G Model

WOMBI-523; No. of Pages 7

Women and Birth xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Women and Birth

journal homepage: www.elsevier.com/locate/wombi

Original Research – Quantitative

Immediate stress reduction effects of yoga during pregnancy:

One group pre–post test
Momoko Kusaka, Masayo Matsuzaki *, Mie Shiraishi, Megumi Haruna
Department of Midwifery and Women’s Health, Division of Health Sciences and Nursing, Graduate School of Medicine, The University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

A R T I C L E I N F O A B S T R A C T

Article history: Background: Excessive stress during pregnancy may cause mental disorders in pregnant women and
Received 15 October 2015 inhibit fetal growth. Yoga may alleviate stress during pregnancy.
Received in revised form 26 January 2016 Aim: To verify the immediate effects of yoga on stress response during pregnancy.
Accepted 4 April 2016
Methods: One group pre–post test was conducted at a hospital in Japan. We recruited 60 healthy
primiparas without complications and asked them to attend yoga classes twice a month and to practice
Keywords: yoga at their homes using DVD 3 times a week from 20 gestational weeks until childbirth. Salivary
Exercise
cortisol and alpha-amylase concentration were measured before and after yoga classes at time 1 (27–32
Pregnant women
Saliva
gestational weeks) and time 2 (34–37 gestational weeks). Subjective mood was assessed using the
Stress, Physiological profile of mood states. Saliva values and mood scores before and after each yoga class were compared
Stress, Psychological using paired t-test and Wilcoxon rank-sum test, respectively.
Findings: We analyzed 44 and 35 women at time 1 and time 2, respectively. The mean salivary cortisol
concentration declined significantly after each yoga class [time 1: 0.36–0.26 mg/dL (p < 0.001), time 2:
0.32–0.26 mg/dL (p = 0.001)]. The mean salivary alpha-amylase concentration also decreased
significantly following each class [time 1: 72.2–50.8 kU/L (p = 0.001), time 2: 70.6–52.7 kU/L
(p = 0.006)]. The scores for negative dimensions of mood (Trait-Anxiety, Depression, Anger-Hostility,
Fatigue, and Confusion) decreased significantly. The scores of Vigor for a positive dimension of mood
significantly increased.
Conclusion: This study indicated the immediate stress reduction effects of yoga during pregnancy.
ß 2016 Published by Elsevier Australia (a division of Reed International Books Australia Pty Ltd) on behalf
of Australian College of Midwives.

Summary of Relevance: What this Paper Adds

Problem
Excessive stress during pregnancy may cause mental disorders
in pregnant women and inhibit fetal growth. Yoga may alleviate
physiological and psychological stress during pregnancy; how-
ever, the effects have yet to be clearly shown.
Participants’ subjective mood improved after participation in
one yoga session. Moreover, physiological stress response mea-
sured using salivary cortisol and a-amylase was alleviated.
Thus, yoga may be useful for immediate stress reduction during
pregnancy.

What is Already Known

Previous studies showed that performing yoga during pregnan-
cy decreases perceived stress. In contrast, its effects on physio-
logical stress response have not been clarified.
* Corresponding author. Tel.: +81 3 5841 3396; fax: +81 3 5841 3396.
E-mail address: msumi-tky@umin.ac.jp (M. Matsuzaki).
1. Introduction
Pregnant women often experience challenges adapting to the
physical, mental, and social changes associated with progressing
pregnancy, which may trigger psychological and physiological
stress responses.1 Psychological stress responses include depres-
sive symptoms and anxiety, which potentially lead to maternal
psychological illness. In addition, excessive maternal physiological
stress responses can increase the risk of obstetric complications,

http://dx.doi.org/10.1016/j.wombi.2016.04.003
1871-5192/ß 2016 Published by Elsevier Australia (a division of Reed International Books Australia Pty Ltd) on behalf of Australian College of Midwives.

Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
G Model
WOMBI-523; No. of Pages 7
2 M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx
including preterm birth and low birth weight.2,3 High physiological
stress may induce preterm birth because high cortisol secretion
can disturb hormonal regulation of placenta and fetus.2,3 It is
suggested that fetal growth is hindered by the reduction of
placental perfusion due to the elevation of cortisol and catechol-
amine levels (norepinephrine, epinephrine, and dopamine).4
Cortisol is secreted during activation of the hypothalamic-
pituitary-adrenal axis (HPA axis) as an endocrine-related physio-
logical stress response. The level of catecholamine is also elevated
during activation of the sympathetic-adrenal-medullary axis (SAM
axis). Following HPA- and SAM-axis endocrine activity, several
changes occur in the circulatory system, such as the elevation of
blood pressure and heart rate.5 Stress reduction methods could
provide a useful way to alleviate HPA- and SAM-axis activation.
It is known that stress responses among pregnant women are
relieved by guided imagery, progressive muscle relaxation6,7 and
massage8. In contrast, the stress reduction effect of yoga has not
been clearly demonstrated in pregnant women. Yoga is suitable for
pregnant women because compared with other physical activity it
has relatively low exercise intensity and poses low risks of
accidental falling and hard contact with other persons. The practice
of yoga combines poses (asana), breathing (pranayama), and
meditation (diyana). Hatha yoga is a style of yoga that focuses on
gentle poses that resemble animals and plants, which stretch and
loosen whole body. Using psychological and physiological
measures, many previous reviews of non-pregnant populations
suggested that yoga has reduced stress.9
Previous studies using psychological measurements reported
that performing yoga during pregnancy decreases perceived stress,
anxiety and depressive symptoms and improves quality of life.10 In
addition, some previous studies suggested that yoga has reduced
the risk of obstetric complication, such as preterm birth and low
birth weight.10,11 The underlying mechanism might be the
alleviation of physiological stress response.4,5 Recently, Ber-
shadsky et al.12 and Newman et al.13 have shown a significant
reduction of salivary cortisol concentration after yoga participation
among healthy pregnant women in the USA and UK, respectively.
However, the effects of yoga on pregnant Japanese women have
not been demonstrated. Moreover, it is not been clarified how yoga
during pregnancy will affect SAM-axis activity in the pregnant
women. By clearing the effects of yoga on HPA- and SAM-axis
activity, yoga can be proposed to pregnant women as one of the
techniques which alleviate the stress responses during pregnancy.
Therefore, this study aimed to verify the immediate effects of yoga
on physiological stress responses, which include HPA- and SAM-
axis activity during pregnancy, combined with psychological
measurements. We hypothesized that physiological stress re-
sponse was alleviated and subjective mood improved after
participation in yoga class.
2. Participants, ethics, and methods
2.1. Design and setting
One group pre–post test design was adopted. Participants
practiced yoga from 18–22 gestational weeks until birth.
Participants’ stress responses were measured before and after
yoga classes at two time points to assess how one yoga session
decreases stress at different phases of their pregnancy. The first
assessment was scheduled for 27–32 gestational weeks (time 1),
in the middle of the study period. The second assessment was
scheduled for 34–37 gestational weeks (time 2), near the
conclusion of the study period. The study was conducted at a
university hospital in Tokyo, Japan, between March and
December 2013.
Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
2.2. Participants
Women at 18–22 gestational weeks were identified from the
list of patients who were scheduled to attend the prenatal check-
up on the day. One researcher evaluated the study eligibility of
328 pregnant women who visited the outpatient section for their
check-up, between March and August 2013. Participants’
eligibility criteria were as follows: women who were primiparas;
with singleton pregnancies; and were permitted to participate in
yoga classes by their doctors. The exclusion criteria were as
follows: women who were less than 20 years old; had inadequate
Japanese literacy skills; were planning to transfer to another
hospital; had previously experienced still birth or repeated
abortion; had a psychiatric disorder before or during pregnancy;
were at risk for premature delivery or abortion; had cervical
incompetency; were diagnosed with low-lying placenta or
placenta previa; or were restricted from performing exercises
for cardiovascular disease or other complications. One researcher
confirmed eligibility of each woman for the study by asking her
or her doctors. Then, the researcher explained the study contents,
schedule, and ethical consideration to each woman by using
written information.
2.3. Study schedule
At recruitment, written consent was obtained after the
explanation about the study. Subsequently, participants filled
out a questionnaire survey, which included demographic data,
such as employment status; educational history; and information
concerning alcohol, tobacco, and medications that could influence
saliva secretion. The questionnaires were collected again at time 1
and time 2 to record the changes in participants’ demographic
data. Other demographic data were obtained from the participants’
medical charts, including age, gestational weeks, and health status.
After the first survey at recruitment, participants started to attend
a yoga class and practice yoga at their home by using a digital video
disc (DVD). They continued to practice yoga until they delivered.
Outcome assessments before and after yoga session at time 1 and
time 2 were conducted between May and December 2013. The
outcome assessment included the measurements of stress
responses; salivary cortisol, salivary alpha-amylase (a-amylase),
and subjective mood which was obtained using the profiles of
mood states (POMS)14.
2.4. Outcome variables and the assessment protocol
2.4.1. Physiological stress response
2.4.1.1. Salivary cortisol. Cortisol is a stress hormone that reflects
the HPA axis activity. Cortisol exists in unbound and protein-
bound forms in serum. Unbound cortisol in saliva reflects both the
total and the free cortisol in serum.15 Salivary cortisol follows the
diurnal circadian rhythm in which it increases rapidly after
waking, and then decreases slowly over the course of the day.16
Salivary cortisol reactivity to stress is maintained during
pregnancy.17 Additionally, salivary cortisol concentration ele-
vates as pregnancy progresses.18
Participants used 5-cm straw to collect 2-ml saliva from under
their tongue and deposited the saliva into a polypropylene tube.
Within 4 h after collection, the saliva samples were frozen at -
80 8C. All samples were assayed for salivary cortisol in duplicate
using a salivary enzyme immunoassay (EIA) kit (Salimetrics, State
College, LA, USA) within 4 months after the collection. The plates
were assayed photometrically at 450 nm using a plate reader (Bio-
Rad, 680XR, CA, USA). Average intra- and inter-assay coefficients of
variation were 3.2% and 6.2%, respectively.
G Model
WOMBI-523; No. of Pages 7
M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx
2.4.1.2. Salivary a-amylase. Salivary a-amylase, which reflects
SAM-axis activity, is often used as a non-invasive and alternative
biomarker to norepinephrine.19 Salivary a-amylase follows a
diurnal course that is opposite to the diurnal course of cortisol;
salivary a-amylase decreases sharply after waking, and then
increases gradually.20 Salivary a-amylase reactivity to stress
attenuates during the third trimester.17 Some cross-sectional
studies have suggested that salivary a-amylase concentration
decreases from the first to third trimester.17,20
Alpha-amylase concentration was measured using a salivary
amylase monitor (NIPRO Corporation, Osaka, Japan). The biosensor
consists of a disposable test strip and the main unit that has a
digital display. The test strip contained a reagent paper infused
with 2-Chloro-4-nitrophenyl-4-galactopyranosylmaltoside (Gal-
G2-CNP). Saliva was collected using the strip by placing it under
the participants’ tongue for 30 s. After the strip was inserted into
the main unit, Gal-G2-CNP began to be metabolized. Subsequently,
the color density of the reagent paper in the strip was measured
photometrically at 430 nm, and the salivary a-amylase value was
displayed.21
2.4.2. Psychological stress response
2.4.2.1. Mood. Mood was assessed to determine various aspects of
psychological stress responses. Momentary mood was measured
using the Japanese version of the POMS-65 scale14. The POMS scale
consists of six mood dimensions: Tension-Anxiety, Depression,
Anger-Hostility, Fatigue, and Confusion, which are negative mood
dimensions, as well as Vigor, which is a positive dimension of
mood. Participants rated each item on a 5-point Likert scale that
ranged from 0 (not at all) to 4 (extremely), based on their feeling at
the assessment. The validity and reliability of the Japanese version
were confirmed.14 The mean values of Cronbach’s a in this study
were as follows: 0.88 for Tension-Anxiety, 0.88 for Depression,
0.92 for Anger-Hostility, 0.93 for Fatigue, 0.76 for Confusion, and
0.91 for Vigor.
2.4.3. Assessment protocol
Before and after yoga classes that were designated for outcome
assessments at time 1 and 2, one researcher first collected the
POMS scales and then the saliva samples. To avoid the influences
of food on stress response, participants were asked to refrain from
consuming food for 1 h before the saliva collection. Additionally,
participants were asked to drink nothing except water after their
last meals. Saliva samples were collected more than 10 min after
the participants had rinsed their mouths with water. Samples
were collected at scheduled times to reduce the influence of
diurnal changes in salivary cortisol and a-amylase. Before the
class, participants were assessed between 13:30 and 14:00.
Afterwards, participants practiced yoga for 60 min (14:00–15:00).
Following the class, the participants were assessed again (15:00–
15:30).
2.4.4. Intervention
For the present study, the prenatal yoga classes were conducted
at the hospital twice a month on weekdays. Researchers and a yoga
instructor held the classes. The instructor for the yoga classes was
certificated by the Japan Maternity Fitness Association, which
developed the Hatha yoga program for pregnant women. Each
class lasted 60 min, consisting of 15 min of warm-up, 40 min of
poses, and 5 min of meditation and breathing. During the warm-
up, participants stretched their whole bodies in a sitting position.
Then, participants practiced 7–8 yoga poses, took a 5-min break,
and subsequently practiced 7–8 more poses with guided instruc-
tion. All poses were coordinated with breathing. To avoid halting
breath, participants bended backward during inhalation and
Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
3
bended forward during exhalation for deeper stretches. During
the meditation, participants closed their eyes in supine or lateral
positions and meditated with the guided imagery provided by the
instructor. Yoga was practiced on yoga mats, at a comfortable
temperature, and accompanied by the relaxation music. Partici-
pants were offered to participate in yoga classes as often as
possible.
When participants attended their first yoga class, they received
DVD which contained the same program as the yoga class. They
were asked to practice yoga 3 times a week at their homes. They
recorded the times of their practices, and noted comments on their
physical conditions and emotions before and after the practice in
the original record sheet provided by researchers. Participants
attended the yoga classes and were given the yoga DVD without
charge. Moreover, as a token of gratitude for participation in the
study, they were provided with a 500-yen (approximately five
dollars) bookstore gift certificate three times; that are at
recruitment, time 1, and time 2.
2.4.5. Ethical considerations
The study protocol was approved by the Ethical Committee of
the Graduate School of Medicine of The University of Tokyo (No.
3812). Additionally, this study was registered as a clinical trial
(study ID: UMIN 000008607). The participants received explana-
tions of the purpose and contents of the study, the voluntary nature
of their participation, the risks and benefits of exercise during
pregnancy, and privacy considerations. Written consent was then
obtained from each participant and her partner or other family.
Participants were approved for yoga by their doctors with written
permissions.
During the study period, the health status of each participant
was monitored by the researchers and medical staffs using the
participant’s medical chart. Participants received blood pressure
checks and were asked about their condition before and after yoga
classes. Further, the heart rate of each participant’s fetus was
checked using a Doppler fetal monitor before and after classes.
Participants were instructed not to practice yoga at their homes
when they were in a poor state of health.
2.4.6. Statistical analysis
Participants’ demographic data were presented as mean and
Standard Deviation or n (%). Total times of yoga practice at home
were noted by using the records of participants. Yoga class
participation was recorded by researchers. Frequency of yoga
practice was calculated as the sum of yoga classes and practice at
home divided by the total intervention period of yoga.
We compared the concentrations of salivary cortisol and
a-amylase before and after each class using the paired t-test. The
POMS scores were compared using the Wilcoxon rank-sum test.
We also calculated the percentage of women who showed a
decrease in salivary cortisol and a-amylase concentrations and
POMS scores of negative dimensions. The percentage of women
who showed an increase in POMS vigor scores was also calculated.
Statistical analyses were performed using the Statistical Package
for Social Sciences for Windows Version 22 (SPSS Japan Inc.,
Japan). For all statistical tests, Two-tailed p-values less than 0.05
were considered statistically significant. Moreover, we calculated
the effect size of one yoga session on salivary cortisol and
a-amylase concentration (effect size d) and POMS scores (effect
size r). Effect size d in this study was Cohen’s d adjusted for the
correlation between values before and after yoga classes, which
was calculated by using G*power22. The reference values for
evaluating effect size d were as follows; 0.20 was ‘‘small’’, 0.50 was
‘‘medium’’, and 0.80 was ‘‘large’’.23 The reference values for
evaluating effect size r were as follows; 0.10 was ‘‘small’’, 0.30 was
‘‘medium’’, and 0.50 was large.23
G Model
WOMBI-523; No. of Pages 7

4 M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx

Using power estimated with values of physiological measure-

ments in previous studies, we calculated required sample size.
Studies examining yoga’s effects on salivary cortisol24 and
a-amylase25 among healthy adults have Cohen’s d values of
0.51 and 0.65, respectively. Thus, the required number of
participants for the analysis would be 33 and 21 women,
respectively (p < 0.05 and 80% power).

3. Results

Fig. 1 shows the flow chart of study recruitment and

participation. Of the 328 women who were screened for the study
eligibility, 247 women were excluded because they did not meet
the eligibility criteria. The remaining 81 women were recruited,
and 60 agreed to participate (participation rate: 74.1%).
Of the 60 initial participants, 53 and 46 women continued
participation at time 1 and time 2, respectively. Seven dropped out
by time 1: two were at risk of premature delivery, 2 had fetal
growth retardation (FGR), 2 transferred to another hospital, and 1
refused to continue to participate. Seven more participants
dropped out by time 2: four were at risk of premature delivery,
1 was restricted from exercising because of forelying umbilical
cord, 1 refused to continue, and 1 delivered her child prematurely.
Of those who continued to participate, 44 and 35 women
participated in the outcome assessments at time 1 (44/53: 83.0%)
and time 2 (35/46: 76.0%), respectively. Thus, the final response Fig. 1. Flow chart of study recruitment and participation.
rate was 54.3% (44/81) and 43.2% (35/81) at time 1 and 2,
respectively. Nine and 11 women could not participate in the
outcome assessment at time 1 and time 2, respectively. The main 66) at time 1 and 2, based on self-report. Thus, the mean (range)
reason for their absenteeism was reported as ‘‘for their work’’ at frequency of yoga practice was 2.7 (0.4–6.0) and 2.4 (0.4–6.2)
time 1 and ‘‘for personal schedule conflicts’’ at time 2. The median times a week, respectively, based on self-report. No woman
number of participants in yoga classes, which included women consumed alcohol, tobacco, and medicines that would influence
who did not participate in outcome assessment at that time, was salivary secretion.
13 [range: 4 (at the end of the study) to 17]. The changes in salivary cortisol and a-amylase concentrations
Table 1 presents the participants’ demographic characteristics. and the POMS scores before and after the yoga classes were
The mean age of participants at recruitment was 34.4 years (35.0 presented in Table 2 (time 1) and Table 3 (time 2). The mean
years at time 1 and 35.1 years at time 2). At recruitment, 75.0% of salivary cortisol concentration declined significantly after each
the participants had graduated from the university or graduate yoga class [time 1: 0.36–0.26 mg/dL (p < 0.001), time 2: 0.32–
university. At least half of the participants worked at time 1 and 2 0.26 mg/dL (p = 0.001)]. The effect size d of yoga on salivary cortisol
(66.7% and 50.0%, respectively). At time 2, 85.8% of the analyzed was large. The mean salivary a-amylase concentration also
participants did not work including 15 (42.9%) women who had decreased significantly following each class [time 1: 72.2–
obtained maternity leave. Participants had attended mean (range) 50.8 kU/L (p = 0.001), time 2: 70.6–52.7 kU/L (p = 0.006)]. The
of 3.2 (2–5) and 5.3 (3–8) yoga classes at time 1 and time 2, effect size d of yoga on salivary a-amylase was medium. The
respectively. Mean (range) total times of yoga practice at home percentage of women who showed a decrease in salivary cortisol
(times) before outcome assessment were 12.9 (1–35) and 23.4 (1– and a-amylase concentrations were 75.0% and 63.6% at time 1 and

Table 1
Participants’ demographic characteristics.

Baseline (n = 60) time 1 (n = 44) time 2 (n = 35)

Variables Mean SD Mean SD Mean SD

Gestational weeks (weeks) 19.9 1.4 28.5 1.5 35.2 0.9

Age (years) 34.4 4.1 35.0 4.4 35.1 4.3
Participation in yoga
Total times of participation in classes (times) – 3.2 0.9 5.3 1.3
Total times of yoga practice at home (times) – 12.9 8.4 23.4 18.1
Total intervention period of yoga (days) – 44.7 13.7 90.1 13.6
Frequency of yoga practice a (times/week) – 2.6 1.3 2.3 1.5

n % n % n %

Working
Yes 40 66.7 22 50.0 5 14.2
No 20 33.3 22 50.0 30 85.8
Education
University, graduated university 45 75.0 32 72.7 25 71.4
High school, technical school, two-year college 15 25.0 12 27.3 10 28.5

SD, standard deviation.

a
The sum of yoga classes and practice at home, divided by the total intervention period of yoga.

Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
G Model
WOMBI-523; No. of Pages 7

M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx 5

Table 2 healthy pregnant women. We found that their salivary cortisol and
The changes in the salivary cortisol and a-amylase concentrations and the profiles
of mood states (POMS) scores before and after the yoga classes at time 1 (n = 44).
a-amylase decreased and their subjective mood improved after
yoga classes.
Pre-test Post-test pa db

Mean SD Mean SD 4.1. The stress reduction effects of yoga

Salivary 0.36 0.16 0.26 0.10 <0.001 0.97
cortisol (mg/dL) 4.1.1. Effects on physiological stress response
Salivary 72.2 45.8 50.8 26.5 0.001 0.60 Some previous studies, which investigated populations other
a-amylase (kU/L) than pregnant women, have shown reductions in salivary
Pre-test Post-test pc r cortisol24,26 and a-amylase25 concentrations following yoga. In
POMS original Median IQR Median IQR
addition, some studies12,13 have recently shown a significant
score reduction of salivary cortisol concentration after yoga participation
among healthy pregnant women in USA and UK, respectively. The
Tension-Anxiety 7.5 4.25–11.0 3.0 1.0–7.75 <0.001 0.75
Depression 3.5 0.3–7.0 0.0 0.0–2.75 <0.001 0.71 reduction in salivary cortisol that was observed in the present
Anger-Hostility 4.0 1.0–8.5 0.0 0.0–4.75 <0.001 0.71 study shows the reproducibility of results of these previous studies
Fatigue 9.0 4.0–13.0 3.5 1.0–9.0 <0.001 0.68 in a different country and race. Moreover, the present study is the
Confusion 7.0 4.25–10.0 3.0 3.0–6.0 <0.001 0.74
first to reveal a reduction of salivary a-amylase concentrations
Vigor 12.0 7.25–16.0 15.0 11.0–19.75 <0.001 0.59
after yoga classes among pregnant women. The results indicate
SD, standard deviation; IQR, inter-quartile range. that both HPA- and SAM-axis activations are alleviated by yoga.
a
Salivary cortisol and a-amylase concentrations were compared using the
Three components of yoga may reduce stress response. First,
paired t-test.
b
Effect size d adjusted with the correlation between pre and post values. body muscles tension could be loosened by various poses. Second,
c
POMS scores were compared using the Wilcoxon-rank sum test. it is suggested that slow and deep breathing lead to control the
autonomic nervous activity. Third, while meditating, participants
were especially conscious of their breathing and mind, allowing
70.6% and 73.5% at time 2, respectively. With respect to the median greater relaxation.
of the POMS scores, Trait-Anxiety, Depression, Anger-Hostility, However, some cautions are warranted regarding the interpre-
Fatigue, and Confusion significantly decreased after the yoga tation of the reduction in salivary cortisol concentration that was
classes at both time 1 and time 2. The percentage of women who observed in this study; indeed, this change may include reductions
showed a decrease in POMS negative mood dimensions at time 1 resulting from both the yoga intervention and the normal, gently
were as follows: 77.2%, tension-anxiety; 65.9%, depression; 77.2%, decreasing diurnal pattern. However, the changes in cortisol
anger-hostility; 81.8%, fatigue; and 79.5%, confusion. The percent- observed in this study were larger than the diurnal decrease
age at time 2 were as follows: 60.0%, tension-anxiety; 57.1%, reported by a previous study16. In addition, the salivary a-amylase
depression; 48.6%, anger; 74.3%, fatigue, and 68.6%, confusion. The concentration in this study reduced against the diurnal increase.
median Vigor scores also significantly increased. The effect size r of Accordingly, our results suggest that the reductions in salivary
yoga on mood scales was medium to large. The percentage of cortisol and a-amylase were mainly caused by yoga.
women who showed an increase in POMS vigor scores were 70.5%
and 62.9% at time 1 and 2, respectively. 4.1.2. Effects on psychological stress response
Few studies have assessed the short-term effects of exercise on
4. Discussion the mood of pregnant women. Polman et al.27 investigated the
mood of pregnant women using the POMS scale, and showed the
This study investigated the immediate physiological and improvement of mood after water- and studio-based exercises.
psychological stress reduction resulting from yoga performed by The present study also showed that the scores of all negative
dimensions decreased, while Vigor improved significantly. These
Table 3 results suggest that yoga is effective for improving mood during
The changes in the salivary cortisol and a-amylase concentrations and the profiles pregnancy. However, women with psychiatric disorders were not
of mood states (POMS) scores before and after the yoga classes at time 2 (n = 35).
included in this study. The participants’ scores are low in some
Pre-test Post-test pa db of the negative mood dimensions, especially in depression and
Mean SD Mean SD anger-hostility. Thus, the changes might not be clinically signifi-
cant. In addition, participants might have filled out the desired
Salivary 0.32 0.12 0.26 0.10 0.001 0.99
cortisol (mg/dL)c
answer on POMS scales because the study objectives could be
Salivary 70.6 37.0 52.7 34.7 0.006 0.74 perceived by them.
a-amylase (kU/L)c
Pre-test Post-test pd r 4.2. Evaluation of baseline concentration of salivary cortisol and
alpha-amylase before yoga classes
POMS original Median IQR Median IQR
score
Salivary cortisol concentrations before yoga classes were
Tension-Anxiety 7.0 4.0–11.0 3.0 2.0–6.0 0.001 0.54
Depression 1.0 0.0–4.0 0.0 0.0–2.0 0.001 0.56
almost as high as those in normal daily life observed in previous
Anger-Hostility 1.0 0.0–5.0 0.0 0.0–1.0 0.003 0.49 studies.16,28,29 In contrast, the salivary a-amylase concentrations
Fatigue 9.0 4.0–12.0 3.0 0.0–7.0 <0.001 0.67 observed in the present study were higher than the concentrations
Confusion 7.0 4.0–8.0 3.0 3.0–6.0 <0.001 0.65 observed in several previous studies of pregnant women.17,20 In
Vigor 11.0 9.0–13.0 13.0 11.0–17.0 0.007 0.46
a Japanese study that used the same amylase monitor as the
SD, standard deviation; IQR, inter-quartile range. present study, mean  SD salivary a-amylase concentrations was
a
Salivary cortisol and a-amylase concentrations were compared using the 42.5  34.1 kU/L among 104 pregnant women before health checkups
paired t-test
b
Effect size d adjusted with the correlation between pre and post values.
by midwives.30 The the high values of a-amylase in this study can
c
n = 34. partly been explained by the characteristics of salivary a-amylase,
d
POMS scores were compared using the Wilcoxon-rank sum test. that is, a large variation in concentration across individuals.

Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
G Model
WOMBI-523; No. of Pages 7
6 M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx
Additionally, the collection times and gestational weeks of partici-
pants differed across studies. Therefore, mean a-amylase concentra-
tions may vary across studies.
4.3. Study limitations
This study has several limitations. First, it was conducted at a
single university hospital. Perhaps in consequence, participants
were older and were more commonly employed the general
Japanese population. Moreover, the participants in this study were
limited to primiparous women because stress response may vary
between primiparas and multiparas. Therefore, caution is war-
ranted when generalizing the results of this study to other clinical
settings or women from different demographic backgrounds.
Second, this study had a one group pre–post test design. Therefore,
the changes in salivary cortisol and a-amylase had to be compared
with previous findings concerning the natural diurnal change.
Further studies using a resting control group instead of yoga
participation are required. Third, although researchers attempted
not to indicate the aims of the study to participants, it is difficult to
completely conceal the aim. Therefore, participants may have
reported desirable values on the POMS scale. Fourth, yoga consists
of 3 components; poses, breathing, and meditation. Thus, the stress
reduction effects of yoga may be induced by the complex
intervention. Although the program focused on poses, it is not
clear which component reduced stress most effectively. Fifth,
before and after yoga class, participants often chatted with other
participants, the yoga instructor, and researchers. Accordingly, the
stress reduction effect of the yoga classes may have partly resulted
from such communication. Finally, many participants could not
attend outcome assessment at yoga class at time 1 and 2 because of
conflicting schedules, as yoga classes were held only twice a month
on weekdays. Thus, the effects of yoga may differ between women
who could attend yoga class and those who could not. In addition,
the frequency of yoga practice was widely different across
participants. Thus, the yoga programs should be modified to suit
the preferences of pregnant women.
4.4. Strength and indications for further study
This study first cleared immediate stress reduction effects of
yoga among pregnant Japanese women by combining salivary
cortisol and a-amylase to measure both HPA- and SAM-axis
activity. The results indicate the immediate stress reduction effects
of yoga during pregnancy. Thus, yoga can be proposed to pregnant
women as one of the stress reduction methods. However, the long-
term effects of yoga on stress during pregnancy are not clearly
known. Therefore, further study is necessary to assess yoga’s long-
term effects on stress response and obstetric outcomes during
pregnancy.
5. Conclusion
This study revealed that pregnant women experienced imme-
diate reductions in salivary cortisol and a-amylase concentrations
after yoga classes. Their subjective mood also improved by yoga.
However, there are many limitations associated the study design.
Thus, it is required to conduct a randomized controlled trial that
compares the natural changes of physiological and psychological
stress in the control group. Moreover, many women cannot have
participated in yoga classes. Thus, in further studies, yoga classes
should be held when participants can attend the classes easily. In
addition, we only targeted healthy primiparous women in this
study. It may be required to assess the effects of yoga on
physiological stress among muitiparous women or women with
psychological disorders.
Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003
Conflicts of interest
All the authors declare no conflicts of interest in this work.
Acknowledgement
This work was supported by the Foundation for Total Health
Promotion, 2012; and the JSPS KAKENHI Grant Numbers 23792638
and 25293451. We thank the staff and instructors of the Japan
Maternity Fitness Association for offering the yoga program.
References
1. Lederman R, Weis K. Psychosocial Adaptation to Pregnancy; Seven dimensions of
maternal role development. 3rd ed. New York, USA: Springer; 2009.
2. Field T, Hernandez-Reif M, Diego M, Figueiredo B, Schanberg S, Kuhn C. Prenatal
cortisol, prematurity and low birthweight. Infant Behav Dev 2006;29(April
(2)):268–75.
3. Vrekoussis T, Kalantaridou SN, Mastorakos G, Zoumakis E, Makrigiannakis A,
Syrrou M, et al. The role of stress in female reproduction and pregnancy: an
update. Ann N Y Acad Sci 2010;1205(September):69–75.
4. de Weerth C, Buitelaar JK. Physiological stress reactivity in human pregnancy –
a review. Neurosci Biobehav Rev 2005;29(April (2)):295–312.
5. Dusek JA, Benson H. Mind-body medicine: a model of the comparative clinical
impact of the acute stress and relaxation responses. Minn Med 2009;92(May
(5)):47–50.
6. Alder J, Urech C, Fink N, Bitzer J, Hoesli I. Response to induced relaxation during
pregnancy: comparison of women with high versus low levels of anxiety. J Clin
Psychol Med Sett 2011;18(March (1)):13–21.
7. Urech C, Fink NS, Hoesli I, Wilhelm FH, Bitzer J, Alder J. Effects of relaxation on
psychobiological wellbeing during pregnancy: a randomized controlled trial.
Psychoneuroendocrinology 2010;35(October (9)):1348–55.
8. Field T, Diego M, Hernandez-Reif M, Deeds O, Figueiredo B. Pregnancy massage
reduces prematurity, low birth-weight and postpartum depression. Infant
Behav Dev 2009;32(December (4)):454–60.
9. Li AW, Goldsmith CA. The effects of yoga on anxiety and stress. Altern Med Rev
2012;17(March (1)):21–35.
10. Jiang Q, Wu Z, Zhou L, Dunlop J, Chen P. Effects of yoga intervention during
pregnancy: a review for current status. Am J Perinatol 2015;32(May (6)):
503–14.
11. Narendran S, Nagarathna R, Narendran V, Gunasheela S, Nagendra HR. Efficacy
of yoga on pregnancy outcome. J Altern Complement Med 2005;11(April
(2)):237–44.
12. Bershadsky S, Trumpfheller L, Kimble HB, Pipaloff D, Yim IS. The effect of
prenatal Hatha yoga on affect, cortisol, and depressive symptoms. Complement
Ther Clin Pract 2014;20(May (2)):106–13.
13. Newham JJ, Wittkowski A, Hurley J, Aplin JD, Westwood M. Effects of antenatal
yoga on maternal anxiety and depression: a randomized controlled trial.
Depress Anxiety 2014;31(August (8)):631–40.
14. Yokoyama K, Araki S, Kawakami N, Takeshita T. Production of the Japanese
edition of profile of mood states (POMS): assessment of reliability and validity.
Nihon Koshu Eisei Zasshi 1990;37(November (11)):913–8.
15. Törnhage CJ. Salivary cortisol for assessment of hypothalamic-pituitary-adre-
nal axis function. Neuroimmunomodulation 2009;16(5):284–9.
16. Giesbrecht GF, Campbell T, Letourneau N, Kooistra L, Kaplan B, APrON Study
Team. Psychological distress and salivary cortisol covary within persons during
pregnancy. Psychoneuroendocrinology 2012;37(February (2)):270–9.
17. 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(April (4)):1329–35.
18. Allolio B, Hoffmann J, Linton EA, Winkelmann W, Kusche M, Schulte HM.
Diurnal salivary cortisol patterns during pregnancy and after delivery: rela-
tionship to plasma corticotrophin-releasing-hormone. Clin Endocrinol (Oxf)
1990 August;33(2):279–89.
19. Nater UM, Rohleder N. Salivary alpha-amylase as a non-invasive biomarker for
the sympathetic nervous system: current state of research. Psychoneuroendo-
crinology 2009;34(May (4)):486–96.
20. Giesbrecht GF, Granger DA, Campbell T, Kaplan B, APrON Study Team. Salivary
alpha-amylase during pregnancy: diurnal course and associations with obstet-
ric history, maternal demographics, and mood. Dev Psychobiol 2013;55(March
(2)):156–67.
21. Shetty V, Zigler C, Robles TF, Elashoff D, Yamaguchi M. Developmental valida-
tion of a point-of-care, salivary a-amylase biosensor. Psychoneuroendocrinology
2011;36(February (2)):193–9.
22. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power
analysis program for the social, behavioral, and biomedical sciences. Behav Res
Methods 2007;39(May (2)):175–91.
23. Cohen J. A power primer. Psychol Bull 1992;112(July (1)):155–9.
24. West J, Otte C, Geher K, Johnson J, Mohr DC. Effects of Hatha yoga and African
dance on perceived stress, affect, and salivary cortisol. Ann Behav Med
2004;28(October (2)):114–8.
G Model
WOMBI-523; No. of Pages 7

M. Kusaka et al. / Women and Birth xxx (2016) xxx–xxx 7

25. Gururaja D, Harano K, Toyotake I, Kobayashi H. Effect of yoga on mental health:
Comparative study between young and senior subjects in Japan. Int J Yoga
2011;4(January (1)):7–12.
26. Michalsen A, Grossman P, Acil A, Langhorst J, Lüdtke R, Esch T, et al. Rapid stress
reduction and anxiolysis among distressed women as a consequence of a three-
month intensive yoga program. Med Sci Monit 2005;11(December (12)). CR555-61.
27. Polman R, Kaiseler M, Borkoles E. Effect of a single bout of exercise on the mood
of pregnant women. J Sports Med Phys Fitness 2007;47(March (1)):103–11.
28. Tsubouchi H, Nakai Y, Toda M, Morimoto K, Chang YS, Ushioda N, et al.
Change of salivary stress marker concentrations during pregnancy: maternal
depressive status suppress changes of those levels. J Obstet Gynaecol Res
2011;37(August (8)):1004–9.
29. Voegtline KM, Costigan KA, Kivlighan KT, Laudenlager ML, Henderson JL,
DiPietro JA. Concurrent levels of maternal salivary cortisol are unrelated to
self-reported psychological measures in low-risk pregnant women. Arch
Womens Ment Health 2013;16(April (2)):101–8.
30. Izumi M, Habuta C, Kabeyama K. View of pregnant women concerning health
checkups involving ultrasonography and its psychological effects – comparison
between midwives and physicians based on the cardiac rate, salivary amylase,
and STAI. J Jpn Acad Midwif 2011;25(1):36–44.

Please cite this article in press as: Kusaka M, et al. Immediate stress reduction effects of yoga during pregnancy: One group pre–post
test. Women Birth (2016), http://dx.doi.org/10.1016/j.wombi.2016.04.003