to the SSRI.

Antidepressants and birth outcome

The recent AHRQ systematic review of studies through July 2013 reported that based on low-strength evidence, neonates of women with depression who took SSRIs during pregnancy did not have a significantly increased risk of PTB compared to neonates of untreated women; the odds ratio (OR) was 1.87 (95% CI 0.89–3.89) . A meta-analysis of studies through 2012 reported that antidepressant use in pregnancy was significantly associated with LBW (1.44 relative risk ratio (95% CI 1.21–1.70)) and PTB (1.69 relative risk ratio (95% CI 1.52–1.88)) compared to no exposure . The significant increased risk was evident even with controlling for exposure to maternal depression. Another recent review of studies through 2012 reported a pooled OR of 1.96 (95% CI 1.62–2.38) for PTB with the third-trimester antidepressant use that remained significant when adjustments were made for a diagnosis of depression . A review of studies through 2010 reported a significant increased risk of PTB (pooled OR of 1.55 (95% CI, 1.38–1.74)) with prenatal exposure to antidepressants . This review commented that the risk of PTB averaged 3 days shorter gestational age, considered a small effect . Exposure to antidepressants was associated with a significantly lower birth weight (mean difference [grams], −74; 95% CI, −117 to −31) compared to no exposure . The 74 g mean difference for birth weight was considered a small effect. This review also reported small, but statistically significant, decreases in gestational age, 1-min, and 5-min Apgar scores with prenatal antidepressant exposure . In summary, multiple studies, several with large registry samples, have reported an increased risk of PTB (OR ranging from 1.5 to 2) with prenatal exposure to antidepressants; however, studies vary in whether or not the increased risk is statistically significant. Overall, the risk is considered modest, with pregnancy perhaps being shortened by 3–5 days .

Antidepressants and congenital malformations

A recent large cohort study reported that there was no substantial increase in the risk of cardiac malformations with first-trimester exposure to antidepressants . A review of studies through 2011 reported that if an increased risk of major congenital malformations does exist, it is a small increase in the absolute risk, and no consistent organ anomaly has been reported with antidepressants as a class . Another recent review of studies reported that there is no increased risk of congenital major malformations with antidepressants, but there may be a statistically significant, but not clinically significant, increased risk of cardiovascular malformations with antidepressants, particularly paroxetine . A recent meta-analysis reported a small, but significant, increased risk of cardiac malformations with paroxetine . Although the recently published cohort study did not find a significant association between the use of paroxetine and the right ventricular outflow tract obstruction , another recent large population-based cohort study in the UK reported an adjusted OR of 1.78 (95% CI 1.09–2.88) of increased heart anomalies with paroxetine use . The recent AHRQ systematic review of studies through July 2013 reported that although there may be a small increased risk of malformations with paroxetine and fluoxetine, direct evidence is insufficient to draw conclusions . An increased risk of left outflow tract heart defects has been reported with bupropion . One small study found that Chiari I malformations at the base of the skull were more common at 1–2 years of age following prenatal SSRI exposure . Further larger studies are needed to determine if this is a genuine observation. The Food and Drug Association (FDA) issued a warning in 2005 that paroxetine exposure during the first trimester may increase the risk of cardiac malformations, and paroxetine was changed to pregnancy category D. Fetal echocardiography with first-trimester paroxetine exposure has been suggested .

Study results continue to be inconsistent, and the small absolute increased risks can be due to the antidepressant, the underlying illness, or other unknown factors . One study reported increased rates of congenital malformations with SSRI exposure, but an increase was also found in women who have “paused” their SSRI throughout pregnancy, suggesting confounding by indication . The mechanism by which antidepressants might increase cardiac malformations is not clear, but altered serotonergic influence on cardiac development is hypothesized. Most studies report increased OR values between 1.2 and 2 with SSRI exposure, considered a small increase in the absolute risk of major congenital malformations, which may not be of clinical significance . Overall, SSRIs are not considered to be major teratogens .

Antidepressants and persistent pulmonary hypertension of the newborn

Persistent pulmonary hypertension of the newborn (PPHN) occurs in 1.9 infants per 1000 live births with a 10–20% mortality risk that can vary with etiology . Risk factors for PPHN include congenital malformations, PTB, meconium aspiration, maternal obesity, and cesarean section . Etiological theories include an increased vasoconstriction of smooth muscle cells and/or a decreased production of nitric oxide, a potent vasodilator. SSRIs may influence serotonin effects on pulmonary vascular smooth muscle endothelium, and they may increase vasoconstriction . PPHN may represent the severe end of a spectrum of respiratory difficulties with SSRI exposure . The increased risk has been noted with several SSRIs, and it is considered a class effect .

A recent meta-analysis of seven studies (five cohort and two case–control studies) reported that exposure to SSRIs in late pregnancy was associated with a 2.5 increased risk ratio (95% CI 1.32–4.73) of PPHN, while exposure to SSRIs in early pregnancy was not associated with PPHN . The absolute risk difference for the development of PPHN after SSRI exposure in late pregnancy was 2.9–3.5 per 1000 live births. Congenital malformations and meconium aspiration were not significant moderator variables, but this meta-analysis was not able to assess for the effects of PTB, cesarean section, obesity, or maternal depression . In July 2006, the US FDA and Health Canada issued a Public Health Advisory advising careful consideration of the potential benefits and risks of SSRIs for each individual pregnant woman based on the reports from one study that PPHN was six times more common in babies whose mothers took an SSRI antidepressant after the 20th week of the pregnancy compared to babies whose mothers did not take an antidepressant. In December 2011, the US FDA issued a safety announcement advising health-care professionals not to alter their current clinical practice of treating depression during pregnancy due to conflicting findings of studies of SSRI use during pregnancy and the development of PPHN.

Antidepressants and poor neonatal adaptation syndrome

Exposure to an antidepressant at the end of the third trimester may lead to a poor neonatal adaptation syndrome (PNAS), also described as a neonatal toxicity or a neonatal behavioral syndrome. Neonatal symptoms including respiratory distress, cyanosis, apnea, seizures, temperature instability, feeding difficulty, vomiting, hypoglycemia, hypotonia, hypertonia, hyperreflexia, tremor, jitteriness, irritability, and persistent crying occur in up to 30% of newborns exposed to SSRIs and serotonin and norepinephrine reuptake inhibitors (SNRIs). The recent AHRQ systematic review of studies through July 2013 reported that based on low-strength evidence, neonates of women with depression who took SSRIs during pregnancy had a higher risk of respiratory distress than neonates of untreated women . Another recent review reported that antidepressant exposure during pregnancy was associated with a five times increased OR (95% CI, 3.25–7.90) of PNAS, respiratory distress, and tremors compared to no exposure . Although an FDA alert in 2004 warned that neonates exposed to SSRIs or SNRIs late in the third trimester may develop complications following delivery requiring prolonged hospitalization, respiratory support, and tube feeding, the symptoms of PNAS are generally mild and usually resolve within days to 2 weeks .

It is unclear what causes PNAS; potential etiologies include serotonergic toxicity due to accumulation of the antidepressant in the neonate and withdrawal signs from the abrupt discontinuation of neonate exposure to the antidepressant . Symptoms of PNAS may differ by antidepressant, maternal factors, or neonatal factors . Some studies have reported that PNAS is more likely with third-trimester exposure to fluoxetine, paroxetine, and venlafaxine . Exposure to prenatal depression without medication also leads to neonatal symptoms , and many studies of PNAS have not controlled for maternal depression or included systematic blinded ratings. To date, there is no evidence that tapering the dose of antidepressant prior to expected delivery lessens PNAS, and this strategy could increase the risk of postnatal depression . Future studies of PNAS need to include the use of a validated neonatal behavioral symptom scale, assessments of infants by evaluators blind to maternal SSRI use, control for maternal psychiatric disorders and other variables, and follow-up to evaluate long-term sequelae .

Tricyclic antidepressants and pregnancy

Although most recent studies have examined the newer antidepressants, SSRIs and SNRIs, tricyclic antidepressants (TCAs) have a role in some perinatal women. Overall, TCAs are not associated with major congenital anomalies or PPHN . TCAs are associated with an increased risk of PTB, LBW, and neonatal symptoms . Due to neonatal symptoms consistent with both withdrawal and toxicity, the least anticholinergic TCAs (nortriptyline and desipramine) are the preferred TCAs in pregnancy . Clomipramine exposure may increase the risk of cardiac defects, and it may be associated with more severe and prolonged neonatal symptoms .

Long-term effects of antidepressants during pregnancy

A recent systematic review of prenatal exposure to antidepressants and developmental outcomes reported inconsistent findings . Some studies suggest delayed motor development and motor control, and possibly delayed language, but no studies have reported a significant association with cognitive problems . When adverse effects are noted, they appear to be within normal limits and resolve with age. It is difficult to separate the effects of prenatal exposure to antidepressants from the exposure to the underlying prenatal illness and the influence of postpartum factors . It is well established that continued exposure to untreated maternal depression during and beyond the first postnatal year continues to be a critical risk factor for developmental problems in childhood, and it contributes to long-term psychopathology .

Recent studies investigating prenatal SSRI exposure effects on infant and child development continue to report mixed results. Prenatal exposure to SSRIs, but not maternal depression, was associated with altered motor movements in 3–4-month-old infants , and lower gross motor, social–emotional, and adaptive behavior in 10-month-old infants . Positive reports with SSRI use during pregnancy compared to no exposure have included an increased infant readiness to interact at 3 months of age , an increased infant attention measured by P50 auditory sensory gating , and an accelerated infant speech perception . A recent study reported that prenatal exposure to untreated depression, not exposure to prenatal antidepressant use, was associated with behavioral or emotional problems in children at 4–5 years of age . Another study reported that children between 3 and 7 years of age exposed in utero to SSRIs or to venlafaxine did not differ from children exposed to untreated maternal depression on cognitive or behavioral measures .

One of the more concerning possible adverse effects from prenatal SSRI exposure is the development of autism or autism spectrum disorder (ASD), but studies report mixed findings. In addition, untreated maternal depression may be a risk factor for ASD. Four studies have reported an increased risk. The first study reported an almost four times greater risk of ASD with exposure to SSRIs in the first trimester , a second study reported a three times greater risk of ASD without intellectual disability , a third study reported a three times greater risk of ASD in boys , and the fourth study reported an increased risk of autistic traits . Some of these studies did not examine the presence or severity of depression during pregnancy, or adjust for some behaviors that might be associated with prenatal depression. Two recent studies have not found an association between SSRI use during pregnancy and increased risk of ASD .

Other potential effects of prenatal antidepressant exposure

Increased infant mortality in the first postdelivery month and year has been reported with prenatal antidepressant use; however, two recent studies reported no association between SSRIs during pregnancy and increased stillbirth, neonatal , or postneonatal mortality . Antidepressants have also been associated with pregnancy-induced hypertension and preeclampsia . The potential confound of underlying depression is important in the interpretation of the results of these studies.