Psychosomatic Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nierop, A.
Right arrow Articles by Ehlert, U.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nierop, A.
Right arrow Articles by Ehlert, U.
Related Collections
Right arrow Neuroendocrine
Right arrow Depression
Right arrow Stress and Coping
Right arrow Pregnancy
Psychosomatic Medicine 68:931-937 (2006)
© 2006 American Psychosomatic Society

ORIGINAL ARTICLES

Are Stress-Induced Cortisol Changes During Pregnancy Associated With Postpartum Depressive Symptoms?

Ada Nierop, PhD, Aliki Bratsikas, PhD, Roland Zimmermann, MD and Ulrike Ehlert, PhD

From the Department of Clinical Psychology and Psychotherapy, University of Zurich, Zurich, Switzerland (A.N., A.B., U.E.); and the Department of Obstetrics, University Hospital of Zurich, Zurich, Switzerland (R.Z.).

Address correspondence and reprint requests to Ulrike Ehlert, PhD, Department of Clinical Psychology and Psychotherapy, University of Zurich, Zurichbergstrasse 43, CH-8044 Zurich, Switzerland. E-mail: u.ehlert{at}psychologie.unizh.ch


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objective: The purpose of this study was to examine the association between psychobiological stress reactivity during healthy pregnancy and depressive symptoms in the early puerperium.

Methods: A sample of healthy nulliparous pregnant women (N = 57) between the ages of 21 and 35 years underwent a standardized psychosocial stress test during pregnancy. Within an average of 13 days after delivery, postpartum depressive symptoms were assessed using the German version of the Edinburgh postnatal depression scale (EPDS). The sample was divided into a group with probable cases (EPDS score >9, N = 16) and a group with probable noncases (EPDS score ≤9, N = 41).

Results: The probable case group showed significantly higher cortisol responses to the stress test compared with the probable noncase group, whereas baseline levels did not differ. Additionally, women in the probable case group showed significantly higher state anxiety and lower mood state throughout the experiment. Furthermore, the probable case group showed higher stress susceptibility, higher trait anxiety, and higher levels in the Symptom Checklist. No differences were found for prior episodes of psychiatric disorders, obstetrical complications, birth weight, or mode of delivery.

Conclusions: Our data provide evidence that healthy pregnant women developing postpartum depressive symptoms might already be identified during pregnancy by means of their higher cortisol reactivity and their higher psychological reactivity in response to psychosocial stress. Further investigations are required to explore whether higher psychobiological stress responses not only precede depressive symptoms within 2 weeks after birth, but might also predict postpartum major depression.

Key Words: cortisol • psychosocial stress • pregnancy • postpartum depressive symptoms

Abbreviations: ANOVA = analysis of variance; ANCOVA = analysis of covariance; AUC = area under the curve; HPA = hypothalamic–pituitary–adrenal; EPDS = Edinburgh Postnatal Depression Scale; MDBF = Multidimensional Mood Questionnaire; MESA = Stress Susceptibility Scale; SEM = standard error of means; SD = standard deviation; SCL90-R = Symptom-Checklist-90-Revised; STAI = State-Trait Anxiety Inventory; TSST = Trier Social Stress Test; VAS=Visual Analog Scale.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Postpartum depressive symptoms occur in approximately 50% of women who have recently given birth, whereas nonpsychotic postpartum major depression occurs in approximately 10% to 20% of women within 6 months postpartum (1). In the case of postpartum depressive symptoms, the mothers’ ability to respond sensitively and competently to their newborn is compromised (2). Postpartum disorders can have long-term adverse effects for both mother and child if it remains untreated (3), and even the presence of mild depressive symptoms has been found to be associated with impaired maternal functioning comparable to that of women with postpartum depression (4). A broad range of research into the etiology of postpartum mood disorders has already been conducted. Although psychosocial factors such as marital disharmony, number of life events in the previous year, and poor social support may be associated with postpartum depression, a number of hypotheses have been formulated to account for postpartum depressive symptoms in terms of neuroendocrine dysfunction. Research in the field of endocrine factors associated with postpartum mood disorders has mainly focused on estradiol and progesterone and has produced conflicting findings (5,6). As yet, only a small number of studies have examined the role of cortisol.

Notably, all studies regarding the role of cortisol in depressive mood changes in the puerperium were based on the examination of unprovoked cortisol baseline levels. During healthy pregnancy, salivary cortisol baseline levels start rising after the 25th to 28th weeks of gestation, reaching concentrations at the end of pregnancy that are twice as high as in nonpregnant women (7). Kammerer et al. (8) found attenuated cortisol reactivity in pregnant women after a physical stress test. However, in contrast to this finding, in a recent study by our work group, the inducement of psychosocial stress in second- and third-trimester healthy pregnant women did not result in a restraint of the hypothalamic–pituitary–adrenal (HPA) compared with nonpregnant controls (9).

A study, conducted by Handley et al. (10), examined a sample of 71 women and measured cortisol at 36 and 38 weeks gestation on days 1 to 5 postpartum during hospital admission and at 6 weeks postpartum. Results revealed significantly elevated cortisol levels during the end of pregnancy (38 weeks) associated with more severe blues after delivery (10). Ehlert et al. (1) measured cortisol levels after delivery and found that women showed significantly higher elevated morning cortisol levels on those days after delivery when the postpartum depressive symptoms occurred compared with days with balanced mood and the group without depressive symptoms. However, other research groups did not find significant associations between elevated postpartum basal cortisol levels and depressive symptoms during the first 5 days after delivery (11,12). Pedersen et al. (13) found significantly higher levels of plasma cortisol after delivery in women with a history of major depression than those with no such history.

In summary, the sparse findings of associations between unprovoked basal cortisol levels, measured either during pregnancy or after delivery, and postpartum depressive symptoms are controversial. Notably, no study has investigated the relation between HPA reactivity of women during the time of their pregnancy and the occurrence of postpartum depressive symptoms.

In contrast, a broad body of research has clearly demonstrated the dysfunction of the unprovoked HPA axis in nonpregnant and nonpuerperal individuals with major depression. Alterations of HPA axis in terms of baseline hypercortisolism in nonpregnant adults with melancholic depression are a common finding and have been demonstrated in several studies (14–16). Besides the findings of elevated baseline cortisol levels and CRH levels at rest, numerous investigations have consistently reported abnormal glucocorticoid feedback in terms of blunted feedback sensitivity of the HPA axis (17–20).

Extending the knowledge regarding HPA functioning in depressed individuals demonstrated by studies of basal HPA activity and pharmacological challenge studies, several studies investigated cortisol in response to psychological stress in depressed individuals. There is evidence that depressed patients show blunted salivary cortisol reactivity compared with nondepressed individuals (21), with one study, for example, showing blunted cortisol responses after naturally occurring events in the daily lives of depressed outpatients measured by event assessment (22). A further study using a naturalistic stressor consisting of an unexpected visit by health professionals at the home of very low-income Mexican women after an in-depth interview and physical assessment showed that women with more severe depressive symptoms exhibited more blunted cortisol responses to stress than those with less severe symptoms (23). Authors using a mental arithmetic task revealed that clinically depressed individuals also showed blunted cortisol responses (24). Burke et al. (21) summarize the literature on HPA reactivity after psychological stressors in depressed patients and outline that depressed individuals exhibit a relatively flat and unresponsive pattern of cortisol release compared with their nondepressed counterparts. The authors also showed that blunted stress reactivity in depressed patients was most pronounced in older and more severely depressed patients. In contrast, cortisol activity in the nondepressed samples was characterized by a more dynamic pattern, including greater stress reactivity and more rapid recovery after stress.

Based on these findings, it can be summed up that there is great evidence for elevated basal cortisol levels, a strong suppression of cortisol after pharmacological challenge, and blunted cortisol reactivity in response to psychological stressors in nonpregnant and nonpuerperal depressed patients.

However, as reported previously, only a small number of studies have actually found evidence for elevated basal cortisol levels during pregnancy and in the puerperium in terms of postpartum depressive symptoms. To our knowledge, there have been no investigations into the association of the psychobiological response to psychosocial stress during pregnancy and postpartum depressive symptoms in clinical and subclinical populations, respectively.

The aim of this study was therefore to examine the predictive value of psychobiological stress reactivity during pregnancy for the development of depressive symptoms in the puerperium. To date, we are the first to investigate the relation between stress reactivity during pregnancy and depressive symptoms in the postpartum. In this context, we decided to assess the HPA axis reactivity of healthy pregnant women in response to a psychosocial stressor (mock job interview and mental arithmetic task performed in front of an audience) and then relate cortisol reactivity during pregnancy to postpartum depressive symptoms assessed after delivery. Moreover, we investigated the psychological stress response to this standardized stressor.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Subjects
The study was conducted between January 2004 and December 2004 in a sample of 57 healthy nulliparous women with a singleton intrauterine pregnancy (range of 13th to 31st weeks of pregnancy). Subjects were recruited through postal announcements, flyers at the University of Zurich, the University Hospital of Zurich, and in obstetricians’ offices in Zurich and the surrounding area. Pregnant women’s physical health was approved by their obstetricians, and psychiatric disorders were screened 2 weeks before the experiment using a structured clinical interview according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition and SKID-II (25). Exclusion criteria for participation were presence of psychiatric disorders such as depression, anxiety disorders, posttraumatic stress disorder, substance abuse or dependency, eating disorders, or psychotic symptoms, alcohol consumption of more than one glass of wine or beer per week, smoking, insufficient knowledge of the German language, low school education, and any medication. Pregnancy-specific exclusion criteria were artificial insemination and known medical maternal or fetal complications, including hypertension, diabetes mellitus, hyperemesis gravidarum, suspected fetal growth restriction, and fetal structural anomaly on ultrasound. The study protocol was approved by the ethics committee of the University of Zurich and the ethics committee of the Canton of Zurich. All pregnant women were informed about the course and aim of the study and gave written informed consent before taking part in the experiment.

Procedures
Psychosocial Stress Test (Time 1)
All pregnant women participated in the Trier Social Stress Test (TSST) (26), which has been repeatedly found to induce profound endocrine and cardiovascular responses in 70% to 80% of the subjects tested (27). The TSST basically consists of an unprepared speech (5 minutes) and mental arithmetic task (5 minutes) performed in front of an audience. After a basal salivary-free cortisol sample was taken (first sample), women were introduced to the TSST (simulated job interview). After a 10-minute preparation phase in a separate room, the women were then taken into the TSST room, which contained a panel of two observers and an ostentatious video camera. Saliva samples were taken 10 minutes (second sample) and immediately before and after the TSST (third and fourth sample) with an additional five samples taken at 10, 20, 30, 45, and 60 minutes after stress exposure. The TSST took place between 2:00 PM and 6:00 PM. Subjects were reimbursed for participation in the study with 150 Swiss francs.

Follow-Up Assessment (Time 2)
At the end of the experiment, the pregnant women were given a flyer including contact information with a request to give a notice of birth by telephoning or mailing. At time 2, when the women gave birth, they were then visited within 13 days postpartum (range, 2–27 days) at the hospital or by home visit. The visit included a semistructured interview and the assessment of the psychological and objective variables (see subsequently). The objective variables were taken from medical protocols given by the obstetricians. Finally, after the interview, the new mothers were given body lotion as a gift.

Sampling Methods and Biochemical Analyses
Salivary cortisol was collected using Salivette collection devices (Sarstedt, Sevelen, Switzerland). After chewing on cotton rolls for approximately 60 seconds, collection devices were kept at room temperature until the end of the session. Saliva samples were then stored at –20°C until biochemical analyses took place. After thawing, saliva samples were centrifuged and spun at 3000 rpm for 5 minutes, resulting in 1.0 mL clear supernatant of low viscosity. To reduce error variance caused by inaccuracy of the intraassay, all samples of one subject were analyzed in the same run of each assay. Salivary-free cortisol concentrations were measured using a commercially obtainable chemiluminescence immunoassay with high sensitivity (IBL, Hamburg, Germany).

Psychometric Measures
To assess postpartum depression symptoms, a German-validated version of the Edinburgh Postnatal Depression Scale (EPDS) was applied (28). This internationally used and well-validated 10-item self-report instrument does not diagnose depression but is used as a screening tool, analyzing depressive symptoms after giving birth (29). Using a cutoff point of a score of 10 or more, the EPDS has a sensitivity of 84% to 100% and a specificity of 82% to 88% when compared with a diagnosis of major postpartum depression assessed through a psychiatric interview (30). This cutoff point has been recommended to have sufficient sensitivity for community screening purposes (31). In addition, the following German-language questionnaires were used to investigate psychological stress factors: The Multidimensional Mood Questionnaire (MDBF) was used to repeatedly assess current mood on three dimensions during the TSST (32). State and trait anxiety was assessed using the State-Trait Anxiety Inventory (STAI) (33). State anxiety was repeatedly assessed during the TSST, whereas trait anxiety was analyzed 2 weeks before the TSST. A 36-item questionnaire was applied to measure stress susceptibility (MESA). The MESA comprises six subscales: sensitivity to failure, tolerance of work overload, tolerance of social conflict, sensitivity to criticism, tolerance of uncertainty, and ability to relax (34). Psychopathological symptoms were assessed using the Symptom-Checklist-90-Revised (SCL90-R) comprising the following 10 subscales: somatization, obsessive–compulsive symptoms, interpersonal sensitivity, depression, anxiety, anger–hostility, phobic anxiety, paranoid ideation, psychoticism, and "additional items" referring to sleep behavior, appetite, and thoughts about dying and death (35). The MESA and the SCL-90R were assessed 40 to 60 minutes after the stress test. The stressfulness of the TSST was repeatedly measured by a visual analog scale (VAS) during the TSST.

Birth Outcomes
For the assessment of objective data postpartum, medical data were obtained from birth protocol. Three birth outcomes were measured: gestational age at the time of delivery, birth weight (as a continuous variable in grams), and mode of delivery. Newborns were weighed in the delivery room immediately after birth. Furthermore, pregnancy complications occurring after the stress exposure and during pregnancy until giving birth were assessed. We also measured the following birth complications: secondary caesarean section, loss of blood >500 mL, second- and third-degree perineal rupture, and preterm delivery (delivery <37 + 0 weeks).

Statistical Analyses
To assess possible time and group effects, analyses of variance (ANOVAs) for repeated measures were computed. Possible effects of weeks of pregnancy on cortisol stress reactivity were controlled by conducting analyses of covariance (ANCOVA). All reported ANOVA and ANCOVA results were corrected by the Greenhouse-Geisser procedure when indicated (violation of sphericity assumption). Cortisol areas under the curve were calculated with respect to increase (AUCi) and to ground (AUCg) using the trapezoid formula (36). All data were tested for homogeneity of variance and for normal distribution using a Kolmogorov-Smirnov and Levene’s test before statistical procedures were applied. For all analyses, the level of significance was {alpha} = 5%. Unless otherwise indicated, all results shown are means ± standard error of means (SEM).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Sample Characteristics
Fifty-seven healthy nulliparous women with a singleton intrauterine pregnancy participated in the TSST during pregnancy. Allocation based on the EPDS score postpartum resulted in two groups with 16 (28.07%) participants showing a score of 10 or more (probable cases) and 41 (71.93%) participants having a score less than or equal to nine (probable noncases). The EPDS was applied within an average of 13.4 (standard deviation [SD] = 5.17) days postpartum. With regard to the highest level of "last completed education," 6 (37.5%) of the probable cases group and 16 (39.02%) of the probable noncases group had visited high school; 5 (31.25%) of the probable cases group and 14 (34.15%) of the probable noncases group had visited schools providing vocational education; and 5 (31.25%) of the probable cases and 11 (26.83%) of the probable noncases group had visited college. Regarding "civil status," 5 (31.25%) of the probable cases group and 19 (46.34%) of the probable noncases group were married, 9 (56.25%) women of the probable cases group and 20 (48.78%) of the probable noncases group were not married but were living together with their partners, and 2 (12.5%) of the probable cases and 2 (4.88%) of the probable noncases group indicated that they were not married or living in a partnership. The two groups did not differ with respect to mean age (probable cases: 30.50, SD = 3.18 versus probable noncases: 29.20, SD = 3.56) or week of gestation at the time of the experiment (probable cases: 21.06, SD = 6.84 versus probable noncases: 22.45, SD = 6.53), complications during pregnancy (probable cases: 1.75, SD = 0.45 versus probable noncases: 1.75, SD = 0.44), and complications during birth (probable cases: 1.50, SD = 0.52 versus probable noncases: 1.38, SD = 0.49). Furthermore, one-way ANOVAs revealed no group differences in terms of gestational length (probable cases: 39.49, SD = 1.52 versus probable noncases: 39.52, SD = 1.62), birth weight (probable cases: 3385.63, SD = 543.23 versus probable noncases: 3310.13, SD = 399.16), and mode of delivery (four women in the group of probable noncases and five women in the group of probable cases delivered by caesarean section). With regard to depressive history, no differences were found. In the group of probable cases with postpartum depressive symptoms, there were three women who had experienced depressive episodes in the past, whereas in the group of probable noncases, five women were found to have a history of depressive symptoms.

Salivary Cortisol Responses
The TSST resulted in a significant increase in salivary cortisol (in nanomoles per liter) (main effect of time: F (2.41,48.09) = 5.58; p < .001) from repeated-measures ANOVAs. There was no significant time x group interaction found over all of the nine cortisol measures (F (1.51,88.36) = 1.73; p = .19), whereas ANOVAs with repeated measures did reveal a significant time x group interaction over seven cortisol measures that included one measure immediately before and the remaining six after the stressor (time x group effect: F (2.41,25.74) = 2.99; p = .04) (Fig. 1). One-way ANOVA did not indicate significant salivary cortisol baseline differences (in nanomoles per liter) between groups (F (1.00,20.01) = 1.74; p = .19) or for the very first cortisol measure after stress exposure (F (1.00,0.90) = 0.33; p = .86). As a result of the known difference in cortisol stress reactivity between women in the second and third trimesters (9), weeks of pregnancy were controlled by ANCOVA with repeated measures and did not differ between the two groups. Similarly, because a history of depressive illness is known to be a risk factor for postpartum depressive symptoms (37), known depressive episodes in the women’s past as well as highest level of education and civil status were controlled by ANCOVA and did not differ between the two groups. Furthermore, results obtained by one-way ANOVA with the calculated area under the response curve of salivary cortisol over the aforementioned seven measures with respect to increase (AUCi) did not show significant differences between groups (F (1.00,2379.67) = 1.82; p = .18), or for the AUCg (F (1.00,305.21) = 0.18; p = .67).


Figure 117
View larger version (10K):
[in this window]
[in a new window]

  		Figure 1. Salivary cortisol reactivity (mean and standard error of mean) in probable cases and probable noncases. Stress exposure (Trier Social Stress Test) took place between minutes –10 to 0. *p < .05.

 

Psychological Responses
The TSST resulted in a significant decrease in mood (MDBF) (F (3.30,60.35) = 17.88; p < .001). Two-way ANOVA with repeated measures indicated a significant interaction of time and group (F (3.30,10.93) = 3.24; p = .02) and a significant group effect (F (1.00,163.31) = 9.83; p < .01) with the lowest values in the probable case group and the highest values in the probable noncase group (Fig. 2).


Figure 217
View larger version (7K):
[in this window]
[in a new window]

  		Figure 2. Mood (Multidimensional Mood Questionnaire; mean and standard error of mean) in probable cases and probable noncases. Stress exposure (Trier Social Stress Test) took place between minutes –10 to 0. *p < .05.

 

Furthermore, statistical analyses revealed that the probable case group showed significantly higher levels of state anxiety (STAI) than the probable noncase group throughout the entire period of the experiment (group effect: F (1.00,3.06) = 6.32; p = .02) and a significant interaction of time and group (F (2.93,0.18) = 3.33; p = .02) (Fig. 3). Changes over time revealed by two-way ANOVA with repeated measures showed a significant increase in state anxiety in response to stress (F (2.93,1.57) = 29.10; p < .001).


Figure 317
View larger version (9K):
[in this window]
[in a new window]

  		Figure 3. State anxiety (State-Trait Anxiety Inventory; mean and standard error of mean) in probable cases and probable noncases. Stress exposure (Trier Social Stress Test) took place between minutes –10 to 0. *p < .05.

 

Results obtained from one-way ANOVA indicated that the two groups differed significantly in trait anxiety (STAI: F (1.00,0.38) = 4.14; p = .05) (Fig. 4A) as well as in stress susceptibility (MESA: F (1.00,574.17) = 9.37; p < .01) (Fig. 4B).


Figure 417
View larger version (9K):
[in this window]
[in a new window]

  		Figure 4. Group differences between probable cases and probable noncases (mean and standard error of mean) in (A) trait anxiety (State-Trait Anxiety Inventory), (B) stress susceptibility (Stress Susceptibility Scale), and (C) additional scale referring to sleep behavior, appetite, and thoughts about dying and death (Symptom-Checklist-90-Revised). *p ≤ .05; **p < .01.

 

In addition, one-way ANOVA revealed significant differences in the "additional scale" referring to sleep behavior, appetite, and thoughts about dying and death (SCL90-R (F (1.00,2.36) = 7.48; p < .01), shown in Figure 4C. No significant differences between groups were found for the VAS.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
This is the first study to prospectively compare healthy women with high and low risk for postpartum depressive symptoms in terms of their psychological and physiological reactivity to a standardized psychosocial stressor during pregnancy. High-and low-risk groups were formed on the basis of a postpartum EPDS cutoff score of nine. The EPDS was applied within an average of 13 days postpartum. Women in the probable case group with an EPDS score greater than nine after delivery showed an increased hormonal response to psychosocial stress during pregnancy. According to this, women in the probable case group responded with a heightened increase in state anxiety and with a stronger decrease in mood to the psychosocial stressor. These data provide evidence that healthy pregnant women developing depressive symptoms postpartum might be identified during the time of pregnancy by means of their physiological and psychological coping with psychosocial stress.

As mentioned previously, the probable case group showed an increased hormonal response to the stress test. With regard to the baseline cortisol levels, measured at the beginning of the experimental procedure during pregnancy, no significant differences were found between the two groups. This is in contrast to the findings of Handley et al. (10), who revealed significantly elevated baseline levels of cortisol during the time of pregnancy in women with more severe blues on the fifth day postpartum compared with women without the experience of postpartum mood changes. These contradictory findings might be the result of the different point in time of cortisol measurement during pregnancy. Handley et al. (10) took baseline cortisol measurements at 38 weeks gestation, whereas in the present study, baseline cortisol measurements were taken in the second trimester (probable cases: 21.06, SD = 6.84 and probable noncases: 22.45, SD = 6.53, respectively). Hence, it might be assumed that possible gestational cortisol baseline differences between women with and without the development of depressive symptoms in the early puerperium become detectable at the end of pregnancy rather than in the second trimester. However, further investigations regarding unprovoked baseline cortisol levels measured during pregnancy, associated with postpartum mood states, should bear in mind that the point in time of baseline cortisol measurement during the period of pregnancy might play a decisive role.

The different endocrine stress response found in the probable case group is represented by an increased cortisol reactivity, whereas gestational age was controlled. The cortisol response to the TSST during pregnancy was significantly higher in the probable case group compared with the probable noncase group without depressive symptoms during early puerperium.

Considering this finding of higher cortisol reactivity in the probable case group in the light of reports of HPA reactivity in nonpregnant and nonpuerperal depressive patients, how can our results be explained? As previously reported, nonpregnant and nonpuerperal depressive patients show a significantly blunted pattern of cortisol release in response to psychosocial stress (21), whereas healthy individuals show a more dynamic pattern. The fact that the pregnant women in the present study were healthy at the time of the experiment may corroborate the assumption that this differing hormonal stress response might be seen as a biological prodromal symptom preceding postpartum depressive mood changes. This premorbid pattern observed in the probable case group of the present study might be interpreted as an allostatic load (38). The higher HPA axis responsiveness may represent an inadequate response of the HPA axis as an allostatic system to potentially stressful challenges such as the TSST. Thus, if the long-term pattern of increased HPA reactivity persists, one might imagine that this may lead to damage in the long run, resulting in a blunted HPA reactivity as observed in patients with melancholic depression (23). This hypothesis is appropriate, because animal studies showed that there is an initial increase of corticosterone secretion to stress, which is followed by a reduced responsiveness over time (39).

Similar findings of biological premorbid patterns associated with subsequent nonpregnant and nonpuerperal major depressive disorders are only found for cortisol activity, but not for cortisol reactivity. One such longitudinal study conducted by Harris et al. (40) examined a total sample of 116 women who were not currently depressed but was comprised of 83 women who were known to be vulnerable to onset of major depression as a result of psychosocial reasons and 33 women who were not. At entry, unprovoked salivary cortisol was measured on 4 consecutive days at 8:00 AM and again at 8:00 PM, whereas onset of major depression was measured during the 13-month follow-up period. Unprovoked morning cortisol levels at entry turned out to be significantly associated with a subsequent onset of major depression, providing evidence that premorbid elevated morning cortisol levels might represent an independent risk factor for the development of nonpregnant and nonpuerperal depression. Another finding was revealed in a sample of 180 adolescents at high risk for psychopathology, in which the occurrence of peaks in morning cortisol emerged as predictive for subsequent nonpuerperal major depression at a 12-month follow up (41). Hence, there is evidence for a specific premorbid pattern of psychoendocrine factors predicting the onset of nonpregnant and nonpuerperal major depressive disorder. According to these findings, there might also be a similar biological premorbid pattern for puerperal depressive symptoms.

However, future research into unprovoked cortisol baseline levels as well as HPA responsiveness to psychosocial stress during pregnancy is needed to illuminate gestational HPA dysregulations in terms of postpartum depressive symptoms in pregnant samples.

With regard to psychological changes to stress, we found a significant change in mood and state anxiety resulting from the TSST. Analyses of variance with repeated measures revealed that women with postpartum depressive symptoms showed increased state anxiety (p < .05) and had a significantly lower mood state throughout the experiment (p < .05). Our findings might be the result of different coping strategies of women who showed postpartum depressive symptoms. However, Murata and colleagues (42) found elevated trait anxiety during pregnancy associated with postpartum blues, which is extended by our results in showing that the probable case group showed higher trait anxiety compared with the probable noncase group (p < .05).

Furthermore, we found differences in the susceptibility to stress between high- and low-risk pregnant women for postpartum depressive symptoms in terms of higher stress susceptibility found in the probable case group. This finding might indicate a general vulnerability to stress for women at high risk for postpartum mood changes.

No differences were found for obstetrical factors, because obstetrical complications and mode of delivery did not differ between the two groups. This result is in line with findings from several studies regarding postpartum depressive symptoms associated with obstetric complications (43–46) and mode of delivery (44), respectively. In contrast, there are some studies that provide an association between birth outcomes and postpartum mood disorders (47,48). These conflicting findings may be the result of methodological problems, because, for instance, definitions of complications may differ between physicians or hospitals (37).

In summary, our data provide evidence that women developing depressive symptoms in the puerperium already show increased reactivity in physiological and psychological measurements in response to a standardized psychosocial stressor during their pregnancy. Our results may lead to the conclusion that the increased psychobiological responses to psychosocial stress shown in healthy pregnant women refers to a potential risk for the development of postpartum depressive symptoms. Further investigations are required to explore whether healthy pregnant women with an increased cortisol reactivity, high levels of general anxiety, and high stress susceptibility are at increased risk not only for the development of depressive symptoms within 2 weeks after birth, but potentially also for the development of postpartum major depression. In conclusion, women at high risk for depressive symptoms after delivery might already be identified during pregnancy through their higher stress reactivity. The health and well-being of the mother and child call for early diagnosis and can be facilitated through an awareness of the psychobiological mechanisms and the risk factors for postpartum depressive symptoms and disorders.

We thank Ariadne Klinkenberg for help in the medical screening of the study participants. We also acknowledge the help of Daniela Klaus, Nicole Fluri, Gabriela Schöbi, Karin Meier-Schick, Christine Wolfinger, and Lea Hauri in conducting the experiments.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 

This study was supported by the Swiss National Science Foundation grant no. 105311 to 101793/1 (U.E., R.Z.).

DOI:10.1097/01.psy.0000244385.93141.3b


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Ehlert U, Patalla U, Kirschbaum C, Piedmont E, Hellhammer DH. Postpartum blues: salivary cortisol and psychological factors. J Psychosom Res 1990;34:319–25.[CrossRef][Medline]
  2. Beardslee WR, MacMillan HL. Preventive intervention with the children of depressed parents. A case study. Psychoanal Study Child 1993;48:249–76.[Medline]
  3. Jacobsen T. The effects of postpartum disorders on offspring and mother-infant interaction. In: Miller LJ, ed. Postpartum Mood Disorders. Washington, DC: American Psychiatric Press; 1999:119–39.
  4. Weinberg MK, Tronick EZ, Beeghly M, Olson KL, Kernan H, Riley JM. Subsyndromal depressive symptoms and major depression in postpartum women. Am J Orthopsychiatry 2001;71:87–97.[CrossRef][Medline]
  5. O’Hara MW. The nature of postpartum depressive disorder. In: Cooper LMPJ, ed. Postpartum Depression and Child Development. New York: Guilford; 1997:3–31.
  6. Harris B. A hormonal component to postnatal depression. Br J Psychiatry 1993;163:403–5.[Free Full Text]
  7. Allolio B, Hoffmann J, Linton EA, Winkelmann W, Kusche M, Schulte HM. Diurnal salivary cortisol patterns during pregnancy and after delivery: relationship to plasma corticotrophin-releasing-hormone. Clin Endocrinol (Oxf) 1990;33:279–89.[Medline]
  8. Kammerer M, Adams D, Castelberg B, Glover V. Pregnant women become insensitive to cold stress. BMC Pregnancy Childbirth 2002;2:8.[CrossRef][Medline]
  9. Nierop A, Bratsikas A, Klinkenberg A, Nater UM, Zimmermann R, Ehlert U. Prolonged salivary cortisol recovery in second trimester pregnant women and attenuated salivary alpha-amylase responses to psychosocial stress in human pregnancy. J Clin Endocrinol Metab 2006;91:1329–35.[Abstract/Free Full Text]
  10. Handley SL, Dunn TL, Waldron G, Baker JM. Tryptophan, cortisol and puerperal mood. Br J Psychiatry 1980;136:498–508.[Abstract/Free Full Text]
  11. Feksi A, Harris B, Walker RF, Riad-Fahmy D, Newcombe RG. ‘Maternity blues’ and hormone levels in saliva. J Affect Disord 1984;6:351–5.[CrossRef][Medline]
  12. Kuevi V, Causon R, Dixson AF, Everard DM, Hall JM, Hole D, Whitehead SA, Wilson CA, Wise JC. Plasma amine and hormone changes in ‘post-partum blues.’ Clin Endocrinol (Oxf) 1983;19:39–46.[Medline]
  13. Pedersen CA, Stern RA, Pate J, Senger MA, Bowes WA, Mason GA. Thyroid and adrenal measures during late pregnancy and the puerperium in women who have been major depressed or who become dysphoric postpartum. J Affect Disord 1993;29:201–11.[CrossRef][Medline]
  14. Carroll BJ, Curtis GC. Neuroendocrine identification of depressed patients. Aust N Z J Psychiatry 1976;10:13–20.[Medline]
  15. Rubin RT, Poland RE, Lesser IM, Winston RA, Blodgett AL. Neuroendocrine aspects of primary endogenous depression. I. Cortisol secretory dynamics in patients and matched controls. Arch Gen Psychiatry 1987;44:328–36.[Abstract/Free Full Text]
  16. Halbreich U, Asnis GM, Shindledecker R, Zumoff B, Nathan RS. Cortisol secretion in endogenous depression. I. Basal plasma levels. Arch Gen Psychiatry 1985;42:904–8.[Abstract/Free Full Text]
  17. Young EA, Haskett RF, Murphy-Weinberg V, Watson SJ, Akil H. Loss of glucocorticoid fast feedback in depression. Arch Gen Psychiatry 1991;48:693–9.[Abstract/Free Full Text]
  18. Young EA, Kotun J, Haskett RF, Grunhaus L, Greden JF, Watson SJ, Akil H. Dissociation between pituitary and adrenal suppression to dexamethasone in depression. Arch Gen Psychiatry 1993;50:395–403.[Abstract/Free Full Text]
  19. Nemeroff CB, Widerlov E, Bissette G, Walleus H, Karlsson I, Eklund K, Kilts CD, Loosen PT, Vale W. Elevated concentrations of CSF corticotropin-releasing factor-like immunoreactivity in depressed patients. Science 1984;226:1342–4.[Abstract/Free Full Text]
  20. Checkley S. The neuroendocrinology of depression and chronic stress. Br Med Bull 1996;52:597–617.[Abstract/Free Full Text]
  21. Burke HM, Davis MC, Otte C, Mohr DC. Depression and cortisol responses to psychological stress: a meta-analysis. Psychoneuroendocrinology 2005;30:846–56.[CrossRef][Medline]
  22. Peeters F, Nicholson NA, Berkhof J. Cortisol responses to daily events in major depressive disorder. Psychosom Med 2003;65:836–41.[Abstract/Free Full Text]
  23. Burke HM, Fernald LC, Gertler PJ, Adler NE. Depressive symptoms are associated with blunted cortisol stress responses in very low-income women. Psychosom Med 2005;67:211–6.[Abstract/Free Full Text]
  24. Trestman RL, Coccaro EF, Bernstein D, Lawrence T, Gabriel SM, Horvath TB, Siever LJ. Cortisol responses to mental arithmetic in acute and remitted depression. Biol Psychiatry 1991;29:1051–4.[CrossRef][Medline]
  25. Wittchen H-U, Gruschwitz S, Zaudig M. Strukturiertes Klinisches Interview für DSM-IV (SKID). Göttingen: Beltz-Test; 1996.
  26. Kirschbaum C, Pirke KM, Hellhammer DH. The ‘Trier Social Stress Test’—a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 1993;28:76–81.[Medline]
  27. Dickerson SS, Kemeny ME. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychol Bull 2004;130:355–91.[CrossRef][Medline]
  28. Bergant AM, Nguyen T, Heim K, Ulmer H, Dapunt O. German language version and validation of the Edinburgh postnatal depression scale. Dtsch Med Wochenschr 1998;123:35–40.[Medline]
  29. Beck CT. Predictors of postpartum depression: an update. Nurs Res 2001;50:275–85.[CrossRef][Medline]
  30. Murray L, Carothers AD. The validation of the Edinburgh Post-natal Depression Scale on a community sample. Br J Psychiatry 1990;157:288–90.[Abstract/Free Full Text]
  31. Zelkowitz P, Milet TH. Screening for post-partum depression in a community sample. Can J Psychiatry 1995;40:80–6.[Medline]
  32. Steyer R, Schwenkmezger P, Notz P, Eid M. Der Mehrdimensionale Befindlichkeitsfragebogen (MDBF): Handanweisung. Göttingen: Hogrefe; 1997.
  33. Laux L, Glanzmann P, Schaffner P, Spielberger CD. Das State-Trait-Angstinventar. Theoretische Grundlagen und Handanweisungen. Weinheim: Beltz; 1981.
  34. Schulz P, Jansen LJ, Schlotz W. Stressreaktivität: Theoretisches Konzept und Messung. Diagnostica.
  35. Franke G. SCL90-R. Die Symptom Checkliste von Derogatis–Deutsche Version. Göttingen: Beltz Test; 1995.
  36. Pruessner JC, Kirschbaum C, Meinlschmid G, Hellhammer DH. Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology 2003;28:916–31.[CrossRef][Medline]
  37. Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry 2004;26:289–95.[CrossRef][Medline]
  38. McEwen BS. Plasticity of the hippocampus: adaptation to chronic stress and allostatic load. Ann N Y Acad Sci 2001;933:265–77.[Medline]
  39. Heim C, Ehlert U, Hellhammer DH. The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology 2000;25:1–35.[CrossRef][Medline]
  40. Harris TO, Borsanyi S, Messari S, Stanford K, Cleary SE, Shiers HM, Brown GW, Herbert J. Morning cortisol as a risk factor for subsequent major depressive disorder in adult women. Br J Psychiatry 2000;177:505–10.[Abstract/Free Full Text]
  41. Goodyer IM, Herbert J, Tamplin A, Altham PM. Recent life events, cortisol, dehydroepiandrosterone and the onset of major depression in high-risk adolescents. Br J Psychiatry 2000;177:499–504.[Abstract/Free Full Text]
  42. Murata A, Nadaoka T, Morioka Y, Oiji A, Saito H. Prevalence and background factors of maternity blues. Gynecol Obstet Invest 1998;46:99–104.[CrossRef][Medline]
  43. Josefsson A, Angelsioo L, Berg G, Ekstrom CM, Gunnervik C, Nordin C, Sydsjo G. Obstetric, somatic, and demographic risk factors for postpartum depressive symptoms. Obstet Gynecol 2002;99:223–8.[CrossRef][Medline]
  44. Johnstone SJ, Boyce PM, Hickey AR, Morris-Yatees AD, Harris MG. Obstetric risk factors for postnatal depression in urban and rural community samples. Aust N Z J Psychiatry 2001;35:69–74.[CrossRef][Medline]
  45. Saisto T, Salmela-Aro K, Nurmi JE, Halmesmaki E. Psychosocial predictors of disappointment with delivery and puerperal depression. A longitudinal study. Acta Obstet Gynecol Scand 2001;80:39–45.[CrossRef][Medline]
  46. Patel RR, Murphy DJ, Peters TJ. Operative delivery and postnatal depression: a cohort study. BMJ 2005;330:879.[Abstract/Free Full Text]
  47. Boyce PM, Todd AL. Increased risk of postnatal depression after emergency caesarean section. Med J Aust 1992;157:172–4.[Medline]
  48. Koo V, Lynch J, Cooper S. Risk of postnatal depression after emergency delivery. J Obstet Gynaecol Res 2003;29:246–50.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Arch Gen PsychiatryHome page
I. S. Yim, L. M. Glynn, C. Dunkel Schetter, C. J. Hobel, A. Chicz-DeMet, and C. A. Sandman
Risk of Postpartum Depressive Symptoms With Elevated Corticotropin-Releasing Hormone in Human Pregnancy
Arch Gen Psychiatry, February 1, 2009; 66(2): 162 - 169.
[Abstract] [Full Text] [PDF]

Home page
 J Gerontol B Psychol Sci Soc SciHome page
M. D. Gayman, R. J. Turner, and M. Cui
Physical Limitations and Depressive Symptoms: Exploring the Nature of the Association
J Gerontol B Psychol Sci Soc Sci, July 1, 2008; 63(4): S219 - S228.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nierop, A.
Right arrow Articles by Ehlert, U.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nierop, A.
Right arrow Articles by Ehlert, U.
Related Collections
Right arrow Neuroendocrine
Right arrow Depression
Right arrow Stress and Coping
Right arrow Pregnancy

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS