Managing antidepressant treatment in pregnancy and puerperium. Careful with that axe, Eugene

Abstract

Until 2005, selective serotonin reuptake inhibitors (SSRIs), the class of antidepressants most frequently used in clinical practice, had been deemed devoid of any teratogenic effects. However, in recent years several concerns have been raised about their reproductive safety, including: disturbed fetal development, increased rates of congenital anomalies, increased risks of neonatal complications, neuro-motor delay and even autism. Specific concerns are also arising about the safety of SSRIs for infants breastfed by mothers who take such medications in puerperium. Such considerations have led to the ‘bad reproductive reputation’ of SSRIs, whose utilization during pregnancy and breastfeeding is deemed incautious. Specific reproductive problems also involve tricyclic antidepressants, especially clomipramine. Thus, any conclusion about what antidepressant should be considered the safest during pregnancy must be stated and read with great caution. However, the risks associated with pharmacological treatment must be balanced with the effects of untreated antenatal maternal depression on the mother-fetus dyad, which are likely to be devastating. During puerperium, it is mandatory to weigh the risks to the infant of antidepressant exposure through breast milk against the disadvantage of not receiving mother’s milk and being exposed to a relapse of maternal mood symptoms (which may also have tragic consequences for the patient).

Keywords:

• antidepressants
• breastfeeding
• pregnancy
• safety

The analysis of the risk/benefit ratio to start or continue antidepressant treatment during pregnancy is a complex clinical challenge. This difficulty derives from a number of factors:

1. Data on the reproductive safety of the most frequently used antidepressant drugs in pregnancy, selective serotonin reuptake inhibitors (SSRIs), mainly emerge from studies based on national prescription registries. Thus, it is impossible to establish whether or not pregnant women who deemed antidepressant prescription really adhered to the pharmacological regimen. Indeed, pregnancy is the physiological condition most frequently associated with drug discontinuation.

2. Most studies analyzed SSRIs as a class. However, such medications are not ‘me-too drugs’, because their own receptorial affinity widely varies among specific compounds of this class. Thus, the potential impact on the developing fetus may be different between each single medication.

3. Moreover, studies which assessed the teratogenic risk of SSRIs show contradictory findings. Such medications have alternatively been associated (or not) with several typologies of major malformations. When investigated individually, the only recurrent pattern of congenital anomalies associated with specific SSRI exposure is a modest increase in the risk of cardiac defects.

4. However, even in the case of studies suggesting structural teratogenicity, it is often unfeasible to ascribe the increased rate of birth defects to iatrogenic effects or to unhealthy lifestyle behaviors associated with the underlying mental disorder.

5. Another problem is that warnings issued by national regulatory agencies do not include recommendations useful in clinical practice. Regardless, in 2004 the US FDA instituted a class labeling change for both SSRI and serotonin-norepinephrine reuptake inhibitor antidepressants warning that third trimester exposure to antidepressants may be associated with signs and symptoms consistent with poor neonatal adaptation syndrome, also labeled Prenatal Antidepressant Exposure Syndrome [1]. According to the label change, “reported clinical findings include respiratory distress, cyanosis, apnea, seizures, temperature instability, feeding difficulty, vomiting, hypoglycemia, hypotonia, hypertonia, hyperreflexia, tremor, jitteriness, irritability, and constant crying”. However, we had to wait for a review published in 2007 to become aware that such “clinical manifestations may require specific or supportive treatments in intensive care units” [2].

6. On the other hand, the effects of untreated antenatal maternal depression on both the mother and her baby are likely to be devastating. Untreated gestational depression may have untoward effects on the developing fetus (hyperactivity, irregular fetal heart rate), the newborn (increased cortisol and norepinephrine levels, decreased dopamine levels, altered electroencephalogram patterns, reduced vagal tone, stress/depressive-like behaviors and increased rates of premature deaths and neonatal intensive care unit admission) and the child (increased salivary cortisol levels, internalizing and externalizing problems and central adiposity). During adolescence, an independent association exists between antenatal exposure to maternal mood symptoms and a slight increase in criminal behaviors.

Given this background, it should be expected that any evaluations of the reproductive safety of SSRIs, albeit systematic, may suffer from the personal cultural background of the authors, and the different weight that they attribute to the strengths (few) or limitations (many) of reviewed studies.

In their article [3], Weisskopf et al. state that SSRIs should be considered as first-line agents during pregnancy. In particular, the authors promote the use of citalopram, escitalopram and sertraline, but discourage fluoxetine treatment. However, all these three compounds considered as the safest have been associated with an increase in the risk of birth defects (sertraline with septal heart defects [4,5], citalopram with other cardiac anomalies [6] and escitalopram with muscle-skeletal malformations [7]). Very recently, Bérard et al. [8] also found that among 18,493 pregnancies, the 366 cases exposed to sertraline during the first trimester were associated with an increased risk of atrial/ventricular defects and craniosynostosis, whereas exposure to SSRIs other than sertraline was associated with craniosynostosis and musculoskeletal defects.

Indeed, the UK the National Teratology Information Service still recommends two tricyclic antidepressants (TCAs), namely amitriptyline and imipramine, as first-line agents for antenatal depression [9]. The clinical guidelines of the National Institute for Health and Clinical Excellence in England and Wales also states that prescribers “should take into account that TCAs, such as amitriptyline, imipramine, and nortriptyline, have lower known risks during pregnancy than other antidepressants” [10]. Obviously, among TCAs too, significant differences between single medications in impacting on the fetal development may exist. Clomipramine is actually suspected to increase the risk of birth defects [11], whereas nortriptyline is the drug with the lowest number of reports of adverse events [11]. Moreover, albeit TCAs can also increase the risk of preeclampsia and induce an usually mild and transient poor neonatal adaptation syndrome (particularly severe just in the case of clomipramine exposure) if used in late pregnancy, until now this class of antidepressants have not been associated with increased risks of persistent pulmonary hypertension of the newborn, QT prolongation and necrotizing enterocolitis (more or less frequently associated with SSRI exposure). For these reasons, there appears to be a slight gain in safety if TCAs (with the exception of clomipramine) are used in late pregnancy [11].

For all these reasons, the urgency to perform well-designed studies to assess the reproductive safety of antidepressants should be further boosted. As previously suggested [12], large, multicenter, prospective studies are mandatory to satisfy this necessity. Such studies should include women of similar psychiatric diagnosis (e.g., monopolar major depression) and should be designed using a four-arm approach: women under treatment who display euthymic conditions prior to conception; exposed women who fail to achieve remission within the first trimester; depressed women refusing any pharmacological approach; healthy pregnant women control group. This is the only design which may allow to distinguish iatrogenic effects from the detrimental consequences of maternal depressive symptoms and behaviors. However, any other study designs and analytical epidemiological approaches (which, together with for more consistent results coming from randomized controlled trials) will contribute to clarify the relationship between antenatal antidepressant exposure and fetal development may be useful.

As regards breastfeeding, the authors state that citalopram, escitalopram and sertraline are also compatible with lactation [3]. However, the authors support this conclusion by using parameters whose usefulness in clinical practice is still doubtful. Indeed, adverse reactions associated with SSRI used during breastfeeding have been reported in clinical cases characterized by milk-to-plasma ratios < 1 (the theoretical limit of concern), with low values of relative infant dose, and even in infants with undetectable serum levels of SSRIs [13]. In particular, citalopram (which is excreted in human milk in higher amounts) have caused a number of suspected adverse effects [14]. Available data are too limited to consider TCAs as a class as first choice agents in nursing mothers but, for instance, nortriptyline shows limited but concordant reassuring information.

Thus, to analyze the possibility that women on antidepressant treatment may breastfeed their infant, the need exists for the institution of an international case-register recording the outcome of the infants exposed to antidepressant agents through maternal milk [15]. The register might represent an indispensable instrument for collecting additional, significant data about this specific safety side of the most widely used psychotropic medications. At least, the register should record information on maternal psychiatric diagnosis, drug daily dose, duration of both treatment and breastfeeding. Conversely, the reports describing negative outcome but characterized by placental exposure and/or by concomitant maternal treatment with other drugs should be included in a different section. In this way, it may be possible: i) to reduce the risk of finding contaminations due to the interference of two relevant confounding factors and ii) it may be less difficult for clinicians to weigh the potential risks to the infant of antidepressant exposure through breast milk against the disadvantage of not receiving mother’s milk and being exposed to a relapse of maternal mood symptoms (which may also have tragic consequences for the patient).

In all cases, utilizing strategies such as planning for pregnancy, educating the mother and family regarding the risk/benefit ratio of treatment versus no treatment, and involving in the decision-making process a multidisciplinary team with good expertise in perinatal psychiatry may improve outcomes for the mother-infant dyad.

Meanwhile, any conclusion about what antidepressant should be considered the safest during pregnancy and puerperium must be stated and read with great caution.

Acknowledgments

S Gentile wishes to acknowledge the Pink Floyd, his favorite rock band, for having inspired the title of this Editorial. Pink Floyd. Careful with that axe, Eugene. Ummagumma. https://www.youtube.com/watch?v=b72I7TwsvBE.

Declaration of interest

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending or royalties.