Introduction

The lifetime prevalence of major depressive disorder (MDD) is about twice as high in women compared with men . About 18–19% of women suffer from perinatal depression . Untreated depression during pregnancy poses risks to the mother and baby, including obstetrical and neonatal complications . Depression during pregnancy is a major risk factor for postnatal depression . Postnatal depression has been associated with a broad negative effect on child development, including difficult infant and childhood temperament , attachment insecurity , and increased risk of developmental delay and lower IQ scores . A substantial rate of suicide occurs postpartum, as maternal suicide accounts for up to 20% of postnatal deaths in depressed women .

Despite the risks of untreated perinatal depression, women often discontinue antidepressant treatment during attempts to conceive or during pregnancy. Safety and risk profiles of antidepressant use during pregnancy and lactation are increasingly being studied; however, women often seek treatment other than standard medications during pregnancy or while breastfeeding. In a recent UK study, more than one-quarter of women reported use of complementary and alternative medicine therapies (CAM) during pregnancy . Therefore, an understanding of CAM therapies is important, as many women may seek CAM treatments for perinatal depression.

Complimentary and alternative medicine

Complementary and alternative medicine therapies refers to a diverse range of healthcare practices used for health promotion, disease prevention, and illness treatment that are not considered standard or established practices in western medicine. In general, women use CAM treatments more frequently than men, and are also more likely to suffer from disorders such as MDD and anxiety disorders, for which CAM treatments are commonly pursued . Despite the growing prevalence of CAM use, the number of adequately powered, well-designed, randomised clinical trials of CAM treatments is limited. As CAM therapies include a large number of diverse modalities that have varying amounts of study, we have reviewed the following seven CAM therapies based on their prevalence of use and the availability of randomised, placebo-controlled data, with considerations for their use in women with unipolar depression during the perinatal period: omega-3 fatty acids, folate, S-adenosyl-methionine, St John’s Wort, bright light therapy, exercise, massage, and acupuncture.

Complimentary and alternative medicine

Complementary and alternative medicine therapies refers to a diverse range of healthcare practices used for health promotion, disease prevention, and illness treatment that are not considered standard or established practices in western medicine. In general, women use CAM treatments more frequently than men, and are also more likely to suffer from disorders such as MDD and anxiety disorders, for which CAM treatments are commonly pursued . Despite the growing prevalence of CAM use, the number of adequately powered, well-designed, randomised clinical trials of CAM treatments is limited. As CAM therapies include a large number of diverse modalities that have varying amounts of study, we have reviewed the following seven CAM therapies based on their prevalence of use and the availability of randomised, placebo-controlled data, with considerations for their use in women with unipolar depression during the perinatal period: omega-3 fatty acids, folate, S-adenosyl-methionine, St John’s Wort, bright light therapy, exercise, massage, and acupuncture.

Omega-3 fatty acids

Omega-3 fatty acids are among the most commonly used CAM treatments in the USA . Omega-3 fatty acids are essential fatty acids with well-established health benefits and particular benefits for obstetrical outcomes and infant development . Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are two important omega-3 fatty acids found in fish. The American Heart Association recommends eating fish, especially fatty fish, at least twice a week. To optimise pregnancy outcomes and fetal health, consensus guidelines have recommended that pregnant women consume at least 200 mg of DHA per day . The UK government recommends that adults consume two or more portions of fish a week, which corresponds to about 450 mg EPA plus DHA daily . Recommendations for pregnant women include one or two portions of oily fish weekly or 200 mg DHA per day . The European Food Safety Authority recommends 250 mg EPA plus DHA daily for adults, with an additional 100–200 mg DHA a day during pregnancy . Despite increased demand for omega-3 fatty acids during pregnancy, dietary intake by perinatal women in the US and UK has been noted as deficient, with dietary intake during pregnancy even more diminished in the US after US Food and Drug Administration issuances of mercury advisories regarding fish intake during pregnancy.

Omega-3 fatty acids have received the most rigorous study to date in randomised-controlled trials (RCTs) for adjunctive treatment of MDD. Meta-analyses of RCT demonstrate a statistically significant antidepressant benefit of omega-3 fatty acids in mood disorders overall, but heterogeneity in study designs and results have been reported, and they are best studied as an augmentation treatment rather than monotherapy . The Omega-3 Fatty Acids Subcommittee, assembled by the American Psychiatric Association, recommends that patients with a mood disorder should consume 1 g EPA plus DHA daily. Current evidence may support the use of 1–9 g supplement of EPA plus DHA daily for people with mood disorders, though use of greater than 3 g daily should be monitored by a clinician owing to the risk of bleeding.

Studies investigating the relationship between seafood intake and perinatal depressive symptoms have been mixed, with some demonstrating an inverse relationship between the two , and others reporting no such relationship . In a large Danish prospective cohort study of more than 54,000 women, participants who were in the lowest quartile of self-reported fish intake during pregnancy were at increased risk of being treated for depression with an antidepressant up to 1-year postnatally . Although relatively low omega-3 fatty acid intake was associated with higher rates of treatment with an antidepressant, the investigators found that fish intake was also strongly associated with sociodemographic characteristics.

Although several studies have shown an inverse relationship between antenatal fish intake and perinatal depressive symptoms, several randomised placebo-controlled trials assessing the effectiveness of omega-3 fatty acid supplementation compared with placebo have not demonstrated a benefit in acute treatment or prevention of perinatal depression . In the largest double-blind, multicentre, RCT, 2399 pregnant women during the last half of pregnancy were randomised to supplementation with fish oil (800 mg/d DHA and 100 mg/d EPA) or vegetable oil (placebo). Depressive symptoms, measured by the Edinburgh Postnatal Depression Scale (EPDS) at 6 weeks and 6 months postnatally, did not differ between groups, nor did neurodevelopmental outcomes of their offspring differ at age 18 months . The DHA group had fewer very preterm births (less than 34 weeks’ gestation) compared with the control group, but more post-term births requiring obstetric intervention in the DHA group compared with the control group.

A small ( n = 36) randomised, double-blind placebo-controlled trial found a significant benefit of omega-3 fatty acids compared with placebo in the treatment of antenatal depression . Other studies have provided promising preliminary data on feasibility, tolerability, and efficacy in perinatal depression . Omega-3 fatty acid supplements have been well tolerated by perinatal women and seem to be free of significant levels of mercury or other contaminants. On the basis of positive findings from RCTs and meta-analyses in non-perinatal depression, we recommend perinatal patients with depression should consume 1 g EPA plus DHA daily. At this time, considering that the studies to date have been small and with inconsistent findings on efficacy of MDD, it is reasonable to augment other treatments with omega-3 fatty acids.

Folate

Folate, available as folic acid, folinic acid and 5-methyltetrahydrofolate (5-MTHF) or l -methylfolate, functions as a coenzyme or co-substrate in single-carbon transfers in the synthesis of nucleic acids and amino acid metabolism. An important folate-dependent reaction is the conversion of homocysteine to methionine in the synthesis of S-adenosyl-methionine. Folate undergoes transformation to l -methylfolate, a biologically active form of folate, which crosses the blood–brain barrier; 5-MTHF is biologically active. Folic acid and folinic acid are synthetic forms of dietary folate, which require the enzyme methylenetetrahydrofolate reductase for conversion into bioactive forms; however, this enzyme is affected by a polymorphism common in people with depression, which impairs transformation to l -methylfolate and is associated with MDD. Some people with depression may exhibit relatively low folate levels and experience impaired methylation and monoamine neurotransmitter metabolism. Most, but not all, studies report an association of low folate levels and an increased risk of depression . Low blood folate has been associated with a poorer response to treatment with antidepressants in MDD and higher folate levels at baseline seem to be associated with a better response.

Folate has been studied in a placebo–controlled trial an adjunctive treatment to fluoxetine, with significantly greater improvement in the folate group, a difference most pronounced in women . Ninety-four per cent of women who received fluoxetine, with the addition of folate 500 mcg per day, were treatment responders, compared with 61% of those who received fluoxetine and placebo. More recently, results were mixed from two multicentre, placebo-controlled RCTs trials examining the use of l -methylfolate with on-going antidepressant therapy for MDD; improvement was found with 15 mg per day but not the 7.5 mg per day dose.

It is recommended that women of reproductive age consume 0.4–1 mg folic acid daily to reduce the risk of neural tube birth defects: in a large UK prospective cohort study of non-pregnant women, less than 6% of women who became pregnant during the observation period reported daily folic acid supplementation of 0.4 mg/day or more . For women with a family history of neural tube defect, or who take antiepileptic medications, the recommendation is to take 5 mg folic acid daily before and during pregnancy . High rates of unplanned pregnancy make folate supplementation important in women of reproductive age, regardless of plans to conceive. No studies have been published on the efficacy of folate monotherapy or augmentation therapy for perinatal depression. Epidemiological data do not demonstrate that higher folate intake during pregnancy mitigates against the development of postnatal depression .

On the basis of the potential decrease in birth defects and RCT data that modestly support an antidepressant effect of augmentation with folate (i.e. folic acid 0.4–5 mg/day or folinic acid 15–30 mg/day), we recommend folate as an important adjunctive strategy for perinatal unipolar depression, as it carries little risk and may be particularly effective in women with low serum folate levels.

S-adenosyl l -methionine

S-adenosyl-methionine (SAMe) naturally occurs in humans, including in the brain. It is produced from the amino acid l -methionine through the one-carbon cycle, a metabolic pathway that requires adequate concentrations of folate and vitamin B-12. Its antidepressant effect may be via methyl donation in neurotransmitter synthesis or effects on anti-oxidative, anti-inflammatory and neuroprotective processes having important biological roles in depression.

Several placebo-controlled RCTs and two meta-analyses found that SAMe, studied in doses of 200–1600 mg per day, were significantly more efficacious than placebo and equivalent to tricyclic antidepressants in the treatment of depression; other studies report it equivalent to placebo . More recently, studies assessing adjunctive SAMe with a serotonin reuptake inhibitor have also been positive, with one study indicating that adjunctive SAMe improves memory-related cognitive symptoms in depressed people . S-adenosyl-methionine is generally well tolerated, with infrequently reported side-effects, which include mild gastrointestinal symptoms, sweating, dizziness, irritability, and anxiety.

No studies to date have evaluated the efficacy of SAMe in antenatal depression. Use of SAMe in pregnancy, however, has been studied as a treatment for cholestasis in pregnancy. In five trials of SAMe for cholestasis of pregnancy, adverse events or side-effects were absent for mothers and their infants . In one placebo-controlled study of SAMe for postnatal depressive symptoms, significant decreases in depressive symptoms with SAMe compared with placebo were reported . Further studies should be conducted before SAMe can be recommended for use in the perinatal period given the lack of efficacy and safety data during pregnancy.

St John’s wort (hypericum perforatum)

Hypericum perforatum, from the plant St John’s Wort, has been used for its medicinal properties since ancient times. Several of its bioactive substances, including hypericin, hyperforin and flavonoids, have affinity for neurotransmitter systems known to be important to the pathophysiology and pharmacotherapy of MDD, including activity at γ-aminobutyric acid (GABA), N-methyl- d -aspartic acid (NMDA), and serotonin receptors.

Two meta-analyses have compared the efficacy of St John’s Wort with placebo or standard antidepressants in the treatment of depressive symptoms or MDD . Results from the individual studies were mixed, with a more robust effect seen in people with mild-to-moderate depressive symptoms and less evidence of effectiveness in severe MDD. St John’s Wort, at a daily dosage of 300–1200 mg, had an advantage over placebo and efficacy similar to tricyclic or selective serotonin reuptake inhibitor antidepressants, with better efficacy than the standard antidepressant in a subgroup of people with mild-to-moderate depression .

St John’s Wort contains high concentrations of the bioactive component hyperforin, which induces the cytochrome P450 system (CYP3A4), and inhibits a membrane-bound transporter that facilitates transport across the intestinal lumen and the blood–brain barrier. St John’s Wort may interact with medications, including oral contraceptives, reducing oestrogen plasma concentration and effectiveness . Studies have not conducted to assess whether women should be on higher doses of oestrogen oral contraceptives; therefore, clinicians should be aware of the herb–drug interaction.

Few studies have investigated the safety of St John’s Wort during the perinatal period, and no published RCTs have evaluated efficacy or safety in perinatal depression. Safety data based on animal studies are mixed, with some studies raising concerns about exposure to hypericum and hypericin . Limited data in 54 human pregnancies indicated no increased risk of major malformations or prematurity rate for infants born to women taking St John’s Wort during pregnancy and matched controls . It seems to be excreted into breast milk at undetectable to low levels, compared with other antidepressants, and its bioactive components at or below the limit of quantification in infant plasma . In one small prospective observational study, increased rates of adverse events in infants (e.g. colic, drowsiness and lethargy) were reported in women ( n = 33) who were treated with hypericum while breastfeeding their infants compared with infants of matched depressed and non-depressed controls . Further studies should be conducted before St John’s Wort can be recommended for use in the perinatal period, given reported infant side-effects during lactation and a limited evidence base for efficacy.

Bright light therapy

Most, but not all , studies have shown that bright light therapy is an efficacious, first-line treatment for both seasonal and non-seasonal MDD . Bright light therapy is generally well tolerated, although induction of mania is a risk in individuals with bipolar disorder . In studies of seasonal depression, predictors of beneficial response included patients of younger age and with symptoms of hypersomnia and increased eating. Three studies that assessed bright light therapy in antenatal depression suggest efficacy. In a small open study, pregnant women ( n = 16) received 1 h of morning bright light therapy . In participants who received at least three weeks of bright light therapy, depression scores improved by a mean of 49%; among participants who completed at least 5 weeks, scores improved by 59%. No significant side-effects were reported, except treatment-related nausea. In a small, double-blind study, pregnant women ( n = 10) were randomised to bright light therapy (7000 lux) or a dim light (500 lux) placebo condition for 5 weeks followed by a 5-week extension phase . Bright light therapy produced a significantly greater antidepressant response than dim light, with significant differences at the end of the extension phase. One participant who received bright light therapy experienced hypomania, which resolved upon reduction of the duration of daily light exposure. In a larger RCT of 27 pregnant women randomised to 7000 lux bright (active) light or 70 lux dim red (placebo) light upon awakening for 60 mins over 5 weeks, bright light was associated with a greater reduction of depressive symptoms, with 81.3% of women receiving bright light experiencing 50% or greater reduction in symptoms compared with 45.5% who received placebo dim light. The treatment was tolerated well, without adverse effects .

In a small RCT of women with postnatal depression ( n = 15), participants were assigned to 30 mins of 10,000 lux bright light or a 600 lux dim light placebo condition for a duration of 6 weeks . Both groups experienced significant improvement from baseline, without differences in response. Light therapy may be an attractive option for some perinatal women as ultraviolet screened light boxes with 10,000 lux illumination are available commercially. On the basis of a growing evidence, we recommend initial dosing of 30 mins beginning within 10 mins of awakening. On the basis of the high prevalence of mood episodes during pregnancy and the postpartum in the course of bipolar disorder, even if not previously diagnosed, patients should be monitored carefully for emergent symptoms of hypomania or mania, sleep disturbance, and agitation when bright light therapy is initiated.

Exercise

Exercise is integral to optimal health in pregnancy, as well as in the prevention of heart disease, obesity and diabetes, comorbid conditions that those with MDD are more likely than the general population to develop. Mechanisms underlying the potential antidepressant effects of exercise are not yet clearly delineated, but exercise may affect neuroendocrine function and neurotransmission. Several trials have shown that aerobic exercise reduces depressive symptoms ; epidemiological data suggest that regular exercise is associated with decreased risk of depressive symptoms , although not all trials have consistently demonstrated benefit . Treatment research is difficult with exercise, as adequate ‘dosing’ (e.g. type, intensity and duration), study control conditions and maintenance to treatment assignment pose challenges in study design .

According to recommendations from both the American College of Obstetricians and Gynecologists and the Royal College of Obstetricians and Gynaecologists, pregnant women without medical contraindications should engage in regular aerobic and strength-conditioning exercise during the perinatal period . Studies of both antenatal aerobic exercise and general physical activity in women without MDD showed that exercise or physical activity was associated with fewer depressive symptoms in pregnancy. No studies to date assess exercise for perinatal MDD, although few data have studied the effects of exercise on depressive symptoms in women without MDD. In one study, investigators evaluated if antenatal exercise consisting of aerobic and strengthening exercises prevented postnatal depressive symptoms. They did not find that antenatal exercise was associated with a lower prevalence of postnatal depressive symptoms as measured by the EPDS . An 8-week ‘mother and baby’ programme, including exercise, was associated with lower EPDS scores in the intervention group at study end, but between-group significance was lost at follow up 1-month later .

Few controlled studies have evaluated the effectiveness of exercise as a treatment for postnatal depression. Two RCTs comparing a 12-week intervention of exercise with a control situation in postnatal women with an EPDS score of 10 or greater (suggestive of postnatal depression, but in itself not diagnostic) showed significantly greater reduction in EPDS during the study with exercise than the control situation , and one study indicated that exercise can reduce physical fatigue in postnatal depressed women . On the basis of limited evidence for exercise in the treatment of perinatal depression, we recommend, for general health, 30 mins a day of exercise most days of the week in the absence of either medical or obstetric complications and after consultation with an obstetrician.