Perinatal depression

Perinatal depression is defined in various ways. At one end of the spectrum, perinatal depression is defined symptomatically as exceeding a threshold on a screening measure, such as the Edinburgh Postnatal Depression Scale . For example, pregnant and postpartum women are often asked to complete this 10-item questionnaire that asks about depression and anxiety-related symptoms over the past 7 days. Scores can range from 0 to 30, and women whose scores exceed a specified threshold (usually between ≥10 to 13) are often referred for further evaluation or treatment for a perinatal depressive or anxiety disorder. More rigorous assessment occurs in the context of a clinical interview, such as the Structured Clinical Interview for DSM-IV . Of course, the clinician can assess for major depression or a range of anxiety disorders without the aid of a structured interview protocol. What is important is a determination of what a pregnant or postpartum woman needs, which might range from simple advice and reassurance, psychological counselling, antidepressant medication, to some combination of these treatments. Treatment might also include complementary and alternative treatments, such as exercise, bright light therapy, and yoga. Treatment is usually on an outpatient basis, although psychiatric emergencies, such as suicidal thoughts with a plan and intent or postpartum psychosis, warrant hospitalisation. Intervention options and settings are addressed in several chapters in this issue.

Prevalence estimates for perinatal depressive disorders vary considerably depending upon the definition of the disorder and the period over which prevalence is determined . For example, the most recent quantitative review of the prevalence of perinatal depression found that the period prevalence (over the first 3 months postpartum) of depression broadly defined was 19.2%, but the prevalence of the more narrowly defined (and more severe) major depression was 7.2% . The prevalence of depression across pregnancy was estimated at 18.4% (12.7% major depression). At about the same time, Bennett et al. conducted a separate meta-analysis of depression in pregnancy based on 21 studies and 19,284 participants, and reported prevalence rates of 7.4% (first trimester), 12.8% (second trimester), and 12.0% (third trimester).

Fisher et al. reported separately on prevalence of perinatal depression in low- and middle-income countries. The mean prevalence of postpartum depression in these countries was estimated to be 19.8% (15.6% during pregnancy). Since the publication of these reviews, several large-scale prevalence studies have been conducted in Europe. For example, investigators reported a 9.6% 1-year prevalence of major and minor depression in Pisa, Italy , which would be relatively low compared with the report of Gavin et al. A large study in Barcelona, however, yielded a 6-week prevalence of 9.2% for major and minor depression and dysthymia . Although these prevalence estimates are quite similar, they cover vastly different postpartum periods. All of these reviews and empirical studies conclude that depression is common during pregnancy and after delivery in developing and developed countries.

Postpartum anxiety disorders

A wide range of anxiety disorders are prevalent in the perinatal period. These include generalised anxiety, obsessive–compulsive, panic, and social anxiety disorders. In many cases, the severity and effect of anxiety symptoms (e.g. worry, avoidance, and obsessions) do not rise to the level of an anxiety disorder diagnosis; nevertheless, they cause at least mild-to-moderate levels of distress and impairment. Similar to the case for perinatal depression, anxiety is detected using self-report measures, such as the Beck Anxiety Inventory .

Rates of anxiety disorders in pregnancy and the postpartum period range widely. For example, in a Nigerian study, Adewuya et al. reported that 10.5% of pregnant women met DSM-IV criteria for generalised anxiety disorder (GAD). In a US study, however, only 1.9% of pregnant women met criteria for GAD . In the first 6 months after delivery, rates of GAD, based on several studies, ranged from 6.1% to 7.7%. Rates of obsessive–compulsive disorder (OCD) during pregnancy ranged from 1.2% to 5.2%, and the rate was about 4.0% at 6 months postpartum. With panic disorder, rates ranged from 1.4% to 9.1% during pregnancy and 0.5% to 2.9% between 6 and 10 weeks postpartum. Finally, rates for social anxiety disorder in pregnancy ranged from 2.0% to 6.4%, and in the early postpartum period prevalence rates ranged from 0.2% to 6.5% . What these prevalence rates suggest is that anxiety ranging from excessive worry to panic attacks is prevalent during pregnancy and the postpartum period. Given the deleterious effects of significant maternal anxiety on the developing fetus, and the way in which it complicates early parenting, the obstetrician should be as alert for anxiety symptoms as for depressive symptoms in the perinatal period.

Anxiety disorders often run a chronic course, and many women will enter pregnancy with an anxiety disorder. As with perinatal depression, the major risk factor for perinatal anxiety disorders is a history of anxiety disorders. The intensity and the degree of impairment associated with any particular anxiety disorder may wax and wane over the course of pregnancy and the postpartum period. Beyond a prior history of an anxiety disorder or depression, research on risk factors for perinatal anxiety disorders is mixed and inconclusive .

In a large-scale postpartum screening study with follow-up diagnostic evaluations , 66% of women with major depression in the postpartum period had comorbid anxiety disorders. This observation may explain the reason for the relatively small group of publications on anxiety disorders across childbearing. Most adults with mood disorders experienced significant anxiety symptoms or an anxiety disorder in childhood or adolescence; however, the same was not true of postpartum women, who developed a recurrent depressive disorder superimposed on an anxiety disorder.

Postpartum psychosis: bipolar disorder

The obstetrician–gynaecologist may be the first healthcare professional that a woman with postpartum psychosis, or more likely her distressed family, contacts. The incidence of postpartum psychosis is one or two per 1000 births . The symptom patterns in women with postpartum psychosis have consistently been reported to differ from women with psychosis not related to childbearing . Postpartum psychosis presents rapidly after birth, with mood fluctuation, confusion, and marked cognitive impairment suggestive of delirium, bizarre behaviour, insomnia, visual and auditory hallucinations, and unusual (i.e. tactile and olfactory) hallucinations . Maternal psychotic symptoms can include the altruistic homicide delusion, in which the mother believes she is saving her baby from a fate worse than death by killing it. Andrea Yates believed that the only way to release her children from condemnation to Hell was to drown them. Fortunately, such occurrences are rare, but the potentially preventable tragedy of such cases compels rapid identification and treatment . Mothers who develop postpartum psychosis usually have a diagnosis consistent with bipolar depression, mania, or mixed state, with psychotic features . Women are vulnerable to affective psychoses after birth more than at any other time during their lives .

Women with bipolar disorder are particularly vulnerable to developing psychotic and non-psychotic episodes in the postpartum period . For women with bipolar disorder, the first 30 days after birth carries a relative risk of 23.3 of having a psychiatric admission compared with women at any other period after birth, and 30–60 days after birth incurs a risk of 6.3 . The causes of post-birth decompensation in women with bipolar disorder include genetic factors and massive gonadal steroid withdrawal after birth in women who are neurobiologically sensitive to hormonal change . Sleep deprivation and interference with circadian rhythms during late pregnancy, labour, and frequent newborn feeding promote mood destabilization . Constant care for the newborn is a major stressor, particularly where psychosocial and physical support is lacking. The psychosocial consequences of psychotic and mood episodes during childbearing include an increased risk for substance use, smoking, and high-risk behaviours that increase exposure to sexually transmitted and other diseases, violence, victimisation, poor nutrition, non-compliance with medical care, and alienation from social support systems. The United Kingdom Confidential Enquiry into All Maternal Deaths (1997–1999) revealed that 10% of postpartum deaths resulted from suicide, and 12% resulted from psychiatric causes (e.g. accidental overdose, deaths from medical conditions due to substance use, and violent deaths).

The effect of pregnancy on the course of bipolar disorder has not been elucidated. Published reports differ in their conclusions, which are likely to be based on populations with dissimilar characteristics. Some researchers have described a benign course or mood stabilisation and improvement for women with bipolar disorder during gestation; that is, a protective effect , whereas others have described pregnancy as having no effect or increased risk for episodes .

Viguera et al. retrospectively compared recurrence rates and survival functions for women with bipolar disorder in two groups receiving tertiary psychiatric care treatment. One group discontinued lithium during childbearing, and the other discontinued lithium but was non-childbearing. Rates of recurrence over the first 40 weeks after lithium discontinuation were similar for pregnant (52%) and non-pregnant women (58%); however, rates for both were comparatively much lower the year before treatment discontinuation (21%).

One of the few screening measures for bipolar disorder, and the most commonly used, is the Mood Disorders Questionnaire (MDQ) . This questionnaire assesses lifetime history of mania and hypomania with 13 yes/no symptoms reflecting DSM-IV criteria, a yes/no question about symptom occurrence during the same time period, and a designation of the resulting problem level. A positive screen requires seven or more symptoms in the same period that caused moderate or serious problems. In the only published screening study for postpartum BD, Sharma and Xie gave women with a diagnosis of MDD ( n = 68) or bipolar disorder ( n = 57) the MDQ at 2–4 weeks after delivery. The MDQ yielded a sensitivity of 75% and specificity of 87% with the original scoring criteria.

The recognition and clinical diagnosis of bipolar disorder is by far the most important prerequisite for adequate treatment . Although many women with bipolar disorder receive treatment for depression or comorbid mental disorders (such as substance abuse or anxiety disorders), lack of recognition of the underlying bipolarity results in few receiving appropriate treatment . Half of women who have ‘treatment resistant’ postpartum depression actually suffer from undiagnosed bipolar disorder . Antidepressant drug monotherapy for bipolar disorder may precipitate rapid cycling, induce mania, or increase treatment resistance . Studies are under way to obtain more information about the usefulness of pairing the Edinburgh Postnatal Depression Scale with the MDQ to address the differential diagnoses of these major perinatal mood disorders .

Differentiating postpartum psychosis from obsessions and compulsions is often challenging, but is crucial for appropriate management . Obsessions include intrusive thoughts and images, which are in the ‘mind’s eye’, and differ from hallucinations, which appear as if in the real world. Common examples are ‘what if I put the baby in the microwave?’ ‘The baby’s head could slip under water in the bath and I might drown her.’ ‘What if I throw the baby over the banister?’ These are also symptoms of OCD, which is classified within anxiety disorders, and also other mental illnesses, which commonly co-occur. In women with postpartum onset major depression, 57% reported obsessional thoughts of harm to their baby, and most had checking compulsions (i.e. highly repetitive night-time checking to make sure the baby is breathing) . In new mothers, checking behaviour might be considered appropriate because of the mother’s concern for the baby’s wellbeing; however, in OCD, it has intensified to a degree that compromises the mother’s ability to provide care .

Exploration for hallucinations in all senses (e.g. visual, auditory, olfactory, gustatory, tactile) and delusions (i.e. the baby is the devil, another person lives in me) is also critical to rule out a psychotic process. Postpartum women with obsessions are distressed by the thoughts and usually do not reveal them without sensitive questioning. These mothers recognise them as bizarre and intrusive, which is another distinction between obsessions and psychosis. The compulsions may not manifest as an active ritual but may involve extreme avoidance of the infant because of fear of actually doing harm.

Women with OCD without psychosis or severe personality disorder have no increased risk of aggressive harm to their infants .

Postpartum blues

In the first week or so after delivery, it is common for women to experience a variety of physical discomforts and symptoms that have been characterised as the ‘postpartum blues’ or ‘baby blues’. Prevalence estimates for the blues range from 26% to 84% , reflecting more or less stringent criteria. Unfortunately, no established criteria for the blues exist, although O’Hara et al. adapted the work of Handley et al. and proposed diagnostic criteria that led to their determination of a 26% prevalence rate. Blues symptoms included dysphoric mood, crying, mood lability, anxiety, insomnia, loss of appetite, and irritability. Participants met Handley Blues criteria if they were rated as having four of seven symptoms of at least mild severity (3 on 6-point scale). A more recent study of the factor structure of the blues during the first week postpartum found that symptoms such as ‘up and down mood’, ‘mentally tense’, ‘overemotional’, ‘oversensitive’, ‘brooding’, ‘low spirited’, ‘tearful’, and ‘irritable’ all loaded above 0.60 on postpartum negative affect . Interestingly, a distinct positive affect factor was reported in the first week postpartum, which included positive emotions such as ‘inspired’, ‘confident’, ‘lively’, ‘enthusiastic’, ‘interested’, ‘alert’, ‘determined’, and ‘calm’, which loaded above 0.60 on postpartum positive affect . The findings that two distinct mood factors exist — negative and positive affect — in the early postpartum period suggest that the structure of mood in the postpartum period is the same as the structure of mood at other times in a woman’s life. These findings are important because they suggest that blues symptoms are on the same dimension as mood symptoms at other times in a woman’s life. Good evidence shows that mood symptoms worsen in the first week postpartum and then gradually improve . Negative mood symptoms have been found to peak between day 3 and day 5 postpartum, perhaps reflecting hormonal readjustments after delivery . It is generally the case that blues symptoms will largely diminish by days 10 to 12 postpartum, except in cases in which they evolve into an episode of postpartum depression. Postpartum depressions are generally independent of the blues, but the postpartum blues do represent a risk factor for postpartum depression .