Recognising and managing depressive disorders in pregnant women presents specific challenges for both the patient and her GP – particularly in terms of understanding and communicating the risks and benefits of antidepressant treatment. Here, the authors discuss the sometimes complex issues.

Dr Sarah Catherine Jones Clinical Research Fellow, National Centre for Mental Health, Cardiff University Higher Trainee, East London NHS Foundation Trust
Dr Liz McDonald Honorary Consultant Perinatal Psychiatrist, East London NHS Foundation Trust

Depression and anxiety are common during pregnancy, with the point prevalence of major depressive disorder in pregnancy estimated to be between 3.1 and 4.9% – and 11% when minor depression is included.1 It is thought that approximately half of all postnatal depressive episodes start in pregnancy. Untreated maternal depression is associated with negative sequelae for both mother and fetus/infant, family relationships, interpersonal functioning 2,3,4 and can also have devastating consequences, with suicide remaining one of the leading causes of maternal deaths in the UK.5

More women are facing the dilemma of whether or not to start or continue with antidepressant medication during pregnancy. Danish data found that the rate of antidepressant use increased from 0.2% in 1997 to 3.2% in 2010.6 In the UK, approximately 3% of pregnant women from 2001 to 2006 received antidepressants at some stage during pregnancy.7 However, not all women who are taking antidepressants have a history of moderate- severe depression, and an assessment of the severity of depression, current or historical, will need to be undertaken when considering whether to start or continue to prescribe an antidepressant in pregnancy. 
This article addresses some of the issues when prescribing antidepressants for pregnant women in the primary care setting. However, other treatment approaches informed by NICE guidance should be borne in mind and women offered the treatment that is most appropriate and effective for them. More than half of pregnancies are unplanned,8 and this figure is much higher in women suffering with mental illness; hence the issues discussed here are pertinent to all women of child bearing potential and not to pregnant women alone.
Identifying women with perinatal mood disorders 
Maternal depression is frequently unrecognised and untreated in pregnancy despite processes having been developed throughout the UK to identify those at risk through raised awareness in midwifery teams, primary care and the development of specialist perinatal mental health teams. Women who are depressed may be reluctant to disclose or discuss their concerns about their mental health because of stigma, shame, the potential negative perception of them as a mother or fear that their infant might be removed from their care. These same factors may also affect their attitude towards and acceptance of medication, even if it is believed to be the most appropriate treatment for that particular woman at the time. Making culturally relevant information on mental health and pregnancy available, while actively involving the woman in decisions about her care and treatment may help to overcome some of these barriers. Displaying sensitivity, warmth and a non-judgmental attitude during the consultation are also characteristics that will enhance the woman’s trust and perception of the treating clinician.9
On a clinical level there can be some overlap in symptomatology between pregnancy and the biological features of depression, which can make diagnosis challenging. There is little difference in diagnostic features between a major depressive episode inside or outside of the perinatal period,10 although the context of being pregnant and parenting alters the risk profile. Pregnant women may complain of sleep and appetite disturbance, with reduced energy and libido. Asking about depressive ideas such as hopelessness, worthlessness, guilt and suicidality will facilitate more accurate diagnosis of depression, allowing women who require psychotropic medication to be identified and treated (See Figure 1).
Suspected mild and moderate cases of uncomplicated non-psychotic depression and anxiety can be assessed and managed in the primary care setting; however those women with more severe depressive episodes will need an assessment by a psychiatrist. A referral to secondary care mental health services/perinatal mental health services should be carried out, but treatment initiation should not be delayed. Women with a history of severe depression or psychotic mood disorders, including bipolar disorder and psychotic depression, should be under the care of mental health services or specialised perinatal mental health services for the duration of the pregnancy and postnatal period. Those women requiring emergency assessment and treatment should be managed appropriately by psychiatric liaison teams in the A&E setting.
Risk profiling for two: a complex issue
Decision-making regarding prescribing in pregnancy is difficult and complex but should not be avoided merely because of the risks, both real and perceived, of exposure of the fetus to psychotropic medication. The prescriber has to consider the welfare of two individuals at all times in the context of an evidence base that is limited and associated with multiple confounders. Untreated depression carries risks to both the fetus (see under Risks of non-treatment) and the mother, as does abrupt cessation of an antidepressant. The doctor has to work in a supportive, collaborative manner that actively involves the woman in decision-making presenting her with balanced and up-to-date evidence. When pregnant, the woman is in contact with a range of healthcare professionals, and communicating treatment plans and decisions made with these colleagues is important for both the welfare of the mother and the fetus.

Figure 1: Key symptoms of depression (NICE, October 2009)
At least one of these, most days, most of the time for at least 2 weeks:  
  • Persistent sadness or low mood; and/or  
  • Marked loss of interests or pleasure
If any of above present, ask about associated symptoms:  
  • Disturbed sleep (decreased or increased compared to usual)  
  • Decreased or increased appetite and/or weight  
  • Fatigue or loss of energy  
  • Agitation or slowing of movements  
  • Poor concentration or indecisiveness  
  • Feelings of worthlessness or excessive or inappropriate guilt  
  • Suicidal thoughts or acts
Note that women may have sleep, appetite and energy changes in pregnancy without being depressed
Figure 2: Risk-benefit analysis of antidepressant treatment in pregnancy
  • The potential benefits of any treatment, taking into account the severity of the disorder and response to any previous treatment  
  • The potential risks and harms of treatment  
  • The risks and harms associated with untreated depression  
  • The need for prompt treatment because of the potential impact of an untreated mental disorder on the fetus or infant  
  • The risk of relapse or deterioration associated with switching or stopping medication  
  • The risks associated with stopping medication abruptly  
  • The possibility that stopping or switching a drug with known teratogenic risk after pregnancy is confirmed may not remove the risk of malformations

Each woman has a unique presentation, history and understanding of her illness. It is important, therfore, to consider the individual risk profile for each patient (see Figure 2). A detailed history focusing on presenting symptoms, past psychiatric history (including a history of deliberate self-harm or suicide attempts) and response to medications and psychological interventions is critical. 
It is essential to ascertain her understanding of her illness and the impact this has on her functioning as well as her engagement with services and compliance with medication to help each woman and her clinician to make the right treatment decision. It is also important to have an understanding of the risks that are acceptable for her as an individual. Some women may decide that any perceived risk of relapse of depressive symptoms will be unacceptable and will favour antidepressant treatment, prioritising the potential benefits of treatment over side-effects or potential risks to the developing fetus. For other women, it may be that any potential clinical or statistical risk of harm to the developing fetus is unacceptable. Therefore, a comprehensive understanding of the patient’s history and attitudes towards her mental health and treatment is essential in approaching the complex task of consulting on medication decisions in pregnancy.
Communicating risk
In approaching the discussions about treatment in pregnancy, a number of general factors need to be considered. Firstly, there is a high degree of uncertainty associated with the risk-benefit analysis of antidepressants in pregnancy, since despite a considerable increase in the number of studies, the messages remain complex, confusing and contradictory (see Figure 3).
Figure 3: Problems with available data
  • Retrospective bias  
  • Little information on severity of the malformation/disorder  
  • Increased vigilance for malformations in women taking medication  
  • Confounding factors: biological effects of underlying depression, behaviours associated with depression, other medications, obesity, smoking, alcohol, overlapping genetic factors between mood disorders and other mental disorders

Secondly, communicating information about risk can be challenging. It is important to consider how statistical significance and clinical significance is understood by the woman. Information about benefits and potential harm needs to be presented at a level that the patient will understand. Absolute values based on a common metric (e.g. 1/100) should be used.11 It may be important to involve partners and/or family in the decision-making process if this is appropriate and acceptable to the woman (see Figure 4). 
Figure 4: Communicating risk
When discussing the likely benefits, risk or harms associated with treatment:  
  • use absolute values based on a common metric (that is numbers out of a 100 or 1000)  
  • acknowledge the uncertainty around any estimate of risk, harm or benefit  
  • provide information in written or other acceptable format  
  • include partner or family if appropriate

Risks of non-treatment 
The risks of an untreated mood disorder needs to be balanced against the risk of antidepressant treatment in pregnancy. Untreated depression during pregnancy has been associated with adverse outcomes for the developing fetus. Some studies have demonstrated an association between untreated maternal depression and spontaneous abortion, lower gestational age, lower infant birth weight and lower APGAR scores, pre-eclampsia, neonatal intensive care unit admissions and breast-feeding initiation.12 Women who have an untreated depressive disorder during pregnancy have also been shown to display impulsive risk- taking behaviours, higher rates of smoking and poor attendance at antenatal clinics. Severe depression also carries the risk of self-harm, impaired social functioning and poor engagement with obstetric services. In the perinatal period, suicide is one of the leading causes of maternal death13, and untreated severe/complex antenatal depression can have tragic consequences. Untreated maternal depression can have a negative impact on the woman’s partner, her other children as well as her parental and occupational function.14 In one of the largest meta-analysis conducted so far, the results show that there may be a link between antenatal depression and an increased risk for premature delivery and lower rates of breast- feeding initiation, although the odds ratios were small putting into doubt the strength of the association.15
Risks associated with antidepressant treatment 
(See Figures 6-8) 
Recently there has been an increase in the number of large studies examining the significance of associations between antidepressant exposure and adverse fetal outcome. Paroxetine has been shown to be associated with an increased risk of cardiac malformations.16,17 Antidepressants do not seem to be associated with an increased risk of congenital malformations in general, although a statistically significant association was observed between cardiovascular malformations and exposure to any antidepressant in utero; however the results may not be clinically significant and the excess may be related to uncontrolled confounders.18 
There is an association between antidepressant exposure in utero and poor neonatal adaptation syndrome (PNAS - see Figure 5), where fetal exposure to antidepressant medication manifests as a syndrome of sleep disturbance, agitation, tremor, restlessness, irritability, poor feeding, vomiting or diarrhoea, hypoglycaemia, poor temperature control, respiratory distress or seizures.19 This has been shown to be statistically and clinically significant, affecting approximately 30% of babies exposed. The syndrome, while common, is self-limiting with minimal clinical or long-term impact. 
Figure 5. Poor Neonatal Adaptation Syndrome
  • Floppy baby  
  • Irritability  
  • Constant crying  
  • Shivering  
  • Tremor  
  • Restlessness  
  • Hypoglycaemia  
  • Poor temperature control  
  • Increased tone  
  • Feeding and sleeping difficulties  
  • (Rarely) seizures 

There is an increased risk of persistent pulmonary hypertension in the neonate (PPHN) when the fetus is exposed in late pregnancy (after 20 weeks gestation), with the general population risk of PPHN of 1.9 per 1000 live births increasing to 5.40 per 1000 births. The absolute risk difference is 0.35% with a number needed to harm (NNTH) of 286.20 This means that for every 286 patients treated with antidepressant medication after 20 weeks gestation, one more would have a baby born with PPHN. The increased risk of PPHN is clinically small compared to the background rate; however, the clinical implications are life-threatening. 
More recently, a number of studies have reported an association between exposure to SSRI medication in pregnancy and a higher risk of autism spectrum disorders (ASDs) in children,21,22 but the cause of this relationship remains uncertain. In a recent paper,23 depressive symptoms in pregnancy that were not treated with antidepressants were also associated with a subsequent risk of autism. Although the risk was higher for the group treated with SSRIs, it is possible that this reflects an increased severity of depression in the women taking medication. The baseline risk in the general population for ASDs in children is around 1%. The papers that find an association suggest the rate in children born to women taking SSRIs in pregnancy may be approximately double at around 2%. It remains difficult to know whether this modest increase in risk is due to the medication being taken, to the mood disorder itself, to an overlap in genetic vulnerability or to other factors associated with mood disorders and antidepressant medication.
Principles of prescribing antidepressants in pregnancy 
The general principles of prescribing, such as adequate dose, efficacy and side-effect profile, should be followed in pregnancy. It is always pertinent to avoid unwarranted use of psychotropic medication in the first trimester, when the major organs develop. 
The decision to prescribe will be guided by a risk-benefit analysis. Only women suffering with a moderate to severe depressive disorder should be treated with medication, and prompt treatment of severe depression is important because of the associated risks. In those women with a history of a severe depression and where the risk of relapse is deemed to be high, continuation of antidepressant treatment will be necessary. Use the established antidepressant medication at the lowest effective dose; switching medication carries a risk of relapse, as well as potential effects on the fetus so an increased level of monitoring will be required during these times. Relapse rates are higher in those with a history of major depression who discontinue medication.24 
An SSRI will be the first-line in women who have a first episode of antidepressant treatment or are known to be responders to SSRIs. Women who have a history of non-responsiveness to SSRIs should take the antidepressant that they are known to respond to. 
No psychotropic medication has a UK marketing authorisation specifically for women who are pregnant or breastfeeding. The prescriber should follow relevant professional guidance, taking full responsibility for the decision and obtaining and documenting informed consent (see Figure 9). 

Figure 6: Risks associated with antidepressants
  • TCAs, SSRIs and (S)NRIs taken in the first trimester may be associated with a small increased risk of fetal cardiac defects  
  • TCAs, SSRIs and (S)NRIs taken after 20 weeks’ gestation may be associated with a small increased risk of Persistent Pulmonary Hypertension in the newborn infant  
  • Venlafaxine may be associated with an increased risk of maternal high blood pressure at high doses and higher toxicity in overdose in the woman than SSRIs  
  • There is a risk of discontinuation symptoms in the woman and neonatal adaptation syndrome in the baby with most TCAs, SSRIs and (S)NRIs  
  • Paroxetine and Venlafaxine are associated with increased severity of discontinuation symptoms in the woman and neonatal adaptation syndrome in the baby  
  • TCAs have a higher fatal toxicity index than SSRIs in overdose

Figure 7: Summary of risks associated with antidepressant treatment
  • Teratogenicity – cardiac and other congenital malformations  
  • Neonatal Adaptation Syndrome – withdrawal/ toxicity  
  • Persistent Pulmonary Hypertension of the Neonate  
  • Neurobehavioural sequelae – Autism Spectrum Disorders

Figure 8: General principles of prescribing antidepressants in pregnancy
  • Establish a clear indication for prescribing an antidepressant  
  • Choose a drug with a lower risk profile for the woman and the fetus  
  • Start at the lowest effective dose and slowly increase  
  • Preferably use monotherapy  
  • Take into account the woman’s known history of responsiveness to particular antidepressants ie. choose an antidepressant that will be effective  
  • Take into account the woman’s preference  
  • Obtain up-to-date advice  
  • Monitor regularly and assess efficacy and side-effects  
  • Communicate and record the antidepressant prescribed and the associated discussions in all versions of the woman’s notes and with all relevant healthcare professionals

Figure 9: Antidepressants and breastfeeding: key considerations 
  • Encourage breastfeeding  
  • Support woman if she chooses not to breastfeed  
  • Acknowledge limited data on safety of antidepressants in breast milk  
  • Monitor infant for adverse effects Most antidepressants are present in breast milk at relatively low levels  
  • Fluoxetine present in breast milk at relatively high level

Suicide remains one of the leading causes of maternal deaths; however prescribing antidepressants in pregnancy is a complex issue. Consideration of each patient’s history, attitudes to medication and acceptable risks of each woman is essential. The prescriber has to carry out a risk-benefit analysis for two individuals, patient and fetus, in the context of a limited evidence base. 
Communicating risk is challenging, and it is important to distinguish between clinical and statistical significance. Antidepressant medication should be prescribed where indicated, since untreated maternal depression is associated with negative sequelae for both mother and fetus and can have tragic consequences.

1. Gavin N, Gaynes B, Lohr K, et al. Obstetrics and Gynaecology. 106(5 Pt 1):1071-83, 2005 Nov. 
2. Alder J, Fink N, Bitzer J, et al. J Matern Fetal Neonatal Med. 2007 Mar;20(3):189-209. 
3. Coverdale J, McCullough L, Chervenak F, Bayer T. Aust NZ J Obstet Gynaecol. 1996;36:424–429. doi: 10.1111/j.1479- 828X.1996.tb02185.x. 
4. Zuckerman B, Amaro H, Bauchner H, Cabral H. Am J Obstet Gynecol. 1989;160:1107–111 
5. Cantwell R, Oates MR. Brit J Obstet Gynaecol. 2011;118(Suppl 1):102032 US 
6. Jimenez-Solem E, Andersen J, Petersen M, et al. PLoS One. 2013 Apr 25;8(4):e63034. doi: 10.1371/journal. pone.0063034. Print 2013. PMID: 23638179 
7. Petersen I, Gilbert R, Evans S, et al. J Clin Psychiatry. 2011 Jul;72(7):979-85. doi: 10.4088/JCP.10m06090blu. Epub 2011 Mar 8 
8. Wellings K, Jones, KG, Mercer C, et al. Lancet2006. 382 (9907): 1807. doi:10.1016/S0140-6736(13)62071-1. 
9. Shakespeare J, Blake F, Garcia J. J Reprod Infant Psych 2006:24:2 
10. Jones I and Cantwell R. The classification of perinatal mood disorders – suggestions for DSMV and ICD-11 
11. NICE guideline 2007. Clinical guideline 45. 
12. Bonari L, Pinto N, Ahn E. Canadian J Psychiatry 49(11), 726-735. 
13. Saving Mothers’ Lives. Reviewing maternal deaths to make motherhood safer: 2006-2008; Centre for Maternal and Child Enquiries (CMACE), Brit J Obstet Gynaec, Mar 2011 
14. Hoffbrand S, Howard L, Crawley H. The Cochrane Library. Issue 2. John Wiley & Sons, Ltd. 
15. Grigoriadis S, VonderPorten EH, Mamisashvili L, et al. J Clin Psychiatry. 2013;74(4):e321–e341 
16. Cole J, Ephross S, Cosmatos I, Walker A. Pharmacoepidemiol Drug Saf. 2007 
17. Kallen B, Otterblad Olausson P. Birth Defects Res A Clin Mol Teratol 2007;79:301-8 
18. Grigoriadis S, VonderPorten E, Mamisashvili L, et al. J Clin Psychiatry. 2013;74(4):e293–e308 
19. Grigoriadis S, VonderPorten EH, Mamisashvili L, et al. J Clin Psychiatry. 2013;74(4):e309–e320 
20. Grigoriadis S, Vonderporten E, Mamisashvili L, et al. Br Med J 2014;348 
21. Croen L, Grether J, Yoshida C, et al. Arch Gen Psychiatry. 2011 Nov;68(11):1104-12. doi: 10.1001/ archgenpsychiatry.2011.73. Epub 2011 Jul 4. PMID: 21727247 
22. Rai D, Lee B, Dalman C, et al. Br Med J. 2013 Apr 19;346:f2059. doi: 10.1136/bmj.f2059. 
23. Marroun H, White T, van der knaap N, et al. Br J Psych August 2014 205:95-102; doi:10.1192/bjp.bp.113.127746 
24. Cohen L, Altshuler L, Harlow B, et al. JAMA. 2006 Feb 1;295(5):499-507.