Abstract
Postpartum depression occurs in at least one in seven new mothers, usually within the first 6 months after delivery. By the time of onset of postpartum depression, the mother has usually long since been discharged from the maternity hospital. Early identification and treatment of these mothers reduces both maternal and infant suffering. Careful risk–benefit decision-making regarding various treatment options in the postpartum should be discussed with the mother. Risks of untreated depression include poor bonding with the infant, lack of self care, infant neglect and infanticide.
Case study
Mrs Jones is a 27-year-old woman, who is seen at her 6-week postpartum obstetrics appointment, and has brought her infant daughter with her. She relates to the clinic nurse that although she is feeling well physically, she has been feeling ‘overwhelmed’ by caring for her infant and does not have the assistance at home from her mother that she had anticipated. She feels intensely guilty about not being ready to return to work and also feels inadequate as a mother. These worries began 1 week after coming home from the hospital. Upon further questioning, she relates that she has been feeling sad most of the time, has low energy and psychomotor retardation. She has had difficulty falling asleep, even when the baby is asleep. She worries that she will need to be up soon to breastfeed. She has not been eating appropriately – although she has been trying to do so for the baby's sake – she is unable to ‘force down much food’. She reports no personal history of bipolar disorder or depression, nor any such family history.
Background
Across the world, approximately one in seven women (13%) suffer from postpartum depression (PPD). From the 39 weeks before pregnancy to the 39 weeks after pregnancy, in a large retrospective study (n = 4398) more than one in seven women (15.4%) received a diagnosis of depressive disorder [1]. At 6 weeks postpartum, 22% of mothers in one sample (n = 60) had depressive symptoms; this remained stable at 6-months postpartum (21%) [2].
Not only is there stigma associated with any mental illness, but admitting that one is depressed during what is supposed to be the happiest time in a woman's life is often difficult. However, in actuality, the time of highest risk in a woman's lifetime for the development of mental illness is in the postpartum. Depression is the most prevalent ‘complication’ of childbirth [3]. Nonetheless, approximately half of the cases of PPD are undiagnosed and untreated.
Risk factors for postpartum depression
Mothers are at particular risk for PPD if they have a past history of depression (especially after other pregnancies) and if their antidepressant medications were discontinued during pregnancy. Other risk factors for PPD include: depression or anxiety in pregnancy, family history of depression, vulnerability and stressful life events (such as deployment of a partner [4]), relationship [5] and financial problems [6], and lack of emotional support [5,7–9]. Hormonal factors (such as receiving progestogen within 48 h of delivery [10]), role changes, the added stress of caring for a newborn, the sleep deprivation associated with lactation and being a new parent may also elevate risk. In addition, women with certain personality traits (e.g., avoidant, dependant and obsessive–compulsive) may be at higher risk [11]. Contrary to popular belief, there is not currently any evidence that lactation status is independently associated with PPD [11].
Diagnosis
According to the Diagnostic and Statistical Manual-IV [12], symptoms of PPD are the same as symptoms of depression that begin outside of the perinatal period. These symptoms include: depressed mood, loss of interest in activities or anhedonia, sleep disturbance, appetite disturbance (either increased or decreased), feelings of worthlessness or excessive guilt, psychomotor retardation, fatigue or low energy, decreased concentration and suicidal thoughts [12]. For a DSM diagnosis of major depression, a person must experience at least five of these nine symptoms, including either depressed mood or loss of interest/pleasure. In addition, depressive symptoms must have lasted for at least 2 weeks, and result in impairment or distress. The DSM allows a diagnosis of a major depressive episode with a postpartum onset specifier, if symptoms began within 4 weeks after delivery. The International Classification of Diseases-10 considers the onset within 6 weeks of delivery [13]. However, other authorities suggest that PPD should be based on the occurrence of symptoms in the first 6 months postpartum [14], or even the entire postpartum year.
Caution must be used in order to not identify normal experiences in the postpartum as pathologic. For example, although the DSM discusses changes in sleep and appetite as criteria for depression, in reality most new mothers will experience changes in sleep patterns and appetite. A helpful screening question is to ask whether the mother is able to sleep when the baby sleeps. This will provide a clue about whether a mother has true insomnia [101].
Depression may range from mild to severe, related to the number of depressive symptoms and impairment that the woman experiences. Women who have experienced severe depression at any time in their lives, including the postpartum, may also have associated psychotic symptoms, such as mood-congruent delusions in which content includes maternal inadequacy, nihilism or feeling deserving of punishment.
Anxiety symptoms commonly co-occur with postpartum depression. Women referred for postnatal depression in a sample from the UK and New Zealand (n = 129) were evaluated for anxiety with the Birmingham Interview [15]. Their postpartum anxiety focused on: morbid anxiety regarding infant health and infant care, pathological fear of cot death (sudden infant death syndrome), fear of criticism or concerns about removal of the child. Some mothers were particularly anxious in their baby's presence, which led to decreased contact with their infant [15]. Early depressive symptoms lead to more anxiety, which in turn predicts more depressive symptoms [16]. Anxiety in late pregnancy predicts both PPD and post-traumatic stress disorder (PTSD). Critical life events and a stressful delivery also predict postpartum depression and PTSD symptoms [2].
The Edinburgh Postnatal Depression Scale (EPDS) screens for depressive symptoms. It is a ten-question scale quickly completed by the mother, which is user-friendly and simple to score. Scores range from 0 (no symptoms) to 30 (severe depression and anxiety). A cutoff threshold of ten is often used when identifying depression; this measure is accurate and effective [3]. The EPDS may also be used on more than one occasion to track a woman's symptoms. In a sample of 49 urban women who scored higher than this threshold on the EPDS at pediatric visits, a third experienced high symptom levels throughout the year, 26% developed high symptom levels after 3 months postpartum and 41% improved after the first 3 months [17]. The EPDS is available online at many websites (e.g., [102]).
Screening for PPD is effective when carried out by professionals in obstetrics/gynecology, pediatrics or by public health nurses [18–21]. Mothers may be more likely to attend pediatrics appointments than their own routine postpartum checks if they have a limited amount of resources. Staff may use a screening tool or ask directly about maternal functioning. However, clinicians must keep in mind that some mothers fear being judged or losing custody of their infants. Written communications (such as pamphlets in the waiting room regarding depression and available services) and emotional support are useful [22–23]. Once screening is completed, it is critical to offer accessible clinical services for treatment and follow-up [24].
Differential diagnosis
The differential diagnosis for PPD includes the baby blues, postpartum psychosis, bipolar disorder, anxiety disorders and ‘organic’ or physical causes for depression-like symptoms.
In contrast to PPD, the baby blues (or postpartum blues) are actually a normative postpartum experience, occurring in approximately 50–80% of new mothers. Related to sudden hormone shifts, role change and lack of sleep, the baby blues occur within the first 2 weeks postpartum. Understandably, the new mother may not feel like her normal self, and may experience symptoms such as irritability, tiredness and excessive tearfulness. However, these mothers are not suicidal, do not feel worthless and are able to care for the baby appropriately [14]. As hormones normalize and sleep improves, these symptoms spontaneously disappear in the baby blues [25] within days to 1 week, rather than lasting for more than 2 weeks, as in cases of PPD. Thus, the treatment for baby blues is reassurance and education – regarding both the importance of sleep, and how to identify symptoms of a true PPD. If symptoms persist or worsen to the level of a PPD, further evaluation and treatment are indicated. A history of depression and possibly elevated levels of estriol are predictors of the baby blues [26]. Clinicians must be careful not to mistake normal emotional changes in the postpartum (crying, change in sleep, libido or appetite alone) for true depression.
In addition, care must be taken to rule out postpartum psychosis (PPP). PPP is relatively rare, occurring in one to two per thousand births. Women with PPP experience a more severe and rapid onset of symptoms, which may include not only psychosis but also mania, depression and confusion [27]. Mothers may experience bizarre behavior, hallucinations, confusion, rapid mood swings, agitation, abnormal sleep and disorganized thinking [27]. The majority of mothers who experience PPP have bipolar disorder. PPP often appears earlier than PPD, usually within the first couple of weeks after childbirth. Although the great majority of mothers with PPD can be treated effectively as outpatients, the rapidly evolving nature of PPP and the risk of infanticide and suicide, often require psychiatric hospitalization. For a more detailed review of PPP, the reader is referred to Friedman
It is important to rule out bipolar disorder in assessing PPD. In one sample (n = 30), over two-thirds of women with bipolar disorder experienced a postpartum mood episode, the vast majority of which were depressed episodes [29]. In bipolar disorder, discrete episodes of depression and mania or hypomania occur. Although a patient with bipolar disorder may experience depression in the postpartum period, she would also have elevated or irritable mood symptoms at other times in her life. In some cases, the first period of abnormal mood may be in the postpartum. Since antidepressant medication can lead to a manic episode in a bipolar woman, it is important to inquire about any past history of elevated mood. Clinicians should screen for family history of bipolar disorder. Collateral information is often quite useful in ruling in or out bipolar disorder. For a review of bipolar disorder, the interested reader is referred to Friedman
Anxiety may occur as part of a generalized anxiety disorder with many worries; due to PTSD (e.g., related to a traumatic delivery), as a postpartum exacerbation of a panic disorder, or owing to an obsessive–compulsive disorder (OCD). Women with OCD may experience both obsessive thoughts and compulsive actions. Ego-dystonic thoughts or worries may include fears that harm will befall the baby or themselves. These are fears rather than plans. As a result of their fear, mothers may limit contact with the baby. In one sample (n = 37), 57% of the mothers reported obsessional thoughts. The median number of such thoughts was seven and some were aggressive thoughts [32].
Finally, organic and medical causes of depression merit consideration. In addition to those medical illnesses that may cause fatigue (which simulates depression), a hypoactive delirium (soon after delivery) may appear similar to depression. However, careful scrutiny will demonstrate that a hypoactive delirium from brain dysfunction is a disorder of consciousness rather than of mood. Therefore, assessing memory and attention is important. Anemia and thyroid disease should also be ruled out.
Treatment
Multiple factors must be considered in the treatment of PPD. The consequences of untreated depression must be balanced against the potential effects of treatments for both the mother and the infant. The ultimate goal of treatment of PPD is to decrease symptoms of depression and consequences of depression, so that infant exposure to both maternal depression and to psychotropic medications is minimized [33].
The clinician should determine the depressed mother's level of distress and the functional impact on her parenting. Does she require inpatient psychiatric treatment (if she is suicidal or homicidal or quite severely impaired) or can she be managed on an outpatient basis with close follow-up? Consideration should also be given to what support system she has in place and her willingness to ask others for needed assistance. Arranging for home visits by a visiting nurse may be very helpful.
The treatment of depression in the postpartum is similar to the treatment of depression in general. Special considerations include the impact of depression on the infant, as well as the impact of medication on lactation. The specific medication choice should be based on the treatment response history and the severity of symptoms. The clinician should also consider side effects of medication that may cause difficulty with parental obligations, such as sleepiness or dizziness.
The treater should be aware that some mothers may self-medicate with St John's Wort, believing that because it is natural, it is safe. However, to date, there is insufficient evidence regarding its safety [34].
Lactation & medication treatment
Breast milk provides ideal nutrition for the infant. According to the CDC, over half of new mothers in the USA breastfeed during the early postpartum [103]. Although most antidepressants pass readily into breast milk, most infants exposed to antidepressants through lactation do not experience adverse events. Some mothers will supplement their breast milk with bottle-feeding. Others may choose to bottle-feed at night, particularly if a partner can be helpful with infant feedings and if sleep deprivation is a concern.
We suggest that if a mother has been doing well on an antidepressant medication throughout her pregnancy, the same medication should usually be continued in the postpartum; the infant's drug exposure
The infant has immature renal and hepatic function, especially in first 2 months of life, as well as an immature blood–brain barrier. This increases passage of medications into the CNS. Particularly vulnerable are those with neonatal disease, metabolic disorders and premature infants [35]. We recommend that the clinician involve the pediatrician and establish the infant's baseline of behavior, sleep, feeding and alertness prior to initiating maternal medication. Although routine clinical monitoring of the infant should be carried out, medication levels in infant serum are generally not performed.
Much of our knowledge regarding medications is based on case reports owing to obvious ethical issues surrounding prospective studies. In some case reports, mothers have not only been exposed to an antidepressant but also to other medications or drugs of abuse. No increased risk of infant problems with maternal tricyclic antidepressant (TCA) use has been found [37]; however, these agents are rarely used today because of their unfavorable side-effect profile. While there have been case reports of adverse events among infants exposed to selective serotonin reuptake inhibitors through lactation (such as irritability, jitteriness, excessive crying, feeding and sleep disturbances), overall, no significant risk of serious adverse effects have been reported over years of use [33,36].
A pooled analysis of 57 studies regarding antidepressants in lactation found low plasma concentrations among infants exposed to paroxetine, sertraline or nortriptyline; however, those exposed to fluoxetine were at higher risk of elevated levels (defined as greater than 10% of the maternal level; <5 ng/l for all but fluoxetine), especially with prenatal exposure [35,38]. Fluoxetine has a metabolite, norfluoxetine and a long half-life, which both lead to its accumulation. Fluoxetine has over 190 reported infant exposures in lactation [36] with no adverse events in 180 cases, and nine reports of colic-like symptoms. Some babies receiving fluoxetine in breast milk have been noted to have increased crying, decreased sleep, gastrointestinal distress and irritability [35]. The side effects are usually transient.
Sertraline has reports of over 90 exposures without infant adverse events [36] and is considered to be relatively safe [35]. Citalopram has reports of over 102 exposures without adverse events [35,36], and paroxetine has reports of over 70 exposures without adverse events [35].
Peak breast milk concentrations of sertraline and fluoxetine occur 8–9 h after maternal use; therefore, some women ‘pump and dump’ a feed. A double-blind comparative trial (n = 109) of nortriptyline versus sertraline found no difference in maternal remission rates [39]. Among breastfed infants, serum levels were low or below the level of quantification for each medication [39].
Paroxetine and sertraline appear to be the most favorable selective serotonin-reuptake inhibitors during lactation with regard to minimizing infant levels [40]. However, if the mother was already taking fluoxetine and responding well, in the author' opinion, it should be continued rather than exposing the infant to another agent, which may or may not be effective for treatment of the mother's depression.
It is difficult to make recommendations regarding the newest antidepressants owing to the limited data available. Should a mother have a history of a past response only to buproprion, for example, she should be restarted on that agent. Similarly, if a mother was treated successfully in pregnancy with venlafaxine (and thus the infant was already exposed), it makes sense to continue venlafaxine during nursing.
Electroconvulsive therapy (ECT) is useful in some cases of severe maternal depression. ECT has a more rapid effect than antidepressants and it does not lead to the presence of antidepressant medication being in breast milk.
Since the treatment benefits of antidepressants usually take several weeks to occur, the as-needed use of lorazepam (a benzodiazepine with a short half-life) may be useful to treat acute anxiety. Lorazepam has not been associated with adverse events among nursing infants [41]. This is important because suicide is associated with psychic anxiety among depressed patients [42]. However, long-acting benzodiazepines may build up in infants due to the immature liver [36]. For discussion of evaluation and treatment of Mrs Jones from the case study, refer to
Other treatments
Nonpharmacological treatment options should be considered as a component of the treatment plan for PPD. The elements of treatment should include: psychoeducation, engagement of social support and psychotherapy when possible.
Since many mothers wish to avoid antidepressants entirely, careful explanation of alternative or adjunctive therapies is important. Some mothers may elect to undergo psychotherapy and add medications only if their depression worsens. Other therapies, such as light therapy, critically timed sleep deprivation, omega-3 fatty acid and estrogen therapy [10,43–45] are being investigated. Music therapy [46], creative arts therapy [47], exercise therapy [48] and stress management skills may also be helpful. Finally, massage and acupuncture are under investigation [49].
Mothers who participate in psychotherapy or psychosocial treatment are less likely to remain depressed [50]. Interpersonal psychotherapy (IPT) shows particular promise in PPD [51–52]. IPT was developed to focus on role transitions and interpersonal disputes common to pregnancy and new motherhood. Clinicians should keep in mind that some mothers will have difficulty attending appointments regularly owing to childcare issues. IPT may be carried out during a child's pediatric visits. Mothers often put their children first and thus may not seek their own mental healthcare [53].
Women with social or professional support are less likely to remain depressed [54]. Even telephone support may, among other benefits, decrease symptoms [55]. Psychoeducational sessions that occur with a partner, rather than with the mother alone, have been shown to help more with depression [56]. Marital therapy can be useful in some cases [57]. In addition, home-based psychological interventions including psychodynamic, cognitive–behavioral and nondirective counseling have been shown to decrease maternal depression [58]. Finally, some treatment centers and even community centers offer group therapy. Various group therapies have proven effective for PPD including support and psychoeducation groups [59–61].
Risks of not treating postpartum depression
Impaired bonding, insecure maternal–baby attachment, decreased cognitive skills, language development and potentially long-term child behavioral problems have all been associated with maternal depression [62–66]. Depressed mothers may not respond their infant's cues. Furthermore, depression may contribute to partner relationship issues. Mothers have much to attend to; not only physical needs, but also the cognitive, emotional and social development of their infants. A depressed mother may neglect her own care or the care of the infant [62]. Exposure to antidepressants through lactation does not affect infant weight negatively; however, exposure to maternal depression does [67].
A growing amount of literature supports the importance of bonding within the mother–infant dyad. A depressed mother's behaviors may include withdrawal, intrusiveness and hostility. She may perceive that infant care is excessively difficult. Her infant may experience emotional/social impairment, growth impairment, temperamental problems and decreased intellect. Negative and disengaged parenting behavior, with less positive interaction, less play, less vocal interaction, a negative perception of infant, and appetite or sleep problems in infants may occur in maternal depression [37].
Evaluation and treatment: case discussion
Gather further information from both the mother and her partner regarding her symptoms and her history.
Rule out bipolar disorder and past episodes of hypomania.
Explore for anxiety symptoms.
Inquire about whether she has had any suicidal thoughts or thoughts of harming the baby; this will help determine whether she needs an urgent evaluation for inpatient psychiatric care, or whether she can be managed as an outpatient.
Encourage her to seek out her social supports and let them know that she is feeling overwhelmed. Encourage her to take some time for herself and to do some activities that she enjoys.
Educate her and her partner regarding treatment options for postpartum depression as well as warning signs of a recurrence.
Document a careful discussion of the risks of untreated depression, and the risks and benefits of medications in lactation and psychotherapy. Consult the baby's pediatrician regarding medication decisions. If she is willing to start an antidepressant medication, one of the selective serotonin-reuptake inhibitors discussed in the text would be a reasonable choice.
Provide close follow-up, and consider arranging visiting nurses to the home.
Mrs Jones fulfils the critera for a diagnosis of postpartum depression.
A meta-analysis of seven studies found that in their early childhood, children of depressed mothers had a lower likelihood of secure attachment, and a higher likelihood of avoidant or disorganized attachment [68]. Of 132 infants of mothers who had depression at 3 months, the children as 11-year-olds had an elevated rate of decreased IQ, attention problems and special educational needs [69–70]. Maternal depression is a risk factor for child psychopathology (in the preschool and school age years) [71–73]. Exposure to maternal depression has been correlated with depression and disruptive disorders among adolescents [73]. Childhood IQ was negatively associated with the duration of depression in pregnancy and language development with the number of depressive episodes after delivery [74] rather than being related to the antenatal exposure to fluoxetine and tricyclic antidepressants.
Finally, suicide and infanticide are risks of PPD that merit serious concern. In fact, suicide is a leading cause of postpartum death, accounting for up to a fifth of the deaths [75]. Mothers who are suicidal are also at risk for infanticide [76]. Such women may plan an ‘extended suicide’ with the altruistic belief that their infant is better off in heaven with their mother than remaining alive and motherless, in what the depressed mother sees as a cruel uncaring world [77]. Some mothers experience recurrent thoughts of harming their infant, which they are hesitant to share with family. Nonetheless, some may realize that they need help [78]. One study of depressed mothers with children up to 3 years of age found that 41% of the mothers had experienced thoughts of harming their child [79]. Physicians should sensitively inquire about thoughts of harming the infant, and also about the suicidal mother's plans for the children if she takes her own life [80]. Emergency hospitalization should be pursued when appropriate.
Executive summary
A total of 10–20% of mothers will experience postpartum depression (PPD). Stigma may prevent them from coming forward for treatment.
Risk of PPD is elevated in women with a personal history of depression (particularly of PPD), depression or anxiety in pregnancy, family history of depression, stressful life events, lack of support and sleep deprivation.
Symptoms of PPD are similar to symptoms of major depression at other times.
Anxiety commonly occurs comorbidly.
The Edinburgh Postnatal Depression Scale is a useful quick screening scale.
‘Baby blue' is a normative experience, which resolves itself spontaneously within 2 weeks postpartum.
Postpartum psychosis is a rare disorder in which a woman is out of touch with reality and needs emergency treatment.
Bipolar disorder should be ruled out.
Physical illness, such as anemia, thyroid problems and hypoactive delirium, should be eliminated as causes of PPD-type symptoms.
Goals of treatment include minimizing exposure of the infant to both maternal depression and antidepressant medications.
A careful medication history should be taken.
Medications that cause maternal sedation and dizziness should be avoided.
Although antidepressants do pass into breast milk, they rarely cause adverse effects.
Some mothers may choose not to breastfeed, after discussion of risks and benefits.
Communication with the significant other and the paediatrician is encouraged.
Psychotherapy as well as other therapies show some promise in PPD.
Risks include poor maternal–infant bonding, slowed infant development, infanticide, child neglect and suicide.
Women may be screened in obstetrics, pediatrics and by visiting nurses.
Mothers with a history of PPD should be educated and followed to reduce risk of recurrence.
Collection of a larger sample of mother–infant pairs will help to provide more authoritative recommendations about medication use in lactation.
The concept of behavioral teratogenesis should be further explored.
Alternative therapies will be further investigated.
Prevention
Primary prevention of PPD involves identifying the at-risk group and secondary prevention requires the early identification of PPD. Community health nurses and pediatricians can play an important role in prevention. Obstetricians, family practitioners and pediatricians all have occasion to see new mothers in their practice. Mothers can be easily screened using the EPDS or with sensitive questioning. Brochures explaining depression and treatment options can be left in physician' waiting rooms. Mental health services should be readily available to women with PPD. Mothers with a personal history of PPD should be educated and monitored closely to prevent relapse. Providing mentors to at-risk mothers can be useful [81]. Larger trials are needed in order to categorically recommend antidepressant treatment prophylactically in the postpartum [82]. However, in individual cases they are certainly warranted.
Future perspective
Further studies about PPD are needed regarding the issues of including prevention, identification and treatment. As time progresses, more mothers and infants will be exposed to antidepressants during pregnancy and lactation. Naturalistic data will therefore be increased and more authoritative guidelines established. Further exploration is needed for long-term ‘behavioral teratogenesi’. This concept addresses whether medication given during pregnancy or lactation has behavioral consequences (such as a higher risk of attention-deficit hyperactivity disorder or other disorders) for the infant in later childhood. In addition, evidence may accumulate for the efficacy of psychotherapy, and that of other alternative treatments, such as estrogen or light therapy.
