Emerging Risk Factors for Postpartum Depression: Serotonin Transporter Genotype and Omega-3 Fatty Acid Status

Abstract

Objective

Depression is a leading cause of disability and hospitalization. Women are at the highest risk of depression during their childbearing years, and the birth of a child may precipitate a depressive episode in vulnerable women. Postpartum depression (PPD) is associated with diminished maternal somatic health as well as health and developmental problems in their offspring. This review focuses on 2 PPD risk factors of emerging interest: serotonin transporter (5-HTT) genotype and omega-3 polyunsaturated fatty acid (n-3 PUFA) status.

Method

The MEDLINE, PubMed, and Web of Science databases were searched using the key words postpartum depression, nutrition, omega-3 fatty acids, and serotonin transporter gene. Studies were also located by reviewing the reference lists of selected articles.

Results

Seventy-five articles were identified as relevant to this review. Three carefully conducted studies reported associations between the 5-HTT genotype and PPD. As well, there is accumulating evidence that n-3 PUFA intake is associated with risk of PPD. Preliminary evidence suggests that there could be an interaction between these 2 emerging risk factors. However, further studies are required to confirm such an interaction and to elucidate the underlying mechanisms.

Conclusions

Evidence to date supports a research agenda clarifying the associations between n-3 PUFAs, the 5-HTT genotype, and PPD. This is of particular interest owing to the high prevalence of poor n-3 PUFA intake among women of childbearing age and the consequent potential for alternative preventive measures and treatments for PPD.

Major depressive disorder is the most common mental disorder in Canada,¹ with a lifetime incidence of 7.9% to 8.6%, a 1-year prevalence of 4% to 5%, and a 6% point prevalence of symptoms consistent with depression.² The rate of outpatient treatment for depression increased more than 3-fold in the United States between 1987 and 1997.³ Among women of childbearing age, depression is the second-leading cause of disability worldwide.⁴ Taken together, depression and other affective disorders constitute the leading cause of nonobstetric hospitalization among women of childbearing age in the United States.⁵ In Canada, the incidence and rate of hospitalization for MDD is about 50% higher for women than for men.^2,6 Finally, it is during the first postpartum year that women are at the highest risk of depression, with 45% to 65% of ever-depressed women having their first episode.⁷

Postpartum Depression

PPD is defined as a nonpsychotic depressive illness of mild-to-moderate severity occurring in a mother during the first postnatal year. Though clinically heterogeneous, PPD is distinct from the less severe postpartum blues or baby blues (a mild depressive reaction within the first few days following birth, occurring in 15% to 84% of mothers depending on timing, number of assessments, and criteria used to establish a case)⁸ and the much less frequent, though more serious, postpartum psychosis (a psychiatric emergency occurring in fewer than 1 of 500 mothers, with rapid onset within the first 4 weeks after delivery, generally associated with bipolar disorder and requiring hospitalization).⁹ The prevalence of PPD is generally reported as between 10% and 15%.^7,10

Consequences of PPD

PPD is associated with reduced maternal functional status,¹¹ chronic disease¹² and diminished physical health-related quality of life.¹³ PPD has also been linked with numerous somatic and psychiatric problems in the children of depressed mothers.^14,15 Research suggests that mothers suffering from PPD tend to be more disengaged, hostile, critical, and less sensitive and responsive toward their children.^16–18 These patterns can lead children to develop an insecure attachment relationship with their mother, resulting in disruptions in sleep patterns, delays in language and cognitive development, poor affect regulation, and other emotional and behavioural problems.¹⁹ For example, 10-year-old children exposed to maternal PPD symptoms since birth have larger left and right amygdala and a heightened cortisol response to stress,²⁰ and other long-term associations have also been observed between maternal PPD and cognitive outcomes including IQ in adolescents.²¹

Clinical Implications

• Nutritional and genetic exposures are emerging as risk factors of interest in the etiology of PPD.

• If confirmed in future studies, these exposures hold potential as part of a screening strategy to identify women at risk of PPD, and to define target populations for evaluating prevention strategies.

• Intakes of n-3 PUFA in pregnant women are well below recommendations and are amenable to improvement.

Limitations

• The number of studies on these risk factors is too small to be able to draw firm conclusions.

• More evidence is required concerning the specific biological mechanisms underlying PPD, and the pathways through which these risk factors could potentially interact.

PPD, Compared With Other Depression

While the diagnostic criteria for PPD are otherwise identical to those for depression occurring at other times, certain biologic markers distinguish PPD from other types of depression. It is likely that hormonal changes associated with parturition contribute to mood alterations in vulnerable women.²² Studies have shown decreased susceptibility to depression in women during times of reproductive hormone stability, suggesting that PPD may stem, in part, from marked hormonal variations associated with childbirth.²³ Over the course of pregnancy, cortisol levels double,²⁴ while progesterone and estradiol levels increase 10 and 50 times, respectively; these hormones then abruptly return to normal levels within the first 2 weeks of the postpartum period.²⁵ Experimental evidence suggests that women who develop PPD may be particularly sensitive to these hormonal fluctuations.^26–28 Further, decreased levels of monoamines, including 5-HT, norepinephrine, and dopamine, are implicated in the pathogenesis of depression,²⁹ and data from animal models suggest that estrogen and other steroid hormones mediate the transcription of genes regulating synthesis and metabolism of neurotransmitters and their receptors,^30,31 supporting the hypothesis that hormonal fluctuations affect the risk of PPD, in part, through their effects on the central nervous system. Finally, the prenatal and postpartum periods involve exceptional social stressors and demands on women that have been found to increase both the risk and the consequences of depression.^32,33

Treatment of PPD

The care and treatment of women with PPD varies widely between countries, owing, in part, to inadequate guidelines and disparities in accessible treatment options.³⁴ Nonetheless, there is evidence supporting pharmacologic and other biological interventions (for example, hormonal interventions or bright light therapy) for the treatment of moderate-to-severe depression in postpartum women.^35,36 While few placebo-controlled trials of ADs have been conducted in women with PPD, 2 trials have yielded positive body of research results,^37,38 and ADs appear to be as effective for PPD as for MDD occurring at other times in the life cycle.³⁶ However, concerns have been raised regarding the robustness of evidence on which these conclusions are based,³⁵ and few studies have compared different classes of medications for the treatment of PPD.³⁹ In addition, evidence suggests that adherence to ADs during the postpartum period may be poor.³⁴ Psychotherapeutic and other nonpharmacologic interventions, including relaxation and massage therapy, infant sleep interventions, and maternal exercise, have also shown promise in the treatment of PPD. However, the evidence base concerning their effectiveness is still limited.^40,41

Risk Factors for PPD

Given the limited knowledge regarding efficient and safe treatment of PPD, one avenue for reducing PPD and need for treatment would be through preventive approaches targeting risk factors. A constellation of risk factors for PPD has been identified that includes social, demographic, obstetric, biological, hormonal, psychiatric, and genetic features, as well as characteristics of the newborn child. Among the key social factors predicting PPD are a strained marital relationship, low social support, and stressful life events.^7,14 Low SES^42,43 and personal or family history of depression or mood disorders¹⁴ have also been identified as significant risk factors for PPD. PPD has been linked with severe obstetrical complications during pregnancy^44,45 and at delivery⁴⁶; with adverse birth outcomes (low birth weight and preterm birth)⁴⁷; and with adverse neonatal outcomes, such as infant irritability and poor motor function.⁴⁸

It has also been proposed that the environmental risk for depression may be moderated by genetic factors. It is estimated that about 40% to 50% of the risk for depression is genetic,⁴⁹ with family studies showing a 3- to 4-fold increased risk of depression for family members with depression, depending on the degree of relation.⁵⁰ However, the specific mechanisms of genetic causality are not well understood,²⁵ and the relative contribution of various combinations of genetic and environmental factors to PPD is as yet undetermined.

Beyond social influences, one key environmental factor may be nutrition. There is indeed evidence that diet quality, dietary intake, and overall nutritional status can affect the risk of PPD.⁵¹ Pregnancy is a period during which nutritional requirements and vulnerability to poor nutritional status are heightened. In fact, requirements for many nutrients in women reach a lifetime peak during pregnancy or lactation. Improved nutritional status during these periods may positively impact on maternal mental health, both directly and by augmenting the effectiveness of ADs.⁵¹

Therefore, our review will focus on 2 risk factors of emerging interest, 5-HTT genotype and n-3 PUFA status. The 5-HTT gene was selected because it has become the most investigated genetic variant in psychiatry, psychology, and neuroscience.⁵² Further, a significant body of research has explored the association between n-3 PUFAs and PPD.^53–56 Because n-3 PUFAs are hypothesized to reduce the risk of depression, in part through the regulation of gene expression,⁵³ studies testing the interaction between n-3 PUFAs and genetic exposures in the prediction of PPD would be warranted. This is particularly important as nutrition is a potentially modifiable environmental risk factor^51,57 that could interact with a genetic predisposition to PDD.

Literature Search

We searched the MEDLINE (1950 to 2011), PubMed (1966 to 2011), and Web of Science (1965 to 2011) databases for articles in English or French using the key words postpartum depression, nutrition, omega-3 fatty acids, and serotonin transporter gene. We included narrative and systematic reviews, original research reports of observational or experimental studies, and editorials. Studies were also located by reviewing the reference lists of selected articles. Our search generated 257 articles. Abstracts from research reports and systematic reviews were assessed for exposure and outcome measures. To be included, studies needed to have PPD or depressive symptoms as an outcome. At least one of the following exposures was also required: 5-HTT genotype, n-3 PUFA dietary intake, supplementation, or biomarker measurement, and fish consumption. Review articles and editorials addressing these exposures, as well as other genetic and nutritional risk factors for PPD, were included. This left 75 articles forming the core of our narrative review.

The 5-HTT Gene and PPD

Following reports of associations between the short allele of the 5-HTTLPR and anxiety-related personality traits,⁵⁸ studies have addressed the influence of the 5-HTT gene on depression, both alone⁵⁹ and in interaction with environmental risk factors.⁶⁰ The 5-HTT gene modulates the reuptake of 5-HT at brain synapses, a principal neurobiological feature of depression and the target of selective serotonin reuptake inhibitor ADs.⁶⁰

Nevertheless, the precise relation between 5-HTT and the risk of depression is somewhat controversial, with a recent meta-analysis concluding no overall effect.⁶¹ However, this finding has been critiqued on several grounds, including heterogeneity in measurement of outcome and environmental exposure, exclusion of studies with high-quality designs,⁶² inadequate measurement of relevant environmental exposures,^62,63 and use of inappropriate interaction models.^64,65 Another meta-analysis concluding that positive results for interactions between 5-HTTLPR and stressful life events in the prediction of depression were compatible with chance findings⁶⁶ has also been critiqued on some of these same grounds.⁶⁷ A third meta-analysis with broader inclusion criteria⁶⁸ concluded that 5-HTTLPR moderates the relation between stress and depression. In addition, significant associations between 5-HTTLPR genotype and depression were found in 2 other meta-analyses that did not take into account stress as a covariable.^69,70 Accordingly, there are reasons to suspect that the 5-HTT gene is related to depression in some subpopulations, including women in the postpartum period.

Biologic evidence suggests a role of the 5-HT system in PPD that may differ from that in other forms of depression.⁷¹ Synthesis of cerebral 5-HT decreases during pregnancy owing to placental catabolism of tryptophan, the precursor to 5-HT.⁷² PPD symptoms are positively correlated with postnatal tryptophan catabolism⁷³ and inversely correlated with maternal plasma tryptophan concentrations.⁷¹ This suggests that the 5-HT system may be of particular importance in the pathophysiology of depression in postpartum women.

Epidemiologic evidence suggests that 5-HTT gene expression patterns may have differential effects for men and women, particularly in the context of psychosocial stress. One study⁷⁴ on 5-HTT, family environmental risk, and depression showed effects for women but not men. Two additional studies^75,76 showed increased depressive symptoms in females carrying the short allele but a protective effect of the short allele in males.

The 5-HTT genotype was linked to PPD in 3 other studies. A study⁷⁷ looking at depressive symptoms at 3 time points after delivery found a significant positive association between depressive symptoms and 5-HTT expression level at 8 weeks into the postpartum period. Another study⁷⁸ of women with a prior history of depression found that short allele carrier status (either 1 or 2 copies of the short allele) of the 5-HTT gene predicted depression at 1- to 8-weeks during the postpartum period (OR 5.13; 95% CI 1.16 to 22.7, P = 0.02). Finally, a recent study⁷⁹ showed the 5-HTT short allele to be associated with increased risk of PPD in low-SES women but with decreased risk in high-SES women. Taken together, these heterogeneous results suggest that null findings from other studies, ^80,81 particularly meta-analyses,^61,66 may mask interactions between 5-HTT genotype and environmental risk factors.

n-3 PUFA and PPD

Found in fish as well as some seeds and nuts,⁸² n-3 PUFAs are essential unsaturated fatty acids, and they merit attention and further study for several reasons. First, n-3 PUFAs directly affect brain activities, including receptor function, neurotransmitter uptake, and signal transmission,⁵¹ and evidence suggests a beneficial role of n-3 PUFAs in the treatment of patients with diagnosed depression.^83,84 Second, dietary intake of n-3 PUFAs is particularly poor, and the ratio of n-6 to n-3 PUFA intake has risen dramatically over the last century.⁸⁵ This ratio is a commonly used marker of dietary fatty acid composition and is positively related to risk for various diseases.⁸⁶ Finally, as n-3 PUFA stores are transferred from the mother to the developing fetus during gestation and later to the infant by lactating mothers, maternal n-3 PUFA levels decrease during pregnancy and remain lowered at least 6 weeks into the postpartum period.⁸⁷

Research on n-3 PUFAs and PPD has been informed by an interest in the interrelations between fatty acids, depression, and cardiovascular disease.^88,89 Patients with depression show increased cardiovascular mortality, and depression is a frequent comorbidity in patients with coronary artery disease and is associated with worse outcomes in these patients.^89–91 Depression and cardiovascular disease may exacerbate each other directly, but it is also hypothesized that these 2 seemingly disparate health problems share common causes.⁹² n-3 PUFAs are understood to modulate both serotonergic neurotransmission and thrombotic and inflammatory mechanisms associated with coronary disease,^90,93 and it is likely that inflammatory markers comprise part of the physiological mechanism of depression as well.^94–97

Evidence linking n-3 PUFAs and depression spans multiple study designs and populations.^85,98,99 Associations have been found in case–control, cross-sectional, and cohort studies; with exposures including blood lipid samples, adipose tissue samples (reflecting long-term or habitual intake), fish consumption, overall dietary fatty acid intake,¹⁰⁰ and postmortem brain cortex analyses;¹⁰¹ and with outcomes including clinical depression, depressive symptoms,⁹⁹ depression during pregnancy, and PPD.¹⁰² Serum levels of DHA, one of the principal n-3 PUFAs associated with depression, have been observed to decline during pregnancy and after delivery, leaving postpartum women vulnerable to DHA deficiency.^55,103 Dietary intake and serum levels of n-3 PUFAs have been inversely associated with PPD^53,55 and with depression in other populations.^99,104 Ecologic and cross-sectional studies⁸⁵ have found inverse associations between consumption of fish (a primary dietary source of n-3 PUFAs) and major depression and PPD.

Evaluating research on n-3 PUFAs and PPD must be done with caution. This research is conducted against a background of robust links between psychosocial exposures and affective disorders and strong demonstrated associations between depression, before or during pregnancy, and PPD. An emerging body of research also shows links between depression and other nutrients whose intake is likely to exhibit some collinearity with n-3 PUFAs. Nevertheless, randomized controlled trials have shown n-3 PUFAs to be effective as AD treatment,^84,105 suggesting a causal role for this nutrient class in the etiology of depression. This claim is supported by evidence linking n-3 PUFAs with efficient neurotransmission¹⁰⁶ and with inflammatory mechanisms connected to depression. Several clinical trials of n-3 PUFA supplementation for patients with MDD have shown large effect sizes.¹⁰⁷ However, meta-analyses suggest it is more realistic to expect moderate effect sizes from supplementation.^83,84 Observational studies of n-3 PUFAs and depression have also shown moderate effect sizes. For example, in a study⁹¹ of patients with recent acute coronary syndromes (representing a high-risk group), the per cent of phospholipid fatty acids represented by n-3 PUFAs was about 12% lower in people with depression, and the percentage of DHA about 14% lower, compared with nondepressed people.

Finally, it needs to be considered that plasma levels of fatty acids are an imperfect measure of dietary intake and also an imperfect predictor of fatty acid levels in brain tissue. Serum fatty acid levels have been shown to be sensitive to recent changes in dietary fatty acid intake in adults.^108–110 This pattern has specifically been observed in pregnant women, with n-3 PUFA supplementation associated with elevated plasma and postpartum breast milk DHA levels.¹¹¹ Findings from animal studies^85,112 suggest a robust relation between serum and brain fatty acid levels, and dietary deficiency in n-3 PUFAs has been associated with observed changes in brain composition and neural functioning in animal models.⁸⁵ Significantly, n-3 PUFA deficiency has been associated with altered metabolism of dopamine and 5-HT,¹⁰⁹ 2 of the key neurotransmitters underlying the neural physiology of depression. However, animal models show the mechanisms through which fatty acids. are absorbed, converted, synthesized, and processed in the brain are complex and change over the life course.^{109,113–118} Dietary fatty acid intake affects brain fatty acid levels most readily during early development,^119–121 and it is unclear how quickly brain fatty acid levels change in relation to dietary intake in mature animals. Nevertheless, significant changes in brain fatty acid levels were observed in adult female rats within a time span of one reproductive cycle following diet modification.¹²²

n-3 PUFA Status and Modification of Intake

Evidence from numerous fronts suggests that intakes of n-3 PUFAs are far below recommended levels and are amenable to improvement. In the US adult population, intake of DHA and EPA in 2000 was more than 70% below recommendations from the National Institutes of Health.¹²³ A 4-fold increase in fish consumption would be required to bring EPA and DHA intake to recommended levels. In Canada and Australia, maternal milk concentrations of DHA appear to have decreased by about 50% over the 15-year period ending in 1999.¹²⁴

Inadequate n-3 PUFA levels are of even greater concern in pregnant women. In a cross-sectional survey¹²⁵ of pregnant women in central Mexico, the median DHA and EPA intakes, as calculated from a food frequency questionnaire, were 55 and 18 mg/day, respectively. This compares with recommendations by the American Dietetic Association and Dietitians of Canada of 500 mg/day DHA and EPA combined.¹²⁶ A Canadian study¹²⁷ of adults in Quebec found that 85% had an EPA and DHA intake lower than this recommendation. Among the women of childbearing age in that study, median intake of DHA was 126 mg/day,¹²⁷ while a study¹²⁴ of pregnant women in British Columbia showed a mean DHA intake of 160 mg/day.

Because maternal plasma n-3 PUFA concentrations decline substantially after delivery,^128,129 maintaining a sufficient intake of n-3 PUFAs is important to ensure adequate fatty acid stores during the postpartum period. In addition to the implications for maternal mental health, n-3 PUFAs are essential for infant neural and visual development.¹³⁰ n-3 PUFA intake is thus critical for lactating mothers. While there has been considerable focus on n-3 PUFA status in adolescent mothers owing to the enhanced nutritional risks associated with adolescence,¹³¹ several studies have also examined n-3 PUFA levels in adult postpartum women. Two studies found DHA intake of 30 to 58 mg/day and concentration in breast milk of about 0.10%, well below recommendations of 0.2% to 0.4%.^132,133 A study comparing lactating and nonlactating women found a DHA intake of 29 to 47 mg/day and EPA intake of 52 to 91 mg/day,¹²⁸ again well below recommendations. These results suggest poor maternal n-3 PUFA intake to be a significant problem, not only during pregnancy but also in the postpartum period.

n-3 PUFA, 5-HTT Genotype, and PPD

A growing body of literature is exploring nutritional aspects of depression¹³⁴ and PPD specifically.^25,51 However, little research has addressed interactions between nutritional and genetic risk factors in the prediction and etiology of depression. There has been considerable focus on interactions between the 5-HTT gene and psychosocial stress^68,135 but little investigation into genetic interactions with nutritional exposures that may exhibit some of the same effects as stress on brain function. Two studies examining the seasonal variation in n-3 PUFAs, plasma tryptophan, and serotonergic markers^136,137 suggest that fatty acid levels in the brain may modulate 5-HT release and reuptake. These findings support research into interactions between n-3 PUFAs and the 5-HTT gene and suggest that these 2 seemingly disparate exposures may affect the risk of PPD through a common neurobiological mechanism. Accordingly, studying their association in the prediction of PPD may help further elucidate the neurobiological underpinnings of this condition while helping to target prevention and treatment efforts.

Conclusion

There is a growing awareness of the importance of nutritional and genetic exposures as risk factors for PPD. The 5-HTT gene is a promising avenue for genetic research, and it appears highly likely that this gene affects the risk of depression and other psychiatric conditions. However, it is unclear which genotypes are associated with elevated risk in which populations, and specifically how associations between 5-HTT genotype and depression may differ during the perinatal period from other time points across the life course. Similarly, increasing evidence links n-3 PUFAs and depression in diverse populations. However, the biological mechanisms through which these links function, and the ways in which they may be modified in pregnancy, are not clearly understood.

One of these mechanisms could operate through a gene–environment interaction. Because it can be reasonably hypothesized that the 5-HTT genotype and n-3 PUFAs impact on the risk of PPD, in part through the same mechanism, studying them jointly would present an opportunity to advance our understanding of how genetic and dietary exposures may interact in the etiology of PPD. Knowledge garnered from this effort has the potential to improve the prediction, prevention, and treatment of this significant public health problem. This is particularly important as current intake of n-3 PUFAs in pregnant women is well below recommendations and is thus amenable to improvement.

Abbreviations

5-HT

serotonin

5-HTT

serotonin transporter

5-HTTLPR

serotonin transporter gene linked polymorphic region

AD

antidepressant

DHA

docosahexaenoic acid

EPA

eicosapentaenoic acid

MDD

major depressive disorder

PPD

postpartum depression

n-3 PUFA

omega-3 polyunsaturated fatty acid

SES

socioeconomic status