Content uploaded by Eynav Elgavish Accortt
Author content
All content in this area was uploaded by Eynav Elgavish Accortt on Sep 29, 2015
Content may be subject to copyright.
ORIGINAL ARTICLE
Lower prenatal vitamin D status and postpartum depressive
symptomatology in African American women: Preliminary
evidence for moderation by inflammatory cytokines
Eynav Elgavish Accortt
1,2
&Christine Dunkel Schetter
1
&
Rosalind M. Peters
3
&Andrea E. Cassidy-Bushrow
4
Received: 5 February 2015 /Accepted: 14 September 2015
#Springer-Verlag Wien 2015
Abstract Vitamin D deficiency and elevated pro-
inflammatory cytokines have each been associated individually
with postpartum depression (PPD). African American women
are at increased risk for prenatal vitamin D deficiency, inflamma-
tion, and prenatal and postpartum depressive symptoms, but
biological risk factors for PPD in this population have rarely been
tested. This prospective study tested whether low prenatal vita-
min D status (serum 25-hydroxyvitamin D, 25[OH]D) predicted
PPD symptomatology in pregnant African American women and
whether high levels of prenatal inflammatory cytokines interacted
with low 25(OH)D in effects on PPD symptoms. Vitamin D
status was measured in the first trimester in a sample of 91
African American pregnant women who had a second trimester
blood sample assayed for inflammatory markers. Depressive
symptoms were assessed at a postpartum visit. An inverse asso-
ciation between prenatal log 25(OH)D and PPD symptomatology
approached significance (β=−0.209, p=0.058), and interleukin-6
and IL-6/IL-10 ratio significantly moderated the effect. Among
women with higher levels of inflammatory markers, lower
prenatal log 25(OH)D was associated with significantly higher
PPD symptoms (p<0.05). These preliminary results are intrigu-
ing because, if replicable, easy translational opportunities, such
as increasing vitamin D status in pregnant women with elevated
pro-inflammatory cytokines, may reduce PPD symptoms.
Keywords Pregnancy .Vitamin D .25(OH)D .Cytokines .
Inflammation .Postpartum depression
Introduction
Postpartum depression is a serious mental health condition
occurring after childbirth characterized by emotional distur-
bance and behavioral changes. It is estimated that 10 to 20 %
of American women experience symptoms of magnitude and
duration to meet criteria for PPD (Centers for Disease Control
and Prevention 2008). A recent meta-analysis (Molyneaux
et al. 2014) estimated rates of diagnosed PPD to be higher
in overweight and obese pregnant women (13 vs. 10 % of
normal-weight women). Depression after childbirth portends
maternal suffering, more likelihood of parenting ineffective-
ness, and possible difficulties for the entire family (Grace et al.
2003; Paulson et al. 2006; Zelkowitz and Milet 1996).
Therefore, it is advantageous to increase our understanding
of the risk factors, etiology, and mechanisms of depression
in women following birth. Independent risk factors include
prenatal depression, high life stress, low social support, low
education, and African American race (Beck 2001; Bennett
et al. 2004). A comprehensive systematic review on post-
partum depressive symptoms and diagnosis highlighted that
most biological factors have been studied in isolation
(Yim et al. 2015). We investigated vitamin D status, inflam-
matory cytokines, and PPD symptoms together in a sample of
AfricanAmericanpregnantwomen,manyofwhomwere
overweight or obese.
*Eynav Elgavish Accortt
eynav.accortt@cshs.org
*Christine Dunkel Schetter
dunkel@psych.ucla.edu
1
Department of Psychology, University of California, 1285 Franz
Hall, Box 951563, Los Angeles, CA 90095-1563, USA
2
Department of Obstetrics and Gynecology, Cedars-Sinai Medical
Center, 8635 West 3rd Street, 160 West Tower, Los
Angeles 90048, CA, USA
3
College of Nursing, Wayne State University, Detroit, MI, USA
4
Department of Public Health Sciences, Henry Ford Hospital,
Detroit, MI, USA
Arch Womens Ment Health
DOI 10.1007/s00737-015-0585-1
Vitamin D is a unique neurosteroid hormone required for
normal brain homeostasis and development. Vitamin D
3
is
produced in the skin when exposed to the sun’s ultraviolet
rays and is also absorbed from various food sources, such as
oily fish (Zhang and Naughton 2010).
1
Low levels of vitamin
D have also been associated with several mental disorders
including depression (McCann and Ames 2008). According
to a recent meta-analysis of 31,424 males and females, vita-
min D was inversely associated with depression (Anglin et al.
2013); however, none of the included studies were on preg-
nancy or the postpartum. In a prospective study of 796 wom-
en, a negative association was found between prenatal vitamin
D levels at 18 weeks gestation and PPD symptoms (Robinson
et al. 2014; see also Murphy et al. 2010). Another recent
prospective study reported that lower maternal 25(OH)D3
levels in the second trimester of pregnancy were associated
with higher levels of PPD symptoms at 1 week, 6 weeks, and
6monthspostpartum(Guretal.2014). Cross-sectional studies
have also been published in support of these prospective ones
(Brandenbarg et al. 2012; Cassidy-Bushrow et al. 2012a).
Thus, the larger literature on low vitamin D levels and depres-
sion has been extended into pregnancy and postpartum. To our
knowledge, no studies have investigated the combined effects
of prenatal vitamin D status and inflammatory markers on
PPD symptoms in any ethnic group.
A complex and incompletely understood relationship be-
tween vitamin D and the immune system has been described
(Arora and Hobel 2010; Liu et al. 2006; McCann and Ames
2008). Expression of the vitamin D receptor has been detected
on a variety of immune cells including monocytes, T lympho-
cytes, dendritic cells, and macrophages (Veldman et al. 2000).
Vitamin D is beneficial, acting as an influential immune mod-
erator of both the adaptive and innate immune systems
through its ability to alter cytokine secretion (Adams and
Hewison 2008; Van Etten and Mathieu 2005), and acts as a
regulator of cell signaling pathways via its effect on toll-like
receptor expression and function (Liu et al. 2006; Thota et al.
2012; Van Etten and Mathieu 2005; Zhang et al. 2012). In
pregnancy, vitamin D regulates placental development and
function and promotes tolerance of the fetus (Arora and
Hobel 2010) and vitamin D deficiency has been associated
with neonatal and birth outcomes (Aghajafari et al. 2011,
2013; Wei et al. 2013; Theodoratou et al. 2014). Further, early
pregnancy vitamin D is inversely associated with inflamma-
tory cytokines in mid-pregnancy (Bobbitt et al. 2014). High
vitamin D status might also be a protective factor in the link
between pro-inflammatory cytokines and depression.
Numerous research studies have demonstrated a strong
positive association between inflammation and depression
(e.g., Dantzer et al. 2008;Howrenetal.2009;Milleretal.
2009), as well as an emerging body of research on the role of
stress, inflammation, and depression in perinatal health
(Christian 2012; Coussons-Read 2012). The role of vitamin
D status in this set of processes is unknown.
Inflammation is of interest in pregnancy because of the
physiological changes that occur during normal gestation. At
the uterine level, early pregnancy (implantation) and late preg-
nancy when approaching delivery are pro-inflammatory
states, whereas mid-pregnancy is an anti-inflammatory state
(Mor et al. 2011). Recent studies have longitudinally profiled
inflammatory markers throughout the perinatal period.
Interleukin-6 (IL-6), for example, increased slightly over ges-
tation in a racially diverse sample of 57 healthy pregnant
women with a significant postpartum increase (Christian and
Porter 2014). Blackmore and colleagues (2014) studied wom-
en at 4 time points in pregnancy and postpartum. They report-
ed elevations in serum pro-inflammatory markers in African
American compared to non-African American women, but
when controlling for BMI, the effect of race on IL-6 levels
was no longer significant. Furthermore, neither IL-6 nor
TNF-αwere associated with depression (Blackmore et al.
2014).
Significant associations between perinatal depression and
inflammation have been reported in several studies. For ex-
ample, depression was associated with inflammation during
pregnancy in three studies (Cassidy-Bushrow et al. 2012b;
Christian et al. 2009; Coussons-Read et al. 2007), and inflam-
mation in the postpartum period was associated with PPD in
three more studies (e.g., Boufidou et al. 2009; Corwin and
Pajer 2008;Maesetal.2000), although not in two other stud-
ies (Okun et al. 2011; Skalkidou et al. 2009). Moreover, pre-
natal depression consistently predicts PPD (Bennett et al.
2004). Nonetheless, some experts feel firm conclusions can-
not be made about the role of inflammatory processes in peri-
natal depression as yet (Osborne and Monk 2013). Firm con-
clusions are complicated by use of a broad range of tools for
mood assessment, different cutoff points, measurement of dif-
ferent cytokines, and racially and socioeconomically homog-
enous samples.
African American women appear to have among the
highest rates of prenatal and postpartum depression compared
to other racial groups in the USA (Howell et al. 2006; Orr et al.
2006; Segre et al. 2006). They are also at increased risk for
vitamin D deficiency (Nassar et al. 2011) because darker skin
limits synthesis of vitamin D and due to lower intake of sup-
plemental vitamin D (Bodnar and Simhan 2010).
Furthermore, higher levels of inflammatory markers have
been documented in African American men and women
(Deverts et al. 2010) and in both non-pregnant and pregnant
African American women in a stress reactivity study
(Christian et al. 2013). Thus, data on African American
1
Approximately 0.5 MED of UVB radiation would be absorbed after an
average of 5 to 10 min of exposure of the arms and legs to direct sunlight
depending on the time of day, season, latitude, and skin sensitivity (Zhang
and Naughton 2010).
E.E. Accortt et al.
women in pregnancy and postpartum is well suited to our
research questions.
The purpose of this study was to examine associations pro-
spectively between prenatal vitamin D status and postpartum
depressive symptomatology in a sample of African American
women. We hypothesized that in pregnant women with higher
prenatal inflammatory markers there would be a stronger
negative association of prenatal 25(OH)D, and postpartum
depressive symptoms as compared to those with lower inflam-
mation, and this association would remain after controlling for
BMI.
Methods
Participants and procedures
We analyzed prospective data obtained during pregnancy and
the postpartum from a previous study to test our study hypoth-
eses (see Cassidy-Bushrow et al. 2012a,bfor methodological
details). The study population consisted of pregnant women
who were patients in the Henry Ford Health System (HFHS),
in Detroit MI. Potential participants were identified by
accessing patient appointment lists in the electronic medical
record (EMR) of nine HFHS obstetrics and gynecology (OB/
GYN) clinics. Clinics were chosen based on the likelihood
that they would have a large number of African American
patients of varying socioeconomic status. African American
women, aged 18–44 years, in the second trimester of pregnan-
cy (13–28 weeks gestation based on self-reported last men-
strual period [LMP] or expected delivery date [EDD] defined
by ultrasound) were identified as potential participants.
Trained interviewers arranged to meet women at an upcoming
clinic appointment. Recruitment spanned February 2009 to
June 2010. A total of 203 women completed the prenatal
study visit. Eleven women were unable to complete the entire
prenatal study visit in person. In addition, three women with
very morbid obesity (BMI>60 kg/m
2
)wereexcluded.
For the current study on postpartum depression, we ex-
cluded two women with ≥1inflammatorybiomarkersex-
ceeding ±3 standard deviations from the mean for compa-
rability with a prior study (Christian et al. 2009). Following
written consent, eligible women provided self-reported de-
mographic information, and a 10-ml blood sample was ob-
tained during the second trimester research visit (13–
28 weeks gestation). The final analytic sample consisted
of 91 women with a first trimester 25(OH)D measurement,
second trimester measure of inflammatory markers, and
who had a postpartum visit during which the depression
screening was completed (N=98 did not return for postpar-
tum visit). The study protocol was approved by the
Institutional Review Boards at the respective institutions.
Vitamin D status
Serum 25-OHD is considered the best measure of overall vi-
tamin D nutrition status and the most useful clinically. As part
of routine obstetrical care, vitamin D status (serum 25-OHD)
was measured during the first prenatal care visit (M=9.7±
3.7 weeks gestation). Serum samples were assayed using a
competitive chemiluminescence immunoassay platform (Rao
1999), similar to other reported methods (Ersfeld et al. 2004;
Hollis 2008). Serum 25-OHD level was used as a continuous
variable in all regression analyses.
Inflammatory biomarkers
All assays were performed in serum collected from women
during the second trimester research visit (stored at −80 °C),
which was scheduled around a pre-existing obstetrics visit
(M=21.3± 3.8 weeks gestation). Time of day for the blood
draw therefore varied, and women did not fast beforehand.
High-sensitivity C-reactive protein (hs-CRP) was measured
by enzyme immunoassay as per the manufacturer’sinstruc-
tions (BioCheck, Inc., Foster City, CA, USA) and concentra-
tions (mg/l) calculated from a standard curve. The limit of
detection (LOD) is 0.1 mg/l. Interleukins, IL-1β, IL-6, IL-
10, and TNF-αconcentrations (pg/ml) were assayed on a
Bio-Plex 200 System using Bio-Plex Pro Cytokine Assay
custom 4-plex kits (Bio-Rad, Hercules, CA, USA). LOD for
these cytokines are 0.6, 2.6, 0.3, and 6.0 pg/ml. Results are
reported only for inflammatory markers with <20 % of the
sample outside of the LOD (J. Carroll personal communica-
tion; Breen et al. 2011). Therefore, results for IL-6, IL-10, and
hs-CRP are reported and those for TNF-α(54 % sample be-
low LOD) and IL-1β(42 % sample below LOD) are not.
The link between depression and inflammatory disorders
may be partially explained by a disruption of the immune-
regulatory balance between pro- and anti-inflammatory cyto-
kines (e.g., IL-6 and IL-10, respectively). In addition to their
absolute concentrations, the relative concentrations of pro- to
anti-inflammatory cytokines may provide a useful index of the
net inflammatory milieu and of immune dysregulation
(Dhabhar et al. 2009). After confirming that IL-6 and IL-10
were not significantly associated (r=0.1; p=0.174), we calcu-
lated the IL-6/IL-10 ratio and hypothesized that this ratio as
well as IL-6 (but not IL-10 alone) would moderate the asso-
ciation between prenatal 25(OH)D and PPD symptomatology.
Postpartum depressive symptoms
The Edinburgh Postnatal Depression Scale (EPDS; (Cox et al.
1987)) is a 10-item scale that assesses the cognitive and affec-
tive components of depressive symptomatology, while ex-
cluding somatic symptoms specific to postpartum. The
EPDS has been validated for use in pregnant and postpartum
Prenatal vitamin D and postpartum depressive symptoms
women and the sensitivity and specificity was 86 and 78 %,
respectively (Cox et al. 1987). The EPDS was administered to
assess depressive symptoms during routine postpartum visits,
which are typically scheduled 4–6 weeks post-delivery
(M=4.5 ± 1.8 weeks). EPDS scores were available electroni-
cally from medical records with participant consent. The
EPDS score was used as a continuous variable in all regres-
sion analyses.
Covariates
Participants reported date of birth, marital status, education,
employment status, and cigarette smoking at the second
trimester research visit. Pre-pregnancy height and weight
were self-reported, and pre-pregnancy body mass index
(BMI) calculated as weight (kg)/height (m
2
). They also re-
ported prenatal depressive symptoms using the 20-item
Center for Epidemiologic Studies Depression Scale (CES-
D; (Radloff 1977)). The prenatal CES-D score was used as
a continuous variable in all regression analyses. A history
of self-reported prior depressive illness (prior to pregnancy)
was abstracted from the OB intake form within the electron-
ic medical record. Additional medical variables (Table 1)
for descriptive purposes were abstracted from the electronic
medical record.
Statistical analysis
We used standard statistical software programs (SPSS 20.0 &
STATA 14 ) to conduct all analyses, statistical significance
was defined as p<0.05, and marginal effects were interpreted
at 0.05< p<0.10. Descriptive statistics were calculated to char-
acterize participants. Linear regression models were fitted to
estimate associations between prenatal 25(OH)D (ng/ml), in-
flammatory markers, and postpartum depressive symptoms.
Models were adjusted for demographic and medical variables
(see tables for details). To reduce non-normality in vitamin D
status, values were log transformed and are referred to as log
(25-OHD) or log vitamin D status. To examine moderation by
inflammatory markers, interaction terms were created com-
bining log vitamin D status with hs-CRP, inflammatory
markers IL-6 and IL-10 (log transformed to reduce non-nor-
mality), as well as the calculated IL-6/IL-10 ratio.
Results
Demographic characteristics
Women in the current study (N=91, Table 1)wereonaverage
26 years old (SD=5.9). The majority of women were unmar-
ried (75 %), currently employed (63 %), and approximately
half had a high school diploma (58 %). Few had a history of
depression (7 %), history of hypertension (9 %), or history of
preterm birth (9 %). A total of 6 of the 91 women delivered
babies before 37 weeks of gestation (7 % preterm birth). Many
women were overweight (N=24, BMI 25–30) or obese
(N=34, BMI> 30); average BMI was 29.2 (SD = 8.2). The
majority did not smoke during this pregnancy (98 %).
We compared the 91 women in the analytic sample to the
85 women with 2nd trimester vitamin D and inflammation
data but no postpartum EPDS score. There were no significant
differences in any of the key variables: maternal age, 2nd
Tabl e 1 Descriptive characteristics for sample
Total s amp le ( N=91)
Demographic variables
Maternal age (years) 26± 6
Married 23 (25 %)
≥High school diploma 53 (58 %)
Currently employed 57 (63 %)
Annual income ($)
a
$36,623±34,609
Medical variables
History of chronic hypertension 4 (4 %)
History of preterm birth 8 (9 %)
Preterm birth in this pregnancy
b
6(7 %)
Low birth weight in this pregnancy
b
7(8 %)
Nulliparous
c
43 (47 %)
Smoking in this pregnancy 2 (2 %)
Pre-pregnancy BMI (kg/m
2
)29±8
Psychological variables
Prenatal CES-D 16.1±11
History of depression 6 (7 %)
EPDS 5.20±4.95
Vitamin D variables
Vitamin D season:
April–December 55 (60 %)
January–March 36 (40 %)
Vitamin D status ng/ml 13.2±9.4
Vit ami n D ≤20 ng/ml
d
77 (85 %)
Inflammatory cytokines
IL-6 (pg/ml)
e
4.0± 2.1
IL-10 (pg/ml) 2.0± 0.5
IL-6/IL-10 2.1 ±1.1
hs-CRP (pg/ml) 4.6±3.0
Data are N(%) or mean ±standard deviation
EPDS Edinburgh Postnatal Depression Scale, CES-D Center for Epide-
miologic Studies Depression Scale
a
Income data for 79 women
b
Birth outcomes for 90 women
c
Defined as no previous pregnancy lasting at least 6 months
d
Inadequate or deficient vitamin D (Institute of Medicine 2011)
e
18 women with inflammatory biomarker IL-6 <LOD (19.8 % were set to
the mid-point between 0 and LOD (Alper et al. 2010)
E.E. Accortt et al.
trimester CESD score, rate of preterm birth or levels of 25-
OHD, hs-CRP, IL-6, or IL-10 (all p>0.05).
Sample characteristics on vitamin D status
Sixty percent had their vitamin D status measured during
April–December whereas 40 % had vitamin D levels mea-
sured during January–March. The average level of 25(OH)D
in the sample was 13.2 ng/ml (SD= 9.4), and 85 % (N=77)
met criteria for vitamin D inadequacy or deficiency (vitamin
D≤20 ng/ml) by the standard cutoff (Ross et al. 2011).
Furthermore, 97 % (N=88) did not meet the level suggested
for pregnant or lactating women (vitamin D ≥30 ng/ml) by the
Endocrine Society’s Clinical Practice Guideline (Holick et al.
2011). An authoritative source on recommended serum levels
of vitamin D for African Americans, or specifically pregnant
African Americans, does not exist. Therefore, we examined
two additional vitamin D cutoffs (risks for cardiovascular dis-
ease and all-cause mortality) to be conservative.
2
Vitamin D status and PPD symptoms
The mean EPDS score was 5.20 (SD=4.95). Twelve percent
of this sample had EPDS scores ≥12 indicating probable PPD,
a rate comparable to rates in other reports (Centers for Disease
Control and Prevention 2008), and EPDS scores were signif-
icantly higher in overweight and obese women (mean =5.98,
SD=5.17) as compared to normal-weight women (mean=
3.82, SD=4.28, p< 0.05), as reported in a recent meta-
analysis (Molyneaux et al. 2014), although rates were still in
the low range of the EPDS in both groups. The entire de-
pressed subgroup (11 of 11) was below the most lenient cutoff
for low vitamin D status (≤30 ng/ml), although the rate was
not statistically significant from the non-depressed subsample
in which 96 % (77 of 80) were vitamin D deficient.
Sample characteristics on inflammatory markers
Levels of prenatal inflammatory biomarkers (Table 1)were
higher than those reported in other studies (Coussons-Read
et al. 2007; Madan et al. 2009; Vassiliadis et al. 1998), poten-
tially due to differing assay modalities or racial composition of
the samples.
3
Associations of 25(OH)D and inflammatory markers
with PPD symptoms
We first examined the association between demographic and
medical variables with postpartum depressive symptoms to
determine whether these should be included in the model as
potential confounders. The significant associations between
history of depression
4
, BMI, and low education with postpar-
tum depressive symptoms were the basis for inclusion of these
three variables in analyses. Additional theoretically derived
confounding variables included maternal age, marital status,
prenatal depressive symptoms (CES-D), and season of vita-
min D measurement. Regression analyses were then conduct-
ed to examine whether vitamin D status and inflammatory
marker levels were associated with depressive symptoms.
Tables 2and 3display results from regression analyses.
Tab le 2shows that log 25(OH)D was not significantly associ-
ated with postpartum depressive symptomatology in unadjusted
regression analyses (β=−0.145, p=0.172). When adjusted for
covariates, the effect was marginally significant (β=−0.209,
p=0.058); higher log 25(OH)D was associated marginally with
lower EPDS symptoms. None of the individual inflammatory
markers were significantly associated with EPDS scores; how-
ever, as shown in Table 3, IL-6 and the IL-6/1 L-10 ratio both
significantly moderated the association between prenatal
25(OH)D levels and PPD symptoms. Figures 1and 2depict
moderation results. The unstandardized simple slope for wom-
en 1 SD above the mean of IL-6 was significant, such that in
women with higher levels of IL-6, lower prenatal 25(OH)D
levels were associated with higher levels of postpartum depres-
sive symptoms. The unstandardized simple slope for women 1
SD above the mean of the IL-6/IL-10 ratio was also significant,
such that in women with greater levels of IL-6/IL-10 ratio,
lower prenatal vitamin D levels were associated with higher
levels of PPD symptoms. A test of moderation by IL-10 alone
was marginally significant (p=0.09), and there was no evidence
that hs-CRP moderated the association (p=0.12). Finally, these
findings persisted when controlling for prenatal depressive
symptoms, using a standardized though different screener mea-
sured in second trimester (CES-D, Radloff 1977)andwhen
controlling for preterm birth (results not shown).
Discussion
Findings from this study suggest that higher levels of
25(OH)D in early pregnancy in African American women
2
The other vitamin cutoffs we considered were vitamin D ≤15 ng/ml for
risk of cardiovascular disease (Wang et al. 2008) and vitamin D ≤10 ng/ml
used for risk for all-cause mortality (Dobnig et al. 2008). Analyses
showed that 69 % of our sample (63 women) met the cardiovascular risk
cutoff, and 48 % (44 women) were at risk of all-cause mortality.
3
A table displaying the Spearman correlation coefficients among the
main study variables, including vitamin D status and each inflammatory
marker can be provided by the authors.
4
History of depression was abstracted from the electronic medical record.
We do not know if any women sought out psychological or complemen-
tary treatments for depression nor how long ago a previous episode oc-
curred. One woman in our study reported taking antidepressants (Zoloft)
during her postpartum visit, and removing her from analyses did not
change our results.
Prenatal vitamin D and postpartum depressive symptoms
may contribute to lower postpartum depressive symptoms,
though the association of vitamin D status and EPDS scores
was only marginally significant (p=0.058). Moreover, when
pregnant African American women were lower in 25(OH)D
and higher in pro-inflammatory cytokine levels (specifically
IL-6, IL-6/IL-10 ratio) in adjusted models, they reported more
postpartum depressive symptoms. Although two recent inves-
tigations reported an association between prenatal vitamin D
status and PPD symptoms similar to ours, neither study
assessed prenatal depression; thus, depression symptoms
may have predated the measurement of 25(OH)D (Gur et al.
2014; Robinson et al. 2014). In contrast, our study found
evidence for a possible inverse association between
25(OH)D and PPD symptoms controlling for prenatal depres-
sive symptoms and also for history of depression.
Previous research has shown that vitamin D appears to
downregulate the Th1 or cellular immunity pathway and stim-
ulate the Th2 or humoral immunity pathway (McCann and
Ames 2008). Vitamin D also regulates placental development
and function and promotes tolerance of the fetus, likely by
enhancing anti-bacterial and anti-inflammatory responses in
both the maternal and fetal components of the placenta
(Arora and Hobel 2010). These regulatory shifts are accom-
plished in part by decreasing the production of pro-
inflammatory cytokines and by increasing the production of
anti-inflammatory cytokines (McCann and Ames 2008).
Tab l e 2 Linear regression model testing the main effects of log-
transformed vitamin D status and cytokines on EPDS
Variable Unadjusted Adjusted
βpβp
Log 25(OH)D −0.145 0.172 −0.209 0.058
Log IL-6 0.084 0.430 0.142 0.180
Log IL-10 −0.004 0.971 0.094 0.349
Log IL-6/IL-10 ratio 0.089 0.401 0.100 0.354
Log hs-CRP 0.174 0.099 0.147 0.209
Dependent variable: Edinburgh Postnatal Depression Scale (EPDS)
score, N=91. Each independent variable in the BUnadjusted^column
was run separately. Each variable in the BAdjusted^column was run
separately with the following covariates: maternal age, education (high
school diploma), marital status, history of depression, season of vitamin
D measurement, and pre-pregnancy BMI. Results pertaining to covariates
were obtained but were omitted from this table. Only two covariates were
consistently statistically significantly associated with EPDS; Beducation^
was negatively associated, and Bhistory of depression^was positively
associated with EPDS in each analysis. Also, Bbeing married^was pos-
itively associated with EPDS in the Log Vitamin D analysis
Tabl e 3 Linear regression models testing the interactive effects of log-
transformed vitamin D and cytokines on EPDS
Vari a b l e βp
a. Log 25(OH)D −0.123 0.271
Log IL-6 0.062 0.559
Log 25(OH)D X Log IL-6 −0.232 0.025
b. Log 25(OH)D −0.208 0.057
Log IL-10 0.061 0.535
Log 25(OH)D X Log IL-10 −0.166 0.093
c. Log 25(OH)D 0.257 0.233
Log IL-6/IL-10 Ratio −0.065 0.580
Log 25(OH)D X Log IL-6/IL-10 ratio −0.531 0.016
d. Log 25(OH)D −0.182 0.097
Log hs-CRP 0.155 0.178
Log 25(OH)D X Log hs-CRP −0.152 0.118
Statistically significant pvalues are shown in italics
Models Ba–d^were adjusted for maternal age, education (high school
diploma), marital status, history of depression, season of vitamin D mea-
surement, and pre-pregnancy BMI. Results pertaining to covariates were
obtained but were omitted from this table. Only two covariates were
consistently statistically associated with EPDS in all above models;
Beducation^was negatively associated, and Bhistory of depression^was
positively associated with EPDS. In models a, c, and d, Bbeing married^
was also positively associated with EPDS
-5 0 5 10 15
Postpartum Depressive Symptoms
(EPDS)
-.7 -.5 -.3 -.1 .1 .3 .5 .7 .9
Log Vitamin D (Centered)
1 SD below mean
Mean IL-6
1 SD above mean
Log IL-6 (Centered)
Interaction Results
Fig. 1 Interleukin-6 (IL-6) significantly moderates the relationship be-
tween log 25(OH)D and EPDS
-10 -5 0 5 10 15
Postpartum Depressive Symptoms
(EPDS)
-.7 -.5 -.3 -.1 .1 .3 .5 .7 .9
Log Vitamin D (Centered)
1 SD below mean
Mean IL-6/IL-10 Ratio
1 SD above mean
Log IL-6/IL-10 Ratio (Centered)
Interaction Results
Fig. 2 The ratio of interleukin-6 (IL-6) to interleukin-10 (IL-10)
significantly moderates the relationship between log 25(OH)D and EPDS
E.E. Accortt et al.
Additionally, cytokine signals from the periphery can be trans-
mitted to the brain via the circumventricular organs, unique
areas of the brain outside the blood-brain barrier (BBB), or
vagal afferents (Buller 2001;Maier2003), potentially leading
to changes in maternal neurotransmission and later depressive
symptomatology. Since inflammatory markers were collected
in the second trimester and may be mechanisms linking first
trimester vitamin D status to PPD symptomatology, we ran an
exploratory analysis on possible mediation by each inflamma-
tory marker of the association between 25(OH)D and PPD
symptoms. None of these analyses were statistically signifi-
cant. We recommend testing this in future studies with larger
samples. We also did not find strong evidence for direct effects
of inflammatory markers on postpartum depressive symptoms
although we note that reviews of this literature show mixed
results (Osborne and Monk 2013). It is possible that higher
levels of cytokines and lower levels of 25(OH)D increase risk
for later depressive symptoms in pregnant women only when
acting synergistically. Whether this is unique to African
American women, we do not know. We also did not find
evidence of significant moderation by hs-CRP, a marker of
low-grade inflammation. Consistent with our results, a recent
study in a non-pregnant sample found an association between
depression and CRP in Caucasian but not African American
women (Amyre et al. 2011).
Levels of prenatal inflammatory markers were higher here
than those reported in other studies (Coussons-Read et al.
2007; Madan et al. 2009; Vassiliadis et al. 1998), perhaps
because the sample was all African American women.
Differences in assay modality (differing LOD) may also have
led to some differences between our study and others. Our
findings, however, were similar to those of Christian et al.
(2009) whose sample was 57 % African American
(Christian et al. 2009). Two recent studies focused on levels
of cytokines across pregnancy in racially diverse samples, and
one attributed differences in levels of cytokines to high BMI,
not race (Blackmore et al. 2014; Christian and Porter 2014).
Notably, our findings remained significant even after control-
ling for BMI.
We suspect that null results in past studies of vitamin D
status and depression in non-pregnant (e.g., Bertone-Johnson
2009) and pregnant (Nielsen et al. 2013)samplesmaybedue
to low levels of inflammatory markers. Our preliminary find-
ings suggest that the effects of low 25(OH)D on PPD symp-
toms may be potentiated only in the presence of inflammation.
Whether inflammation moderates the relationship between
low 25(OH)D and depression in non-pregnant samples is
therefore worth testing.
Strengths, limitations, and future directions
Compared to prior research, this prospective study utilized a
relatively large sample which permitted us to test interactions
of study variables. In addition, women in our sample were
studied both in pregnancy and postpartum, and having medi-
cal and psychosocial data enabled us to control for many
relevant confounders. The sample was also solely African
Americans who are at higher risk of inflammation and low
vitamin D status. Finally, vitamin D measures that were ob-
tained as part of HFHS clinical care were assayed with a
competitive chemiluminescence immunoassay platform, con-
sistent with other large clinical laboratories. Importantly,
HFHS participates in the Vitamin D External Quality
Assessment Scheme, as recommended by experts in the field
(e.g., Hollis 2008).
Limitations of this secondary data analysis include some
missing data on depressive symptoms 6 weeks postpartum
because a portion of women either did not return for their
routine postpartum visits or their clinician did not record
the EPDS result. Future research would benefit from
assessing depressive symptoms repeatedly rather than only
once in the first 6 weeks at their obstetrics postpartum
follow-up visit, e.g., at 12 weeks. Additionally, the percent-
age of women with elevated postpartum depression scores
in our sample is consistent with other reports of low income
African American women (Dolbier et al. 2013). Further,
participants dropped from analyses due to lack of EPDS
scores were of similar age and 25(OH)D levels and had
comparable prenatal CES-D scores to those included.
Thus, underestimates of depressive symptoms due to attri-
tion seem less probable.
Finally, 74 % of women in this sample reported use of a
vitamin D supplement (50,000 IU/week of vitamin D
2
)which
might have raised mean levels in the sample and also may
have favorably affected levels of inflammatory markers.
However, serum levels of 25(OH)D were measured between
9 and 13 weeks gestation with supplementation prescribed
thereafter in only a portion of women.
5
Inflammatory markers
were measured later between 13–28 weeks of gestation, and
they are normally lower at this point during pregnancy and
higher in early and late pregnancy. However, if a short-term
benefit of vitamin D supplementation occurred, it would most
likely have attenuated the association we detected with in-
flammatory markers and PPD symptoms. In a small subset
of women in the sample (n=38 of 203), we measured the same
inflammatory cytokines again in the 3rd trimester. There were
high paired sample correlations among the cytokines from the
2nd to 3rd trimester: IL-6 (r=0.57; p<0.001), IL-10 (r=0.53;
p=0.001). This suggests that our results were not necessarily
influenced by large trimester-specific changes in these
cytokines.
5
IU refers to international units, which equal 25 ng. Vitamin D supple-
ments were prescribed to 87 % of the women with 74 % reporting use of
the supplement at a prenatal care visit. Most vitamin D supplements
prescribed were 50,000 IU, taken once weekly.
Prenatal vitamin D and postpartum depressive symptoms
In addition to obtaining concurrent measures of 25(OH)D,
inflammatory markers, and depressive symptoms throughout
pregnancy and postpartum, it would be ideal to track adher-
ence of vitamin D use and determine whether levels of
25(OH)D influence the trajectory of depression during preg-
nancy and from pregnancy to postpartum. Future research
would also benefit from collecting dietary intake data and
sun exposure information from participants. We studied four
inflammatory markers/marker combinations, thus our findings
may be subject to multiple testing issues. Because exploratory
studies such as this rarely correct for multiple comparisons
and Bonferroni correction is considered conservative
(Gelman et al. 2012), it is not necessarily appropriate to hold
the findings to that standard; nonetheless, our significant mod-
eration results were very close to significance when tested by
Bonferroni criteria.
6
While these results are preliminary and
do not establish causation, they are novel and particularly
important given the prevalence of low 25(OH)D in this pop-
ulation of African American women. We hope that they may
spark interest in further research on this topic including ran-
domized controlled studies. Furthermore, this study does not
involve confirmed cases of PPD and future research can in-
clude diagnostic interviews.
An important clinical implication is to determine if in-
creased prenatal vitamin D supplementation can reduce PPD
symptoms using a randomized controlled trial.
Supplementation is particularly relevant for overweight and
obese pregnant women, as they are more likely to have insuf-
ficient 25(OH)D levels (Bodnar et al. 2007) and higher rates
of PPD (Molyneaux et al. 2014). Arguably, supplementation
might also be recommended only for those women with high
levels of inflammatory cytokines in pregnancy and those with
clinically significant depression. Indeed, two recent systemat-
ic reviews of vitamin D supplementation on depressive symp-
toms in the non-pregnant state suggested no overall effect on
depressive symptoms. In one review, the evidence suggested
that supplementation might be effective in reducing depres-
sive symptoms in individuals with clinically significant pre-
existing depression (Shaffer et al. 2014). The other, a meta-
analysis, separated randomized control trials into those with
and without methodological flaws and concluded that studies
without flaws demonstrated a statistically significant improve-
ment in depression with vitamin D supplements (Spedding
2014). Vitamin D supplementation is a cost-effective and safe
intervention during pregnancy (Hollis et al. 2011)thatmay
reduce prenatal and postpartum symptoms of depression and
benefit the mother’s health and that of the developing fetus
(Kalra et al. 2012; Morales et al. 2012). A recent randomized,
double-blind, placebo-controlled clinical trial of vitamin D
supplementation in 48 pregnant women resulted in significant
decreases in serum hs-CRP, fasting plasma glucose, systolic
blood pressure, and diastolic blood pressure compared with
placebo (Asemi et al. 2013). However, depressive symptoms
were not measured in this study. Randomized controlled trials
of prenatal vitamin D supplementation to reduce postpartum
depression, especially in overweight women with high levels
of prenatal inflammation, may be merited.
Conclusion
These findings provide the first evidence that low prenatal
25(OH)D and high prenatal inflammation together might pre-
dict future postpartum depressive symptomatology in African
American women. We know of no other studies that investi-
gated inflammatory moderation of the relationship between
25(OH)D levels and later depressive symptomatology in
non-pregnant women or men. These results may therefore
elucidate inflammatory moderators linking vitamin D status
to depressive symptoms, not only in pregnant African
American women, but potentially in other subgroups of the
population. Future research on the synergistic relationship of
25(OH)D and inflammation on depression is worthwhile.
Understanding how vitamin D alters the immune system
may shed new light on the emerging links between inflamma-
tion and depression.
Acknowledgments R.M.P. and A.C.B. conducted the study. E.E.A.
developed the research questions for secondary analyses, analyzed the
data, and wrote the article. C.D.S, A.C.B., and R.M.P. assisted in writing.
All authors contributed to the study design, interpretation of data, and
approved the final manuscript. The authors have no financial gain related
to the outcome of this research, and there are no potential conflicts of
interest. This project was supported by funding from the Institute for
Population Sciences, Health Assessment, Administration, Services, and
Economics (INPHAASE) to A.C.B. and R.M.P. and National Institutes of
Mental Health (T32MHI5750) postdoctoral fellowship to E.E.A. We
thank the participants in this study for contributing to this research and
increasing our knowledge about the experiences of African American
pregnant women. We acknowledge the support of the research team
members, Dayna Johnson, Project Manager, HenryFord Hospital Depart-
ment of Public Health Sciences and Christine Wells, Ph.D., Statistician,
UCLA Statistical Consulting Group. We also thank Judith E. Carroll,
Ph.D. of the UCLA Semel Institute for Neuroscience and Human Behav-
ior and the Dunkel Schetter lab for consultation in this work.
References
Adams JS, Hewison M (2008) Unexpected actions of vitamin D: new
perspectives on the regulation of innate and adaptive immunity. Nat
Clin Pract Endocrinol Metab 4:80–90. doi:10.1038/ncpendmet0716
Aghajafari N, Halligan GH, Roberts CL, Morris JM, Ashton AW (2011)
Systematic review of first-trimester vitamin D normative levels and
outcomes of pregnancy. Am J Obstet Gynecol 205:208. doi:10.
1016/j.ajog.2011.03.058,e1-7
Aghajafari F, Nagulesapillai T, Ronksley PE, Tough SC, O’Beirne M,
Rabi DM (2013) Association between maternal serum 25-
hydroxyvitamin D level and pregnancy and neonatal outcomes:
6
Bonferroni critical pof 0.0125 (IL-6 p= 0.025; IL-6/IL-10 p=0.016).
E.E. Accortt et al.
systematic review and meta-analysis of observational studies. BMJ
(Clin ResEd) 346:f1169. doi:10.1136/bmj.f1169
Alper CM, Li-Korotky HS, Lo CY, Doyle AP, Winther B, Doyle WJ
(2010) Nasal secretion concentrations of IL-5, IL-6, and IL-10 in
children with and without upper respiratory tract viruses. Arch
Otolaryngol Head Neck Surg 136:281–286
Amyre MA, Zhao L, Ahmed Y, Stoyanova N, De SC, Hooper W et al
(2011) Association between depression and inflammation—differ-
ences by race and sex: the META-Health study. Psychosom Med 73:
462–468
Anglin RES, Samaan Z, Walter SD, McDonald SD (2013) Vitamin D
deficiency and depression in adults: systematic review and meta-
analysis. Br J Psychiatry 202(2):100–107. doi:10.1192/bjp.bp.111.
106666
Arora CP, Hobel CJ (2010) Vitamin D—a novel role in pregnancy.
Bioploym Cell 26:97–104
Asemi Z, Samimi M, Tabassi Z, Shakeri H, Esmaillzadeh A (2013)
Vitamin D supplementation affects serum high-sensitivity C-reac-
tive protein, insulin resistance, and biomarkers of oxidative stress in
pregnant women. J Nutr 143:1432–1438. doi:10.3945/jn.113.
177550.stress
Beck CT (2001) Predictors of postpartum depression. Nurs Res 50(5):
275–285. doi:10.1097/00006199-200109000-00004
Bennett HA, Einarson A, Taddio A, Koren G, Einarson TR (2004)
Prevalence of depression during pregnancy: systematic review.
Obstet Gynecol 103(4):698–709. doi:10.1097/01.AOG.
0000116689.75396.5f
Bertone-Johnson ER (2009) Vitamin D and the occurrence of depression:
causal association or circumstantial evidence? Nutr Rev 67(8):481–
492. doi:10.1111/j.1753-4887.2009.00220.x
Blackmore ER, Groth SW, Chen DGD, Gilchrist MA, O’Connor TG,
Moynihan JA (2014) Depressive symptoms and proinflammatory
cytokines across the perinatal period in African American women.
J Psychosom Obstet Gynaecol 35(1):8–15. doi:10.3109/0167482X.
2013.868879
Bobbitt KR, Peters RM, Li J, Rao SD, Woodcroft KJ, Cassidy-Bushrow
AE (2014) Early pregnancy vitamin D and patterns of antenatal
inflammation in African–American women. J Reprod Immunol:1–
7. doi:10.1016/j.jri.2014.09.054
Bodnar LM, Simhan HN (2010) Vitamin D may be a link to black-white
disparities in adverse birth outcomes. Obstet Gynecol Surv 65:273–
284. doi:10.1097/OGX.0b013e3181dbc55b
Bodnar LM, Catov JM, Roberts JM, Simhan HN (2007) Prepregnancy
obesity predicts poor vitamin D status in mothers and their neonates.
J Nutr 137(11):2437–2442
Boufidou F, Lambrinoudaki I, Argeitis J, Zervas IM, Pliatsika P,
Leonardou AA, Nikolaou C (2009) CSF and plasma cytokines at
delivery and postpartum mood disturbances. J Affect Disord 115:
287–292. doi:10.1016/j.jad.2008.07.008
Brandenbarg J, Vrijkotte TGM, Goedhart G, van Eijsden M (2012)
Maternal early-pregnancy vitamin D status is associated with mater-
nal depressive symptoms in the Amsterdam born children and their
development cohort. Psychosom Med 74(7):751–757. doi:10.1097/
PSY.0b013e3182639fdb
Breen EC, Reynolds SM, Cox C, Jacobson LP, Magpantay L, Mulder
CB, Norris PJ (2011) Multisite comparison of high-sensitivity mul-
tiplex cytokine assays. Clin Vaccine Immunol 18(8):1229–1242
Buller K (2001) Role of circumventricular organs in pro-inflammatory
cytokine-induced activation of the hypothalamic-pituitary-adrenal
axis. Clin Exp Pharmacol Physiol 28:581–589
Cassidy-Bushrow AE, Peters RM, Johnson DA, Li J, Rao DS (2012a)
Vitamin D nutritional status and antenatal depressive symptoms in
African American women. J Women's Health 21(11):1189–1195.
doi:10.1089/jwh.2012.3528
Cassidy-Bushrow AE, Peters RM, Johnson DA, Templin TN (2012b)
Association of depressive symptoms with inflammatory biomarkers
among pregnant African-American women. J Reprod Immunol
94(2):202–209. doi:10.1016/j.jri.2012.01.007
Centers for Disease Control and Prevention (2008) Morbidity and mor-
tality weekly report (MMWR). Retrieved from http://www.cdc.gov/
mmwR/preview/mmwrhtml/mm5714a1.htm
Christian LM (2012) Psychoneuroimmunology in pregnancy: immune
pathways linking stress with maternal health, adverse birth out-
comes, and fetal development. Neurosci Biobehav Rev 36(1):350–
61. doi:10.1016/j.neubiorev.2011.07.005
ChristianLM, Porter K (2014) Longitudinal changes in serum proinflam-
matory markers across pregnancy and postpartum: effects of mater-
nal body mass index. Cytokine 70(2):134–140. doi:10.1016/j.cyto.
2014.06.018
Christian LM, Franco A, Glaser R, Iams JD (2009) Depressive symptoms
are associated with elevated serum proinflammatory cytokines
among pregnant women. Brain Behav Immun 23(6):750–4. doi:
10.1016/j.bbi.2009.02.012
Christian LM, Glaser R, Porter K, Iams JD (2013) Stress-induced inflam-
matory responses in women: effects of race and pregnancy.
Psychosom Med 75(7):658–69. doi:10.1097/PSY.
0b013e31829bbc89
Corwin EJ, Pajer K (2008) The psychoneuroimmunology of postpartum
depression. J Women's Health 17(9):1529–34. doi:10.1089/jwh.
2007.0725
Coussons-Read ME (2012) The psychoneuroimmunology of stress in
pregnancy. Curr Dir Psychol Sci 21(5):323–328. doi:10.1177/
0963721412453720
Coussons-Read ME, Okun ML, Nettles CD (2007) Psychosocial stress
increases inflammatory markers and alters cytokine production
across pregnancy. Brain Behav Immun 21(3):343–50. doi:10.1016/
j.bbi.2006.08.006
Cox JL, Holden JM, Sagovsky R (1987) Detection of postnatal depres-
sion. Development of the 10-item Edinburgh Postnatal Depression
Scale. Br J Psychiatry 150(6):782–786. doi:10.1192/bjp.150.6.782
Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley KW (2008)
From inflammation to sickness and depression: when the immune
system subjugates the brain. Nat Rev Neurosci 9(1):46–56. doi:10.
1038/nrn2297
Deverts DJ, Cohen S, DiLillo VG, Lewis CE, Kiefe C, Whooley M,
Matthews KA (2010) Depressive symptoms, race, and circulating
C-reactive protein: the Coronary Artery Risk Development in
Young Adults (CARDIA) study. Psychosom Med 72(8):734–41.
doi:10.1097/PSY.0b013e3181ec4b98
Dhabhar FS, Burke HM, Epel ES, Mellon SH, Rosser R, Reus VI,
Wolkowitz OM (2009) Low serum IL-10 concentrations and loss
of regulatory association between IL-6 and IL-10 in adults with
major depression. J Psychiatr Res 43(11):962–9. doi:10.1016/j.
jpsychires.2009.05.010
Dobnig H, Pilz S, Scharnagl H, Renner W, Seelhorst U, Wellnitz B,
Maerz W (2008) Independent association of low serum 25-
hydroxyvitamin d and 1,25-dihydroxyvitamin d levels with all-
cause and cardiovascular mortality. Arch Intern Med 168:1340–
1349. doi:10.1001/archinte.168.12.1340
Dolbier C, Rush T, Sahadeo L, Shaffer M, Thorp J (2013) Relationships
of race and socioeconomic status to postpartum depressive symp-
toms in rural African American and Non-Hispanic White women.
Matern Child Health J 17:1277–1287. doi:10.1007/s10995-012-
1123-7
Ersfeld DL, Rao DS, Body JJ et al (2004) Analytical and clinical valida-
tion of the 25 OH vitamin D assay for the LIAISON automated
analyzer. Clin Biochem 37:867–874
Gelman A, Hill J, Yajima M (2012) Why we (usually) don't have to worry
about multiple comparisons. J Res Educ Eff 5(2):189–211
Grace SL, Evindar A, Stewart DE (2003) The effect of postpartum de-
pression onchild cognitive development and behavior: a review and
Prenatal vitamin D and postpartum depressive symptoms
critical analysis of the literature. Arch Womens Ment Health 6(4):
263–274. doi:10.1007/s00737-003-0024-6
Gur EB, Gokduman A, Turan GA, Tatar S, Hepyilmaz I, Zengin EB,
Eskicioglu F, Guclu S (2014) Mid-pregnancy vitamin D levels and
postpartum depression. Eur J Obstet Gynecol ReprodBiol 179:110–
116
Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA,
Heaney RP, Weaver CM (2011) Evaluation, treatment, and preven-
tion of vitamin D deficiency: an Endocrine Society clinical practice
guideline. J Clin Endocrinol Metab 96:1911–1930. doi:10.1210/jc.
2011-0385
Hollis BW (2008) Measuring 25-hydroxyvitamin D in a clinical
environment: challenges and needs. Am J Clin Nutr 88:507S–
510S
Hollis BW, Johnson D, Hulsey TC, Ebeling M, Wagner CL (2011)
Vitamin D supplementation during pregnancy: double-blind, ran-
domized clinical trial of safety and effectiveness. J Bone Miner
Res 26:2341–57. doi:10.1002/jbmr.463
Howell EA, Mora P, Leventhal H (2006) Correlates of early postpartum
depressive symptoms. Matern Child Health J 10:149–157. doi:10.
1007/s10995-005-0048-9
Howren MB, Lamkin DM, Suls J (2009) Associations of depression with
C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med
71(2):171–186. doi:10.1097/PSY.0b013e3181907c1b
Institute of Medicine (2011) Dietary reference intakes for calcium and
vitamin D. National Academy Press, Washington, DC
Kalra P, Das V, Agarwal A, Kumar M, Ramesh V, Bhatia E, Bhatia V
(2012) Effect of vitamin D supplementation during pregnancy on
neonatal mineral homeostasis and anthropometry of the newborn
and infant. Br J Nutr 108(06):1052–1058. doi:10.1017/
S0007114511006246
Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, Modlin RL
(2006) Toll-like receptor triggering of a vitamin D-mediated human
antimicrobial response. Science 311(5768):1770–1773. doi:10.
1126/science.1123933
Madan JC, Davis JM, Craig WY, Collins M, Allan W, Quinn R,
Dammann O (2009) Maternal obesity and markers of inflammation
in pregnancy. Cytokine 47(1):61–4. doi:10.1016/j.cyto.2009.05.004
Maes M, Lin A, Ombelet W, Stevens K, Kenis G, De Jongh R, Bosmans
E (2000) Immune activation in the early puerperium is related to
postpartum anxiety and depressive symptoms.
Psychoneuroendocrinology 25(2):121–137. doi:10.1016/S0306-
4530(99)00043-8
Maier SF (2003) Bi-directional immune-brain communication: implica-
tions for understanding stress, pain, and cognition. Brain Behav
Immun 17(2):69–85. doi:10.1016/S0889-1591(03)00032-1
McCann JC, Ames BN (2008) Is there convincing biological or behav-
ioral evidence linking vitamin D deficiency to brain dysfunction?
FASEB J 22(4):982–1001. doi:10.1096/fj.07-9326rev
Miller AH, Maletic V, Raison CL (2009) Inflammation and its discon-
tents: the role of cytokines in the pathophysiology of major depres-
sion. Biol Psychiatry 65(9):732–741. doi:10.1016/j.biopsych.2008.
11.029
Molyneaux E, Poston L, Ashurst-Williams S, Howard LM (2014)
Obesity and mental disorders during pregnancy and postpartum: a
systematic review and meta-analysis. Obstet Gynecol 123(4):857–
67. doi:10.1097/AOG.0000000000000170
Mor G, Cardenas I, Abrahams V, Guller S (2011) Inflammation and
pregnancy: the role of the immune system at the implantation site.
Ann N Y Acad Sci 1221:80–87. doi:10.1111/j.1749-6632.2010.
05938.x
Morales E, Guxens M, Llop S, Rodriguez-Bernal CL, Tardon A, Riano I,
Sunyer J (2012) Circulating 25-hydroxyvitamin D3 in pregnancy
and infant neuropsychological development. Pediatrics 130(4):
e913–e920. doi:10.1542/peds.2011-3289
Murphy PK, Mueller M, Hulsey TC, Ebeling MD, Wagner CL (2010) An
exploratory study of postpartum depression and vitamin D. J Am
Psychiatr Nurses Assoc 16(3):170–177
Nassar N, Halligan GH, Roberts CL, Morris JM, Ashton AW (2011)
Systematic review of first-trimester vitamin D normative levels
and outcomes of pregnancy. Am J Obstet Gyneco 205:208.e1–7
Nielsen NO, Strøm M, Boyd HA, Andersen EW, Wohlfahrt J, Lundqvist
M, Melbye M (2013) Vitamin D status during pregnancy and the
risk of subsequent postpartum depression: a case-control study.
PLoS One 8(11):e80686. doi:10.1371/journal.pone.0080686
Okun ML, Luther J, Prather AA, Perel JM, Wisniewski S, Wisner KL
(2011) Changes in sleep quality, but not hormones predict time to
postpartum depression recurrence. J Affect Disord 130:378–384.
doi:10.1016/j.jad.2010.07.015
Orr ST, Blazer DG, James SA (2006) Racial disparities in elevated pre-
natal depressive symptoms among black and white women in east-
ern North Carolina. Ann Epidemiol 16:463–468. doi:10.1016/j.
annepidem.2005.08.004
Osborne LM, Monk C (2013) Perinatal depression—the fourth inflam-
matory morbidity of pregnancy?: Theory and literature review.
Psychoneuroendocrinology 38(10):1929–52. doi:10.1016/j.
psyneuen.2013.03.019
Paulson JF, Dauber S, Leiferman JA (2006) Individual and combined
effects of postpartum depression in mothers and fathers on parenting
behavior. Pediatrics 118:659–668. doi:10.1542/peds.2005-2948
Radloff LS (1977) The CES-D scale: a self-report depression scale for
research in the general population. Appl Psychol Meas 1(3):385–
401. doi:10.1177/014662167700100306
Rao DS (1999) Perspective on assessment of vitamin D nutrition. J Clin
Densitom 2:457–464
Robinson M, Whitehouse AJO, Newnham JP, Gorman S, Jacoby P, Holt
BJ, Kusel MMH (2014) Low maternal serum vitamin D during
pregnancy and the risk for postpartum depression symptoms. Arch
Womens Ment Health 17(3):213–9. doi:10.1007/s00737-014-
0422-y
Ross A, Manson J, Abrams S, Aloia J, Brannon P, Clinton S, Shapses S
(2011) The 2011 report on dietary reference intakes for calcium and
vitamin D from the Institute of Medicine: what clinicians need to
know. J Clin Endocrinol Metabo 96(1):53–8. doi:10.1210/jc.2010-
2704
Segre LS, O’Hara MW, Losch ME (2006) Race/ethnicity and perinatal
depressed mood. J Reprod Infant Psychol 24(02):99–106. doi:10.
1080/02646830600643908
Shaffer JA, Edmondson D, Wasson LT, Falzon L, Homma K, Ezeokoli N,
Davidson KW (2014) Vitamin D supplementation for depressive
symptoms: a systematic review and meta-analysis of randomized
controlled trials. Psychosom Med 76(3):190–6. doi:10.1097/PSY.
0000000000000044
Skalkidou A, Sylvén SM, Papadopoulos FC, Olovsson M, Larsson A,
Sundström-Poromaa I (2009) Risk of postpartum depression in as-
sociation with serum leptin and interleukin-6 levels at delivery: a
nested case-control study within the UPPSAT cohort.
Psychoneuroendocrinology 34:1329–1337. doi:10.1016/j.
psyneuen.2009.04.003
Spedding S (2014) Vitamin D and depression: a systematic review and
meta-analysis comparing studies with and without biological flaws.
Nutrients 6(4):1501–1518. doi:10.3390/nu6041501
Theodoratou E, Tzoulaki I, Zgaga L, Ioannidis JPA (2014) Vitamin D and
multiple health outcomes: umbrella review of systematic reviews
and meta-analyses of observational studies and randomised trials.
BMJ (Clin Res Ed) 348:g2035. doi:10.1136/bmj.g2035
Thota C, Farmer T, Garfield RE, Menon R, Al-Hendy A (2012) Vitamin
D elicits anti-inflammatory response, inhibits contractile-associated
proteins, and modulates toll-like receptors in human myometrial
cells. Reprod Sci 20(4):463–75. doi:10.1177/1933719112459225
E.E. Accortt et al.
Van Etten E, Mathieu C (2005) Immunoregulation by 1,25-
dihydroxyvitamin D3: basic concepts. J Steroid Biochem Mol Biol
97(1-2):93–101. doi:10.1016/j.jsbmb.2005.06.002
Vassiliadis S, Ranella A, Papadimitriou L, Makrygiannakis A,
Athanassakis I (1998) Serum levels of pro- and anti-
inflammatory cytokines in non-pregnant women, during preg-
nancy, labour and abortion. Mediat Inflamm 7(2):69–72. doi:
10.1080/09629359891199
Veldman CM, Cantorna MT, DeLuca HF (2000) Expression of 1,25-
dihydroxyvitamin D(3) receptor in the immune system. Arch
Biochem Biophys 374(2):334–338. doi:10.1006/abbi.1999.1605
Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K,
Vasan RS (2008) Vitamin D deficiency and risk of cardiovascular
disease. Circulation 117(4):503–511. doi:10.1161/
CIRCULATIONAHA.107.706127
Wei S-Q, Qi H-P, Luo Z-C, Fraser WD (2013) Maternal vitamin D status
and adverse pregnancy outcomes: a systematic review and meta-
analysis. J Matern Fetal Neonatal Med 26:889–99. doi:10.3109/
14767058.2013.765849
Yim IS, Tanner Stapleton LR, Guardino CM, Hahn Holbrook J, Dunkel
Schetter C (2015) Biological and psychosocial predictors of post-
partum depression: systematic review and call for integration. Annu
Rev Clin Psychol 11:99–137
Zelkowitz P, Milet TH (1996) Postpartum psychiatric disorders: their
relationship to psychological adjustment and marital satisfaction in
the spouses. J Abnorm Psychol 105:281–285. doi:10.1037/0021-
843X.105.2.281
Zhang R, Naughton D (2010) Vitamin D in health and disease. Nutr J
9(65):65. doi:10.1186/1475-2891-9-65
Zhang Y, Leung DYM, Richers BN, Liu Y, Remigio LK, Riches DW,
Goleva E (2012) Vitamin D inhibits monocyte/macrophage proin-
flammatory cytokine production by targeting MAPK phosphatase-1.
J Immunol 188(5):2127–35. doi:10.4049/jimmunol.1102412
Prenatal vitamin D and postpartum depressive symptoms
A preview of this full-text is provided by Springer Nature.
Content available from Archives of Women's Mental Health
This content is subject to copyright. Terms and conditions apply.