+Model

EUJIM-392; No. of Pages 9 ARTICLE IN PRESS

Available online at www.sciencedirect.com

ScienceDirect
European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9

Original article

Post-operative exercises after breast cancer surgery: Results of a RCT

evaluating standard care versus standard care plus additional yoga exercise
Helena Harder a,∗ , Carolyn Langridge a , Ivonne Solis-Trapala a , Charles Zammit b , Mokshini Grant c ,
Diane Rees d , Lynn Burkinshaw e , Valerie Jenkins a
a Sussex Health Outcomes Research & Education in Cancer (SHORE-C), Brighton and Sussex Medical School, University of Sussex, England, UK
b Park Centre for Breast Care, Brighton & Sussex University Hospitals NHS Trust, Brighton, England, UK
c Triratna Buddhist Community, Brighton, England, UK
d Nuffield Health Ltd., Brighton, England, UK
e Brighton & Sussex University Hospitals NHS Trust, Brighton, England, UK

Received 24 October 2014; received in revised form 27 January 2015; accepted 10 February 2015

Abstract
Introduction: There is a lack of standardisation in the guidelines for post-operative exercises following breast cancer surgery. Adherence to exercise
programmes is low, and complementary therapies such as yoga often appeal to patients and may encourage practise. A step-by-step guide to yoga
DVD was evaluated in addition to the standard care exercises (SC) compared to SC alone.
Methods: Women with early-stage breast cancer were randomised to SC plus or minus a yoga DVD for 10-weeks. Patient-reported outcomes were
collected at baseline, 10 weeks and 6 months. The primary study-endpoint was the Trial Outcome Index (TOI) of the Functional Assessment of
Cancer Treatment-Breast; a recognised quality of life (QoL) tool with an arm morbidity subscale (FACT-B+4).
Results: 92/103 (89%) women were randomised to the study. The SC group reported practising post-operative exercises more often than the yoga
DVD group. There was a 69% improvement from baseline in FACT-B+4 TOI, which included an arm subscale, at 10 weeks and 6 months in the SC
group. This was 62% and 81% respectively for the yoga DVD group. Numbness in the affected arm was greater in the SC group (OR = 2.5, 95%
CI: 1.1, 5.6) and in patients receiving chemotherapy (OR = 2.17, 95% CI: 1, 4.6). Despite no group differences, 74% of women would definitely
recommend following the yoga DVD after surgery.
Conclusions: Practising post-operative exercises does improve arm and shoulder morbidity following breast cancer surgery. The addition of a
self-practise general yoga programme was well received and appeared to improve QoL at 6 months.
© 2015 Published by Elsevier GmbH.

Keywords: Breast cancer; Yoga; Patient reported outcomes; Arm morbidity; Randomised controlled trial

Introduction radiotherapy to the axilla. These treatments often damage lym-

phatic drainage from the arm and women may subsequently
Each year, more than 48,000 women in the UK are diagnosed develop shoulder and arm dysfunction including lymphoedema.
with breast cancer and the majority (80%) undergo surgical As well as discomfort, there can be restricted movement, pain,
treatment [1], which also involves examination of lymph glands numbness and other sensory problems [2–6]. These side effects
under the arm (axilla). Some women require extensive treat- interfere with daily activities, impair quality of life (QoL), can
ment to the glands in the axilla if they contain cancer cells; this be distressing, and are often irreversible [7,8]. Symptoms are
can take the form of further surgery to remove the glands or costly to the health services in terms of rehabilitative treatments
and lymphoedema clinics.
Management of the axilla and post-operative rehabilitation
∗ Corresponding author. Tel.: +44 1273 873029; fax: +44 1273 873022.
(e.g. upper extremity exercises) are key components in current
E-mail address: H.Harder@sussex.ac.uk (H. Harder). standard post-operative care. Unfortunately, one disadvantage

http://dx.doi.org/10.1016/j.eujim.2015.02.002
1876-3820/© 2015 Published by Elsevier GmbH.

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
xxx.e2 H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9

of early discharge from hospital following breast surgery is that in a 10-week course of general yoga with BBCSG members as
patients in the UK are less likely to be taught specific post- part of a feasibility study. The DVD has 2 parts:
operative arm exercises by a physiotherapist. There is little
consensus in the instructions for arm and shoulder mobilisa- • Disc 1 comprises an introduction from a specialist in breast
tion provided to women [9], and many are only given a leaflet cancer QoL explaining why women have surgery to the arm
or an information pack with instructions to follow at home (e.g. pit as well as the breast. A short introduction to yoga from a
Breast Cancer Care) [10]. Also, there are few specific train- certified yoga teacher, plus a section with her demonstrating
ing programmes available, and prevention of arm and shoulder 16 poses in a graduated way from basic to advanced level
impairments depends on self-care and self-management. practice (levels 1–3).
Yoga as a stress reduction intervention or complementary • Disc 2 features a 1-h yoga class (based on Iyengar Yoga©
treatment is increasingly popular among breast cancer patients and restorative yoga) led by the yoga teacher with BBCSG
and also offered at some cancer centres. Reviews and meta- members of different ages and physical abilities.
analyses evaluating the effect of yoga in randomised controlled
trials (RCT) in women with breast cancer conclude there is mod- Participants were shown how to use the DVD and follow
erate to good evidence that yoga may be a useful practice during the poses (at level 1) by the yoga teacher prior to participation.
recovery from treatment [11–13]. Despite an emerging number They were asked to use the DVD at least once per week for 10
of large RCTs measuring the effectiveness of yoga on fatigue weeks at level 1 and were given yoga materials to use during the
[14,15], sleep quality [16], QoL [17], and during radiotherapy intervention period.
[18], there is scope for more investigation into its physical ben-
efits, in particular its role following breast cancer surgery. Participants and recruitment method
Preliminary results of yoga interventions for arm and shoul-
der morbidity demonstrate improvements in physical function Women aged 18–80 years of age with early-stage breast can-
and symptom relief [19], in shoulder abduction and flexion [20], cer (stages I–III), scheduled for axillary surgery and fluent in
and decreased arm volume in women with breast cancer related English were eligible for the study. They were given study infor-
lymphoedema [21]. However, these studies were observational mation by the clinical team during their pre-operative assessment
and have small sample sizes (6–18 participants). Here we report appointment. Women who showed an interest in the study (by
the results of a RCT that examines the use of a specially devel- returning an expression of interest form) were contacted by the
oped self-practise yoga DVD on QoL and arm and shoulder researchers to provide further information. A home-visit was
morbidity in women who had breast cancer surgery. arranged to obtain informed consent, demographics, level of pre-
vious yoga experience, and details of the hospital post-operative
Methods exercises [SC]. After this visit participants were randomised and
informed about group-allocation.
Study design
Yoga teachers and physiotherapists
A RCT of the standard care post-operative exercises (SC)
alone versus SC plus a 10-week self-practise general yoga Two trained yoga teachers and 2 registered physiotherapists
programme (i.e. yoga DVD). Primary and secondary outcome were involved in the study. Prior to the start of the study they
measures were collected at baseline (after surgery), 10 weeks met with the investigators to devise a protocol of yoga practice
(at completion of the intervention) and at 6 months. The study (based on the pilot study and contents of the DVD), and standard
received local ethical approval (REC: H10/1111/57) and written operating procedures for method of data collection and measure-
informed consent was provided by participants. ments by the physiotherapists. Regular research meetings (2–3
times a year) were held to assess practice.
Standard care
Randomisation and blinding
SC comprised of post-operative exercise materials distributed
by the hospital prior to surgery. This usually is written instruc- Randomisation was conducted using a computer-generated
tions for arm and shoulder mobilisation or an exercise leaflet, programme for producing variable sized, balanced permuted
poster or DVD (e.g. Breast Cancer Care publications [10]). blocks. Women were stratified by age (<50, 50–69, 70+) and
Women allocated to SC were offered the yoga-DVD after the previous experience of yoga (yes, no) and randomised in a
last follow-up assessment (i.e. at 6 months). ratio of 1:1. Randomisation was undertaken by an indepen-
dent researcher (IST). Participants were asked not to reveal their
The yoga DVD group allocation to the physiotherapists.

The self-practise yoga DVD was developed with members Outcome measures
of the Brighton Breast Cancer Support Group (BBCSG), breast
surgeons, specialist nurses, physiotherapists and yoga practi- The primary study endpoint was the efficacy of the inter-
tioners [22]. The DVD incorporated 16 postures that were used vention to reduce self-reported arm and shoulder morbidity

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9 xxx.e3

post-surgery and whether the effect was maintained over a period total volume of the arm was determined by the sum of the
of 6 months. This was assessed using the Trial Outcome Index segment volumes using the following formula for segmental
(TOI) score of the Functional Assessment of Cancer Therapy- volume (V): V = 3C2 /␲ = 3(C12 + C22 + C32 )/␲, where C is the
Breast+4 (FACT-B+4) [23–25]. The FACT-B+4 is a 41-item measured circumference [30]. All physiotherapy measurements
questionnaire with four primary subscales: physical well-being were performed on the operated and non-operated sides for
(PWB, 7 items), social well-being (SWB, 7 items), emotional comparison. Each physiotherapist measured the same patient at
well-being (EWB, 6 items), functional well-being (FWB, 7 both time-points. In addition, levels of pain or discomfort with
items), along with breast cancer additional concerns plus the sum usual activities were assessed by asking participants to rate their
of four questions relating to upper limb swelling and function pain or discomfort intensity on a 10-point scale (with ‘0’ rep-
(arm-specific subscale) (14 items). Responses to each item use resenting no pain and ‘10’ worst possible pain). Relevant data
a 5-point scale ranging from 0 (not at all) to 4 (very much). High concerning medical and surgical history and levels of exercise
FACT-B+4 scores indicate high-level functioning or better QoL. (formal/informal) were also collected by the physiotherapists
The TOI is an efficient summary index of physical/functional during the home visit.
outcomes, and a common endpoint used in clinical trials because
it is responsive to change in physical and functional outcomes. Data analyses
The TOI score is the sum of the scores of the 28 items included
in the PWB, FWB and breast cancer concerns subscales (range The aim of the statistical analysis was to assess changes in
0–112). A change of at least 5 points from baseline in TOI score TOI, FACT-B+4, QuickDASH and Oxford shoulder scores at
(calculated for each individual participant) is considered to be a 10 weeks and 6 months, contrasting between the yoga DVD and
clinically relevant minimally important difference and indicative SC groups. An estimated sample size of 47 participants per arm
of arm or shoulder morbidity [26,27]. provided 80% power with significance level set at 5% to detect a
Secondary endpoints were changes in patient reported out- clinically significant difference of 5 units in mean change scores
come measures (PROMs) for global QoL (FACT-B+4 total of the FACT-B+4 TOI between the yoga DVD and SC groups
score), and self-reported pain or disability in upper limb func- [27].
tion, and objective changes in shoulder mobility, hand grip Changes in scores were assessed using random effects regres-
strength and arm circumference. sion models which extend standard regression analyses to
The Oxford Shoulder Score (OSS), a 12 item questionnaire account for the correlation amongst responses for each indi-
was used to assess the degree of pain (4 items) and disabil- vidual to yield valid inferences on the size of the regression
ity/impairment to activities of daily living (ADL) (8 items) coefficients.
caused by shoulder surgery [28]. Each item is scored 1 (no Changes in the single items of the FACT-B+4 arm subscale
pain/no problem with ADL) to 5 (unbearable pain/impossible were assessed using logistic regression models for the probabil-
to do ADL). The maximum score is 60; higher scores represent ity of reporting symptoms (i.e. the proportion of participants who
greater disability. reported ‘somewhat’, ‘quite a bit’ and ‘very much’ for the items)
The QuickDASH was used as a self-report questionnaire using a generalised estimating equations approach to account for
to address symptoms and physical function of the upper limb the correlation amongst repeated observations.
[29]. The questionnaire has 11 items with scores ranging from Standard linear regression models for the differences at 10
1 (no difficulty) to 5 (unable). At least 10 items must be com- weeks in the secondary outcomes were used. In all the analy-
pleted to calculate a total score, which is then transformed into a sis, difference in response by participants characteristics were
0–100 scale. Higher scores on the QuickDASH represent greater explored by adding age, adjuvant chemo (yes/no), previous
limitations. yoga experience (yes/no), mastectomy (yes/no) and axillary
All PRO measures were completed at home and returned by surgery (axillary lymph node dissection or axillary clearance)
pre-paid post. Home practise (use of the yoga DVD, and practise as explanatory variables in the regression models. All analyses
of the standard hospital exercises) was recorded in a diary which were conducted using the statistical software R [31].
was returned at the end of weeks 5 and 10. Those allocated the
yoga DVD rated its usefulness, user-friendliness and satisfac- Results
tion at the end of the intervention. At 6 months, all participants
completed a follow up questionnaire that assessed current lev- Accrual and participant characteristics
els of physical activity, further treatments since surgery, joint
problems and degree of shoulder and/or arm pain. Fig. 1 displays recruitment, randomisation, and participant
Post-operative shoulder mobility measurements (extension, flow by group. 145 eligible women were approached about the
flexion and abduction) were conducted by the physiothera- study between April 2011 and May 2013. A total of 103 women
pists using a goniometer. Hand grip strength was measured expressed interest in the study; 92 (89%) were randomised.
using a hand-held dynamometer; each measurement was taken Table 1 shows the characteristics of the study sample. Groups
3 times and the average was used for analysis. Evidence of lym- were balanced on demographic and disease-related character-
phoedema was gauged by arm circumference measurements istics, and there were no differences between groups on type
at proximal wrist, 12 cm from wrist and 12 cm from elbow, of axillary surgery, yoga experience, and levels of formal and
and from these measures arm volumes were calculated. The informal exercise.

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
xxx.e4 H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9

Eligible patients given study information (n=145)

No response (n=28)

Not interested (n=14)

Enrolment
Patients approached (n=103)
Excluded (n=11)
Not meeting inclusion criteria (n=3)
Declined participation (n=3)
Declined randomisation (n=2)
Not responding (n=2)
Relocation (n=1)
Randomised 1:1 (n=92)
Allocation

Yoga (n=46) SC (n=46)

Received allocated intervention (n=45) Received allocated intervention (n=45)
Withdrew (n=1) (anxious) Withdrew (n=1) (had fall/frailty)

T0 - baseline post-surgery (n=45) T0 - baseline post-surgery T0 (n=45)

T1- 10 weeks post-intervention (n=40) T1 - 0 weeks post-intervention (n=40)

Discontinued (n=4) (unwell n=3; Discontinued (n=5) (unwell n=4;
Follow up

preference hospital exercises n=1) working/no time n=1)

Lost to follow (n=1) (reason unknown)

T2 - 6 months follow-up (n=39) T2 - 6 months follow-up (n=39)

Discontinued (n=1) (unwell) Discontinued (n=1) (unwell)
Analysis

Analysed (n=39) Analysed (n=39)

Excluded from analysis (n=0) Excluded from analysis (n = 0)

Fig. 1. CONSORT flow diagram.

Protocol adherence and weekly practise (post-surgery) to 10 weeks and 81% from baseline to 6 months,
compared with 69% in the SC group at both time-points. The
Seventy-eight (85%) participants completed the study (39 mixed-effects model showed no significant between-group dif-
in each group). Most study withdrawals (12/14) were during ferences for mean change from baseline TOI score at 10 weeks
the intervention period, predominantly because of illness (9/14, or 6 months (see also Fig. 2a). The estimated differences in mean
including 5 who started chemotherapy). There were no differ- changes between SC plus yoga DVD and SC alone were −0.57
ences between women who completed the study and those with (95% confidence interval (CI): −7.2, 6) and 1.72 (95% CI: −8.5,
baseline data only. 5.1) at 10 weeks and 6 months respectively.
The practise diary was returned by 35 participants in the yoga Overall, there were clinically significant improvement in
DVD group and 38 participants in the SC group. The proportion symptoms, the estimated mean changes in TOI score were 11.3
of women reporting to practise the standard post-operative hos- (95% CI: 7.4, 15.1) and 14.9 (95% CI: 10.9, 18.9) at 10 weeks
pital based exercises ≥5 times per week at week 1–5 was 75% and 6 months respectively, P < 0.001. Significant effects over
in the SC group and 50% in the yoga DVD group (P = 0.055). the 2 time-points were observed for women who had adjuvant
During week 6–10 this was and 68% in the SC group and 32% chemotherapy. The estimated mean TOI score for these patients
in the yoga DVD group (P = 0.011). However, 59% of yoga par- (b = −8.9 95% CI: −13.9, −3.9; P < 0.001) was lower than the
ticipants also reported following the DVD at least 3 times per mean score for patients who did not have chemotherapy.
week in the first half of the study; during week 6–10 this was
42%. No adverse effects (exercise/yoga related) were reported. Intervention effects on secondary outcome measures

Primary outcome TOI There were no significant group differences for the
physiotherapist assessments (shoulder extension, flexion and
Table 2 provides an exploratory analysis of score changes. abduction, hand grip strength of the operated and non-operated
In the yoga DVD group, 62% improved from baseline side, ratio of arm volume). Self-reported pain/discomfort

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9 xxx.e5

Table 1 functioning. Total FACT-B+4 scores were significantly lower in

Demographic and clinical characteristics by group (n, %). patients who had adjuvant chemotherapy compared with those
Yoga (n = 46) Standard care who did not receive chemotherapy, b = −11.3 (95% CI: −17.5,
(n = 46) −5.1; P < 0.001).
Age Arm function improved significantly over time for both
Mean (SD) 54.6 (10.9) 55.8 (11.6) groups (Fig. 2b). There were no significant effects of mastec-
Range, years 33–77 31–77 tomy or previous yoga experience on the arm function items.
Previous yoga experiencea 25 (54.4) 27 (58.7) A significant effect of age showed that older patients were less
Surgery
likely to report stiffness and pain when moving the operated arm
Wide local excision 32 (69.6) 29 (63) compared to younger patients, the odds ratio of reporting symp-
Mastectomy (no reconstruction) 11 (23.9) 15 (32.6) toms comparing 2 patients with a difference of 1 year in age was
Mastectomy (with reconstruction) 3 (6.5) 2 (4.4) OR = 0.96 (95% CI: 0.93, 0.99), P = 0.015.
Axillary surgery Patients in the SC alone group (OR = 2.5, 95% CI: 1.1, 5.6;
Sentinel lymph node biopsy 36 (78.3) 35 (76.1) P = 0.033), and patients receiving chemotherapy reported more
Axillary lymp node clearance 10 (21.7) 11 (23.9) symptoms (OR = 2.17, 95% CI: 1, 4.6; P = 0.045), of numbness
Surgery on side of dominant hand in the operated arm. Patients who had axillary lymph node dis-
Yes 23 (50) 25 (54.4) section or axillary clearance were more likely to report greater
No 20 (43.5) 19 (41.3) limitations (i.e. higher scores on QuickDASH, b = 7, 95% CI:
Unknown 3 (6.5) 2 (4.4)
0.4, 13.7; P = 0.039), and stiffness in the operated arm (OR = 4.4,
Received post-operative exercises 43 (93.5) 39 (84.8) 95% CI: 2, 9.8; P < 0.001).
Medical history
Arthritis 7 (15.2) 9 (19.6) Yoga programme evaluation
Trauma to shoulder/armb 7 (15.2) 10 (21.7)
Joint replacements 0 2 (4.4)
No medical history 32 (69.6) 25 (54.3)
Most women in the yoga DVD group that completed the study
(78%; 25/32) followed the yoga class on the DVD and found it
Adjuvant treatmentc
useful during their post-operative recovery. The majority (81%)
Chemotherapy 24 (54.5) 19 (43.2)
Radiotherapy 29 (67.4)d 29 (65.9)e watched the introduction and 69% rated it very useful, includ-
No chemo/radiotherapy 6 (13.0) 7 (15.2) ing the explanation about breast cancer and lymph glands. Some
Hormone therapy at 6 months 17 (53.1) 17 (51.5) women commented on the physical and emotional benefits of the
Unknown 2 (4.3) 2 (4.3) programme and cited the gentle exercises and relaxation as pos-
Previous level formal exercise itive elements of the study. One woman said: ‘It certainly taught
Daily 7 (6.5) 4 (8.7) me how to relax and breathe whilst doing exercises’. Negative
Weekly 13 (28.3) 11 (23.9) experiences were related to technical problems with the DVD,
Rarely 3 (15.2) 4 (8.7)
None 21 (45.7) 24 (52.2)
the demonstration of the postures (i.e. having to select the levels),
Unknown 2 (4.3) 3 (6.5) and preference for group-based yoga. One participant reported:
‘It would be much more useful to undertake the yoga in a class
Previous level informal exercise
Daily 22 (47.8) 19 (41.3) with other women, using the DVD was a bit tiresome’. However,
Weekly 6 (13.0) 10 (21.7) 74% (23/31) stated that they would definitely recommend the
Rarely 4 (8.7) 5 (10.9) DVD to other women with breast cancer.
None 4 (8.7) 4 (8.7)
Unknown 10 (21.7) 8 (17.4)
Discussion
a Yoga experience varied from 4 months to 10 years.
b Trauma to shoulder/arm: e.g. fractures, sports injuries, dislocation, carpal
This study is the first RCT in which yoga was used as an
tunnel syndrome.
c Sums add up to more than 100% or sample size because multiple treatment intervention for arm and shoulder morbidity in women with
options are possible. breast cancer. Previous observational studies in women fol-
d 15/29 received also chemotherapy. lowing breast cancer surgery with or without lymphoedema
e 11/29 received also chemotherapy.
demonstrated some preliminary effects in favour of the yoga
intervention, but studies were not representative due to small
sample sizes [19–21,32,33].
intensity scores decreased for both groups at 10 weeks how- The results from our RCT showed no significant differences
ever, at 6 months the mean pain score was higher in the SC in favour of the yoga DVD group in arm morbidity com-
group (2.8 versus 1.5; P = 015). pared with standard post-operative exercises. Regular practice
No significant group differences at any time-point were of either standard care alone or in combination with yoga
found for FACT-B+4, QuickDASH and OSS (Table 3). Com- improved arm and shoulder function and QoL significantly at
pared to baseline, the average total scores on the FACT-B+4 10 weeks post-surgery, and this effect was maintained at 6
increased significantly and decreased on the QuickDASH and months for both study groups. As reported in previous studies
OSS in both groups, indicating that patients reported improved [34–37], improvement was greater in women who did not have

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
xxx.e6 H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9

Table 2
Changes in TOI scores over time.
Assessment interval

Baseline to week 10a Baseline to month 6a Week 10 to month 6a

Yoga DVD group n = 34 n = 32 n = 31

Improvement 21 (62) 26 (81) 9 (29)
Deterioration 5 (15) 2 (6) 3 (10)
No change 8 (24) 4 (12) 19 (61)
SC group n = 36 n = 32 n = 33
Improvement 25 (69) 22 (69) 15 (45)
Deterioration 5 (14) 5 (16) 9 (27)
No change 6 (17) 5 (16) 9 (27)
a In numbers and proportions.

axillary clearance or chemotherapy. In addition, older women noted in the ALMANAC study; a prospective RCT of sentinel
(>65yrs) were less likely to complain about stiffness and pain node biopsy (SNB) compared with standard axillary treatment.
in the operated arm and also reported higher QoL than younger Researchers noted for the first 6 months post-surgery younger
women (<50yrs). The finding of better QoL in older women was patients reported less favourable QOL scores, and suggested

Fig. 2. Mean changes in FACT-B+4 scores. (a) TOI by intervention and age group. (b) Arm functioning by intervention and age group

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9 xxx.e7

Table 3
Mean baseline and follow-up PROs scores by group (mean, SD).
Measures Yoga DVD group Standard care group

Post-surgery 10 weeks 6 months Post-surgery 10 weeks 6 months

Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD

TOI score 74.3 (16.1) 84.0 (21.1) 88.7 (19.7) 72.5 (13.6) 83.5 (18.0) 85.6 (17.1)
PWB 19.7 (6.0) 20.7 (7.6) 23.6 (5.6) 20.2 (4.0) 21.9 (5.5) 23.2 (5.3)
FWB 18.7 (5.0) 20.7 (6.3) 21.6 (6.1) 17.4 (5.3) 20.0 (6.7) 21.9 (5.5)
EWB 18.3 (3.6) 19.0 (4.4) 18.9 (4.7) 17.5 (4.5) 18.8 (4.7) 19.2 (4.0)
SWB 25.3 (3.5) 25.3 (3.0) 24.1 (5.3) 23.1 (6.8) 24.1 (4.7) 23.3 (5.4)
Arm function (5 item) 12.2 (4.7) 17.5 (3.7) 17.5 (3.1) 12.0 (4.1) 16.7 (4.2) 15.4 (4.3)
QuickDash 41.2 (20.4) 10.8 (15.8) 9.9 (17.2) 43.2 (18.3) 15.2 (19.1) 15.4 (16.3)
OSS 25.7 (9.1) 16.1 (6.8) 15.0 (6.1) 27.0 (8.6) 17.0 (7.7) 17.7 (7.3)
Pain score 3.0 (2.6) 1.0 (2.0) 1.5 (1.7) 2.9 (2.0) 1.4 (2.1) 2.8 (2.5)

Abbreviations: EWB, emotional well-being; FWB, functional well-being; OSS, Oxford Shoulder Score; PRO, patient reported outcome measures; PWB, physical
well-being; SD, standard deviation; SWB, social well-being; TOI, trial outcome index. Range scores: FACT-B+4 0–164, PWB 0–28, FWB 0–28, EWB 0–24, SWB
0–28, TOI 0–112, BCS 0–36, 5-item arm morbidity (items B3, B10–B13) 0–20: higher scores reflect better functioning; QuickDash 0–100 and OSS 0–60: higher
scores represent poorer functioning or greater limitations; Pain scores 0–10.

that younger women may have different expectations from their cancer showed that active exercise is found to be more effec-
surgery and be more sensitive to the discomfort resulting from tive compared to no physical therapy following breast cancer
axillary surgery [27]. treatment or compared to information on the treatment of impair-
The current study is one of the few to include an objective ments of upper limb [38]. Differences in outcome may therefore
assessment of arm and shoulder function by physiotherapists, as be attributed to exercise frequency rather than specifically to the
most RCTs use only self-reported outcome measures [11]. There yoga component.
were no objective differences noted between groups. Being In addition to this, many women were already active and
assessed twice by physiotherapists however, may have influ- motivated to exercise and therefore may not be representative of
enced women in the SC group to perform their hospital based the general public. Finally, the design of the study did not allow
exercises regularly and more often than women in the yoga DVD us to measure the effect of yoga practise alone because it was
group, as they had no additional exercise to follow. Importantly combined with the routine post-operative exercises.
the results from the objective assessments showed that there
were no adverse effects during study participation, suggesting Conclusions
that it is safe for women with breast cancer to perform gentle
yoga postures post-operatively in addition to standard care. Regular practise of both standard hospital exercises plus
Physical function and QoL of women in both study groups or minus general yoga appears to improve arm and shoulder
improved significantly over time, but our study failed to demon- function and general well-being in patients who had breast can-
strate that this was related to self-guided additional yoga cer surgery. The results of this study do suggest that gentle
practise. A possible explanation for an improvement in phys- yoga exercises following breast cancer surgery are acceptable
ical functioning is that many of the women were keen exercisers to women and can be safely used during the early post-operative
prior to surgery which may reflect in exercise compliance. In recovery period, and it can therefore be considered as a potential
addition, women were not barred from accessing yoga, Pilates therapeutic intervention for breast cancer patients.
or other exercise classes during their time in the study, which An interesting angle for future research is to evaluate the use
may have positively influenced performance in both groups. of behavioural change techniques such as prompts to encour-
Our RCT had several limitations. Unfortunately the study age women to do their post-operative exercises or any other
was underpowered with 92 women randomised and data avail- exercise-based intervention [39,40]. This could take the form
able only for 82 women at all time-points. The lack of a of either telephone calls from a physiotherapist or allied health
pre-surgery baseline assessment of arm and shoulder function care worker, or text alert on a mobile or smartphone.
was restrictive; potential group differences in function before
randomisation may have been overlooked because of this. Also,
the post-operative hospital based exercises were less time con- Disclaimer
suming than familiarising oneself to the poses in the yoga DVD,
which may have been an incentive for women in the SC group. All research was conducted by the authors.
It is also possible that the intervention group performed more
exercise (both standard care and yoga) than the SC group. A Conflicts of interest
recently published review of the effectiveness of postoperative
physical therapy for arm and shoulder morbidity after breast None declared.

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
xxx.e8 H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9

Financial support [16] Mustian KM, Sprod LK, Janelsins M, Peppone LJ, Palesh OG, Chandwani
K, et al. Multicenter, randomized controlled trial of yoga for sleep quality
among cancer survivors. J Clin Oncol 2013;31:3233–41.
None.
[17] Siedentopf F, Utz-Billing I, Gairing S, Schoenegg W, Kentenich H, Kollak
I. Yoga for patients with early breast cancer and its impact on quality of life
– a randomized controlled trial. Geburtshilfe Frauenheilk 2013;73:311–7.
Acknowledgements [18] Chandwani KD, Perkins G, Nagendra HR, Raghuram NV, Spelman A,
Nagarathna R, et al. Randomized, controlled trial of yoga in women with
The women who supported the study, the clinicians and spe- breast cancer undergoing radiotherapy. J Clin Oncol 2014;32:1058–65.
[19] Thomas R, Shaw RM. Yoga for women living with breast cancer-related
cialist nurses at the Park Centre for Breast Care, including, arm morbidity: findings from an exploratory study. Int J Yoga Ther
Lisa Thomas Brooks, Caroline Foster Sarah Newbury & Teresa 2011;21:39–48.
Hoskins. Also, Jo Tytherleigh, Lisa Christie, Professor Fallow- [20] Sudarshan M, Petrucci A, Dumitra S, Duplisea J, Wexler S, Meterissian
field & SHORE-C staff who supported the study and the women S. Yoga therapy for breast cancer patients: a prospective cohort study.
who participated. Complement Ther Clin Pract 2013;19:227–9.
[21] Fisher MI, Donahoe-Fillmore B, Leach L, O’Malley C, Paeplow C, Prescott
T, et al. Effects of yoga on arm volume among women with breast
cancer related lymphedema: a pilot study. J Bodyw Mov Ther 2014,
References http://dx.doi.org/10.1016/j.jbmt.2014.02.006.
[22] Jenkins V, Zammitt C, Grant M, Rees D, Paine T, Fallowfield L.
[1] Cancer Research UK. http://www. cancerresearchuk org/cancer-info/ Development of a structured yoga DVD for women following breast
cancerstats/types/breast/2013 and axillary surgery. EJC Suppl 2010;8:80, http://dx.doi.org/10.1016/
[2] Hayes S, Johansson K, Stout N, Prosnitz R, Armer J, Gabram S, et al. Upper- S1359-6349(10)70105-1.
body morbidity after breast cancer: incidence and evidence for evaluation, [23] Coster S, Poole K, Fallowfield LJ. The validation of a quality of life
prevention, and management within a prospective surveillance model of scale to assess the impact of arm morbidity in breast cancer patients post-
care. Cancer 2012;118:2237–49. operatively. Breast Cancer Res Treat 2001;68:273–82.
[3] Smoot B, Wong J, Cooper B, Wanek L, Topp K, Byl N, et al. Upper extrem- [24] Cella D, Tulsky D, Gray G, Sarafian B, Linn E, Bonomi A, et al. The Func-
ity impairments in women with or without lymphedema following breast tional Assessment of Cancer Therapy scale: development and validation of
cancer treatment. J Cancer Surviv 2010;4:167–78. the general measure. J Clin Oncol 1993;11:570–9.
[4] Fu MR, Ridner SH, Hu SH, Stewart BR, Cormier JN, Armer JM. Psychoso- [25] Brady MJ, Cella DF, Mo F, Bonomi A, Tulsky DS, Lloyd SR, et al. Reli-
cial impact of lymphedema: a systematic review of literature from 2004 to ability and validity of the Functional Assessment of Cancer Therapy-Breast
2011. Psycho-Oncology 2013;22:1466–84. quality-of-life instrument. J Clin Oncol 1997;15:974–86.
[5] Dominick SA, Natarajan L, Pierce JP, Madanat H, Madlensky [26] Eton DT, Cella D, Yost KJ, Yount SE, Peterman AH, Neuberg DS, et al. A
L. The psychosocial impact of lymphedema-related distress among combination of distribution- and anchor-based approaches determined min-
breast cancer survivors in the WHEL Study. Psycho-Oncology 2014, imally important differences (MIDs) for four endpoints in a breast cancer
http://dx.doi.org/10.1002/pon.3510. scale. J Clin Epidemiol 2004;57:898–910.
[6] Johansen S, Fosså K, Nesvold IL, Malinen E, Fosså SD. Arm and shoulder [27] Fleissig A, Fallowfield LJ, Langridge CI, Johnson L, Newcombe RG, Dixon
morbidity following surgery and radiotherapy for breast cancer. Acta Oncol JM, et al. Post-operative arm morbidity and quality of life. Results of
2014;53:521–9. the ALMANAC randomised trial comparing sentinel node biopsy with
[7] Hayes S, Rye S, Battistutta D, DiSipio T, Newman B. Upper body morbidity standard axillary treatment in the management of patients with early breast
following breast cancer treatment is common, may persist longer-term and cancer. Breast Cancer Res Treat 2006;95:279–93.
adversely influences quality of life. Health Qual Life Outcomes 2010;8:92, [28] Olley LM, Carr AJ. The use of a Patient-Based Questionnaire (The Oxford
http://dx.doi.org/10.1186/1477-7525-8-92. Shoulder Score) to assess outcome after rotator cuff repair. Ann R Coll
[8] Nesvold IL, Fosså SD, Holm I, Naume B, Dahl AA. Arm/shoulder problems Surg Engl 1993;90:326–31.
in breast cancer survivors are associated with reduced health and poorer [29] Beaton DE, Wright JG, Katz JN. Development of the QuickDASH:
physical quality of life. Acta Oncol 2010;49:347–53. comparison of three item-reduction approaches. J Bone Joint Surg Am
[9] Todd J, Topping A. A survey of written information on the use 2005;87:1038–46.
of post-operative exercises after breast cancer surgery. Physiotherapy [30] Mansel RE, Fallowfield L, Kissin M, Goyal A, Newcombe RG, Dixon JM,
2005;91:87–93. et al. Randomized multicenter trial of sentinel node biopsy versus standard
[10] Breast Cancer Care. Exercise after breast cancer surgery; 2013 http:// axillary treatment in operable breast cancer: the ALMANAC trial. J Natl
www2.breastcancercare.org.uk/sites/default/files/bcc6-exercises-after- Cancer Inst 2006;98:599–609.
breast-cancer-surgery.pdf [31] R. Development Core Team. R: a language and environment for statistical
[11] Harder H, Parlour L, Jenkins V. Randomised controlled trials of yoga computing. Vienna, Austria: R Foundation for Statistical Computing; 2013
interventions for women with breast cancer: a systematic literature review. http://www.R-project.org
Support Care Cancer 2012;20:3055–64. [32] Douglass J, Immink M, Piller N, Ullah S. Yoga for women with
[12] Zhang J, Yang KH, Tian JH, Wang CM. Effects of yoga on psy- breast cancer-related lymphoedema: a preliminary 6-month study. J Lym-
chologic function and quality of life in women with breast cancer: a phoedema 2012;7:30–8.
meta-analysis of randomized controlled trials. J Altern Complement Med [33] Loudon A, Barnett T, Piller N, Immink MA, Williams AD. Yoga
2012;18:994–1002. management of breast cancer-related lymphoedema: a randomised
[13] Cramer H, Lange S, Klose P, Paul A, Dobos G. Yoga for breast cancer controlled pilot-trial. BMC Complement Altern Med 2014;14:214,
patients and survivors: a systematic review and meta-analysis. BMC Cancer http://dx.doi.org/10.1186/1472-6882-14-214.
2012;12:412, http://dx.doi.org/10.1186/1471-2407-12-412. [34] Peintinger F, Reitsamer R, Stranzl H, Ralph G. Comparison of quality of
[14] Bower JE, Garet D, Sternlieb B, Ganz PA, Irwin MR, Olmstead R, et al. life and arm complaints after axillary lymph node dissection vs sentinel
Yoga for persistent fatigue in breast cancer survivors: a randomized con- lymph node biopsy in breast cancer patients. Br J Cancer 2003;89:648–52.
trolled trial. Cancer 2012;118:3766–75. [35] Dabakuyo T, Fraisse J, Causeret S, Gouy S, Padeano M, Loustalot C, et al. A
[15] Kiecolt-Glaser JK, Bennet JM, Andridge R, Peng J, Shapiro CL, Malarkey multicenter cohort study to compare quality of life in breast cancer patients
WB, et al. Yoga’s impact on inflammation, mood, and fatigue in breast can- according to sentinel lymph node biopsy or axillary lymph node dissection.
cer survivors: a randomized controlled trial. J Clin Oncol 2014;32:1040–9. Ann Oncol 2009;20(8):352–1361.

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002
+Model
EUJIM-392; No. of Pages 9 ARTICLE IN PRESS
H. Harder et al. / European Journal of Integrative Medicine xxx (2015) xxx.e1–xxx.e9 xxx.e9

[36] Belmonte R, Garin O, Segura M, Pont A, Escalada Ferrer M. Quality- impairments following breast cancer treatment: a systematic review. Arch
of-life impact of sentinel lymph node biopsy versus axillary lymph node Phys Med Rehab 2015, http://dx.doi.org/10.1016/j.apmr.2015.01.006.
dissection in breast cancer patients. Value Health 2012;15:907–15. [39] Cole-Lewis H, Kersha T. Text messaging as a tool for behavior change in
[37] Hwang S, Chang S, Park B. Does chemotherapy really affect the quality of disease prevention and management. Epidemiol Rev 2010;32:56–69.
life of women with breast cancer? J Breast Cancer 2013;16(2):229–35. [40] Fry JP, Neff RA. Periodic prompts and reminders in health promotion
[38] De Groef A, Van Kampen, Dieltjens, Christiaens M, Neven P, Geraerts and health behavior interventions: systematic review. J Med Internet Res
I, et al. Effectiveness of postoperative physical therapy for upper limb 2009;11:e16, http://dx.doi.org/10.2196/jmir.1138.

Please cite this article in press as: Harder H, et al. Post-operative exercises after breast cancer surgery: Results of a RCT evaluating standard
care versus standard care plus additional yoga exercise. Eur J Integr Med (2015), http://dx.doi.org/10.1016/j.eujim.2015.02.002