Recognizing and Treating Peripartum Depression

Symptoms of depression are found in 18.4% of all pregnant women (95% confidence interval [CI]: 14.3 to 23.3) and 19.2% of all mothers during the first 3 months post partum (95% CI: 10.7 to 31.9).

Severe depression requiring treatment (major depression) with an overall prevalence similar to that in the general female population afflicts (1)

• 12.7% of women during pregnancy (95% CI: 7.1 to 20.4)
• 7.1% of mothers post partum (95% CI: 4.1 to 11.7).

In Great Britain the rate of suicide in the context of depressive disease among pregnant women and mothers within 6 months after giving birth is 0.27 per 100 000 (2). Only 20% to 40% of depressed women seek professional advice (e1, e2).

Depressive diseases are not only a leading cause of illness in women of child-bearing age worldwide (e3), they are the most frequent psychiatric affliction before and after birth. The consequences are serious and are not limited to the women themselves.

This article reviews recent studies on the prevalence, etiology, risk factors, and treatment of peripartum depression, and provides addresses for further information. Relevant publications up to December 2011 were identified by searching on Medline (search terms “peri/pre/post,” “partum/partal/natal”, “maternal/motherhood/pregnancy,” and “depression/affective disorder”), in the Cochrane Library, and in German library catalogs online. Cited publications were also analyzed.

Peripartum depression: consequences for mother and child

Antenatal depression is associated with an elevated risk of premature birth (relative risk [RR] = 1.13), lower birth weight (RR = 1.18), and delayed intrauterine growth (RR = 1.03) (3). The possible causes include dysregulation of the maternal–fetal hypothalamic–pituitary–adrenal axis and a disordered intrauterine milieu owing to fluctuations in arterial blood flow (4). In their review, Kinsella and Monk (4) report lower variability in heart rate in the fetuses of stressed pregnant women. The fetuses of women with depression display a higher baseline heart rate, prolonged reaction time and pulse recovery, and increased motor activity. The affected women themselves show inadequate weight gain, less frequent attendance for prenatal examinations, and increased substance abuse (e4). Furthermore, Kozhimannil et al. (e5) found a higher frequency of depressive symptoms in women with (gestational) diabetes: this group had a 1.85-fold risk of peripartum depression, corresponding to an increase of almost 7% in absolute risk.

The behavior of depressed mothers of babies aged up to 6 months is characterized by reduced verbal and visual communication. The children more frequently display sleep and breastfeeding problems, avoidance behavior (aversion of gaze, turning the body away), decreased affect regulation, feeding disturbances, and failure to thrive (5). In the long term, the children of mothers with peripartum depression show insecure-avoidant attachment, and reduced cognitive, emotional, verbal, and social skills can be observed right up to puberty (6). Pawlby et al. (7) found that 16-year-olds are at a fourfold risk of themselves developing an affective disease.

Symptoms

Neither DSM-IV nor ICD-10 has a specific classification for peripartum depression. It is coded as F32.xx or F33.xx (8), and the time of occurrence can be specified under O99.3 (Figure). Postpartum depression must be distinguished from the “baby blues” that afflicts 50% to 80% of new mothers (e6) and from the rarely occurring postpartum psychosis (PP), which is found in 0.1% to 0.2% of cases and usually requires immediate (inpatient) treatment because of the danger for both mother and child (e7). PP is characterized by delusions, audible thoughts, thought deprivation, hearing voices, and other hallucinations, usually within the first 2 weeks after giving birth. However, so-called negative symptoms such as conspicuous apathy or blunted or inappropriate affect may also occur. Nosologically, PP seems often to be a manifestation of bipolar disorder: 25% to 50% of new mothers with a history of bipolar illness suffer from PP (e7).

Figure

Diagnostic algorithm for unipolar depression according to the German S3 guideline

Enlarge All figures

Depressive symptoms in the peripartum period are often strongly influenced by concern about the child and the demands of motherhood. The women suffer from fear of failure and feelings of inadequacy. They experience themselves as “bad mothers” who can’t even manage to meet the needs of their child. They often report that the child is “difficult and demanding.” Avoidance behavior on the part of the child and any existing problems with breastfeeding are interpreted as confirmation of their own failure, reinforcing the vicious circle and their ever-increasing exhaustion. Because of the personal and societal expectations of undiluted joy, the taboo against depression is even greater than at other times. The women are afraid to express their negative feelings towards their child and their perceived failure as mothers.

The affected women are particularly troubled by obsessive thoughts or impulses to harm the child. In the study by Chandra et al. (e8), 60.7% of severely depressed women, compared with 27.6% of those with psychoses or bipolar disorder, reported infanticidal thoughts; corresponding behavior was shown primarily by delusional patients. Hornstein et al. (9), however, are of the opinion that the risk of children being harmed by depressed mothers is underestimated. Infanticide is carried out from altruistic motives or because of fear of separation from the child in the context of extended suicide or in the presence of concurrent maternal attachment disorder. Infants in the first year of life are at the highest risk of infanticide (e9).

Overall, suicide and attempted suicide during pregnancy and lactation are rare (10)—the extensive network of support, the closer contact with the healthcare system, and concern for the (unborn) child seem to exert protective effects. Nevertheless, women with psychiatric illness are particularly at risk of suicide in the postpartum period: Appleby et al. (11) found a 72-fold suicide risk during the child’s first year of life among women who received postpartum inpatient psychiatric treatment. A conspicuous feature is the high rate of so-called “hard”, more frequently lethal forms of suicide attempt, e.g., hanging and jumping from a great height, which are otherwise not usually adopted by women (2, 10).

What are the predisposing factors?

Beck’s meta-analysis identified 13 significant predictors for postpartum depression, among them 10 with at least moderate effect strength (Table) (12). The remaining three factors, marital status, socioeconomic status, and unplanned/unwanted pregnancy, displayed low effect strength. Overall, the risk of postpartum depression seems to be increased in the presence of psychopathological problems, whether in the past or during pregnancy, or lack of support from the woman’s partner or her wider social environment. A study by Ludermir et al. underlines the benefit of an intact relationship with the partner (13). They found that after adjustment for the above-mentioned risk factors, including antenatal psychiatric illness, mental abuse by the partner increased the likelihood of postpartum depression 1.6-fold, raising the absolute risk by 6%. Boyce and Hickey discuss the role played by the lack of resilience, which if present could moderate both the perceived support from the woman’s partner and the perceived stress from the child (e10).

Table

Risk factors for peripartum depression in descending order of effect strength (mean effect strength at least r ≥ 0.3 in meta-analyses)

Enlarge All figures

Moreover, Bloch et al. found an association between postpartum depression and hormone-induced mood disturbances such as premenstrual dysphoric disorder (PMDD) or mood swings while taking oral contraceptives (e11). They assumed that a subgroup of women with postpartum depression is characterized by elevated vulnerability to hormonal changes, with particular regard to estrogen and progesterone. Furthermore, a recent study by Skrundz et al. showed a correlation between low plasma oxytocin concentration in the second trimester and suspicion of postpartum depression on the Edinburgh Postnatal Depression Scale two weeks after delivery (e12).

The risk factors for depressive illness in pregnancy have not been studied for long, but on the basis of the data published so far they differ hardly at all from the known risk factors for postpartum depression. The review by Lancaster et al. identified eight risk factors with moderate to high effect strength at least on bivariate analysis (Table) (14). Routine daily stress, socioeconomic status, unemployment, abuse of illegal substances, and obstetric history had no effect.

Psychosocial and psychotherapeutic prevention

Dennis and Creedy conducted a review of psychological and psychosocial measures designed to prevent postpartum depression and came to the conclusion that these measures are unpromising (15): the women in the intervention and control groups had the same risk (RR = 0.81) of becoming depressed after giving birth. The risk was reduced in mothers who received intensive support from a midwife after delivery (RR = 0.68). In general, the successful interventions were those that specifically targeted high-risk women (RR = 0.67), started post partum (RR = 0.76), and were carried out on an individual basis (RR = 0.76).

Milgrom et al. have designed a nine-unit workbook for use by mothers and fathers from around 26 weeks of gestation to 6 weeks post partum (16). The parents follow the workbook without external guidance, with the exception of a weekly telephone conversation with a psychologist (16). The focus is on reducing risk factors, increasing the parents’ competence and problem-solving skills, and facilitating treatment for existing symptoms. Preliminary RCT results show significantly lower levels of depression and anxiety in the intervention group, with moderate effect strengths.

Treatment of peripartum depression

Treatment for depression in the peripartum period is also based on psychoeducation, inclusion of relatives, psychopharmaceutical treatment, and psychotherapy. As might be expected, pregnant women and new mothers who want to breastfeed prefer psychotherapeutic interventions (e13, e14).

Despite the urgent need for studies on the effects of psychotherapeutic and psychosocial interventions in the antenatal period, up to 2007 only one published study had satisfied the requirements of a Cochrane review (17): Interpersonal psychotherapy (IPT) achieved risk reduction (RR = 0.46), but the sample was small (38 women).

In the postpartum period, all evaluated psychotherapeutic and psychosocial interventions, such as peer support, supportive therapy, cognitive behavioral therapy, IPT, and psychodynamic therapy, were significantly more effective than standard aftercare—at least for the first year post partum (18). IPT, as adapted for postpartum depression (e15), concentrates for example on the woman’s changing role and on the demands and expectations associated with the role of mother, as well as on interpersonal, familial conflicts. After delivery, however, interventions in the mother’s home and telephone- and internet-based support are much more practicable—also for reasons of time flexibility, respect for privacy, low degree of stigma, lack of socioeconomic barriers, and universal availability (18, 19).

Lengthy separation of mother and child should generally be avoided. Some psychiatric hospitals offer special mother-and-child units for inpatient care, but the supply has been estimated to meet no more than one-fifth of the demand (e16). An effective 6-week program for the inpatient treatment of mothers with psychiatric illness accompanied by their young children (up to 2 years old) has been designed by Hornstein et al. (e17). This program integrates a wide variety of therapeutic approaches and features interactional support of the mother–child relationship. Evaluation of the long-term outcome remains to be carried out.

Depending on the severity of the mother’s illness, low-threshold support measures to reinforce the parent–child relationship can be contemplated, e.g., baby massage or PEKiP programs, which promote early family formation through play and motor and sensory stimulation in a group setting (e18). Furthermore, one should always consider domestic help or childcare (covered by health insurance) to relieve the strain on the mother.

Pharmacotherapy during pregnancy and lactation

There is a lack of randomized trials on psychopharmacotherapy during pregnancy and lactation and those that have been published are marred by methodological limitations. Nevertheless, the data suffice for evaluation of the older tricyclic antidepressants (TCA) and the frequently prescribed selective serotonin reuptake inhibitors (SSRI) such as fluoxetine, paroxetine, sertraline, and citalopram (Box) (20, e19). All of these substances permeate the placenta or are detectable in breast milk and can lead to central nervous, gastrointestinal, and respiratory adjustment disorders in newborn children (e20). With regard to pregnancy, the embryo is particularly susceptible to toxins in the first trimester. Substances with low metabolism, high protein-binding capacity, and low interaction potential should generally be preferred (21).

Box

Antidepressive psychopharmacotherapy during pregnancy and lactation

Enlarge All figures

Up-to-date treatment recommendations in German can be found at www.embryotox.de, a site maintained by the Pharmacovigilance Center for Embryonal Toxicology at the Charité in Berlin in cooperation with the Department of Gynecological Psychosomatics at Bonn University Hospital. Personal advice can also be obtained from the Institute for Reproductive Toxicology in Ravensburg (www.reprotox.de). The German-language book “Psychopharmakotherapie in Schwangerschaft und Stillzeit” by Rohde and Schaefer (e21) is a comprehensive pharmacological reference with case studies.

Psychopharmacotherapy should generally be started with a single agent at the lowest possible dose. Treatment should be preceded by painstaking analysis of the benefits and risks, and the mother’s wishes with regard to breastfeeding should be respected. The mother should not be left to bear the main burden of decision, especially since her judgment will often be impaired by her illness. It is particularly likely that breastfeeding will have to be discontinued if dosages are high or multiple medications are prescribed.

Other options for prophylaxis and treatment

In the review by Chabrol and Callahan (23), progestogen, omega-3 fatty acids, and thyroxine showed no significant preventive effects when taken during pregnancy. One trial, however, found a preventive effect of daily calcium intake. To date there is no evidence that antenatal depression can be effectively treated by massage or acupuncture (e23).

Also post partum there is no evidence for significant effects of treatment with estrogen or thyroxine or the use of omega-3 fatty acids (23). There are early positive reports of the effect of light therapy, albeit in small study populations (21, e24).

Screening—or how does treatment reach the patient?

Although pregnancy and early motherhood are characterized by regular contact with the healthcare system (24), only 18% of pregnant women with a psychiatric illness receive a corresponding diagnosis (e25). Johanson et al. (e26) found that 12% of depressed pregnant woman and 26% of depressed new mothers were correctly identified. Marcus et al. (e27) found depressive symptoms in 20% of the pregnant women attending a gynecologist’s office, and only 13.8% of them were receiving treatment.

According to a web-based education program, both gynecologists and pediatricians are inadequately trained in the recognition of depression (e28) and are not viewed by new mothers as suitable persons to approach with psychiatric problems (e29). In the USA and Australia, this deficiency is being addressed by routine screening of pregnant women with the Edinburgh Postnatal Depression Scale (EPDS), a ten-question self-report questionnaire. A German version of the EDPS with information on the quality criteria can be found at www.marce-gesellschaft.de/materialien.html (e30). In Great Britain, consistent with the general recommendations of the German National Clinical Practice Guideline (S3) for Unipolar Depression (25), use of the two Whooley questions (e31) is recommended at the first prenatal consultation and again 4 to 6 weeks post partum (e32):

• During the past month, have you often been bothered by feeling down, depressed, or hopeless?
• During the past month, have you often been bothered by little interest or pleasure in doing things?

If the answer to both questions is “Yes,” clinical investigation of the formal diagnostic criteria is required (25).

Ideally, every pregnant woman attending a gynecologist’s office should be examined with regard to her psychiatric status or at least given a self-screening questionnaire, so that she can be referred to her family doctor or an appropriate specialist. Psychiatric outpatient departments can generally provide an appointment at short notice and usually also offer psychotherapeutic services. Low-threshold, usually rapid psychosocial support is readily available up to the child’s third year of life via information centers and pregnancy counseling centers. In Germany, help can often be provided by the local representatives of the German Alliance against Depression (Deutsche Bündnis gegen Depression; www.buendnis-depression.de). Further information in German can be found on the homepage of the self-help group Schatten und Licht e.V. (www.schatten-und-licht.de).

Conflict of interest statement

Dr. Hübner-Liebermann declares that no conflict of interest exists.

Dr Hausner has received reimbursement of costs for attending a congress or training courses from Astra Zeneca, Servier, Janssen-Cilag, GlaxoSmithKline, and Pfizer.

Dr. Wittmann has received consultancy fees from Bristol Myers Squibb; reimbursement of costs for congress participation and accommodation and fees for the preparation of scientific training courses from AstraZeneca, Servier, Wyeth, Lilly, Janssen-Cilag, GlaxoSmithKline, Pfizer, EISEI, and Lundbeck; and fees for performing commissioned clinical studies from Servier.

Manuscript received on 20 May 2011, revised version accepted on
19 March 2012.

Translated from the original German by David Roseveare.

Corresponding author
Dr. rer. medic. Bettina Hübner-Liebermann
Klinik und Poliklinik für Psychiatrie und Psychotherapie der
Universität Regensburg am Bezirksklinikum Regensburg
93053 Regensburg, Germany
bettina.huebner-liebermann@medbo.de

@For eReferences please refer to:
www.aerzteblatt-international.de/ref2412