| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-Amylase Responses to Psychosocial Stress in Human Pregnancy
Department of Clinical Psychology and Psychotherapy (A.N., A.B., U.M.N., U.E.), University of Zurich, CH-8044 Zurich, Switzerland; and Department of Obstetrics (A.K., R.Z.), University Hospital of Zurich, CH-8006 Zurich, Switzerland
Address all correspondence and requests for reprints to: Ulrike Ehlert, Ph.D., 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 IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The aim of this study was to identify endocrine, autonomic, and psychological responses to standardized psychosocial stress at different stages of pregnancy.
Design: Ninety healthy women (aged 21–37 yr), including 30 pregnant women at the beginning of the second trimester and 30 women at the beginning of the third trimester, as well as 30 nonpregnant women in the follicular phase of their menstrual cycle, underwent a psychosocial stress test. Salivary free cortisol, 
-amylase, heart rate, and psychological parameters were repeatedly measured.
Results: Salivary cortisol recovery was significantly prolonged in second-trimester pregnant women (P = 0.04). Cortisol and heart rate increases of both pregnant groups were shown to be comparable with those of the controls. -Amylase increases of both pregnant groups were markedly attenuated compared with nonpregnant women (P = 0.008).
Conclusions: From these data, we conclude that, in contrast to pregnancy in rats, pregnancy in women does not result in a restraint of the hypothalamic-pituitary-adrenal axis to psychosocial stress. Furthermore, attenuated -amylase stress response might reflect protective processes within the autonomic nervous system during pregnancy, whereas prolonged cortisol recovery during the beginning of second-trimester pregnancy might be associated with the vulnerability to stress-related pregnancy complications during this period of time.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
From research in animals, it is known that stressors provoke altered hormonal release of hypothalamic-pituitary-adrenal (HPA) and sympathoadrenal medullary axes in pregnant compared with nonpregnant animals under the same conditions. There is evidence that, at the end of pregnancy, plasma catecholamines show only moderate increases after restraint stress or forced immobilization (5) and the HPA axis becomes hyporeactive after stressful experiences with attenuated releases of ACTH (6, 7). These results indicate that reduced reactivity of biological systems to stress may protect both the pregnant animal and her offspring from harmful excessive levels of stress hormones.
Recent human research on pharmacological or physical provocation procedures in pregnant women has resulted in inhomogeneous findings regarding the extent of alterations of the HPA axis and the autonomic nervous system (ANS) with respect to the progression of pregnancy and the type of stressor. The variety of experimental designs, the lack of adequate control groups, and different forms and low reliability of hormonal assessment make it difficult to draw conclusions (3).
Because research in this field is still in its infancy and there appear to be no studies examining the effects of standardized psychosocial stress at different stages of pregnancy, the immediate objective of the present study was to identify, first of all, psychological, endocrine, and autonomic responses to standardized psychosocial stress, induced by the Trier Social Stress Test (TSST) (8). Numerous studies indicate that the TSST enables a naturalistic exposure to a socioevaluative stressful situation and induces a significant activation of physiological and psychological response to stress in healthy females and patients (9, 10, 11).
Inducing stress in pregnant women requires safe and noninvasive methods. Unlike venipuncture procedures for collecting plasma, collection of saliva samples is particularly suitable because it is painless, noninvasive, virtually stress free, can be used for repeated sampling, and is known for its reliable and sensitive assays (12). It has been established that detectable levels of cortisol in saliva accurately reflect circulating plasma cortisol levels (13). In the search for noninvasive, valid and reliable stress markers, the enzyme -amylase seems to be suited for the investigation of the ANS in pregnancy (14). Increases of -amylase have been found when the organism is exposed to either physiological (15) or psychological (16) stressors. Several studies in animals and humans suggest that activation of the ANS with a combination of both sympathetic and parasympathetic innervations of the salivary glands leads to high activity of -amylase (17).
Elucidating stress response in pregnant women may foster the development of targeted hypotheses about dysregulations of stress-related biological systems and provide essential information about predictors for pregnancy complications, which, in consequence, could be used to develop innovative early prevention programs for high-risk populations of pregnant women.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 90 women, including 60 healthy nulliparous women with a singleton intrauterine pregnancy (aged 21–35 yr) and a group of 30 nonpregnant healthy women (aged 21–37 yr) participated in the study. Two groups of pregnant women were recruited: one group consisted of 30 women at the beginning of the second trimester (13–18 wk), and the second group comprised 30 women at the beginning of the third trimester (26–31 wk) of their pregnancy. Subjects were recruited through posted announcements, signs at the University of Zurich, the University Hospital of Zurich, and in obstetricians’ offices in Zurich and the surrounding area. Exclusion criteria for participation were medical or psychiatric illness, substance abuse, medication, alcohol consumption of more than one glass of wine or beer per week, low level of school education, insufficient knowledge of the German language, artificial insemination for the pregnant women, and oral contraceptives and irregular menstrual cycle for the control group. The experimental sessions of the nonpregnant women were related to the subjects’ follicular phase of the menstrual cycle (because of stable estrogen and progesterone levels in this phase) to minimize mood and cognitive impairments caused by premenstrual symptoms (18). All participants were informed about the course and aim of the study and provided written, informed consent before participation. The study protocol was approved by the ethics committees of the University of Zurich, as well as of the Canton of Zurich.
Study design
Interviews and experiments took place at the University of Zurich between March 2004 and February 2005. Psychosocial stress was induced by the TSST. All subjects were interviewed to assess anamnestic data and to screen for psychiatric disorders using a structured clinical interview, DIA-X/M-CIDI (19, 20) and SKID-II (21) In addition, the subjects underwent a medical checkup. All participants were asked to abstain from brushing their teeth for 1 h and avoid food intake for 2 h before the session. In addition, they were instructed to refrain from exercise, caffeine, and alcohol during the 24 h before the experiment. The stress sessions lasted for 2 h, and all took part in the afternoon (1400–1700 h). After an acclimation phase (10 min), each participant (tested individually) was given a 10-min preparation period for the speaking task. After this, she then participated in the stress test, consisting of a 5-min public speaking task (mock job interview) and a subsequent 5-min mental arithmetic task (serial subtraction) performed in front of a conspicuous video camera and a panel of two evaluators. After the stress test, a 60-min recovery phase began, followed by a debriefing.
Outcome measures
A total of nine saliva samples were taken. The first three samples were taken before the stress test (–20, –10, and –1 min), and six additional saliva samples were collected after stress (+1, +10, +20, +30, +45, and +60 min). Heart rate was measured using a portable heart-rate monitoring device (Polar S810; Polar Electro Oy, Kempele, Finland), with aggregated beat-to-beat analyses.
Socioeconomic data, personal living situation, and current critical life events were assessed according to a semi-standardized interview protocol. Mood and state anxiety were repeatedly assessed (a total of five measurements), as well as a visual analog scale (stressfulness of the situation; a total of six measurements) before and after stress exposure. Validated German versions of the following questionnaires were included: Mood Questionnaire [Mehrdimensionale Befindlichkeitsfragebogen (MDBF)] (22), the State-Trait Anxiety Inventory (STAI) (23), the Global Symptom Index derived from the Hopkins Symptom-Checklist-90-Revised (SCL90-R) (24), and the Trier Inventory for the Assessment of Chronic Stress (TICS) (25). All of these questionnaires are widely used and have shown high internal consistency and validity.
Assays
Saliva was collected by using salivette collection devices (Sarstedt, Sevelen, Switzerland). Samples were stored at –20 C after completion of the session until biochemical analysis took place. After thawing, saliva samples were centrifuged at 3000 rpm for 5 min. Salivary free cortisol was analyzed by using a commercial chemiluminescence immunoassay (IBL, Hamburg, Germany). Intraassay and interassay coefficients of variation were less than 10%. To reduce error variance caused by imprecision of the intraassay, all samples of one subject were analyzed in the same run. For the determination of -amylase activity, an assay was used as described previously (14). An automatic analyzer (Cobas Mira) and assay kits obtained from Roche Products (Rotkreuz, Switzerland) were used. The reagents in the kit contain the enzyme -amylase and -glucosidase, which convert the substrate ethyliden nitrophenyl to p-nitrophenol. The rate of formation of p-nitrophenol is directly proportional to the amylase activity. The activity is determined by measuring the absorbance at 405 nm. The assay is a kinetic colorimetric test. Interassay and intraassay variance was less than 1%.
Data analysis
Cortisol, -amylase, heart rate, and psychological data were analyzed using two-way ANOVA with repeated measurement (group by time). All reported results were corrected by the Greenhouse-Geisser procedure when appropriate (violation of sphericity assumption). The areas under the individual response curves with respect to the increase (AUCIncrease) of cortisol, heart rate, and -amylase were calculated with the trapezoid formula (26). The AUCIncrease is related to the sensitivity of the biological system, pronouncing changes over time, and is characterized by accumulation of the error of the baseline because the formula is based on the difference between the baseline and the subsequent measures (26). Recovery over time was calculated with AUCRecovery, reflecting the area under the curve from the highest level and the last recovery time point, corrected by the last measurement. One-way ANOVA was computed for comparison of means between groups of AUCRecovery, AUCIncrease, and baseline. Significance of post hoc comparisons between groups was calculated with the Bonferroni’s test. All analyses were two tailed, with the level of significance set at P < 0.05. Unless indicated, all results shown are means ± SEM.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There were no statistical differences in age and body mass index among the groups, with weight values of the state shortly before pregnancy included in the calculation. Subjects of the three comparison groups did not differ with respect to trait anxiety or chronic stress, with psychopathological symptoms in the normal range of the general population (Table 1
).
|
Results obtained by two-way ANOVA with repeated measures indicated that the stress paradigm induced significant decreases in mood (MDBF, F(3.07,264.25) = 20.00; P < 0.0001) and increases in state anxiety (STAI, F(2.59,222.45) = 40.54; P < 0.0001) and perceived stressfulness (VAS, F(2.82,245.02) = 54.76; P < 0.0001) over time. There were no significant time x group interactions found with the factor group and six measurements for VAS or for the five measurements of MDBF and STAI.
Cortisol stress response
Results obtained by one-way ANOVA indicate significant salivary free cortisol baseline (–20 min) differences (in nanomoles per liter) between groups (F(2,87) = 17.12; P < 0.0001). Post hoc analyses revealed significantly higher baseline levels in third-trimester (mean ± SEM, 11.6 ± 0.5) compared with second-trimester (mean ± SEM, 7.5 ± 0.5; P < 0.0001) pregnant women and nonpregnant women (mean ± SEM, 7.6 ± 0.7; P < 0.0001).
Results obtained by two-way ANOVA with repeated measures indicated that the stress protocol induced significant increases in cortisol levels (main effect of time, F(1.64,142.9) = 32.75; P < 0.0001) and revealed a significant time by group interaction (time x group effect, F(4.18,181.8) = 2.85; P = 0.02). Post hoc analyses revealed significant differences between third-trimester and second-trimester pregnant women (P = 0.001) and between the third-trimester and the nonpregnant group (P = 0.006) (Fig. 1A).
|
).
|
-Amylase stress response
Results obtained by one-way ANOVA indicate significant salivary free -amylase baseline differences (in units/milliliter) between the groups (F(2,87) = 4.13; P = 0.019). Post hoc analyses revealed that baseline levels (–20 min) were significantly higher in third-trimester (mean ± SEM, 36.8 ± 5.5) compared with second-trimester (mean ± SEM, 56.1 ± 7.7; P = 0.04) women and nonpregnant women (mean ± SEM, 70.7 ± 11.0; P = 0.01).
A two-way ANOVA with repeated measures revealed a significant -amylase response to stress (main effect of time, F(2.56,223.03) = 21.77; P < 0.0001) and a significant group by time effect (F(5.13,223.03) = 4.00; P = 0.002), with the lowest responses in third-trimester pregnant women and the highest responses in nonpregnant women (Fig. 1B
).
For the total amount of AUCIncrease, a significant main effect of group was obtained (F(2,87) = 5.09; P = 0.008), indicating an increase in the nonpregnant group and a decrease in both pregnant groups attributable to the fact that lower values than baseline values were obtained (Fig. 2B) (26). No significant differences of AUCRecovery between groups were obtained (Fig. 3B).
|
Results obtained by one-way ANOVA indicate significant heart-rate baseline differences (in beats per minute) between groups (F(2,87) = 9.34; P < 0.0001). Post hoc analyses revealed that basal heart-rate levels (–20 min) were significantly higher in third-trimester (mean ± SEM, 93.5 ± 1.9) than in second-trimester (mean ± SEM, 85.1 ± 1.9; P = 0.01) pregnant women, as well compared with nonpregnant women (mean ± SEM, 82.0 ± 2.0; P < 0.0001).
A two-way ANOVA with repeated measures revealed the expected significant heart-rate response to stress (main effect of time, F(4.20,361.24) = 130.04; P < 0.0001) across all groups (Fig. 1C). No significant differences of AUCRecovery and AUCIncrease levels between groups were obtained (Figs. 2C and 3C).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The TSST activated the HPA axis as well as the ANS and worsened perceived stress, mood, and anxiety in the three groups examined. The results of the present study demonstrate that, compared with nonpregnant women, -amylase stress response in both pregnant groups was markedly attenuated and cortisol recovery in second-trimester pregnant women was significantly prolonged as a result of psychosocial stress. However, no group differences in heart-rate responses were observed. Thus, habituation processes to psychosocial stress appear not only to be different between the groups but also in the HPA axis compared with the ANS, as well as within the ANS, which is in line with findings of other studies (14, 27). Inconsistencies in responses of the investigated stress parameters are most likely related to the diversity of the underlying mechanisms, i.e. the more predominant involvement of the sympathicus, parasympathicus, and/or of different brain structures. Our findings point to the importance of investigating the ANS as well as the HPA axis with preferably miscellaneous stress indicators at different stages of pregnancies.
Pregnancy-related prolonged cortisol recovery in humans after stress exposure has, so far, not been reported in the literature. Research has focused mainly on the overall magnitude of hormonal changes in response to stressors in late pregnancy in rodents (6, 28) and humans (29). Recently, however, an increasing number of researchers have underlined the importance of following a stress reaction for a longer period of time and determining the physiological recovery from the stressor (3, 10, 30). With respect to this issue, Sapolsky et al. (31) suggest that a prolonged return to the baseline after stress exposure results in longer overall exposure to stress hormones and, as a consequence, could heighten the risk for a negative health outcome. Findings from the few studies that examined how different contexts or individual difference factors predict cortisol recovery provide evidence that cortisol recovery is indeed associated with variations in metabolic risk (32, 33). Taking these considerations into account, we could hypothesize that, compared with healthy subjects, a more prolonged cortisol recovery after stress exposure might be a potential risk factor for pathological processes during pregnancy. Prolonged cortisol recovery found in the first half of pregnancy might be of relevance because there is evidence that, in relation to the impact of stress, this period of time seems to be the most vulnerable period, especially for poor outcome, i.e. preterm birth and low birth weight (2, 34, 35, 36). The underlying mechanisms of this phenomenon are unknown. However, in prenatally stressed animals, a prolonged elevation of corticosteroids was found to be caused by a dysregulation of the HPA axis characterized by decreased feedback inhibition of CRH and involvement of glucocorticoid receptors in the hippocampus, frontal cortex, and amygdala (2). Prolonged cortisol recovery was found in second-trimester women compared with nonpregnant women, but no significant differences were observed between third-trimester and nonpregnant women. The causes may include the influence of CRH, progesterone, and estradiol, which all appear to be more elevated in third-trimester than in second-trimester pregnancy (37). Evidence from animal research indicates that the aforementioned hormones indeed seem to be potent modulators of HPA axis stress regulation (2, 6). Oxytocin, which has been suggested to affect the human HPA axis responsiveness (38), might also play a role. This potential modulating effect of oxytocin on HPA axis responsiveness is likely to be a more central process and could, in fact, be localized in the paraventricular nucleus and the septum, as shown in rats (39).
In contrast to pregnancy in rats (6), pregnancy in women did not result in a restraint of the HPA axis to psychosocial stress in the present study. Notably, the same inconsistency of HPA axis responses was found in research with lactating rats compared with lactating women exposed to psychosocial stress (11). Furthermore, our finding does not accord with a recent study of Kammerer et al. (29), who found attenuated salivary cortisol stress responses in pregnant women compared with nonpregnant women after a cold pressor test. This inconsistency might be caused by the nature of the stressor (physical vs. mental stress). The amygdala is likely to play an important moderating role under psychosocial stress, because this brain structure is known for being activated particularly under psychological stress by corticosteroids, which increase CRH in amygdala at a site that is implicated in the production of fear and anxiety, as shown in animal studies (2).
Another finding in this study, which so far has not been discussed in the literature, was the significantly blunted -amylase levels of pregnant women caused by psychosocial stress compared with nonpregnant women. According to our recent work on biological markers of psychosocial stress, an association of peripherally measured norepinephrine, estrogen, and salivary -amylase seems to be questionable. As a result of psychological stress, peripheral catecholamines and -amylase seem to be part of different biological pathways. However, it seems clear that the ANS plays a predominant role in the secretion process of -amylase (14, 17). Proceeding on the assumption that changes in salivary -amylase after psychosocial stress reflect autonomic changes, the question arises as to what our findings indicate. Clearly, the benefits of a sharp -amylase increase caused by stress, as shown in nonpregnant women, are likely to be of minor significance compared with the biological profit and protection of blunted stress reactivity, which occurs in healthy pregnant women under psychosocial stress conditions. We could therefore hypothesize that deviations from this biological protection, such as high levels of -amylase, might be harmful for the pregnant woman and her unborn child. This hypothesis seems to be appropriate, because examples of potential danger of heightened levels of other stress-related autonomic parameters in association with pregnancy complications are well known (40). Future research should therefore focus on the role of this stress marker to search for possible associations of heightened levels of -amylase with pathological processes.
Blood plasma samples might have provided useful information about other important parameters known for their association with biological stress response (3, 14, 18). This might have yielded a better understanding of the biological processes leading to the stress responses of the HPA axis and the ANS during pregnancy. However, we limited our research particularly to saliva stress markers, because they are easily applicable, not stressful in themselves, noninvasive, and most importantly enabled us to keep to ethical guidelines.
From our data, we conclude that, in contrast to pregnancy in rats, pregnancy in women does not result in a restraint of the HPA axis to psychosocial stress. Furthermore, attenuated -amylase stress response to mental stress might be interpreted as a reflection of protective processes within the ANS during pregnancy. In addition, prolonged cortisol recovery during the beginning of second-trimester pregnancy might be associated with the vulnerability to stress-related pregnancy complications during this period of time. Although the reported results need to be replicated to allow firm conclusions to be drawn, our findings could provide essential basic information for future studies. Additional elucidation of the underlying psychobiological mechanisms involved in psychosocial stress responses during pregnancy may lead to new insights into improved health of pregnant women and their fetuses.
|
Footnotes |
|---|
There are no conflicts of interest.
First Published Online January 24, 2006
Abbreviations: ANS, Autonomic nervous system; AUCIncrease, area under the individual response curves with respect to the increase; AUCRecovery, area under the curve from the highest level and the last recovery time point; HPA, hypothalamic-pituitary-adrenal; MDBF, the mood questionnaire Mehrdimensionale Befindlichkeitsfragebogen; SCL90-R, Global Symptom Index derived from the Hopkins Symptom-Checklist-90-Revised; STAI, State-Trait Anxiety Inventory; TICS, Trier Inventory for the Assessment of Chronic Stress; TSST, Trier Social Stress Test.
Received August 11, 2005.
Accepted January 17, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-amylase. J Sports Sci 17:129–134[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  U. Ehlert, K. Erni, G. Hebisch, and U. Nater Salivary {alpha}-Amylase Levels after Yohimbine Challenge in Healthy Men J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 5130 - 5133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Nierop, A. Bratsikas, R. Zimmermann, and U. Ehlert Are Stress-Induced Cortisol Changes During Pregnancy Associated With Postpartum Depressive Symptoms? Psychosom Med, November 1, 2006; 68(6): 931 - 937. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, second-trimester vs. nonpregnant women; *, +, 
