Key words
breast cancer - yoga - oncological care - quality of life
Schlüsselwörter
Brustkrebs - Yoga - onkologische Versorgung - Lebensqualität
Introduction
Breast cancer is the most common malignancy in women worldwide. It accounts for 14 %
of all cancer diagnoses in Europe [1]. In the
industrialized countries of the northern hemisphere, between one in eight and one
in
ten women will be diagnosed with breast cancer in their lifetime [2], [3]. Due to the enormous
advances in adjuvant and palliative treatment, it has, in many cases, become a
controllable chronic disease [4].
Health-related quality of life (QoL) is very important for patients. QoL is defined
as self-perceived well-being related to, or affected by the presence of disease or
treatment [5]. As a multifactorial construct, it includes
perceptions, both positive and negative, of many different dimensions such as
physical, emotional, social and cognitive functional status. It also includes the
negative aspects of symptoms caused by disease and/or its treatment [6]. Breast cancer is one of the oncologic diseases in which
QoL has been studied most [7]. Recent studies have
suggested that QoL in the months after diagnosis and breast cancer surgery may be
a
predictor of psychological well-being later on [8].
Within the population of cancer patients, there is a growing interest in
complementary therapies. Compared to other oncologic entities, e.g. gynecological
cancer, women with breast cancer frequently make use of complementary therapies
[9]. The overall rate has been reported to be as high
as 75 % [10], [11], [12].
As a mind–body discipline, yoga is claimed to offer physical, mental and spiritual
benefits [13]. Studies have shown beneficial physical and
psychological effects in different cancer patients [13], [14], [15], [16]. Studies of yoga programs for breast
cancer patients showed improvements in physical well-being, social function,
emotional health and functional adaption [17]. Yoga
programs can reduce distressing symptoms in patients undergoing treatment for breast
cancer [18], [19]. After hatha
yoga, constraints on physical activity were reduced and fitness was found to be
improved in breast cancer survivors [20]. Yoga has also
been shown to be beneficial in reducing chemotherapy-induced side effects compared
with other supportive therapies [21]. Women with breast
cancer showed a cognitive improvement after participating in yoga programs [22], [23]. They reported less
fatigue when undergoing breast cancer treatment [24], [25], [26].
Studies of yoga programs for breast cancer patients found improvements in emotional
outcomes and in quality of life [23], [25], [27], [28]. Anxiety and depression was reduced with yoga
interventions compared with supportive therapy following surgery, radiation therapy
and chemotherapy [18], [29].
Several authors have described an increase in positive affect and a decrease in
negative affect in patients with breast cancer who participated in yoga programs
[22], [25].
Women who were prescribed a 6-week daily yoga program reported improved confidence
in
social settings relative to controls who had only a brief supportive therapy [22]. Moadel et al. described that participants in a yoga
group did not report a significant change in social well-being but the 12-week
waiting-list control group noted a significant decrease in support [30].
The aim of our study was to test the hypothesis that newly diagnosed patients with
early breast cancer could benefit from yoga in the early postoperative period. This
benefit would be expressed primarily by a higher QoL and secondarily by an
improvement in physical activity. Furthermore, we hypothesized that QoL is
influenced by sociodemographic factors and the type of surgery.
Material and Methods
The study was approved by the Ethics Committee of the Humboldt University of Berlin.
It was conducted at the breast center of DRK Kliniken Westend, Berlin, Germany.
Patients were allocated on the ward after admission before their first operation for
early breast cancer.
Initially, 93 patients were randomized into an intervention group (IG) and a control
group (waiting group, WG). The IG started yoga immediately after the operation. The
WG started yoga 5 weeks after surgery. Both groups attended yoga classes twice
weekly for 5 weeks. For various reasons 17 patients dropped out in the IG, and 13
dropped out in the WG. The average number of attended yoga classes was 7.62 (SD
± 3.78) in the IG and 7.25 (SD ± 4.21) in the WG with a range of 0–11 in both groups
(p = 0.89, Mann-Whitney U-test). No statistical difference was found in the numbers
attending classes.
Various reasons were cited for canceling classes, but most patients gave no reason
(7
women in the IG, 8 women in the WG). The second most frequently cited reason was
ongoing adjuvant treatment. No difference was found for both groups (p = 0.50,
likelihood ratio test).
After randomization, 2 patients withdrew their consent and 2 patients were excluded
when the inclusion criteria were reassessed. Patient flow is shown in [Fig. 1].
Fig. 1 Flow chart.
The study cohort consisted of patients with breast cancer who underwent surgery
(mastectomy, breast-conserving therapy (BCT), sentinel node biopsy (SLN) or axillary
lymph node dissection (ALND). Inclusion criteria were “being able to attend 10 yoga
classes over a period of 14 weeks” and “being able to start yoga two to three days
after surgery” (IG). All patients who underwent axillary operations had either SLN
or ALND. Patients also had to be able to start yoga classes 5 weeks after surgery
(WG). The Karnofsky Index was > 70 %.
Patients who did not undergo surgery or who underwent procedures other than those
listed above, patients with distant metastases, patients who already regularly
attended yoga classes, patients with limited German language skills, patients with
alcohol and illicit drug abuse and psychotic disorders were all excluded from the
study.
Sample size calculation
Sample size calculation was performed using G*Power3.1. The estimated number of
patients needed to assume a large effect in the study sample (effect size
d = 0.8) was 35 to 37 persons per group, depending on the statistical test used.
Because of the expected dropout rate in a longitudinal trial, a larger number of
participants were randomized to each group.
To evaluate quality of life (QoL), the German version of the European
Organization of Research and Treatment of Cancer QoL questionnaire EORTC QLQ-C30
and its breast-cancer-specific module EORTC QLQ-BR23 were administered at
baseline, after the intervention and 3 months after the intervention. These are
self-reported questionnaires with a good reliability, validity and sensitivity
to change. Validation studies have been completed [31], [32]. The EORTC QLQ-C30 includes
functional and symptom scales, a QoL scale, and six single items. All scales and
single-item measures range in score from 0 to 100. A high score for a functional
scale represents a high level of functioning, a high score for the global QoL
represents a high QoL, but a high score for a symptom scale represents a high
level of symptoms [31]. The EORTC QLQ-BR23
questionnaire assesses areas related to treatment modalities. It comprises 23
questions and also assesses specific aspects of the QoL in breast cancer
patients [33]. Responses are rated on four-point
Likert scales.
After 3 months the patients were additionally asked whether practicing yoga had
improved their subjective level of physical activity (Did yoga improve the way
you move?) and whether they wished to continue yoga after the study (Do you
intend to continue yoga classes after the study?).
The study was designed as a randomized, prospective, controlled trial in
accordance with the CONSORT guidelines. Written informed consent was obtained
from each study participant before randomization. Sealed envelopes were used to
individually randomize patients into one of two groups (IG and WG) using a
lottery randomization process. With this design we hoped to improve recruitment
and acceptance of the study by the patients.
Participants completed the questionnaires at baseline (prior to intervention),
after yoga classes and 3 months after ending the classes. The WG received no
additional interventions beyond adjuvant breast cancer therapy and routine
healthcare.
Classes were held in a gym in the hospital and the program was especially created
for breast cancer patients [34]. Yoga classes were
held twice a week with each class lasting around 75 minutes – not a single yoga
class was cancelled during the period of yoga intervention (June 2008 until
November 2009). Each participant was asked to attend ten classes over a period
of approximately five weeks. Classes were conducted by two certified yoga
teachers, one demonstrating the asana together with variations tailored to the
needs of patients recovering from surgery, the other teacher assisting the
participants with props and finding individual variations of the asana shown.
The term “asana” (Sanskrit for “sitting down”) refers to the sitting body
position in which yoga is practiced. The teachers were trained gitananda and
vini yoga teachers who agreed to a defined set of asana (see [Table 1]) and teaching style. The teachers were
careful to ensure that participants performed each exercise according to their
own optimal degree of intensity.
Table 1 Synopsis of yoga asana used in yoga classes [34].
|
Steps for each yoga class
|
Yoga postures (asana)
|
|
Supine postures: initial posture to “ground” the patient
|
Shanti asana posture (Shavasana)
Dorsal tree posture
(Vrikshasana)
Hip-joint rotations (Pada
Sanchalanasana)
Knees to chest (Apanasana)
|
|
Pranayama postures to promote stretching and breathing
|
Breathing and moving (Yoga Mudra)
Camel posture
(Ushtrasana)
Tiger breathing (Chakravakasana)
|
|
Postures for the neck, shoulders, and arms for the relief of
tension
|
Arm and head coordination (Skandha Chakra)
Head
rotations (Greeva Chakra)
Head Bows (Greeva
Sanchalana)
|
|
Eye exercises: alternate tightening and relaxing of eye
muscles
|
Focus near and far (Nasikagra Drishti)
The lying
eight
Cupping and blinking
|
|
Standing postures: encourage patients to “stand tall” as
before
|
Hero 1 and 2 (Virabhadrasana)
Knee and hip rotations
(Janu Chakra)
Stargazer (Hasta
Utthanasana)
Triangle in motion (Utthita
Trikonasana)
|
|
More supine postures: lying on the mat
|
Boat posture (Paripurna Navasana)
Shoulder bridge
(Dvipada Pitham)
Knees to chest (Apanasana)
|
|
Relaxing postures to feel the effects of yoga
|
Kaya Kriya
Yoga Nidra
|
|
Energizing postures to end yoga practice
|
Alternate nostril breathing (Nadi
Shodana)
Chanting
Seated posture
(Virasana/Sukhasana)
|
Each lesson started with lying postures and the gradual mobilization of arms and
legs to encourage flexibility and strength. Various breathing exercises followed
to allow participants to feel their chest wall expanding. During the dynamic
part of the exercises, a range of standing and sitting positions were shown.
These postures aimed to enhance mobility of the arms and shoulders and
strengthen the legs and feet as well as improving balance. Eye exercises were an
important element of the session to help alleviate the vision problems that can
be caused by chemotherapy, although more research is needed to validate these
effects.
At the end of the class, participants were led through a series of concentration
exercises to help them relax and become aware of the effects of the yoga
exercises. The yoga class encompassed a large variety of movements, all of which
were easily doable in their modified versions. The modifications were created to
meet post-surgical needs and to help during the period following therapy when
the women were less flexible.
Outcome measures
The primary outcome measure was the QoL of the patients before, immediately
after, and three months after the intervention. The influence of yoga on
physical activity served as a secondary outcome measure.
Statistical analysis
Scoring of the questionnaires was done according to the guidelines given in the
test manuals. A missing data analysis was performed and all variables were
analyzed for normality. Means and standard deviations for interval data were
obtained, and independent two-sample t-tests were performed for normally
distributed data. Mann-Whitney U-test was used for abnormally distributed data.
The χ2 test was used to assess the significance of differences in
proportions for nominal and ordinal variables. The likelihood ratio test was
used if the conditions for chi-square testing were not fulfilled. Friedman test
was used for nonparametric measures and one-way repeated measures analysis of
variance (ANOVA) for longitudinal measures in dependent samples. Two-sided
p-values < 0.05 were considered statistically significant. Data analysis was
done using SPSS version 17.0 (Chicago, IL, USA).
Results
Baseline characteristics
Sociodemographic and clinical sample characteristics are shown in [Table 2]. Due to missing values, some of the variables
do not sum to n = 49 (intervention group, IG), n = 44 (waiting group, WG) or
n = 93 (total sample). Mean patient age was 55.82 (SD ± 10.72, R 37–78) for the
IG and 58.41 (SD ± 9.91, 40–84) for the WG (p = 0.23, independent samples
t-test). All sociodemographic parameters were comparable between the groups.
Table 2 Sample characteristics.
|
Parameter
|
IG n (%)
|
WG n (%)
|
Total n (%)
|
p
|
|
a Likelihood ratio test
b
χ2 test
c Mann-Whitney
U-test
M = mean; R = range; SD = standard
deviation
IG = intervention group; WG = waiting
group; BCT = breast-conserving therapy; SLN = sentinel lymph
node biopsy; ALND = axillary lymph node dissection
*
Multiple answers possible
|
|
Nationality
|
|
|
|
0.71a
|
|
|
46 (96)
|
36 (97)
|
82 (97)
|
|
|
|
2 (4)
|
1 (3)
|
3 (3)
|
|
|
Living in a partnership
|
|
|
|
0.37b
|
|
|
30 (64)
|
27 (73)
|
57 (68)
|
|
|
|
17 (36)
|
10 (27)
|
27 (32)
|
|
|
Marital status
|
|
|
|
0.93c
|
|
|
13 (27)
|
9 (24)
|
22 (24)
|
|
|
|
22 (46)
|
19 (51)
|
41 (44)
|
|
|
|
7 (15)
|
4 (11)
|
11 (12)
|
|
|
|
6 (12)
|
5 (14)
|
11 (12)
|
|
|
Level of education
|
|
|
|
0.15c
|
|
|
22 (51)
|
22 (67)
|
44 (58)
|
|
|
|
14 (33)
|
9 (27)
|
23 (30)
|
|
|
|
7 (16)
|
2 (6)
|
9 (12)
|
|
|
Employment status
|
|
|
|
0.52a
|
|
|
32 (71)
|
25 (66)
|
57 (70)
|
|
|
|
13 (29)
|
12 (33)
|
25 (30)
|
|
|
Type of surgery*
|
|
|
|
|
|
|
29 (62)
|
24 (60)
|
53 (61)
|
|
|
|
18 (38)
|
16 (40)
|
34 (39)
|
|
|
|
37 (71)
|
32 (78)
|
69 (74)
|
|
|
|
15 (29)
|
9 (22)
|
24 (26)
|
|
|
Adjuvant treatment*
|
|
|
|
|
|
|
|
|
|
0.14a
|
|
|
17 (53)
|
7 (30)
|
24 (44)
|
|
|
|
15 (47)
|
16 (70)
|
31 (56)
|
|
|
|
|
|
|
0.65b
|
|
|
23 (70)
|
16 (64)
|
39 (67)
|
|
|
|
10 (30)
|
9 (36)
|
19 (33)
|
|
Due to the high number of participants who dropped out for various reasons (16 in
the IG, 16 in the WG; [Fig. 1]) only 31 women
completed the intervention in the IG and 28 women in the WG. After 3 months, 30
women of the IG and 23 women of the WG answered the questionnaires.
Over time, the quality of life (QoL) scores changed in both groups. Using the
EORTC scales, global QoL and functional status increased while symptoms
decreased ([Table 3]). Symptoms that improved over
time were fatigue, nausea and vomiting, pain, dyspnea, insomnia, appetite loss
and other gastrointestinal symptoms. Breast-specific symptoms that became less
troublesome over time were systemic therapy side effects, breast and arm
symptoms and distress at hair loss.
Table 3 EORTC QoL scores.
|
EORTC scales
|
|
M (SD) baseline
|
M (SD) after yoga classes
|
M (SD) after 3 months
|
P
|
|
a Friedman test
b One-way
repeated measures ANOVA
M = mean; SD = standard
deviation
IG = intervention group; WG = waiting
group
|
|
Overall QoL
|
IG
|
51.33 (22.23)
|
59.11 (25.08)
|
66.95 (18.96)
|
F(2) = 12.17
p = 0.002
a
|
|
|
WG
|
52.78 (23.35)
|
57.72 (20.53)
|
70.83 (15.59)
|
F(2) = 3.96
p = 0.14a
|
|
Functional scales
|
IG
|
58.39 (18.12)
|
67.40 (18.60)
|
68.26 (15.50)
|
F(2) = 6.67
p = 0.005
b
|
|
|
WG
|
57.48 (20.70)
|
62.32 (14.85)
|
67.37 (19.18)
|
F(2) = 0.39
p = 0.69b
|
|
Symptoms
|
IG
|
29.76 (18.10)
|
26.01 (17.19)
|
21.99 (10.23)
|
F(2) = 0.98
p = 0.42b
|
|
|
WG
|
26.51 (13.89)
|
28.15 (15.57)
|
22.83 (13.76)
|
F(2) = 7.43
p = 0.01
b
|
Women in the IG had a higher overall QoL and functional status after 5 weeks of
yoga classes than women in the WG at the same point in time (5 weeks after the
operation) before the start of yoga classes. The improvement of overall QoL and
functional status was only statistically significant in the IG (p = 0.002 and
p = 0.005, resp.). Symptoms decreased in the IG and in the WG over a period of 3
months. Patients in the IG had fewer symptoms after 5 weeks of yoga classes.
Women in the WG had more symptoms after yoga classes; these symptoms decreased 3
months later. The decrease in symptoms was statistically significant for the WG
(p = 0.01). There was no difference in symptoms when women in the IG were
compared 5 weeks after surgery and after finishing yoga classes with women in
the WG 5 weeks after surgery before the start of yoga.
86 % (24/28) of the IG affirmed that their physical activity had improved through
their participation in the study, compared to only 59 % (10/17) of the WG
(p = 0.04) ([Fig. 2 a]). More patients in the IG
(97 %) intended to continue with yoga classes after completing the study
(p = 0.03) ([Fig. 2 b]).
Fig. 2 a and b a Improvement of physical activity;
b Intention to continue with yoga.
We found that EORTC scores differed significantly depending on the type of
surgery performed and on patient age. Functional status after mastectomy (t[73]
= − 2.00; p = 0.05) was lower than after BCT (t[73] = 2.20; p = 0.03). Patients
over 60 years of age (t[46] = 2.09; p = 0.04) and patients with SLN scored
significantly lower on symptom scales than those with ALND (t[45] = − 3.01;
p = 0.00) ([Fig. 3]).
Fig. 3 EORTC functional and symptom scales, depending on the type of
surgery and patient age.
Discussion
In our study we examined the impact of yoga on the quality of life (QoL) of newly
diagnosed patients with early breast cancer early after surgery. Our hypothesis was
that patients would benefit from the early start of yoga classes.
In both groups the QoL of patients improved over time. In the intervention group (IG)
the improvement of the QoL as shown by the overall and functional scales was highly
significant (p = 0.002 and p = 0.005, resp.), whereas in the waiting group (WG) the
improvement was much smaller and reached significance only with a decrease of
symptoms (p = 0.01). Women in the IG had a higher overall QoL and functional status
after 5 weeks of yoga than women in the WG at the same time (5 weeks after the
operation) before the onset of yoga. These results show the advantage of starting
yoga classes at an early stage after surgery. The baseline of symptoms was higher
in
the IG compared with the WG. In the IG, symptoms decreased with yoga. In the WG,
symptoms increased during the intervention. This result was most likely due to
ongoing adjuvant treatment in the WG. Compared to the results reported by Eyigor et
al. [35], the EORTC scores in our sample were lower, but
improvements over time and after the intervention were also detectable.
Women in the IG reported significantly more often that they intended to continue with
yoga classes and that their physical activity had improved, thus indicating that
their overall motivation was higher. We assumed that women need more support
immediately after surgery. If additional help is offered at this time, women are
more open to continuing with yoga after the intervention. Based on this, it appears
helpful to offer yoga classes at an early stage of disease.
The type of surgery performed and patient age also influenced the QoL. Patients who
underwent mastectomy had a significantly lower functional status. Older patients
(> 60 years) and patients with SLN scored significantly lower on the symptom
scales than those with ALND. Shortly after surgery, the impact of the type of
axillary operation on the other scales was not (yet) measurable; we would need a
follow-up investigation of the QoL to clarify this point. We found that
breast-conserving therapy (BCT) led to less impairment of functional status, with
modern operating techniques probably contributing to this result.
The increased burden of symptoms in younger patients has implications for their care.
Like Kwan et al. [36], we assume that early measurement
of the QoL during the course of disease measures the distress caused by the
diagnosis of cancer and the subsequent operation as a very invasive procedure
impacting on a womanʼs body image and self-confidence. The distress is higher in
younger women, so they need special attention and supportive strategies. Our finding
is also supported by the data of Høyer et al. [37].
The limitations of the study also have to be mentioned. The patientsʼ acceptance of
yoga was very high. At the same time, the study was underpowered and the drop-out
rate was high. Sample attrition that occurs between the initial randomization and
the outcome measurement is a major problem for the validity of randomized controlled
trials. It can result in an incomparability between the intervention group and the
controls. For this reason we are aware that the statistical significance of the
results could be limited. For some patients, the beginning of adjuvant treatment
interfered with the study and made them miss some of the yoga sessions. This was
also the main reason for sample attrition. We performed a missing data analysis to
be certain that both groups were still comparable.
Another important limiting factor is that, with the given study design, we cannot
prove that the changes in QoL scales are not simply due to the different times which
elapsed since surgery for the IG and the WG. Moreover, funding and personal
resources were limited. Nevertheless, the results are promising for the care of
breast cancer patients.
Conclusion
-
A good QoL indicates the patientʼs ability to cope with the burden of cancer
and somatic therapy.
-
To support a patientʼs coping strategies and capacity to heal, encouraging
patient self-management is paramount. In our study we assessed the impact of
yoga for this.
-
Early initiation of yoga classes contributes to improved physical activity
and leads to a high motivation to continue training.
Acknowledgements
This study was made possible through funding from the DRK-Kliniken Berlin, the Alice
Salomon University of Applied Sciences and the AOK (Allgemeine Ortskrankenkasse).
We
are indebted to Dr. Katrin Schüttpelz-Brauns for her support in processing the
data.
