Clinical questions

1. 1. In women with a history of a depressive disorder who wish to or have recently conceived, does discontinuing psychiatric medication (i) increase the risk of recurrence of illness; and (ii) does discontinuation promote poor maternal and fetal outcomes?

About 5–8% of women are receiving antidepressants at the time of fertilization [9]. The patient, psychiatrist, and obstetrician should collectively discuss whether to continue or discontinue antidepressant medication during pregnancy. Psychiatrists and obstetricians are concerned about both the health of the patient and offspring’s health and hence would like to avoid a potential maternal relapse. Clinicians and patients may also have concerns about possible adverse effects of medication. These concerns should be discussed in a thorough and un‐biased manner and clinicians should conduct a risk assessment.

Of those who discontinue treatment during pregnancy, depression history, illness severity, and treatment history are factors that may influence risk for recurrence of a depressive disorder. Characteristics of the course of illness, including greater number of previous depressive episodes, younger age of onset, and presence of concurrent psychiatric conditions, increase the likelihood that a pregnant woman will experience a recurrence during her prenatal period.

1. 2. In pregnant patients with first trimester complaints of emotional symptoms, what are the sensitivity and specificity of various historical risk factors and symptoms in the diagnosis of major depressive disorder, or bipolar I disorder?

Symptoms for depressive disorders include depressed mood, anhedonia (loss of interest), significant weight loss or gain, sleep changes, psychomotor changes, fatigue, feelings of worthlessness or guilt, inability to concentrate, and thoughts of death or self‐harm. A unipolar major depressive episode includes five of these symptoms, with one being either depressed mood or anhedonia, for at least two consecutive weeks. A minor depressive episode will include dysphoria or anhedonia, and one or two other symptoms, also during a continuous two‐week period [2].

Women are often screened for antenatal depression with questionnaires and surveys such as the Center for Epidemiologic Studies Depression Scale (CES‐D), Beck Depression Inventory (BDI), and Patient Health Questionnaire 9 (PHQ‐9) [10]. However, diagnosing depression during pregnancy can be difficult for providers because of similarities between physiological changes of pregnancy and symptoms of a depressive episode. The Edinburgh Postnatal Depression Scale (EPDS) takes this into account by focusing on emotional symptoms [10]. The cut‐off score in validity studies is 12/13 [11]. Using this threshold in postnatal women, the sensitivity for detected a woman with a major depressive episode was 0.75 and the positive predictive value as 0.24 [11]. A study of pregnant and postpartum women (n = 91) found that at a score of greater than 12, the sensitivity was 0.78, specificity 0.77, positive predictive value of 0.66, and negative predictive value of 0.86 [12].

A patient’s social history before pregnancy can indicate risk for development of a major depressive episode during pregnancy. Marital status, relationship with partner, history of partner violence, and educational level completed, are predictive factors in multiple studies [13–15]. Age during pregnancy may have some relevance but studies vary on whether specific age groups, such as adolescence or advanced maternal age elevate risk [14, 16]. Pregnancy‐related risk factors, such as fear of childbirth and unplanned pregnancy are associated with increased risk of developing a depressive disorder [14].

A thorough history and physical at the initial prenatal visit would be sufficient to identify past psychiatric, social, and pregnancy‐related factors to evaluate for risk for antenatal depression. Patients who identify as high‐risk for depression should be followed throughout their pregnancy for signs and symptoms for mood disorders. Given that typical physiologic changes of pregnancy can be confused for a depressive disorder, reliance on non‐physiological symptoms can help, make a diagnosis.

1. 3. In pregnant women with a major depressive episode, does treatment with antidepressants associate with an increased risk of adverse perinatal outcomes?

In the United States, it is estimated that 11–13% of pregnant women have been prescribed an antidepressant at some point in their pregnancy [17, 18]. Prevalence of antidepressant use during pregnancy has increased since the 1990s, but has been declining recently as warnings emerge of specific selective serotonin reuptake inhibitor (SSRI)‐associated adverse outcomes during pregnancy [19, 20]. Many pregnant women discontinue medication on their own when they become pregnant; others do so with the recommendation or approval of their treating physician. Accordingly, studies find lower rates of antidepressant use in the third trimester of pregnancy compared to the first trimester [20].

Women may elect to cease treatment with just an intention to become pregnant. Despite earlier reports to the contrary, periconceptional use of antidepressants does not appear to be related to spontaneous abortion. In a clever analysis of 1 279 840 pregnancies, Anderson et al. found that women who were undergoing treatment with an SSRI had a higher rate of spontaneous abortion. However, when they looked at women who took the same medication but stopped treatment in the 3–12 months prior to pregnancy, the rate of spontaneous abortion was as high as in women who continued medication [21]. This study suggests that factors related to the need for antidepressant treatment, but not that treatment itself is linked with spontaneous miscarriage. These results were in line with an earlier study that found women who had a history of treatment with antidepressants had similar rates of spontaneous abortion as women who underwent antidepressant treatment in pregnancy [22].

Antidepressant use has not been shown to influence quantitative outcomes of in vitro fertilization (IVF) treatment. randomized control trials (RCTs) and retrospective reviews have not found a difference in serum estradiol levels, oocyte number, quality during retrieval and their subsequent maturation [23, 24]. However, there was a negative trend in pregnancy rates for SSRI users in two studies. In one study, overall the SSRI users had poorer pregnancy rates; per cycle started – 17.1% vs 28.9%; per embryo transfer – 23.3% vs 32.1%; live birth rate 14.6% vs 21.2%; the difference was not statistically significant – possibly due to the small number of subjects [24, 25]. Notably, cycle cancellation rates have been shown to be significantly higher among SSRI users (26.8% versus 10.0% in non‐SSRI users) [25] due to a poor ovarian response to fertility treatment. The authors stated that while the direct impact of SSRI’s on the canceled cycles was unclear, it is possible there was an interaction with the gonadal axis in some fashion [25].

Depression has already been linked to a higher incidence of preterm birth (PTB), but an additional increased risk of PTB has been associated with both mixed antidepressant and SSRI use [26–29], although there is some disagreement among studies [30]. A meta‐analysis attributed a 1.74 relative risk increase of PTB in patients who have used SSRI’s at any point during pregnancy [26]. Further analysis shows that a majority of these births are primarily late preterm (34–37 weeks) [29]. Additionally, there is limited evidence that spontaneous PTB, but not indicated PTB is associated with antidepressant use in pregnancy [31].

Researchers have also focused on associations between antenatal antidepressant use and various birth defects. Conclusions have been on both sides of an association due to difficulties comparing studies and cofounding factors [32–34]. Meta‐analyses show that the rate of malformations associated with antidepressant use in pregnancy is a magnitude smaller than the rates that one sees with teratogens such as tretinoin [35].

A meta‐analysis in 2010 established paroxetine use and cardiac defects had a prevalence odds ratio of 1.46, or one more case of a cardiac defect per 200 newborns exposed to paroxetine as a fetus [36]. GlaxoSmithKline themselves have acknowledged the risk, labeled the drug as Category D [37].

Other congenital birth defects have been associated with such as omphalocele and neural tube defects have also been cited in the literature, but studies have several limitations [38–42]. A recent study that used a large multi‐site case‐control study to replicate previous findings in the literature found associations between first trimester exposure to paroxetine and cardiac defects, anencephaly, gastroschisis and, omphalocele, as well as fluoxetine and right ventricular flow defects and craniosynostosis [43]. This lack of consistency has not further clarified risks and one is left assuming a small risk associated with a variety of the detected malformations.

Postnatal adaptation syndrome (PNAS) is a pattern of symptoms found in infants exposed to antidepressants at any time during gestation [44, 45]. The infant may suffer from symptoms such as tachypnea, hypothermia, hypoglycemia, and irritability for days to weeks – similar to those found in adult SSRI‐discontinuation syndrome and serotonin syndrome [46]. Research estimates of PNAS in up to 30% of exposed infants, prompted the FDA to issue a statement in 2004 encouraging providers to adjust dosage during the third trimester of pregnancy [45–47]. The problem with this strategy is that it can increase the risk of relapse in mother as she approaches delivery. Furthermore, it is not clear whether this tact will decrease rates of PNAS.

Long‐term developmental data regarding intrauterine exposure to antidepressants is limited, but studies are currently being conducted. Of note, several new case‐control cohort studies found a significant increased risk of developmental disorders such as autism‐spectrum disorder and attention‐deficit hyperactive disorder if a depressed pregnant woman used antidepressants during their pregnancy [48–53]. Conversely, the largest population‐based study did not find a significant association between SSRI use during pregnancy and risk of autism spectrum disorder [54]. A causal relationship is difficult to determine, but if there is one, the effect is small.

In conclusion, antidepressant use has been linked to a variety of maternal, fetal, and obstetrical complications. Studies on the topic have their limitations typically arising from confounding. Many women who take antidepressants are also undergoing treatment with other psychotropic and non‐psychotropic medication that can cause adverse birth outcomes [55]. As well, women who require antidepressant treatment may also be engaging in unhealthy habits such as licit or illicit substance use. Observational data can control for these factors if information about confounders is accurately collected but many women are reluctant to disclose these problems. Several medications, most notably paroxetine, are associated with various congenital anomalies if used during organogenesis. PTB has a consistent association with antidepressant use, but most these agents appear to shorten gestation by only a few days. As research of antidepressant ages, more data will surface regarding its effect on fertilization and long‐term sequelae, such as child development.

1. 4. In pregnant women with a major depressive episode, are there non‐pharmacological treatments that can reduce depressive symptoms significantly when compared to the antidepressants?

Management of depression during pregnancy primarily relies on severity of presenting symptoms [9, 56]. Patients with mild depression, or simply depressive symptoms, may be appropriate candidates to trial non‐pharmacological management options or use them as adjunct therapy. Psychotherapy, such as cognitive behavioral therapy

Stay updated, free articles. Join our Telegram channel

Join