REVIEW

published: 11 June 2021

doi: 10.3389/fpsyt.2021.620371

Postpartum Depression: Current

Status and Possible Identification
Using Biomarkers
Yi Yu 1,2 , Hong-Feng Liang 1 , Jing Chen 1,3 , Zhi-Bin Li 1,3 , Yu-Shuai Han 1,3 , Jia-Xi Chen 1,3 and
Ji-Cheng Li 1,3*
1
Central Laboratory, Yangjiang People’s Hospital, Yangjiang, China, 2 Center for Analyses and Measurements, College of
Chemical Engineering, Zhejiang University of Technology, Hangzhou, China, 3 Institute of Cell Biology, Zhejiang University,
Hangzhou, China

Postpartum depression (PPD) is a serious health issue that can affect about 15% of
the female population within after giving birth. It often conveys significant negative
consequences to the offsprings. The symptoms and risk factors are somewhat similar to
those found in non-postpartum depression. The main difference resides in the fact that
PPD is triggered by postpartum specific factors, including especially biological changes
in the hormone levels. Patients are usually diagnosed using a questionnaire onsite or in a
clinic. Treatment of PPD often involves psychotherapy and antidepressant medications.
In recent years, there have been more researches on the identification of biological
Edited by:
markers for PPD. In this review, we will focus on the current research status of PPD,
Gianluca Rosso, with an emphasis on the recent progress made on the identification of PPD biomarkers.
University of Turin, Italy
Keywords: postpartum depression, postnatal depression, biological markers, biomarkers, multi-omics
Reviewed by:
Giuseppe Tavormina,
Independent Researcher, Provaglio
d’Iseo, Italy INTRODUCTION
Gayatri Saraf,
University of British Columbia, Canada Postpartum depression (PPD) has raised a major public health concern. It has been estimated
*Correspondence:
that about 15% of women within 1 year after childbirth may suffer from PPD (1). Like major
Ji-Cheng Li depression, PPD is a disabling disorder. Significant negative effects of PPD on children, even after
[email protected] they grow into adulthood, have been documented in the literature. PPD-related suicide has become
the second-leading cause of death for women in the postpartum period (2). While antidepressants
Specialty section: have been effective in treating PPD in many cases, possible side effects of antidepressant medication
This article was submitted to have been of great concern. Therefore, it is crucial to identify PPD at an early stage.
Mood and Anxiety Disorders, Although, known since 400 BC, PPD it did not catch wide attention until about half a century
a section of the journal ago. Over the past decades, there have been many research efforts and reviews in the field of PPD.
Frontiers in Psychiatry
Grace et al. (3) reviewed the literature and found that PPD confers small effects on cognitive
Received: 22 October 2020 development such as language and IQ. Behavioral effects may persist up to 5 years post partum
Accepted: 19 May 2021 and beyond. Dennis and McQueen (4) reported that women with depressive symptoms at the
Published: 11 June 2021
early stage of the postpartum period were associated with increased risk for negative infant feeding
Citation: outcomes. The review by Blum (5) focused on the psychodynamics of PPD, and found that a triad of
Yu Y, Liang H-F, Chen J, Li Z-B,
three common, specific emotional conflicts (dependency conflicts, anger conflicts, and motherhood
Han Y-S, Chen J-X and Li J-C (2021)
Postpartum Depression: Current
conflicts) was typical of many women who develop PPD. Yawn et al. (6) found that among those
Status and Possible Identification evaluated programs between 2000 and 2010, only four studies included patient outcomes, and only
Using Biomarkers. two reported success in improving outcomes. O’Hara and McCabe (7) found that the results of most
Front. Psychiatry 12:620371. studies did not seem to converge, and the majority had a small sample size or suffered from a lack of
doi: 10.3389/fpsyt.2021.620371 proper controls. Anderson and Maes (8) reviewed the biological aspects of PPD, and suggested that

Frontiers in Psychiatry | www.frontiersin.org 1 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

tryptophan catabolites, indoleamine 2,3-dioxygenase, serotonin, is used as the cutoff for being PPD positive. Brief subscales of
and autoimmunity play a powerful role in immuno- EPDS have also been designed (26), such as 3-item, 7-item and
inflammation and oxidative and nitrosative stress. Furthermore, 2-item subscales. Other screening tools include the Hamilton
decreased level of endogenous anti-inflammatory compounds Rating Scale for Depredession (HAM-D) (27), which was not
together with decreased ω-3 poly-unsaturated fatty acids (PUFA) designed for PPD specifically. The reliability of HAM-D varies
in the post-partum period may be a central cause for PPD. significantly in different evaluations, ranging from 0.46 to 0.98
Kim et al. (9) conducted a review on the role of oxytocin in the (28). Other scales for diagnosing related mood disorders, such as
treatment of PPD, and found that the results was inconsistent. the Bipolar Spectrum Diagnostic Scale (BSDS) (29) for bipolar
Yim et al. (10) conducted a review on researches published disorders (BD), may also become relevant when such disorders
between 2000 and 2013 on the predictors for PPD and found occur during the perinatal period.
that the biological and psychosocial literatures were largely
disconnected, and integrative analyses were rare to find. They ETIOLOGICAL MODELS OF PPD
reported that the strongest biological predictors for PPD risk
were hypothalamic-pituitary-adrenal (HPA) axis dysregulation, The exact causes of PPD are still unknown. Models of PPD can
inflammatory processes, and genetic vulnerabilities, while the mainly be divided into two categories: biological vs. psychological
strongest psychosocial predictors were severe life events, chronic models. Integrated models of both are rare. However, many
strain, poor relationship quality, and family support. There existing studies suffer from relatively small sample sizes or lack
are many other reviews on PPD in the literature, which can of control (or both), and none of the models mentioned here
be classified mainly into two distinct categories: biological vs. is conclusive.
psychosocial approaches. The former addressed the endocrine
system, the immune system, and genetic factors (11–14), while Biological Models
the latter addressed stressors and interpersonal relationships It is well-known that childbirth is accompanied by a dramatic
(5, 15–17). Reviews that covered both categories (18, 19) are decreases in several hormones, such as estradiol, progesterone,
relatively rare. and cortisol. In withdrawal models, reproductive hormones
In the present review, we shall emphasize on the literature (30) and stress hormones (31) rise dramatically during pregnancy
on PPD in the past several years. There has been an increasing and then drop suddenly upon delivery, and thus leads to system
number of researches on the screening and diagnosis of dysregulation and hence PPD (32). These models cannot explain
PPD using biological markers. Several biomarkers have been how the hormonal withdrawal acts to cause depression in
identified using the modern technology of multi-omics. The women, nor can they explain those depressive symptoms that
omics-based biomarkers can provide a more quantitative and begin during pregnancy before parturition.
objective criterion for the diagnosis of PPD, compared with the In depression models, PPD is associated with dysregulation
questionnaire-based diagnosis. of stress hormones, particularly cortisol (12). Several recent
reviews suggested that dysregulation of the HPA axis plays a
main role in the development of PPD (33, 34). Diminished
DIAGNOSIS OF PPD dopaminergic function may also play a role in PPD (35).
Sudden estradiol withdrawal could lead to dysregulation
There remains a controversy regarding the criterion for the onset in brain dopaminergic pathways and hence PPD. Multiple
time of PPD (1). The Diagnostic and Statistical Manual of Mental neuroendocrine changes caused by pregnancy may also play
Disorders revision 5 (DSM-5) of the US encompasses episodes a role in PPD development, including dysfunctional gamma-
that sets in during pregnancy (20) and that begin within 6 months aminobutyric acid (GABA) signaling (36, 37). PPD has also been
of delivery. In clinical practice and in various studies in the found to be associated with low allopregnanolone levels during
literature, the onset time for PPD has been generalized to up to 1 pregnancy (38). The involvement of GABA and allopregnanolone
year post-partum. in PPD has also been hypothesized in other models of PPD
A clinical interview can be used to diagnose PPD, such as the pathophysiology (39).
Structured Clinical Interview for the DSM-IV (21). Alternatively,
easy-to-use self-report measures such as questionnaires have also Psychological Models
been widely used for clinical assessment. The most prominent Psychological models emphasize the deleterious role of
and widely used is the Edinburgh Postnatal Depression Scale psychological stressors and underlying cognitive vulnerabilities
(EPDS) (22), which has been reported to be reliable, well- and the ameliorating role of psychosocial resources. In these
validated, and often more practical and cost-effective in wide- theories, pregnancy, childbirth, and new parenthood are
scale screenings for PPD risk (23). EPDS lays an emphasis on stressors that cause women to develop PPD symptoms. One can
psychic symptoms of depression, so as to reduce the weight of find consistent support for these models in the psychological
common symptoms of most new mothers. Other commonly used literature (32, 40, 41).
questionnaire-based screening tools include two-item Patient
Health Questionnaire (PHQ-2) (24) and 9-item Patient Health Integrated Models
Questionnaire (PHQ-9) (25). PHQ-2 contains the first two items An integrated model may bridge the biological and psychological
of the PHQ-9. A typical EPDS or PHQ-9 score of 10 or above theories. For example, in the stress vulnerability model, stress can

Frontiers in Psychiatry | www.frontiersin.org 2 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

cause PPD symptoms in women that have genetic, hormonal, forming part of a unique mood spectrum (54, 55). Accordingly,
and cognitive vulnerabilities (32). The bio-psycho-social-cultural these common features may cause misdiagnosis. Indeed,
model of Halbreich (42) is a combination of the stress patients with BD may sometimes be misdiagnosed as having
vulnerability model with biological and cultural factors. There MDD, leading to inappropriate treatment with antidepressant
exists limited evidence in the literature that supports these medication (29, 56). The misuse of antidepressants in patients
integrated models. with BD without mood-regulators may induce (hypo)mania or
rapid cycling and may increase the risk of illness recurrence (57–
Evolutionary Models 60). Therefore, to ensure proper treatment, extra caution must be
Models from an evolutionary perspective regard PPD as a excised to ensure proper diagnosis.
consequence of modern civilization, due to psychological
adaptation during the human evolution. Hagen (43)
proposed that PPD could cause parents to reduce or
PSYCHOSOCIAL RISKS FACTORS FOR
eliminate investment in infants that may have health PPD
and development problems, and it also may help them
negotiate greater levels of investment from others. Recently, A very large body of literature has addressed risk factors for
Hahn-Holbrook and Haselton (44) proposed a “mismatch PPD based on cross-sectional and prospective studies (45).
hypothesis” of PPD that dramatic cultural changes that According to several meta-analyses of risk factors for PPD (15,
have occurred over the past century leads to significant 16), risk factors can be categorized based on the strength of their
divergences from the typical lifestyles throughout human association with PPD. Depression and anxiety during pregnancy,
evolutionary history, and gives rise to the current high postpartum blues, history of depression, neuroticism, stressful
incidence rate of PPD, suggesting that PPD may be a “disease life events, poor marital relationship, and poor social support,
of civilization.” low self-esteem, as well as some cognitive emotion regulation
strategies (61) have been found to have a strong or moderately
strong association. A large-scale population based study (62),
TREATMENT OF PPD using data on more than 700,000 deliveries in Sweden between
1997 and 2008, reported that the risk of PPD for women with
Psychological treatment usually happens in the form of
a depression history was over 20 times higher than women
counseling (or psychotherapy), either one-on-one with a
without. On the other hand, low socioeconomic status (SES),
psychologist or in a group setting (45), and has been extremely
single marital status, unwanted pregnancy, obstetrical stressors,
beneficial for many women. Some women can effectively recover
and difficult infant temperament (63, 64) have been reported
from their depression through counseling alone, while others
to exhibit a relatively weaker association. Maternal attitudes
may have to undergo counseling in conjunction with the use of
(65), women’s experience of a various related complications
antidepressants. Thus far, strong support can be found in the
such preterm birth, prenatal hospitalization, emergency cesarean
literature that a variety of psychological treatments of PPD can
section, pre-eclampsia, and poor infant health (66), can also cause
be effective.
an elevated risk of developing PPD (67–69). These risk factors are
Medical treatment for PPD includes pharmacotherapy with
more closely related to the social and psychological aspects rather
antidepressants. In fact, antidepressant medication is the most
than biological aspects.
common treatment for PPD (46). There have been extensive
investigations of a broad spectrum of antidepressants in the
treatment of PPD, and many have been found to be associated BIOLOGICAL PREDICTORS AND
with symptomatic improvement (46), and to be as effective as BIOMARKERS FOR PPD
usual care plus counseling (47). Recently, an allopregnanolone-
based treatment for PPD, brexanolone, now commercially In recent years, more development has been made to identify
R
called Zulresso , has been approved by the Food and Drug the biological predictors for PPD. Substantial biological changes
Administration (FDA) in the United States as a fast-acting, can be associated with pregnancy. Such changes are necessary
long-lasting antidepressant (48). In double-blind, randomized, in order to maintain normal pregnancy and fetal development,
controlled clinical studies, brexanolone injection has been as well as successful labor and lactation. Upon parturition, the
observed to be associated with a rapid reduction at 60 h in intricate balance that has developed during gestation to sustain
depressive symptoms compared to placebo, and the reduction the maternal-placental-fetal unit is suddenly no longer needed.
could sustain no <30 days with broad responses (49). Furthermore, the maternal system has to undergo a dramatic
Regarding the drug exposure of the infant through biological changes into the lactation phase within a short time. It
breastfeeding, a few reviews (50–53) concluded that may days or even months to re-establish a new biological balance.
nortriptyline, paroxetine, and sertraline have the strongest It is conceivable that failure to re-establish the balance properly
safety performance during lactation. and promptly may cause maternal mental health issues in return.
It should be noted that PPD may share some common genetic
and biological risk factors and biomarkers with other mood Genetic and Epigenetic Studies
disorders such as the major depressive disorder (MDD) and the It is generally hoped that investment of possible genetic causes
bipolar disorder (BD). In this sense, PPD may be viewed as a of a psychiatric disorder may help to reveal the underlying

Frontiers in Psychiatry | www.frontiersin.org 3 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

pathophysiological mechanism, which can help to find cures or may be associated with higher risk of developing PPD. However,
improved treatments. Studies revealed that there may exist an the long allele variant of the 5-HTTLPR (serotonin transporter,
underlying genetic cause for PPD (70). Viktorin et al. (71) found i.e., 5-HTT, linked polymorphic region) was also reported to be
that the heritability of perinatal depression was estimated at 54 associated with PPD symptoms at 6 weeks (80, 81) or within 1
and 44%, respectively, in twin and sibling samples, which means year post-partum (82).
that about half of the variability in perinatal depression can be A weak link between estrogen receptor gene (ESR1) and PPD
explained by genetic factors. This is substantially higher than the has been reported (76, 83), which, however, failed to remain
heritability of non-perinatal depression at 32%. Forty et al. (72) statistically significant when corrections for multiple tests were
and Murphy-Eberenz et al. (73) reported that PPD with onset taken into account. It was also suggested that the role for ESR1
within 4 weeks post-parturm exhibits familiality in families with in the etiology of PPD could possibly be mediated through
MDD. These studies suggest that, while it also has its own unique the modulation of serotonin signaling (83). Mehta et al. (84)
features, the genetic basis for PPD may partially overlap with that found that women with PPD (with onset within 7 weeks after
for other mood disorders. delivery) displayed an increased sensitivity to estrogen signaling
Unlike MDD, there have been relatively fewer studies in comparison with controls. While these studies indeed support
addressing genetic contribution to PPD. A partial summary of the idea that estrogen plays a role in the development of PPD,
genetic association studies of PPD was tabulated by Payne (70). they are far from being conclusive, partly because of the relatively
In these studies, various genes were investigated for their roles small sample size in these studies.
in PPD symptoms, such as those associated with the regulation Catechol-Omethyltransferase (COMT) and monoamine
of the HPA axis, sex hormones, and the effects of stress on the oxidase-A (MAO-A) are allelic gene variations in the
prefrontal cortex (1). Mahon et al. (74) examined the genetic monoaminergic system. They have been shown to be related to
etiology of postpartum mood disorders using genome-wide data. MDD. The former is involved in dopamine and noradrenalin
They found that genetic variations on chromosomes 1 and 9 metabolism, while the latter plays a role in the degradation of
may increase susceptibility to postpartum mood symptoms, for serotonin and noradrenaline in the brain (85). A few studies
women who had a history of pregnancy and any best-estimate revealed that MAO-A and COMT may be associated with PPDs
mood disorder diagnosis. Specifically, the genes HMCN1 and that have an onset within 8 weeks postpartum (75, 81, 86).
METTL13 may contain polymorphisms that confer susceptibility Sacher et al. (87) found an association between PPD and greater
to postpartum mood symptoms. Nevertheless, these associations MAO-A VT (an index of MAO-A density) in the prefrontal and
were not significant enough to sustain statistical corrections from anterior cingulate cortex, compared to healthy controls.
multiple testing. Alvim-Soares et al. (75) found that a HMCN1 Oxytocin has been found to plays an important role in
polymorphism (rs2891230) is associated with PPD symptoms, physiological and genetic systems that permit the evolution
and the heterozygosity for this single nucleotide polymorphism of the human nervous system and allow the expression of
(SNP) was associated with an increased risk of PPD, in a contemporary human sociality, and stress reactivity. It acts to
sample of 110 randomly selected, unrelated Brazilian women of allow the facilitation of birth, lactation, and maternal behavior
European descent, assessed at 8 weeks postpartum. Their result (88, 89). Decrease in the oxytocin level in plasma has been
seems to support the finding of Mahon et al., however, future associated with PPD (90, 91). Possible roles in PPD have also
studies with a larger sample size are certainly needed. Costas been investigated for polymorphisms of the oxytocin receptor
et al. (76) reported a significant association (with p = 0.002) (OXTR) gene, the oxytocin peptide gene (92), the glucocorticoid
between the SNP rs11924390 between SNP at the transcriptional receptor gene and the CRH receptor 1 gene (93). However, only
start site of kininogen 1 and PPD during the first 32 weeks after some weak, suggestive links have been reported. Jonas et al.
delivery. Clear signatures of gene expression of mononuclear cells (94) found that polymorphisms in OXT rs2740210 interacted
were found in woman with PPD symptoms as compared with with early life adversity to predict PPD. Bell et al. (95) found
healthy controls (77, 78). The usefulness of this study, however, that for women who do not show depression during pregnancy,
suffered from its small sample size. Further studies with a larger but possess the rs53576_GG genotype and exhibit high levels of
sample size are warranted in order to confirm these associations. methylation in OXTR, the risk of developing PPD was nearly
Nonetheless, while these results differ, they are not inconsistent three times that for women of lower methylation levels. Kimmel
with one another. et al. (96) found that the cytosine-guanines (CpGs) located
The serotonin transporter gene (SER T) has been one of the on chr3 at positions 8810078 and 8810069 were associated
most widely studied candidate gene associated with PPD (70). It with PPD scores significantly for a cohort of 240 women
has two primary polymorphisms, 5-HTTLPR and STin2VNTR without a psychiatric history. They also found a PPD specific
(10). The former contains a 44-bp deletion or insertion in negative correlation between DNA methylation in the region
the promoter region, corresponding to the short and long and serum estradiol levels. In addition, estradiol levels and
allele variants, respectively. The latter polymorphism involves a OXTR DNA methylation exhibited a significant interaction
variable number of tandem repeats (VNTR) in the second intron, to associate with the ratio of allopregnanolone (ALLO) to
among which the longer VNTRs have been found to be associated progesterone. Recently, King et al. (97) found that mothers
with mental health issues and depressive disorders. So far, studies with persistent perinatal depression (depressive symptoms both
have shown mixed results on the role of SER T polymorphisms prenatally and postpartum) exhibited significantly higher overall
in PPD. Lesch and Mössner (79) found that the short allele OXTR methylation at 16/22 individual CpG sites. While these

Frontiers in Psychiatry | www.frontiersin.org 4 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

studies seem to support the link between (the epigenetic DNA been mostly on DNA methylation, which can be modified by
methylation of) OXTA and the risk of developing PPD, it should medication and stress, as well as reproductive hormones. There
be noted that these findings are mostly derived from small has been some progress in the identification of biomarkers for
sample sizes, with variable depression rating scales and a lack of DNA methylation. Recent studies have implicated epigenetic
prospective measures of DNA methylation, and thus should be processes in the pathophysiology of MDD (108). Guintivano
interpreted with caution. et al. (109) and Kaminsky and Payne (110) observed enhanced
The brain-derived neurotrophic factor (BDNF) system is sensitivity to estrogen-based DNA methylation reprogramming
known to play an important role in many neuronal functions in those at risk for PPD and identified two potential biomarker
(15). Serum BDNF was found to decline considerably across loci at the HP1BP3 and TTC9B genes that predicted PPD.
pregnancy from 1st through 3rd trimesters (p ≤ 0.008) and Using blood drawn during pregnancy, DNA methylation at two
subsequently to increase at postpartum (p < 0.001), and lower genomic locations with an area under the receiver operator
serum BDNF in late pregnancy was reported to be associated characteristic (ROC) curve [area under the curve (AUC)] of
with higher depressive symptoms (98), although low BDNF levels 0.87 in antenatally euthymic women and 0.12 in a replication
were found to persist even 2 months after birth (99). Aydemir sample of antenatally depressed women, along with complete
et al. (100) and Gazal et al. (101) found that the BDNF levels blood count data, produced an AUC of 0.96 across both
in serum of PPD patients were lower than in healthy control prepartum depressed and euthymic women (109). Osborne et
subjects. Serum BDNF levels in PPD patients that had a suicide al. (111) found that TTC9B and HP1BP3 DNA methylation
risk were significantly lower than those of women that did not in early antenatal stage showed moderate association with
show a suicide risk (102). Recently, Fung et al. (103) found the change in estradiol and ALLO levels over the course of
an association between lower levels of serum BDNF in early pregnancy, suggesting that epigenetic variation at these loci
pregnancy and antepartum depression. However, Figueira et al. may be important for mediating hormonal sensitivity, and that
(104) and Comasco et al. (80) did not find an association between PPD is mediated by differential gene expression and epigenetic
PPD and the BDNF polymorphism Val66Met. Overall, clear sensitivity to pregnancy hormones and thus modeling proxies of
evidence for the association between BDNF and PPD symptoms this sensitivity may enable accurate prediction of PPD. Osborne
is yet to be found (10). This may be partly due to the fact that et al. (38) further found an association between lower ALLO
the normal serum level of BDNF is a non-monotonic function levels in the second trimester of pregnancy and an elevated
of time in the perinatal period. Thus the precising timing and risk of developing PPD, which seem to have been confirmed by
its dynamics may be important. The lack of consensus in the latest studies (39, 49, 112–114), and can be traced back to the
literature may partly reflect the difference in experiment design two genes identified above. Indeed, this seems to have gathered
in terms of timing and sampling strategy, besides the often small more support than many other biomarkers. Very recently,
sample sizes. Payne et al. (115) found that antenatal TTC9B and HP1BP3
Katz et al. (105) collected maternal RNA longitudinally from DNA methylation may be used to predict both antenatal and
preconception through the third trimester of pregnancy in 106 postpartum depression.
women with a lifetime history of mood or anxiety disorders. They A prospective study (109) of 93 pregnant women with a
reported that mRNA expression of a number of glucocorticoid history of either MDD or bipolar disorder found significant
receptor (GR)-complex regulating genes was up-regulated over correlation between PPD risk and 17β-estradiol (E2)-induced
pregnancy, and women with depressive symptoms showed DNA methylation change, suggesting that an enhanced
significantly smaller increases in mRNA expression of four of sensitivity to estrogen-based DNA methylation reprogramming
these genes. They also found that GR sensitivity diminished exists in women at risk for PPD. Estradiol increases the rate of
with increasing maternal depressive symptoms. A prospective transcription of the OXTR gene (116), resulting in elevations
pregnancy cohort study of 56 healthy women with singleton of oxytocin levels in the uterus (117) and in numerous brain
term pregnancies found that altered placental genes expression regions (118) while heterozygosity for the OXTR rs2254298
involved in glucocorticoid and serotonin transfer may function polymorphism can interact with early life adversity to yield
as potential gestational-age-specific marker of PPD risk (106). the highest levels of symptoms of depression, physical anxiety,
Further studies with a larger sample size are needed to replicate and social anxiety (119). Association between higher DNA
these findings. methylation of the OXTR gene and decreased expression of
Recently, by RNA sequencing the whole transcriptomes of the gene was also observed (120). A case control study (95)
peripheral blood mononuclear cells, Pan et al. (107) found that on the OXTR gene DNA methylation at CpG site-934 and
PPD was positively correlated with multiple genes involved in genotype rs53576 and rs2254298 found that women with GG
energy metabolism, neurodegenerative diseases and immune genotype had higher risk of developing PPD with increasing
response, while negatively correlated with multiple genes in methylation level. The finding of King et al. (97) regarding
mismatch repair and cancer-related pathways. In addition, OXTR methylation seem to suggest that the onset timing of PPD
genes associated with appetite regulation and nutrient response is also an important factor.
were differentially expressed between PPD (n = 56) and Overall, the above findings suggest that polymorphic
control subjects (n = 27). variations in candidate genes within the monoaminergic system
Epigenetics refers to changes in gene function that do not can have an effect on the estrogen receptor, the oxytocin peptide,
alter the DNA sequence itself. The main focus in this area has the glucocorticoid receptor, and the CRH receptor 1 genes, and

Frontiers in Psychiatry | www.frontiersin.org 5 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

may act as potential biomarkers for PPD. However, further lactation periods, especially how they respond to the onset and
investigation is needed in order to determine whether the short development of PPD symptoms. It has been suggested that
or long allele of the 5-HTTLPR is associated with PPD risk and lower levels of oxytocin in both the gestation and postpartum
under what conditions. Nevertheless, the lack of consensus in periods may imply an elevated risk for developing PPD (9,
these data from the literature highlights the complex relationship 90, 91). Jobst et al. (133) found that plasma oxytocin levels
between epigenetics and PPD related neuroendocrine changes. significantly increased from week 35 of gestation to 6 months
postpartum in all women. However, levels decreased from the
Reproductive hormones 38th week of gestation to 2 days after delivery in women with
The association between the reproductive hormones and PPD, whereas, they increased continuously in the healthy control
PPD has been studied and reviewed, in terms of estrogens, group. This suggests that the time evolution pattern of oxytocin
progesterone, prolactin, oxytocin, and testosterone. Bloch et al. may be a predictor of PPD in the immediate postpartum period
(121) found evidence that the reproductive hormones estrogen (within 2 weeks). In comparison, Massey et al. (134) found
and progesterone play a role in the development of PPD. that oxytocin level interacted with past MDD to predict PPD
However, the data by Klier et al. (122) did not support the symptom severity in the third trimester; a higher oxytocin level
hypothesis of a role of sex hormones in the etiology of PPD. predicted greater PPD symptom severity in women with past
A review of about 200 studies by Serati et al. (123) found MDD, but not in women without. Thul et al. (135) reviewed
little evidence that supports estrogen withdrawal theories, or the literature on the relationship between both endogenous
suggests that progesterone in late pregnancy or postpartum and synthetic oxytocin and PPD, and found that out of the
period predicts PPD symptoms. 12 studies that focused on endogenous oxytocin, eight studies
Recent studies seem to suggest strong association between suggested an inverse correlation between plasma oxytocin levels
the progesterone level and PPD. The levels of ALLO was and depressive symptoms.
found to increase progressively throughout gestation, and drop There is mixed evidence for the association between
rapidly upon parturition (37, 124, 125). As a metabolite of testosterone levels in late pregnancy and PPD symptoms in
progesterone, ALLO is a neuroactive steroid measurable in the early postpartum stage. Women with PPD symptoms were
peripheral circulation. Therefore, its levels vary proportionally reported to have higher serum testosterone levels in the late
with progesterone levels throughout gestation and after the third trimester (136) and around 24 h postpartum than healthy
delivery, and an association between ALLO and PPD were controls (137). However, other studies have failed to find such an
found, suggesting that hormonal regulation plays an important association (138).
role in the development of PPD (124). Previously, Bloch et al. The role of thyroid hormone in the development of perinatal
(121) found that women with a PPD history are more sensitive mood disorder has also been investigated (139). It has been
to mood-destabilizing effects of gonadal steroids than healthy suggested that timing may be critical in the correlation between
controls. Rather than progesterone withdrawal upon delivery, it thyroid hormones and PPD, since the function of thyroid is
has been found that a low ALLO level during pregnancy predicts to respond to the constant changes in other hormones across
PPD (38, 112, 114). Such an association was also found in earlier gestation. Kuijpens et al. (140) reported that the presence
studies (126, 127). Timing may be an important factor when of thyroperoxidase antibody (TPOAb) during gestation was
assessing potential biomarkers. Epperson et al. (128) reported associated with the occurrence of subsequent depression during
that cortex GABA levels and plasma ALLO concentrations were the postpartum period and as such can be regarded as a marker
reduced in two different groups of postpartum women, regardless for depression. Elevations in thyroid stimulating hormone (TSH)
of PPD diagnosis at 9 weeks or 6 months postpartum, compared upon delivery have been proposed to be a predictor for PPD 6
to healthy follicular phase women, and that no correlation was month post-parturm (141). However, in an earlier large cohort
found between cortical GABA concentrations and estradiol, study, Albacar et al. (142) examined 1,053 postpartum Spanish
progesterone, or ALLO levels. Smith et al. (129) found that women without a previous history of depression, and concluded
the effect of neuroactive steroids on inhibition, which influence that thyroid function at 48 h after delivery does not predict PPD
anxiety state and seizure susceptibility, depends not only on the susceptibility. Groer and Vaughan (143) found that pregnant
subunit composition of the receptor but also on the direction of TPO-positive women were more likely to develop PPD 6 months
Cl− current generated by these target receptors. after delivery. A review (144) suggested that TPOAb in early
Prolactin has physiological functions that are especially to mid-pregnancy was associated with concurrent depression
relevant in the peripartum period, and may act as an attenuation and may be predictive of PPD. Recently, Wesseloo et al. (145)
for behavioral and neuroendocrine stress responses during both found that women with an increased TPOAb titer during
pregnancy and lactation (130). Despite mixed evidence from early gestation were at increased risk for self-reported first-
different studies, two studies with a big sample size indeed onset depression, which suggested an overlap in the etiology of
suggested an negative correlation between PPD and prolactin first-onset PPD and autoimmune thyroid dysfunction. Li et al.
(131, 132). (146) found that PPD patients showed elevated serum levels of
The oxytocin signaling network has been of great interest triiodothyronine, thyroxine, free triiodothyronine, free thyroxine
as it can play an important role in mother-infant bonding along with diminished estradiol, progesterone, and TSH levels. It
and interactions. Recently, there have been investigations on was proposed that it may not just be thyroid hormones alone,
the trajectories of oxytocin throughout the gestation and but rather thyroid in conjunction with other factors such as

Frontiers in Psychiatry | www.frontiersin.org 6 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

estrogens (10) or trauma history (147), that are implicated in PPD (157). Labad et al. (158) found that women with postpartum
etiology. A consensus regarding the role of thyroid hormones as thoughts of harming the infant had higher ACTH levels, when
a biomarker is yet to be reached. compared to those women without intrusive thoughts, and a
While strong evidence seems to have been found to support dysregulation of the HPA axis may play a role in the etiology
that a low ALLO level during pregnancy predicts PPD, it of postpartum thoughts of harming the infant. Jolley et al. (159)
can be seen, however, that for most reproductive hormones, found higher ACTH and lower cortisol levels in women with PPD
a consensus regarding their association with PPD is yet to at 6 and 12 weeks postpartum when compared with controls.
be reached. Many studies are limited by small sample sizes. Women with PPD were found to have greater ACTH stress
Furthermore, the dynamical changes of reproductive hormones reactivity to cold pressor test (CPT), and a significantly elevated
during the perinatal period should also be taken into account. ACTH concentration level at 8 weeks postpartum in response to
This will require that the timing for sampling and depression CPT (160), as well as a markedly blunted plasma ACTH response
assessment be arranged in a more systematic and consistent to serial ovine CRH tests at 3, 6, and 12 weeks postpartum
manner, hopefully across different studies. (161), and it was suggested that the suppressed ACTH response
to ovine CRH might serve as a biochemical marker of the
Stress Hormones postpartum “blues” or depression (161). Together, these findings
Stress hormones, particularly those of the HPA axis, have been again suggested that dysregulation of the HPA axis may be
implicated in non-puerperal depression (148). It was suggested associated with PPD. However, a comparative study found no
that the hyporesponsiveness of the HPA axis may persist for differences in the HPA axis reactivity in terms of the cortisol and
several months postpartum (149). There is strong evidence that ACTH response, e.g., the cortisol/ACTH ratio, to pharmacologic
the HPA axis plays an important role in perinatal depression, test and psychological challenges during the luteal phase between
both during gestation and postpartum. Groer and Morgan (131) current euthymic postpartum women with a history of either
reported that depressed mothers had a down-regulated HPA axis, PPD or MDD and controls (162). We note that the sample sizes
in that the salivary cortisol level was lower in PPD patients than for the last four studies were fairly small, with 12, 34, 17, and 15
in healthy controls. Similar to reproductive hormones, the levels × 3 (15 in each group), respectively. Further investigations with
of stress hormones also increase during pregnancy from the first larger sample sizes and better designs are needed to reconcile
to third trimester and then decrease abruptly upon parturition. these findings.
In contrast, the neuropeptide corticotropin-releasing hormone As for β-endorphin, Yim et al. (163) found that among women
(CRH) often increases exponentially in the process of pregnancy who were euthymic at 25 weeks’ GA, those who developed
(150), mainly because CRH is also produced by the placenta PPD had higher β-endorphin levels throughout pregnancy than
(151). This also leads to an increased level of adrenocorticotropic women without PPD symptoms, and suggested that β-endorphin
hormone (ACTH) and cortisol over the course of pregnancy may play an important role in the pathophysiology of PPD and
(152). The CRH level drops quickly upon parturition, when the may thus be a useful early predictor of PPD symptoms in women
placenta is discharged. Yim et al. (153) found that at a critical who show no depressive symptoms in mid-pregnancy. They also
period in midpregnancy, placental CRH (pCRH) is a sensitive reported that around 25 weeks’ gestation was a crucial time for
and specific early diagnostic test for PPD; at 25 weeks’ GA, assessing PPD symptoms. However, postpartum blood samples
pCRH was a strong predictor of PPD, and the trajectories of were taken only at 9 weeks for assessment of β-endorphin levels,
pCRH in women with PPD are significantly accelerated from and self-report was used to evaluate the depressive symptoms.
23 to 26 weeks’ GA. Hahn-Holbrook et al. (154) found that Parcells (164) found that cortisol levels directly correlated
steeper increases in placental CRH from 29 to 37 weeks’ gestation with maternal depression, anxiety, and stress. Recent studies
predicted more depressive symptoms postpartum. Iliadis et al. found that women with PPD had higher salivary evening cortisol
(155) reported an association between high CRH levels in at 6 weeks postpartum (165), elevated hair cortisol levels in
gestational week 17 and the development of PPD symptoms, the first to third trimesters (166), or lower levels of evening
among women without depressive symptoms during pregnancy. cortisol in the immediate peripartum period (167), compared to
On the contrary, Meltzer-Brody et al. (34) found that higher mid- healthy controls. Corwin et al. (168) found that cortisol levels,
pregnancy placental CRH was not associated with an increased together with family history of depression and interleukin (IL)-
risk of PPD. Glynn and Sandman (156) showed that depressive 8/IL-10 ratio, were significant predictors of PPD symptoms.
symptoms at 3 months postpartum were associated with elevated However, a recent survey (169) showed that most studies
mid-gestational pCRH levels and also accelerated trajectories reported no association between maternal cortisol and antenatal
of pCRH, but pCRH was not predictive of PPD at 6 months depression, and that among studies that reported an association,
postpartum. They concluded that elevated pCRH level during second-trimester and third-trimester cortisol assessments more
pregnancy may act as a marker of risk of developing PPD. consistently reported an association. The link between cortisol
Overall, the result on CRH as a viable biomarker is mixed so levels and postpartum or perinatal depressions is far from
far, possibly due to limitations of small sample sizes and different conclusive, and thus more future investigations are warranted.
timing for sampling and assessment of the PPD symptoms. Among stress hormones, it is less controversial that alteration
Alteration in the HPA axis is a robust biomarker of anxiety and of the HPA axis is a robust biomarker for PPD. However, results
depression, and significant mood symptoms in pregnancy was on CRH, ACTH and cortisol levels are rather mixed, and further
shown to be associated with altered diurnal cortisol in pregnancy studies are needed on these hormones as well as β-endorphin.

Frontiers in Psychiatry | www.frontiersin.org 7 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

Immunological/Inflammatory Studies was not found between serum IL-6 levels at delivery and
The function of the immune system is to protect the body from later PPD symptoms.
foreign substances (as well as other pathogenic organisms) (10). Other proinflammatory cytokines (and their receptors) have
However, the fetus should not be attacked during pregnancy also been investigated, as potential biomarkers for PPD. Groer
even though it is necessarily genetically distinct and carries and Morgan (131) found that women with PPD had lower serum
paternal antigens that are foreign to the maternal immune levels of Interferon-gamma (IFN-γ) and a lower IFN-γ/IL-10
system. Thus this constitutes a big challenge for the maternal ratio in both serum and in whole blood stimulated cultures.
immune system. Although, it is not yet clear how it maintains Fransson et al. (182) found that mothers with depression had
the proper balance of proinflammatory cytokines [e.g., IL-6, IL- higher TGF-β2 concentrations in their breast milk than mothers
1β, tumor necrosis factor-alpha (TNF-α)] and anti-inflammatory without depression. They also found associations between
cytokines (e.g., IL-10), it can be expected that the immune maternal IL-6, IL-8 and cord IL-6, IL-8, IL-10, IL-13, and IL-18
system will have more activities during the perinatal period. A levels and depressive symptoms in the first 5 days postpartum
growing body of literature suggests that inflammatory responses in women who delivered preterm. Clara Cell Protein (CC16),
have an important role in the pathophysiology of depression an endogenous anticytokine, may also be related to PPD. Maes
(170). In addition, prolonged HPA axis hyperactivity activated by et al. (183) found that parturients who developed PPD had
proinflammatory cytokines (IL-1, IL-6, and TNF-α) is one of the significantly lower serum CC16 concentrations than women who
mechanisms underlying cytokine-induced depression (171), even did not. Bränn et al. (184) found that among 70 inflammatory
in perinatal episodes (172). markers, five were significantly elevated in women with
It has been found that puerperal women usually have PPD, including TNF ligand superfamily member (TRANCE),
significantly higher levels of proinflammatory cytokines during hepatocyte growth factor (HGF), IL-18, fibroblast growth factor
the last trimester of pregnancy, and are also at higher risk for 23 (FGF 23), and C-X-C motif chemokine 1 (CXCL1).
depression (173). Other stressors also cause proinflammatory Mixed evidence has been found for the linkage between the
cytokine levels to rise at such a time. Breastfeeding may attenuate C-reactive protein (CRP) and PPD. The review by Lambert and
stress and modulate the inflammatory response. Overall, the Gressier (185) suggested that the dosage of some inflammation
proinflammatory state during the late pregnancy and the early biomarkers, including CRP, at the very end of pregnancy or
postpartum period plays an important role in the development immediately after delivery could predict PPD. Bränn et al.
of PPD (173). IL-6 is the most commonly studied cytokine, and (186) found that the signal transducing adaptor molecule-
has been consistently identified as being elevated in depression. binding protein (STAM-BP), axin-1, adenosine deaminase
An earlier study (174) found that the levels of serum IL-6 (ADA), sulfotransferase 1A1 (ST1A1), and IL-10 were lower
and its receptor (IL-6R) were significantly higher in the early in late pregnancy among women with PPD, and proposed a
puerperium than before delivery, and women who developed summary inflammation variable for predicting PPD. So far,
depressive symptoms in the early puerperium had significantly further studies are needed to address inconsistencies in different
higher serum IL-6 and IL-6R concentrations than those without. results regarding the role of inflammatory processes in the
Corwin et al. (175) reported an increase in IL-1β on Day 14 development of PPD.
and Day 28 postpartum in women with PPD, compared to Overall, evidence for (anti-)proinflammatory cytokines as a
levels in euthymic women, suggesting an association between diagnostic and predictive biomarker for PPD is mixed, which
symptoms of PPD and elevated levels of IL-1β during the first calls for more systematic studies in the future.
month postpartum. Boufidou et al. (176) found that cytokine
IL-6 and TNF-α levels in the cerebrospinal fluid (CSF) and Biomarkers From Biochemical Studies
TNF-α levels in serum were positively associated with depressive Identification of nutritional and biochemical markers for PPD
mood during the first 4 days postpartum and also at sixth diagnosis has gained attention lately. Wójcik et al. (187)
week postpartum. Krause et al. (177) found that regulatory T revealed a correlation between the severity of depressive
cells in pregnancy strongly predicted PPD. Corwin et al. (168) symptoms and decreased serum zinc concentration third
found that family history of depression and cortisol AUC and day after delivery in PPD patients. Roomruangwong et
the IL8/IL10 ratio both on day 14 were significant predictors al. (188) found that lower serum zinc (and higher CRP)
of PPD. Liu et al. (178) found that elevated serum IL-6 at levels strongly predicted prenatal depression and physio-
delivery was associated with development of PPD during the somatic symptoms, which all together predicted postnatal
6 months post partum. In a longitudinal study (179), the IL- depressive symptoms.
6 and IL-10 levels measured in the third trimester were found The role of vitamin D in the development of depression
to be a significant predictor of PPD. However, the evidence is has been studied in recent years, because it has regulatory
mixed for the link between PPD and proinflammatory cytokines functions in the immune system, and thus may act effectively as
(10). For example, serum leptin levels at delivery were found a neurosteroid. Christesen et al. (189) conducted a review on the
earlier (180) to be negatively associated with self-reported impact of vitamin D on pregnancy, and found that a decreased
depression during the first 6 months after delivery, but were vitamin D level during pregnancy may lead to PPD in one study
significantly greater in women with PPD than in healthy controls and preeclampsia in several studies. In an exploratory study
at 3 months post-partum (181). However, a similar association (190), a significant relationship over time was found between low

Frontiers in Psychiatry | www.frontiersin.org 8 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

25-hydroxyvitamin D (25(OH)D) levels and high EPDS scores. Lactobacillus rhamnosus HN001 in pregnancy and postpartum
Brandenbarg et al. (191) found that low early-pregnancy vitamin period reduces the prevalence of PPD.
D status was associated with elevated depressive symptoms in In summary, biochemical studies seem to suggest that low
pregnancy. Gur et al. (192) found that lower maternal 25(OH)D3 serum 25(OH)D levels during pregnancy may be a biomarker
levels were associated with higher levels of PPD at all time points, for PPD. Evidence for other biochemical markers for PPD is
and thus may be a factor affecting the development of PPD. either mixed, very weak or negative, including levels of serum
Similarly, Robinson et al. (193) also reported that low vitamin cholesterol, PUFA status, vitamin B12 and folate, kynurenine
D during pregnancy is a risk factor for the development of level, and gut microbiome.
PPD symptoms. A cross-sectional study (194) found that higher
dietary vitamin D intake was significantly associated with a lower
prevalence of depressive symptoms during pregnancy. A recent Omics-Based Biomarker Studies
prospective study found an association between lower prenatal Biomarker identification in neuropsychiatric disorders such as
log 25(OH)D and significantly more severe PPD symptoms, PPD and MDD can have important advantages and benefits,
among women with higher levels of inflammatory markers (195). in terms of prediction and accurate diagnosis of a disease,
In a controlled study, Fu et al. (196) found an association between and may provide more accurate and reliable information
lower serum 25(OH)D levels measured 24 h after delivery and which can guide the selection and development of a cure or
PPD. Fortunately, these findings seem to be in agreement with treatment. Gadad et al. (210) conducted a review on various
each other. omics approaches for identifying biomarkers of neuropsychiatric
In a prospective cohort study of 238 pregnant women, Teofilo disorders. Many of the biomarkers mentioned in the review
et al. (197) found that HDL-cholesterol concentrations were can be identified using multi-omics, which includes genomics,
inversely associated with EPDS scores during pregnancy. In a epigenomics, transcriptomics, proteomics, metabolomics, and
prospective study of 266 Dutch women, Van Dam et al. (198) did lipidomics. These omics technologies have been actively applied
not find an association between rapid serum cholesterol decline in studies on MDD. See, e.g., the review by Sethi and Brietzke
and the risk of developing PPD. (211). Given the strong similarity between PPD and MDD, these
The PUFA status in late pregnancy was studied in a large omics technologies can in principle be quickly applied to the
sample of women, and only a weak link between PUFA in late identification of biomarkers for PPD.
pregnancy and PPD risk was reported (199). In a community- Metabolomics has recently been applied to unravel the
based prospective cohort, Markhus et al. (200) found that the serum metabolomic profile of PPD (212). Serum metabolomes
DPA content, DHA content, ω-3 index, ω-3/ω-6 ratio, total of a group of women (n = 10) with PPD and a healthy
PUFA score, and the ω-3 PUFA score were all inversely correlated control group (n = 10), all from Greece, were analyzed for
with the EPDS score. Sallis et al. (201) found a weak positive targeted metabolomics using mass spectrometry. In the PPD
correlation between ω-3 fatty acids and PPD. In another study, group, increased levels of five metabolites were found, such
no association between plasma PUFAs and PPD was found (202). as glutathione-disulfide, adenylosuccinate, and ATP. The data
For vitamin B12 and folate, a cross-sectional study reported showed that molecular changes related to PPD were indeed
no association between depressive symptoms and blood levels detectable in peripheral material, and thus these changes may
of vitamin B12 and folate (203). Lewis et al. (204) did not find serve as diagnostic biomarkers.
strong evidence that folic acid supplementation can reduce the A metabolomic profiling of morning urine samples of women
risk of depression during first 8 months of pregnancy. Another with PPD, postpartum women without depression (PPWD),
study found significantly higher homocysteine levels in women and healthy controls (HCs) was recently characterized using
with PPD than in healthy controls, suggesting that the level of gas chromatography-mass spectroscopy (213). Twenty two (22)
serum homocysteine might be a risk biomarker for PPD (205). differential metabolites (14 up regulated and 8 down regulated)
An increased kynurenine level has also been reported to were found to separate PPD subjects from HCs and PPWD.
be associated with the induction of depression. Depressive Meanwhile, a panel of five potential biomarkers – formate,
and anxiety symptoms in the early puerperium are succinate, 1-methylhistidine, α-glucose and dimethylamine –
associated with increased catabolism of tryptophan into was identified, which could be used to effectively distinguish
kynurenine (206), and the increases in plasma kynurenine PPD subjects from HCs and PPWD. Recently, using Liquid
and the kynurenine/tryptophan (K/T) ratio were positively Chromatography Coupled to Quadrupole Time-of-Flight Mass
correlated with the anxiety and depression scores in Spectrometry, Zhang et al. (214) found that the urine
the puerperium. metabolomic profiles of patients with PPD were different from
There have been some studies on gut microbiome as those of HCs. Ten differentiating metabolites were found as main
potential biomarkers recently (207). The coordination between contributors to this difference.
the gut, the central nervous system, and the neuroendocrine So far, there have been far fewer studies on the identification
and neuroimmune axes is referred to as the gut-brain axis of biomarkers for PPD than for MDD. Nevertheless, we
(208). There are not many studies and thus not much has expect that multi-omics technologies will be widely used in
been known, regarding what role the gut-brain axis may play identifying biomarkers for PPD in future studies. Indeed, given
in the development of PPD. In a study of nearly 400 pregnant the enormous advantages of the omics, they should be a major
women (209), it was found that probiotic supplementation with direction of PPD research in the future.

Frontiers in Psychiatry | www.frontiersin.org 9 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

It should be cautioned that, due to possible partial overlap in research. Despite that the application of multi-omics to the study
genetic and biological risk factors and biomarkers among PPD, of PPD has just begun recently, we strongly believe that modern
MDD and BD, a panel of multiple biomarkers may be needed to multi-omics technologies will have a great potential in this arena.
avoid misdiagnosis. While many studies have found associations or correlations
between certain risk factors, predictors, or biomarkers and PPD,
CONCLUSIONS these correlations do not necessarily tell whether these risk
factors, predictors, and markers are consequences or causes
PPD is a serious health issue for new mothers and has of PPD. A correct etiological model which is subject to
negative consequences on both the mothers and the children. comprehensive testing is crucial for uncovering the underlying
Its high prevalence rate raises strong public health concerns. cause of PPD and for developing the right medication and cure
PPD is likely to be influenced by a multitude of risk factors, for PPD.
including biological, psychosocial, and even environmental Finally, we end this review by presenting a brief speculative
factors. There have been various etiological models for PPD. model for the etiology of PPD, based on the findings summarized
However, no consensus has been reached so far. Typical hereinabove. The hormonal withdrawal theory has largely been
treatments for PPD include psychotherapy and phamacotherapy, proved to be wrong, as the hormonal withdrawal is a normal
in the form of psychotherapy/counseling and antidepressant process every pregnant woman has to undergo upon parturition.
medications. While the US FDA has recently approved This is a desired and necessary biological change that is
the first antidepressant medication for PPD, this medicine, adapted to pregnancy and parturition, as a result of human
Zulresso, has turned out to be extremely expensive, and thus evolution. However, for a portion of women the biological
is out of the reach for most PPD patients. This makes system may not perform as perfectly as expected, due to
preventive measures more important, to protect one from the high complexity of human body and a wide range of
developing PPD. genetic and epigenetic variations, as well as environmental,
There has been increasing effort in the diagnosis of PPD psychosocial and biological factors. In other words, for those the
using predictors, from both psychosocial and biological aspects. body does not fulfill and cope with the hormonal withdrawal
Furthermore, there are a variety of researches on PPD biomarkers in a perfect manner, depressive disorders to various degrees
so far, using different methods and approaches for characterizing may develop. Thus, the ultimate goal of studying various
and assessing PPD. While biomarker identification has shown predictors and biomarkers for PPD will be to catch such an
a lot of promise for PPD research, nevertheless no biomarker imperfection at an early stage so that it can be remedied or
is ready for clinical use as of today. From etiologic point of prevented in time.
view, (epi)genetics and hormones may play a more fundamental
role than biochemicals in the development of PPD. Nonetheless, AUTHOR CONTRIBUTIONS
biochemicals may as well be the right signatures or indicators
that can be used for diagnosing and predicting PPD. Biomarkers YY completed the literature survey and manuscript writing.
in genetics and epigenetics may have a big potential for J-CL initiated and oversaw the project. H-FL, JC, Z-BL, Y-SH,
personal risk prediction. Several hormones, neurosteroids, and and J-XC participated in the discussion and the manuscript
biochemicals have been identified in preliminary studies as preparation. All authors contributed to the article and approved
potential biomarkers for predicting PPD, but further studies the submitted version.
and substantiation are needed before they can be put into
clinical use. So far, there are strong inconsistencies in various FUNDING
findings regarding predictors and biomarkers of PPD. These
inconsistencies presumably have to do with the limited sample This work was supported in part by the Science
sizes, inconsistent depressive rating scales and timing for Technology Department of Zhejiang Province,
sampling, and inconsistent designs across different studies, China (Project for Applications of Technology,
as well as the high complexity of PPD. Further large-scale, Grant No. 2017C37004), National Natural Science
integrative studies are needed to fully understand PPD. Given Foundation of China (Grant No. 81772266),
the objectiveness of biomarkers, we expect that the identification and Guangzhou Science and Technology Project
of biomarkers of PPD will be an important subject in future (Grant No. 201804010369).

REFERENCES International Publishing. (2019) p. 155–70. doi: 10.1007/978-3-319-91

557-9_9
1. Yim IS, Schetter CD. Biopsychosocial predictors of perinatal depressive 3. Grace SL, Evindar A, Stewart DE. The effect of postpartum
symptoms: moving toward an integrative approach. Biol Psychol. (2019) depression on child cognitive development and behavior: a
147:107720. doi: 10.1016/j.biopsycho.2019.107720 review and critical analysis of the literature. Arch Womens.
2. Serati M, Carnevali G. Perinatal depression. In: Clinical Cases in Psychiatry: Ment Health. (2003) 6:263–74. doi: 10.1007/s00737-003-
Integrating Translational Neuroscience Approaches. Cham: Springer 0024-6

Frontiers in Psychiatry | www.frontiersin.org 10 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

4. Dennis CLC-L, McQueen K. The relationship between infant-feeding 27. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry.
outcomes and postpartum depression: a qualitative systematic review. (1960) 23:56–62. doi: 10.1136/jnnp.23.1.56
Pediatrics. (2009) 123:e736–51. doi: 10.1542/peds.2008-1629 28. Sharp R. The Hamilton rating scale for depression.
5. Blum LD. Psychodynamics of postpartum depression. Psychoanal Psychol. Occup Med. (2015) 65:340. doi: 10.1093/occmed/
(2007) 24:45–62. doi: 10.1037/0736-9735.24.1.45 kqv043
6. Yawn BP, Olson AL, Bertram S, Pace W, Wollan P, Dietrich AJ. Postpartum 29. Nassir Ghaemi S, Miller CJ, Berv DA, Klugman J, Rosenquist KJ, Pies RW.
depression: screening, diagnosis, and management programs 2000 through Sensitivity and specificity of a new bipolar spectrum diagnostic scale. J Affect
2010. Depress Res Treat. (2012) 2012:363964. doi: 10.1155/2012/363964 Disord. (2005) 84:273–7. doi: 10.1016/S0165-0327(03)00196-4
7. O’Hara MW, McCabe JE. Postpartum depression: current 30. Douma SL, Husband C, O’Donnell ME, Barwin BN, Woodend
status and future directions. Annu Rev Clin Psychol. (2013) AK. Estrogen-related mood disorders. Adv Nurs Sci. (2005)
9:379–407. doi: 10.1146/annurev-clinpsy-050212-185612 28:364–75. doi: 10.1097/00012272-200510000-00008
8. Anderson G, Maes M. Postpartum depression: psychoneuroimmunological 31. Glynn LM, Davis EP, Sandman CA. New insights into the role of perinatal
underpinnings and treatment. Neuropsychiatr Dis Treat. (2013) 9:277– HPA-axis dysregulation in postpartum depression. Neuropeptides. (2013)
87. doi: 10.2147/NDT.S25320 47:363–70. doi: 10.1016/j.npep.2013.10.007
9. Kim S, Soeken TA, Cromer SJ, Martinez SR, Hardy LR, Strathearn L. 32. O’Hara MW, Schlechte JA, Lewis DA, Varner MW. Controlled
Oxytocin and postpartum depression: delivering on what’s known and what’s prospective study of postpartum mood disorders: psychological,
not. Brain Res. (2014) 1580:219–32. doi: 10.1016/j.brainres.2013.11.009 environmental, and hormonal variables. J Abnorm Psychol. (1991)
10. Yim IS, Tanner Stapleton LR, Guardino CM, Hahn-Holbrook J, Dunkel 100:63–73. doi: 10.1037/0021-843X.100.1.63
Schetter C. Biological and psychosocial predictors of postpartum depression: 33. Workman JL, Barha CK, Galea LAM. Endocrine substrates of cognitive and
systematic review and call for integration. Annu Rev Clin Psychol. (2015) affective changes during pregnancy and postpartum. Behav Neurosci. (2012)
11:99–137. doi: 10.1146/annurev-clinpsy-101414-020426 126:54–72. doi: 10.1037/a0025538
11. Bloch M, Daly RC, Rubinow DR. Endocrine factors in the 34. Meltzer-Brody S, Stuebe A, Dole N, Savitz D, Rubinow D, Thorp J. Elevated
etiology of postpartum depression. Compr Psychiatry. (2003) Corticotropin Releasing Hormone (CRH) during pregnancy and risk of
44:234–46. doi: 10.1016/S0010-440X(03)00034-8 Postpartum Depression (PPD). J Clin Endocrinol Metab. (2011) 96:40–
12. Brummelte S, Galea LAM. Depression during pregnancy and postpartum: 7. doi: 10.1210/jc.2010-0978
contribution of stress and ovarian hormones. Prog Neuro-Psychopharmacol 35. Byrnes EM, Byrnes JJ, Bridges RS. Increased sensitivity of dopamine systems
Biol Psychiatry. (2010) 34:766–76. doi: 10.1016/j.pnpbp.2009.09.006 following reproductive experience in rats. Pharmacol Biochem Behav. (2001)
13. Corwin EJ, Kohen R, Jarrett M, Stafford B. The heritability of postpartum 68:481–9. doi: 10.1016/S0091-3057(01)00449-X
depression. Biol Res Nurs. (2010) 12:73–83. doi: 10.1177/1099800410362112 36. Licheri V, Talani G, Gorule AA, Mostallino MC, Biggio G, Sanna E. Plasticity
14. Skalkidou A, Hellgren C, Comasco E, Sylvén S, Sundström Poromaa I. of GABAA receptors during pregnancy and postpartum period: from gene to
Biological aspects of postpartum depression. Womens Heal. (2012) 8:659– function. Neural Plast. (2015) 2015:1–11. doi: 10.1155/2015/170435
72. doi: 10.2217/WHE.12.55 37. Pennell KD, Woodin MA, Pennell PB. Quantification of neurosteroids
15. O’Hara MW, Swain AM. Rates and risk of postpartum during pregnancy using selective ion monitoring mass spectrometry.
depression — a meta-analysis. Int Rev Psychiatry. (1996) Steroids. (2015) 95:24–31. doi: 10.1016/j.steroids.2014.12.007
8:37–54. doi: 10.3109/09540269609037816 38. Osborne LM, Gispen F, Sanyal A, Yenokyan G, Meilman S, Payne
16. Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for JL. Lower allopregnanolone during pregnancy predicts postpartum
postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry. depression: an exploratory study. Psychoneuroendocrinology. (2017)
(2004) 26:289–95. doi: 10.1016/j.genhosppsych.2004.02.006 79:116–21. doi: 10.1016/j.psyneuen.2017.02.012
17. O’Hara MW. Postpartum depression: what we know. J Clin Psychol. (2009) 39. Meltzer-Brody S, Kanes SJ. Allopregnanolone in postpartum depression:
65:1258–69. doi: 10.1002/jclp.20644 role in pathophysiology and treatment. Neurobiol Stress. (2020)
18. Hopkins J, Marcus M, Campbell SB. Postpartum depression: a critical review. 12:100212. doi: 10.1016/j.ynstr.2020.100212
Psychol Bull. (1984) 95:498–515. doi: 10.1037/0033-2909.95.3.498 40. Beck CT. Theoretical perspectives of postpartum depression and their
19. Meltzer-Brody S, Howard LM, Bergink V, Vigod S, Jones I, Munk-Olsen treatment implications. MCN Am J Matern Nurs. (2002) 27:282–
T, et al. Postpartum psychiatric disorders. Nat Rev Dis Prim. (2018) 4:1– 7. doi: 10.1097/00005721-200209000-00008
18. doi: 10.1038/nrdp.2018.22 41. Cutrona CE, Troutman BR. Social support, infant temperament, and
20. American Psychiatric Association. Diagnostic and Statistical Manual parenting self-efficacy: a mediational model of postpartum depression. Child
of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Dev. (1986) 57:1507. doi: 10.2307/1130428
Association (2013). doi: 10.1176/appi.books.9780890425596 42. Halbreich U. Postpartum disorders: multiple interacting
21. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview underlying mechanisms and risk factors. J Affect Disord. (2005)
for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition (SCID- 88:1–7. doi: 10.1016/j.jad.2005.05.002
I/P). New York, NY: New York State Psychiatric Institute (2002). 43. Hagen EH. The functions of postpartum depression. Evol Hum Behav. (1999)
22. Cox JL, Holden JM, Sagovsky R. Edinburgh postnatal depression scale 20:325–59. doi: 10.1016/S1090-5138(99)00016-1
(EPDS). Br J Psychiatry. (1987). 150:782–6. doi: 10.1192/bjp.150.6.782 44. Hahn-Holbrook J, Haselton M. Is postpartum depression a
23. Eberhard-Gran M, Eskild A, Tambs K, Opjordsmoen S, Ove disease of modern civilization Curr Dir Psychol Sci. (2014)
Samuelsen S. Review of validation studies of the Edinburgh 23:395–400. doi: 10.1177/0963721414547736
postnatal depression scale. Acta Psychiatr Scand. (2008) 104:243– 45. Green SM, Haber E, Frey BN, McCabe RE. Cognitive-behavioral group
9. doi: 10.1111/j.1600-0447.2001.00187.x treatment for perinatal anxiety: a pilot study. Arch Womens Ment Health.
24. Arroll B, Goodyear-Smith F, Crengle S, Gunn J, Kerse N, Fishman (2015) 18:631–8. doi: 10.1007/s00737-015-0498-z
T, et al. Validation of PHQ-2 and PHQ-9 to screen for major 46. Pearlstein TB, Zlotnick C, Battle CL, Stuart S, O’Hara MW, Price AB, et
depression in the primary care population. Ann Fam Med. (2010) 8:348– al. Patient choice of treatment for postpartum depression: a pilot study.
53. doi: 10.1370/afm.1139 Arch Womens Ment Health. (2006) 9:303–8. doi: 10.1007/s00737-006-
25. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a 0145-9
brief depression severity measure. J Gen Intern Med. (2001) 16:606– 47. Sharp D, Chew-Graham C, Tylee A, Lewis G, Howard L, Anderson
13. doi: 10.1046/j.1525-1497.2001.016009606.x I, et al. A pragmatic randomised controlled trial to compare
26. Venkatesh KK, Zlotnick C, Triche EW, Ware C, Phipps MG. antidepressants with a community-based psychosocial intervention
Accuracy of brief screening tools for identifying postpartum for the treatment of women with postnatal depression: the
depression among adolescent mothers. Pediatrics. (2014) RESPOND trial. Health Technol Assess. (2010) 14:1–181. doi: 10.3310/
133:45–53. doi: 10.1542/peds.2013-1628 hta14430

Frontiers in Psychiatry | www.frontiersin.org 11 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

48. FDA. FDA Approves First Treatment for Post-Partum Depression. 66. Blom E, Jansen P, Verhulst F, Hofman A, Raat H, Jaddoe V, et al. Perinatal
(2019). Available online at: https://www.fda.gov/news-events/press- complications increase the risk of postpartum depression. The Generation R
announcements/fda-approves-first-treatment-post-partum-depression Study. BJOG. (2010) 117:1390–8. doi: 10.1111/j.1471-0528.2010.02660.x
(accessed July 12, 2020). 67. Vigod S, Villegas L, Dennis C-L, Ross L. Prevalence and risk
49. Meltzer-Brody S, Colquhoun H, Riesenberg R, Epperson CN, Deligiannidis factors for postpartum depression among women with preterm
KM, Rubinow DR, et al. Brexanolone injection in post-partum depression: and low-birth-weight infants: a systematic review. BJOG. (2010)
two multicentre, double-blind, randomised, placebo-controlled, phase 3 117:540–50. doi: 10.1111/j.1471-0528.2009.02493.x
trials. Lancet. (2018) 392:1058–70. doi: 10.1016/S0140-6736(18)31551-4 68. Ihongbe TO, Masho SW. Do successive preterm births increase the
50. Weissman AM, Levy BT, Hartz AJ, Bentler S, Donohue M, Ellingrod risk of postpartum depressive symptoms J Pregnancy. (2017) 2017:1–
VL, et al. Pooled analysis of antidepressant levels in lactating mothers, 10. doi: 10.1155/2017/4148136
breast milk, and nursing infants. Am J Psychiatry. (2004) 161:1066– 69. Helle N, Barkmann C, Bartz-Seel J, Diehl T, Ehrhardt S, Hendel A, et al.
78. doi: 10.1176/appi.ajp.161.6.1066 Very low birth-weight as a risk factor for postpartum depression four to
51. di Scalea TE, Wisner KL. Antidepressant medication use six weeks postbirth in mothers and fathers: cross-sectional results from
during breastfeeding. Clin Obstet Gynecol. (2009) 52:483– a controlled multicentre cohort study. J Affect Disord. (2015) 180:154–
97. doi: 10.1097/GRF.0b013e3181b52bd6 61. doi: 10.1016/j.jad.2015.04.001
52. Fortinguerra F, Clavenna A, Bonati M. Psychotropic drug use during 70. Payne JL. Genetic basis for postpartum depression. In: Biomarkers
breastfeeding: a review of the evidence. Pediatrics. (2009) 124:e547– of Postpartum Psychiatric Disorders. London: Elsevier. (2020) p. 15–
56. doi: 10.1542/peds.2009-0326 34. doi: 10.1016/B978-0-12-815508-0.00002-3
53. Davanzo R, Copertino M, De Cunto A, Minen F, Amaddeo A. Antidepressant 71. Viktorin A, Meltzer-Brody S, Kuja-Halkola R, Sullivan PF, Landén M,
drugs and breastfeeding: a review of the literature. Breastfeed Med. (2010) Lichtenstein P, et al. Heritability of perinatal depression and genetic
6:89–98. doi: 10.1089/bfm.2010.0019 overlap with non-perinatal depression. Am J Psychiatry. (2016) 173:158–
54. Akiskal HS. Toward a clinical understanding of the relationship of anxiety 65. doi: 10.1176/appi.ajp.2015.15010085
and depressive-disorders. In: Maser JD, Cloninger CR, editors. Conference 72. Forty L, Zammit S, Craddock N. Genetic risk and familial transmission of
on Comorbidity of Mood and Anxiety Disorders. Tuxedo, NY: NIMH Bristol depression. In: Dobson KS, Dozois DJA, editors. Risk Factors in Depression.
Myers Sandoz Pharm Upjohn (1990). p. 597–607. San Diego: Elsevier (2008). doi: 10.1016/B978-0-08-045078-0.00002-2
55. Chessick CA, Dimidjian S. Screening for bipolar disorder during pregnancy 73. Murphy-Eberenz K, Zandi PP, March D, Crowe RR, Scheftner WA,
and the postpartum period. Arch Womens Ment Health. (2010) 13:233– Alexander M, et al. Is perinatal depression familial J Affect Disord. (2006)
48. doi: 10.1007/s00737-010-0151-9 90:49–55. doi: 10.1016/j.jad.2005.10.006
56. Tavormina G. Bipolar disorders and bipolarity: the notion of the “mixity.” 74. Mahon PB, Payne JL, MacKinnon DF, Mondimore FM, Goes FS,
Psychiatr Danub. (2019) 31:S434–7. Available online at: http://www. Schweizer B, et al. Genome-wide linkage and follow-up association study
psychiatria-danubina.com/UserDocsImages/pdf/dnb_vol31_noSuppl%203/ of postpartum mood symptoms. Am J Psychiatry. (2009) 166:1229–
dnb_vol31_noSuppl%203_434.pdf 37. doi: 10.1176/appi.ajp.2009.09030417
57. Goldberg JF, Ernst CL. Features associated with the delayed initiation of 75. Alvim-Soares AM, Miranda DM, Campos SB, Figueira P,
mood stabilizers at illness onset in bipolar disorder. J Clin Psychiatry. (2002) Correa H, Romano-Silva MA. HMNC1 gene polymorphism
63:985–91. doi: 10.4088/JCP.v63n1105 associated with postpartum depression. Rev Bras Psiquiatr. (2014)
58. Schneck CD, Miklowitz DJ, Miyahara S, Araga M, Wisniewski S, 36:96–7. doi: 10.1590/1516-4446-2013-3507
Gyulai L, et al. The prospective course of rapid-cycling bipolar 76. Costas J, Gratacòs M, Escaramís G, Martín-Santos R, de Diego Y, Baca-
disorder: findings from the STEP-BD. Am J Psychiatry. (2008) García E, et al. Association study of 44 candidate genes with depressive and
165:370–7. doi: 10.1176/appi.ajp.2007.05081484 anxiety symptoms in post-partum women. J Psychiatr Res. (2010) 44:717–
59. Rihmer Z, Akiskal KK, Rihmer A, Akiskal HS. Current research 24. doi: 10.1016/j.jpsychires.2009.12.012
on affective temperaments. Curr Opin Psychiatry. (2010) 77. Segman RH, Goltser-Dubner T, Weiner I, Canetti L, Galili-
23:12–8. doi: 10.1097/YCO.0b013e32833299d4 Weisstub E, Milwidsky A, et al. Blood mononuclear cell gene
60. Tavormina G. An approach to treat bipolar disorders mixed states. expression signature of postpartum depression. Mol Psychiatry. (2010)
Psychiatr Danub. (2016) 28(Suppl. 1):9–12. Available online at: http://www. 15:93–100. doi: 10.1038/mp.2009.65
psychiatria-danubina.com/UserDocsImages/pdf/dnb_vol28_sup1/dnb_ 78. Licinio J. Potential diagnostic markers for postpartum depression
vol28_sup1_9.pdf point out to altered immune signaling. Mol Psychiatry. (2010)
61. Haga SM, Ulleberg P, Slinning K, Kraft P, Steen TB, Staff A. A 15:1. doi: 10.1038/mp.2009.142
longitudinal study of postpartum depressive symptoms: multilevel growth 79. Lesch K-P, Mössner R. Genetically driven variation in serotonin
curve analyses of emotion regulation strategies, breastfeeding self- uptake: is there a link to affective spectrum, neurodevelopmental,
efficacy, and social support. Arch Womens Ment Health. (2012) 15:175– and neurodegenerative disorders Biol Psychiatry. (1998) 44:179–
84. doi: 10.1007/s00737-012-0274-2 92. doi: 10.1016/S0006-3223(98)00121-8
62. Silverman ME, Reichenberg A, Savitz DA, Cnattingius S, Lichtenstein 80. Comasco E, Sylvén SM, Papadopoulos FC, Oreland L, Sundström-Poromaa
P, Hultman CM, et al. The risk factors for postpartum depression: a I, Skalkidou A. Postpartum depressive symptoms and the BDNF Val66Met
population-based study. Depress Anxiety. (2017) 34:178–87. doi: 10.1002/da. functional polymorphism: effect of season of delivery. Arch Womens Ment
22597 Health. (2011) 14:453–63. doi: 10.1007/s00737-011-0239-x
63. Goyal D, Gay C, Lee KA. How much does low socioeconomic 81. Comasco E, Sylvén SM, Papadopoulos FC, Sundström-Poromaa I,
status increase the risk of prenatal and postpartum depressive Oreland L, Skalkidou A. Postpartum depression symptoms: a case–control
symptoms in first-time mothers Womens Heal Issues. (2010) study on monoaminergic functional polymorphisms and environmental
20:96–104. doi: 10.1016/j.whi.2009.11.003 stressors. Psychiatr Genet. (2011) 21:19–28. doi: 10.1097/YPG.0b013e32834
64. Wassif OM, Abdo AS, Elawady MA, Abd Elmaksoud AE, Eldesouky RS. 1a3c1
Assessment of postpartum depression and anxiety among females attending 82. Zhang X, Wang L, Huang F, Li J, Xiong L, Xue H, et al. Gene-environment
primary health care facilities in Qaliubeya Governorate, Egypt. J Environ interaction in postpartum depression: a Chinese clinical study. J Affect
Public Health. (2019) 2019:1–9. doi: 10.1155/2019/3691752 Disord. (2014) 165:208–12. doi: 10.1016/j.jad.2014.04.049
65. Sockol LE, Epperson CN, Barber JP. The relationship between maternal 83. Pinsonneault JK, Sullivan D, Sadee W, Soares CN, Hampson E, Steiner
attitudes and symptoms of depression and anxiety among pregnant and M. Association study of the estrogen receptor gene ESR1 with postpartum
postpartum first-time mothers. Arch Womens Ment Health. (2014) 17:199– depression—a pilot study. Arch Womens Ment Health. (2013) 16:499–
212. doi: 10.1007/s00737-014-0424-9 509. doi: 10.1007/s00737-013-0373-8

Frontiers in Psychiatry | www.frontiersin.org 12 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

84. Mehta D, Newport DJ, Frishman G, Kraus L, Rex-Haffner M, Ritchie 102. Pinheiro RT, Pinheiro KAT, da Cunha Coelho FM, de Ávila Quevedo L, Gazal
JC, et al. Early predictive biomarkers for postpartum depression point M, da Silva RA, et al. Brain-derived neurotrophic factor levels in women
to a role for estrogen receptor signaling. Psychol Med. (2014) 44:2309– with postpartum affective disorder and suicidality. Neurochem Res. (2012)
22. doi: 10.1017/S0033291713003231 37:2229–34. doi: 10.1007/s11064-012-0851-9
85. Mandelli L, Serretti A. Gene environment interaction studies in depression 103. Fung J, Gelaye B, Zhong Q-Y, Rondon MB, Sanchez SE, Barrios Y V.,
and suicidal behavior: an update. Neurosci Biobehav Rev. (2013) 37:2375– et al. Association of decreased serum brain-derived neurotrophic factor
97. doi: 10.1016/j.neubiorev.2013.07.011 (BDNF) concentrations in early pregnancy with antepartum depression.
86. Doornbos B, Dijck-Brouwer DAJ, Kema IP, Tanke MAC, van Goor BMC Psychiatry. (2015) 15:43. doi: 10.1186/s12888-015-0428-7
SA, Muskiet FAJ, et al. The development of peripartum depressive 104. Figueira P, Malloy-Diniz L, Campos SB, Miranda DM, Romano-Silva MA,
symptoms is associated with gene polymorphisms of MAOA, 5-HTT De Marco L, et al. An association study between the Val66Met polymorphism
and COMT. Prog Neuro-Psychopharmacol Biol Psychiatry. (2009) 33:1250– of the BDNF gene and postpartum depression. Arch Womens Ment Health.
4. doi: 10.1016/j.pnpbp.2009.07.013 (2010) 13:285–9. doi: 10.1007/s00737-010-0146-6
87. Sacher J, Rekkas PV, Wilson AA, Houle S, Romano L, Hamidi J, et al. 105. Katz ER, Stowe ZN, Newport DJ, Kelley ME, Pace TW, Cubells JF, et al.
Relationship of monoamine oxidase-A distribution volume to postpartum Regulation of mRNA expression encoding chaperone and co-chaperone
depression and postpartum crying. Neuropsychopharmacology. (2015) proteins of the glucocorticoid receptor in peripheral blood: association
40:429–35. doi: 10.1038/npp.2014.190 with depressive symptoms during pregnancy. Psychol Med. (2012) 42:943–
88. Carter CS. Oxytocin pathways and the evolution of human behavior. Annu 56. doi: 10.1017/S0033291711002121
Rev Psychol. (2014) 65:17–39. doi: 10.1146/annurev-psych-010213-115110 106. Reynolds RM, Pesonen A-K, O’Reilly JR, Tuovinen S, Lahti M, Kajantie E, et
89. Feldman R. Oxytocin and social affiliation in humans. Horm Behav. (2012) al. Maternal depressive symptoms throughout pregnancy are associated with
61:380–91. doi: 10.1016/j.yhbeh.2012.01.008 increased placental glucocorticoid sensitivity. Psychol Med. (2015) 45:2023–
90. Skrundz M, Bolten M, Nast I, Hellhammer DH, Meinlschmidt G. Plasma 30. doi: 10.1017/S003329171400316X
oxytocin concentration during pregnancy is associated with development 107. Pan D, Xu Y, Zhang L, Su Q, Chen M, Li B, et al. Gene expression profile in
of postpartum depression. Neuropsychopharmacology. (2011) 36:1886– peripheral blood mononuclear cells of postpartum depression patients. Sci
93. doi: 10.1038/npp.2011.74 Rep. (2018) 8:1–9. doi: 10.1038/s41598-018-28509-4
91. Stuebe AM, Grewen K, Meltzer-Brody S. Association between maternal 108. Sun H, Kennedy PJ, Nestler EJ. Epigenetics of the depressed brain: role
mood and oxytocin response to breastfeeding. J Womens Heal. (2013) of histone acetylation and methylation. Neuropsychopharmacology. (2013)
22:352–61. doi: 10.1089/jwh.2012.3768 38:124–37. doi: 10.1038/npp.2012.73
92. Mileva-Seitz V, Steiner M, Atkinson L, Meaney MJ, Levitan R, Kennedy 109. Guintivano J, Arad M, Gould TD, Payne JL, Kaminsky ZA. Antenatal
JL, et al. Interaction between oxytocin genotypes and early experience prediction of postpartum depression with blood DNA methylation
predicts quality of mothering and postpartum mood. PLoS ONE. (2013) biomarkers. Mol Psychiatry. (2014) 19:560–7. doi: 10.1038/mp.2013.62
8:e61443. doi: 10.1371/journal.pone.0061443 110. Kaminsky Z, Payne J. Seeing the future: epigenetic biomarkers
93. Engineer N, Darwin L, Nishigandh D, Ngianga-Bakwin K, Smith of postpartum depression. Neuropsychopharmacology. (2014)
SC, Grammatopoulos DK. Association of glucocorticoid and type 1 39:233–4. doi: 10.1038/npp.2013.238
corticotropin-releasing hormone receptors gene variants and risk for 111. Osborne LM, Clive M, Kimmel M, Gispen F, Guintivano J, Brown T,
depression during pregnancy and post-partum. J Psychiatr Res. (2013) et al. Replication of epigenetic postpartum depression biomarkers and
47:1166–73. doi: 10.1016/j.jpsychires.2013.05.003 variation with hormone levels. Neuropsychopharmacology. (2016) 41:1648–
94. Jonas W, Mileva-Seitz V, Girard AW, Bisceglia R, Kennedy JL, Sokolowski M, 58. doi: 10.1038/npp.2015.333
et al. Genetic variation in oxytocin rs2740210 and early adversity associated 112. Osborne LM, Betz JF, Yenokyan G, Standeven LR, Payne JL. The role of
with postpartum depression and breastfeeding duration. Genes Brain Behav. allopregnanolone in pregnancy in predicting postpartum anxiety symptoms.
(2013) 12:681–94. doi: 10.1111/gbb.12069 Front Psychol. (2019) 10:1033. doi: 10.3389/fpsyg.2019.01033
95. Bell AF, Carter CS, Steer CD, Golding J, Davis JM, Steffen AD, et al. 113. Matrisciano F, Pinna G. PPAR and functional foods: rationale for natural
Interaction between oxytocin receptor DNA methylation and genotype is neurosteroid-based interventions for postpartum depression. Neurobiol
associated with risk of postpartum depression in women without depression Stress. (2020) 12:100222. doi: 10.1016/j.ynstr.2020.100222
in pregnancy. Front Genet. (2015) 6:00243. doi: 10.3389/fgene.2015.00243 114. Pinna G. Allopregnanolone, the neuromodulator turned
96. Kimmel M, Clive M, Gispen F, Guintivano J, Brown T, therapeutic agent: thank you, next Front Endocrinol. (2020)
Cox O, et al. Oxytocin receptor DNA methylation in 11:236. doi: 10.3389/fendo.2020.00236
postpartum depression. Psychoneuroendocrinology. (2016) 115. Payne JL, Osborne LM, Cox O, Kelly J, Meilman S,
69:150–60. doi: 10.1016/j.psyneuen.2016.04.008 Jones I, et al. DNA methylation biomarkers prospectively
97. King L, Robins S, Chen G, Yerko V, Zhou Y, Nagy C, et al. predict both antenatal and postpartum depression. Psychiatry
Perinatal depression and DNA methylation of oxytocin-related genes: Res. (2020) 285:112711. doi: 10.1016/j.psychres.2019.1
a study of mothers and their children. Horm Behav. (2017) 96:84– 12711
94. doi: 10.1016/j.yhbeh.2017.09.006 116. Mamrut S, Harony H, Sood R, Shahar-Gold H, Gainer H,
98. Christian LM, Mitchell AM, Gillespie SL, Palettas M. Serum Shi Y-J, et al. DNA methylation of specific CpG sites in the
brain-derived neurotrophic factor (BDNF) across pregnancy promoter region regulates the transcription of the mouse oxytocin
and postpartum: associations with race, depressive symptoms, receptor. PLoS ONE. (2013) 8:e56869. doi: 10.1371/journal.pone.00
and low birth weight. Psychoneuroendocrinology. (2016) 74:69– 56869
76. doi: 10.1016/j.psyneuen.2016.08.025 117. Kaminsky Z, Payne JL. Epigenetic biomarkers of postpartum depression. In:
99. Lommatzsch M, Hornych K, Zingler C, Schuffwerner P, Hoppner Biomarkers of Postpartum Psychiatric Disorders. London: Elsevier. (2020). p.
J, Virchow J. Maternal serum concentrations of BDNF and 35–43. doi: 10.1016/B978-0-12-815508-0.00003-5
depression in the perinatal period. Psychoneuroendocrinology. (2006) 118. Quiñones-Jenab V, Jenab S, Ogawa S, Adan RAM, Burbach PH, Pfaff
31:388–94. doi: 10.1016/j.psyneuen.2005.09.003 DW. Effects of estrogen on oxytocin receptor messenger ribonucleic
100. Aydemir C, Yalcin ES, Aksaray S, Kisa C, Yildirim SG, Uzbay T, et acid expression in the uterus, pituitary, and forebrain of the female rat.
al. Brain-derived neurotrophic factor (BDNF) changes in the serum of Neuroendocrinology. (1997) 65:9–17. doi: 10.1159/000127160
depressed women. Prog Neuro-Psychopharmacol Biol Psychiatry. (2006) 119. Thompson RJ, Parker KJ, Hallmayer JF, Waugh CE, Gotlib IH.
30:1256–60. doi: 10.1016/j.pnpbp.2006.03.025 Oxytocin receptor gene polymorphism (rs2254298) interacts with
101. Gazal M, Motta LS, Wiener CD, Fernandes JC, Quevedo LÁ, Jansen K, et familial risk for psychopathology to predict symptoms of depression
al. Brain-derived neurotrophic factor in post-partum depressive mothers. and anxiety in adolescent girls. Psychoneuroendocrinology. (2011)
Neurochem Res. (2012) 37:583–7. doi: 10.1007/s11064-011-0647-3 36:144–7. doi: 10.1016/j.psyneuen.2010.07.003

Frontiers in Psychiatry | www.frontiersin.org 13 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

120. Kusui C, Kimura T, Ogita K, Nakamura H, Matsumura Y, Koyama M, 139. Ijuin T, Douchi T, Yamamoto S, Ijuin Y, Nagata Y. The relationship between
et al. DNA methylation of the human oxytocin receptor gene promoter maternity blues and thyroid dysfunction. J Obstet Gynaecol Res. (1998)
regulates tissue-specific gene suppression. Biochem Biophys Res Commun. 24:49–55. doi: 10.1111/j.1447-0756.1998.tb00052.x
(2001) 289:681–6. doi: 10.1006/bbrc.2001.6024 140. Kuijpens JL, Vader HL, Drexhage HA, Wiersinga WM, Van Son MJ,
121. Bloch M. Effects of gonadal steroids in women with a history Pop VJ. Thyroid peroxidase antibodies during gestation are a marker
of postpartum depression. Am J Psychiatry. (2000) 157:924– for subsequent depression postpartum. Eur J Endocrinol. (2001). 145:579–
30. doi: 10.1176/appi.ajp.157.6.924 84. doi: 10.1530/eje.0.1450579
122. Klier CM, Muzik M, Dervic K, Mossaheb N, Benesch T, Ulm B, et al. The role 141. Sylvén SM, Elenis E, Michelakos T, Larsson A, Olovsson M,
of estrogen and progesterone in depression after birth. J Psychiatr Res. (2007) Poromaa IS, et al. Thyroid function tests at delivery and risk for
41:273–9. doi: 10.1016/j.jpsychires.2006.09.002 postpartum depressive symptoms. Psychoneuroendocrinology. (2013)
123. Serati M, Redaelli M, Buoli M, Altamura AC. Perinatal major depression 38:1007–13. doi: 10.1016/j.psyneuen.2012.10.004
biomarkers: a systematic review. J Affect Disord. (2016) 193:391– 142. Albacar G, Sans T, Martín-Santos R, García-Esteve L, Guillamat R,
404. doi: 10.1016/j.jad.2016.01.027 Sanjuan J, et al. Thyroid function 48h after delivery as a marker
124. Luisi S, Petraglia F, Benedetto C, Nappi RE, Bernardi F, Fadalti M, et al. Serum for subsequent postpartum depression. Psychoneuroendocrinology. (2010)
allopregnanolone levels in pregnant women: changes during pregnancy, 35:738–42. doi: 10.1016/j.psyneuen.2009.10.015
at delivery, and in hypertensive patients. J Clin Endocrinol Metab. (2000) 143. Groer MW, Vaughan JH. Positive thyroid peroxidase antibody
85:2429–33. doi: 10.1210/jcem.85.7.6675 titer is associated with dysphoric moods during pregnancy
125. Paoletti AM, Romagnino S, Contu R, Orrù MM, Marotto MF, Zedda and postpartum. J Obstet Gynecol Neonatal Nurs. (2013)
P, et al. Observational study on the stability of the psychological status 42:E26–32. doi: 10.1111/j.1552-6909.2012.01425.x
during normal pregnancy and increased blood levels of neuroactive steroids 144. Dama M, Steiner M, Lieshout R Van. Thyroid peroxidase autoantibodies
with GABA-A receptor agonist activity. Psychoneuroendocrinology. (2006) perinatal depression risk: a systematic review. J Affect Disord. (2016)
31:485–92. doi: 10.1016/j.psyneuen.2005.11.006 198:108–21. doi: 10.1016/j.jad.2016.03.021
126. Hellgren C, Åkerud H, Skalkidou A, Bäckström T, Sundström- 145. Wesseloo R, Kamperman AM, Bergink V, Pop VJM. Thyroid peroxidase
Poromaa I. Low serum allopregnanolone is associated with symptoms antibodies during early gestation and the subsequent risk of first-onset
of depression in late pregnancy. Neuropsychobiology. (2014) postpartum depression: a prospective cohort study. J Affect Disord. (2018)
69:147–53. doi: 10.1159/000358838 225:399–403. doi: 10.1016/j.jad.2017.08.058
127. Nappi R. Serum allopregnanolone in women with postpartum “blues.” Obstet 146. Li D, Li Y, Chen Y, Li H, She Y, Zhang X, et al. Neuroprotection of reduced
Gynecol. (2001) 97:77–80. doi: 10.1016/S0029-7844(00)01112-1 thyroid hormone with increased estrogen and progestogen in postpartum
128. Epperson CN, Gueorguieva R, Czarkowski KA, Stiklus S, Sellers E, Krystal depression. Biosci Rep. (2019) 39:1–11. doi: 10.1042/BSR20182382
JH, et al. Preliminary evidence of reduced occipital GABA concentrations in 147. Pedersen C, Leserman J, Garcia N, Stansbury M, Meltzer-
puerperal women: a 1H-MRS study. Psychopharmacology. (2006) 186:425– Brody S, Johnson J. Late pregnancy thyroid-binding globulin
33. doi: 10.1007/s00213-006-0313-7 predicts perinatal depression. Psychoneuroendocrinology. (2016)
129. Smith SS, Shen H, Gong QH, Zhou X. Neurosteroid regulation of GABAA 65:84–93. doi: 10.1016/j.psyneuen.2015.12.010
receptors: focus on the α4 and δ subunits. Pharmacol Ther. (2007) 116:58– 148. Bao A-M, Meynen G, Swaab DF. The stress system in depression and
76. doi: 10.1016/j.pharmthera.2007.03.008 neurodegeneration: focus on the human hypothalamus. Brain Res Rev.
130. Torner L, Neumann ID. The brain prolactin system: involvement (2008) 57:531–53. doi: 10.1016/j.brainresrev.2007.04.005
in stress response adaptations in lactation. Stress. (2002) 149. Lightman SL, Windle RJ, Wood SA, Kershaw YM, Shanks N, Ingram
5:249–57. doi: 10.1080/1025389021000048638 CD. Chapter 8 peripartum plasticity within the hypothalamo-pituitary-
131. Groer MW, Morgan K. Immune, health and endocrine characteristics of adrenal axis. In: Progress in Brain Research. Elsevier. (2001). p. 111–
depressed postpartum mothers. Psychoneuroendocrinology. (2007) 32:133– 29. doi: 10.1016/S0079-6123(01)33009-1
9. doi: 10.1016/j.psyneuen.2006.11.007 150. Sandman CA, Glynn L, Schetter CD, Wadhwa P, Garite T, Chicz-
132. Groër MW. Differences between exclusive breastfeeders, formula-feeders, DeMet A, et al. Elevated maternal cortisol early in pregnancy
and controls: a study of stress, mood, and endocrine variables. Biol Res Nurs. predicts third trimester levels of placental corticotropin releasing
(2005) 7:106–17. doi: 10.1177/1099800405280936 hormone (CRH): priming the placental clock. Peptides. (2006)
133. Jobst A, Krause D, Maiwald C, Härtl K, Myint A-M, Kästner R, et al. 27:1457–63. doi: 10.1016/j.peptides.2005.10.002
Oxytocin course over pregnancy and postpartum period and the association 151. Atsushi S, Anthony SL, Marjorie ML, Andrew NM, Toshihiro
with postpartum depressive symptoms. Arch Womens Ment Health. (2016) S, Dorothy TK. Immunoreactive corticotropin-releasing factor is
19:571–9. doi: 10.1007/s00737-016-0644-2 present in human maternal plasma during the third trimester of
134. Massey SH, Schuette SA, Pournajafi-Nazarloo H, Wisner KL, Carter CS. pregnancy. J Clin Endocrinol Metab. (1984) 59:812–4. doi: 10.1210/jcem-
Interaction of oxytocin level and past depression may predict postpartum 59-4-812
depressive symptom severity. Arch Womens Ment Health. (2016) 19:799– 152. Mastorakos G, Ilias I. Maternal and fetal hypothalamic-pituitary-adrenal
808. doi: 10.1007/s00737-016-0616-6 axes during pregnancy and postpartum. Ann N Y Acad Sci. (2003) 997:136–
135. Thul TA, Corwin EJ, Carlson NS, Brennan PA, Young LJ. Oxytocin 49. doi: 10.1196/annals.1290.016
and postpartum depression: a systematic review. Psychoneuroendocrinology. 153. Yim IS, Glynn LM, Schetter CD, Hobel CJ, Chicz-DeMet A,
(2020) 120:104793. doi: 10.1016/j.psyneuen.2020.104793 Sandman CA. Risk of postpartum depressive symptoms with elevated
136. Kallak TK, Hellgren C, Skalkidou A, Sandelin-Francke L, Ubhayasekhera K, corticotropin- releasing hormone in human pregnancy. Arch Gen
Bergquist J, et al. Maternal and female fetal testosterone levels are associated Psychiatry. (2009) 66:162–9. doi: 10.1001/archgenpsychiatry.20
with maternal age and gestational weight gain. Eur J Endocrinol. (2017) 08.533
177:379–88. doi: 10.1530/EJE-17-0207 154. Hahn-Holbrook J, Dunkel Schetter C, Arora C, Hobel CJ.
137. Aswathi A, Rajendiren S, Nimesh A, Philip RR, Kattimani S, Jayalakshmi Placental corticotropin-releasing hormone mediates the association
D, et al. High serum testosterone levels during postpartum period are between prenatal social support and postpartum depression.
associated with postpartum depression. Asian J Psychiatr. (2015) 17:85– Clin Psychol Sci. (2013) 1:253–65. doi: 10.1177/21677026124
8. doi: 10.1016/j.ajp.2015.08.008 70646
138. Chatzicharalampous C, Rizos D, Pliatsika P, Leonardou A, Hasiakos 155. Iliadis SI, Sylvén S, Hellgren C, Olivier JD, Schijven D, Comasco E, et
D, Zervas I, et al. Reproductive hormones and postpartum mood al. Mid-pregnancy corticotropin-releasing hormone levels in association
disturbances in Greek women. Gynecol Endocrinol. (2011) 27:543– with postpartum depressive symptoms. Depress Anxiety. (2016) 33:1023–
50. doi: 10.3109/09513590.2010.501886 30. doi: 10.1002/da.22529

Frontiers in Psychiatry | www.frontiersin.org 14 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

156. Glynn LM, Sandman CA. Evaluation of the association between placental 174. Maes M, Lin A, Ombelet W, Stevens K, Kenis G, De Jongh R, et al.
corticotrophin-releasing hormone and postpartum depressive symptoms. Immune activation in the early puerperium is related to postpartum anxiety
Psychosom Med. (2014) 76:355–62. doi: 10.1097/PSY.0000000000000066 and depressive symptoms. Psychoneuroendocrinology. (2000) 25:121–
157. O’Connor TG, Tang W, Gilchrist MA, Moynihan JA, Pressman EK, 37. doi: 10.1016/S0306-4530(99)00043-8
Blackmore ER. Diurnal cortisol patterns and psychiatric symptoms in 175. Corwin EJ, Johnston N, Pugh L. Symptoms of postpartum
pregnancy: short-term longitudinal study. Biol Psychol. (2014) 96:35– depression associated with elevated levels of interleukin-1 beta
41. doi: 10.1016/j.biopsycho.2013.11.002 during the first month postpartum. Biol Res Nurs. (2008)
158. Labad J, Vilella E, Reynolds RM, Sans T, Cavallé P, Valero J, et al. Increased 10:128–33. doi: 10.1177/1099800408323220
morning adrenocorticotrophin hormone (ACTH) levels in women with 176. Boufidou F, Lambrinoudaki I, Argeitis J, Zervas IM, Pliatsika
postpartum thoughts of harming the infant. Psychoneuroendocrinology. P, Leonardou AA, et al. CSF and plasma cytokines at delivery
(2011) 36:924–8. doi: 10.1016/j.psyneuen.2010.11.006 and postpartum mood disturbances. J Affect Disord. (2009)
159. Jolley SN, Elmore S, Barnard KE, Carr DB. Dysregulation of the 115:287–92. doi: 10.1016/j.jad.2008.07.008
hypothalamic-pituitary-adrenal axis in postpartum depression. Biol Res 177. Krause D, Jobst A, Kirchberg F, Kieper S, Härtl K, Kästner R, et al. Prenatal
Nurs. (2007) 8:210–22. doi: 10.1177/1099800406294598 immunologic predictors of postpartum depressive symptoms: a prospective
160. Lara-Cinisomo S, Grewen KM, Girdler SS, Wood J, Meltzer-Brody study for potential diagnostic markers. Eur Arch Psychiatry Clin Neurosci.
S. Perinatal depression, adverse life events, and hypothalamic–adrenal– (2014) 264:615–24. doi: 10.1007/s00406-014-0494-8
pituitary axis response to cold pressor stress in latinas: an exploratory study. 178. Liu H, Zhang Y, Gao Y, Zhang Z. Elevated levels of Hs-CRP and IL-6 after
Womens Heal Issues. (2017) 27:673–82. doi: 10.1016/j.whi.2017.06.004 delivery are associated with depression during the 6 months post partum.
161. Magiakou MA, Mastorakos G, Rabin D, Dubbert B, Gold PW, Chrousos Psychiatry Res. (2016) 243:43–8. doi: 10.1016/j.psychres.2016.02.022
GP. Hypothalamic corticotropin-releasing hormone suppression during 179. Simpson W, Steiner M, Coote M, Frey BN. Relationship between
the postpartum period: implications for the increase in psychiatric inflammatory biomarkers and depressive symptoms during late pregnancy
manifestations at this time. J Clin Endocrinol Metab. (1996) 81:1912– and the early postpartum period: a longitudinal study. Rev Bras Psiquiatr.
7. doi: 10.1210/jcem.81.5.8626857 (2016) 38:190–6. doi: 10.1590/1516-4446-2015-1899
162. Ferguson EH, Di Florio A, Pearson B, Putnam KT, Girdler S, Rubinow DR, 180. Skalkidou A, Sylvén SM, Papadopoulos FC, Olovsson M, Larsson A,
et al. HPA axis reactivity to pharmacologic and psychological stressors in Sundström-Poromaa I. Risk of postpartum depression in association with
euthymic women with histories of postpartum versus major depression. Arch serum leptin and interleukin-6 levels at delivery: a nested case–control study
Womens Ment Health. (2017) 20:411–20. doi: 10.1007/s00737-017-0716-y within the UPPSAT cohort. Psychoneuroendocrinology. (2009) 34:1329–
163. Yim IS, Glynn LM, Schetter CD, Hobel CJ, Chicz-DeMet A, Sandman 37. doi: 10.1016/j.psyneuen.2009.04.003
CA. Prenatal β-endorphin as an early predictor of postpartum 181. Chen C, Gao J, Zhang J, Jia L, Yu T, Zheng Y. Serum leptin level measured
depressive symptoms in euthymic women. J Affect Disord. (2010) 48 h after delivery is associated with development of postpartum depressive
125:128–33. doi: 10.1016/j.jad.2009.12.009 symptoms: a 3-month follow-up study. Arch Womens Ment Health. (2016)
164. Parcells DA. Women’s mental health nursing: depression, anxiety, and 19:1001–8. doi: 10.1007/s00737-016-0647-z
stress during pregnancy. J Psychiatr Ment Health Nurs. (2010) 17:813– 182. Fransson E, Dubicke A, Byström B, Ekman-Ordeberg G, Hjelmstedt
20. doi: 10.1111/j.1365-2850.2010.01588.x A, Lekander M. Negative emotions and cytokines in maternal and
165. Iliadis SI, Comasco E, Sylvén S, Hellgren C, Sundström Poromaa I, cord serum at preterm birth. Am J Reprod Immunol. (2012) 67:506–
Skalkidou A. Prenatal and postpartum evening salivary cortisol levels 14. doi: 10.1111/j.1600-0897.2011.01081.x
in association with peripartum depressive symptoms. PLoS ONE. (2015) 183. Maes M, Ombelet W, Libbrecht I, Stevens K, Kenis G, De Jongh R,
10:e0135471. doi: 10.1371/journal.pone.0135471 et al. Effects of pregnancy and delivery on serum concentrations of
166. Caparros-Gonzalez RA, Romero-Gonzalez B, Strivens-Vilchez H, Clara Cell Protein (CC16), an endogenous anticytokine: lower serum
Gonzalez-Perez R, Martinez-Augustin O, Peralta-Ramirez MI. CC16 is related to postpartum depression. Psychiatry Res. (1999) 87:117–
Hair cortisol levels, psychological stress and psychopathological 27. doi: 10.1016/S0165-1781(99)00073-6
symptoms as predictors of postpartum depression. PLoS ONE. (2017) 184. Bränn E, Fransson E, White RA, Papadopoulos FC, Edvinsson Å,
12:e0182817. doi: 10.1371/journal.pone.0182817 Kamali-Moghaddam M, et al. Inflammatory markers in women with
167. Harris B, Lovett L, Smith J, Read G, Walker R, Newcombe R. Cardiff postpartum depressive symptoms. J Neurosci Res. (2020) 98:1309–
puerperal mood and hormone study. III. postnatal depression at 5 to 6 weeks 21. doi: 10.1002/jnr.24312
postpartum, and its hormonal correlates across the peripartum period. Br J 185. Lambert M, Gressier F. Inflammatory biomarkers and postpartum
Psychiatry. (1996) 168:739–44. doi: 10.1192/bjp.168.6.739 depression: a systematic review of literature. Can J Psychiatry. (2019)
168. Corwin EJ, Pajer K, Paul S, Lowe N, Weber M, McCarthy DO. Bidirectional 64:070674371982897. doi: 10.1177/0706743719828970
psychoneuroimmune interactions in the early postpartum period influence 186. Bränn E, Papadopoulos F, Fransson E, White R, Edvinsson Å, Hellgren C, et
risk of postpartum depression. Brain Behav Immun. (2015) 49:86– al. Inflammatory markers in late pregnancy in association with postpartum
93. doi: 10.1016/j.bbi.2015.04.012 depression—a nested case-control study. Psychoneuroendocrinology. (2017)
169. Orta OR, Gelaye B, Bain PA, Williams MA. The association between maternal 79:146–59. doi: 10.1016/j.psyneuen.2017.02.029
cortisol and depression during pregnancy, a systematic review. Arch Womens 187. Wójcik J, Dudek D, Schlegel-Zawadzka M, Grabowska M, Marcinek A,
Ment Health. (2018) 21:43–53. doi: 10.1007/s00737-017-0777-y Florek E, et al. Antepartum/postpartum depressive symptoms and serum
170. Raison CL, Capuron L, Miller AH. Cytokines sing the blues: inflammation zinc and magnesium levels. Pharmacol Rep. (2006) 58:571–6. Available
and the pathogenesis of depression. Trends Immunol. (2006) 27:24– online at: http://rabbit.if-pan.krakow.pl/pjp/pdf/2006/4_571.pdf
31. doi: 10.1016/j.it.2005.11.006 188. Roomruangwong C, Kanchanatawan B, Sirivichayakul S, Mahieu
171. Schiepers OJG, Wichers MC, Maes M. Cytokines and major B, Nowak G, Maes M. Lower serum zinc and higher crp strongly
depression. Prog Neuro-Psychopharmacol Biol Psychiatry. (2005) predict prenatal depression and physio-somatic symptoms,
29:201–17. doi: 10.1016/j.pnpbp.2004.11.003 which all together predict postnatal depressive symptoms.
172. Osborne LM, Monk C. Perinatal depression-The fourth Mol Neurobiol. (2017) 54:1500–12. doi: 10.1007/s12035-016-
inflammatory morbidity of pregnancy. Theory and 9741-5
literature review. Psychoneuroendocrinology. (2013) 38:1929– 189. Christesen HT, Falkenberg T, Lamont RF, JØrgensen JS. The impact of
52. doi: 10.1016/j.psyneuen.2013.03.019 vitamin D on pregnancy: a systematic review. Acta Obstet Gynecol Scand.
173. Kendall-Tackett K. A new paradigm for depression in new mothers: the (2012) 91:1357–67. doi: 10.1111/aogs.12000
central role of inflammation and how breastfeeding and anti-inflammatory 190. Murphy PK, Mueller M, Hulsey TC, Ebeling MD, Wagner CL. An
treatments protect maternal mental health. Int Breastfeed J. (2007) exploratory study of postpartum depression and vitamin D. J Am Psychiatr
2:6. doi: 10.1186/1746-4358-2-6 Nurses Assoc. (2010) 16:170–7. doi: 10.1177/1078390310370476

Frontiers in Psychiatry | www.frontiersin.org 15 June 2021 | Volume 12 | Article 620371

Yu et al. Postpartum Depression: Status, Identification, Biomarkers

191. Brandenbarg J, Vrijkotte TGM, Goedhart G, van Eijsden M. Maternal early- 204. Lewis SJ, Araya R, Leary S, Smith GD, Ness A. Folic acid supplementation
pregnancy vitamin d status is associated with maternal depressive symptoms during pregnancy may protect against depression 21 months after
in the amsterdam born children and their development cohort. Psychosom pregnancy, an effect modified by MTHFR C677T genotype. Eur J Clin Nutr.
Med. (2012) 74:751–7. doi: 10.1097/PSY.0b013e3182639fdb (2012) 66:97–103. doi: 10.1038/ejcn.2011.136
192. Gur EB, Gokduman A, Turan GA, Tatar S, Hepyilmaz I, Zengin EB, et al. 205. Huang J, Zhang L, He M, Qiang X, Xiao X, Huang S, et al. Comprehensive
Mid-pregnancy vitamin D levels and postpartum depression. Eur J Obstet evaluation of postpartum depression and correlations between postpartum
Gynecol Reprod Biol. (2014) 179:110–6. doi: 10.1016/j.ejogrb.2014.05.017 depression and serum levels of homocysteine in chinese women. J Cent South
193. Robinson M, Whitehouse AJO, Newnham JP, Gorman S, Jacoby P, Holt Univ. (2015) 40:311–6. doi: 10.11817/j.issn.1672-7347.2015.03.013
BJ, et al. Low maternal serum vitamin D during pregnancy and the risk 206. Maes M, Verkerk R, Bonaccorso S, Ombelet W, Bosmans E, Scharpé S.
for postpartum depression symptoms. Arch Womens Ment Health. (2014) Depressive and anxiety symptoms in the early puerperium are related
17:213–9. doi: 10.1007/s00737-014-0422-y to increased degradation of tryptophan into kynurenine, a phenomenon
194. Miyake Y, Tanaka K, Okubo H, Sasaki S, Arakawa M. Dietary vitamin D which is related to immune activation. Life Sci. (2002) 71:1837–
intake and prevalence of depressive symptoms during pregnancy in Japan. 48. doi: 10.1016/S0024-3205(02)01853-2
Nutrition. (2015) 31:160–5. doi: 10.1016/j.nut.2014.06.013 207. Hantsoo L. Potential Hormonal and Neurochemical Biomarkers for
195. Accortt EE, Schetter CD, Peters RM, Cassidy-Bushrow AE. Lower Postpartum Depression. In: Biomarkers of Postpartum Psychiatric Disorders.
prenatal vitamin D status and postpartum depressive symptomatology London: Elsevier (2019). doi: 10.1016/B978-0-12-815508-0.00004-7
in African American women: preliminary evidence for moderation by 208. Foster JA, McVey Neufeld K-A. Gut–brain axis: how the microbiome
inflammatory cytokines. Arch Womens Ment Health. (2016) 19:373– influences anxiety and depression. Trends Neurosci. (2013) 36:305–
83. doi: 10.1007/s00737-015-0585-1 12. doi: 10.1016/j.tins.2013.01.005
196. Fu C-W, Liu J-T, Tu W-J, Yang J-Q, Cao Y. Association between 209. Slykerman RF, Hood F, Wickens K, Thompson JMD, Barthow
serum 25-hydroxyvitamin D levels measured 24 hours after delivery and C, Murphy R, et al. Effect of lactobacillus rhamnosus HN001 in
postpartum depression. BJOG. (2015) 122:1688–94. doi: 10.1111/1471-0528. pregnancy on postpartum symptoms of depression and anxiety: a
13111 randomised double-blind placebo-controlled trial. EBioMedicine. (2017)
197. Teofilo MMA, Farias DR, Pinto T de JP, Vilela AAF, Vaz J 24:159–65. doi: 10.1016/j.ebiom.2017.09.013
dos S, Nardi AE, et al. HDL-cholesterol concentrations are 210. Gadad BS, Jha MK, Czysz A, Furman JL, Mayes TL, Emslie MP, et al.
inversely associated with Edinburgh Postnatal Depression Scale Peripheral biomarkers of major depression and antidepressant treatment
scores during pregnancy: results from a Brazilian cohort study. response: current knowledge and future outlooks. J Affect Disord. (2018)
J Psychiatr Res. (2014) 58:181–8. doi: 10.1016/j.jpsychires.2014. 233:3–14. doi: 10.1016/j.jad.2017.07.001
07.030 211. Sethi S, Brietzke E. Omics-based biomarkers: application of metabolomics
198. Van Dam RM, Schuit AJ, Schouten EG, Vader HL, Pop VJM. in neuropsychiatric disorders. Int J Neuropsychopharmacol. (2015) 19:1–
Serum cholesterol decline and depression in the postpartum period. 13. doi: 10.1093/ijnp/pyv096
J Psychosom Res. (1999) 46:385–90. doi: 10.1016/S0022-3999(98) 212. Papadopoulou Z, Vlaikou AM, Theodoridou D, Komini C, Chalkiadaki G,
00100-7 Vafeiadi M, et al. Unraveling the serum metabolomic profile of post-partum
199. Parker G, Hegarty B, Granville-Smith I, Ho J, Paterson A, Gokiert A, depression. Front Neurosci. (2019) 13:1–9. doi: 10.3389/fnins.2019.00833
et al. Is essential fatty acid status in late pregnancy predictive of post- 213. Lin L, Chen XM, Liu RH. Novel urinary metabolite signature for
natal depression Acta Psychiatr Scand. (2015) 131:148–56. doi: 10.1111/acps. diagnosing postpartum depression. Neuropsychiatr Dis Treat. (2017)
12321 13:1263–70. doi: 10.2147/NDT.S135190
200. Markhus MW, Skotheim S, Graff IE, Frøyland L, Braarud HC, 214. Zhang L, Zou W, Huang Y, Wen X, Huang J, Wang Y, et al. A preliminary
Stormark KM, et al. Low omega-3 index in pregnancy is a possible study of uric metabolomic alteration for postpartum depression based
biological risk factor for postpartum depression. PLoS ONE. (2013) on liquid chromatography coupled to quadrupole time-of-flight mass
8:e67617. doi: 10.1371/journal.pone.0067617 spectrometry. Dis Markers. (2019) 2019:1–9. doi: 10.1155/2019/4264803
201. Sallis H, Steer C, Paternoster L, Davey Smith G, Evans J. Perinatal depression
and omega-3 fatty acids: a Mendelian randomisation study. J Affect Disord. Conflict of Interest: The authors declare that the research was conducted in the
(2014) 166:124–31. doi: 10.1016/j.jad.2014.04.077 absence of any commercial or financial relationships that could be construed as a
202. Chong MFF, Ong Y-L, Calder PC, Colega M, Wong JXY, Tan CS, potential conflict of interest.
et al. Long-chain polyunsaturated fatty acid status during pregnancy
and maternal mental health in pregnancy and the postpartum Copyright © 2021 Yu, Liang, Chen, Li, Han, Chen and Li. This is an open-access
period. J Clin Psychiatry. (2015) 76:e848–56. doi: 10.4088/JCP.14m article distributed under the terms of the Creative Commons Attribution License (CC
09191 BY). The use, distribution or reproduction in other forums is permitted, provided
203. Lukose A, Ramthal A, Thomas T, Bosch R, Kurpad A V., Duggan C, et al. the original author(s) and the copyright owner(s) are credited and that the original
Nutritional factors associated with antenatal depressive symptoms in the publication in this journal is cited, in accordance with accepted academic practice.
early stage of pregnancy among Urban South Indian women. Matern Child No use, distribution or reproduction is permitted which does not comply with these
Health J. (2014) 18:161–70. doi: 10.1007/s10995-013-1249-2 terms.

Frontiers in Psychiatry | www.frontiersin.org 16 June 2021 | Volume 12 | Article 620371