1 s2.0 S0885392416000555 Main

728 Journal of Pain and Symptom Management Vol. 51 No.
4 April 2016
Original Article
Effects of Acupuncture, Tuina, Tai Chi, Qigong, and Traditional

Chinese Medicine Five-Element Music Therapy on Symptom
Management and Quality of Life for Cancer Patients: A Meta-Analysis
Wei-Wei Tao, MS, Hua Jiang, MS, Xiao-Mei Tao, PhD, Ping Jiang, MS, Li-Yan Sha, MS, and Xian-Ce Sun, MD
College of Nursing (W.-W.T., L.-Y.S.); Graduate School (P.J.), and College of Public Health (X.-C.S.), Dalian Medical University, Dalian;
School of Nursing (H.J.), Peking University, Beijing; and Beijing Shijitan Hospital (X.-M.T.), Beijing, People’s Republic of China
Abstract
Context. Most cancer patients suffer from both the disease itself and symptoms induced by conventional treatment.
Available literature on the clinical effects on cancer patients of acupuncture, Tuina, Tai Chi, Qigong, and Traditional Chinese
Medicine Five-Element Music Therapy (TCM-FEMT) reports controversial results.
Objectives. The primary objective of this meta-analysis was to evaluate the effect of acupuncture, Tuina, Tai Chi, Qigong,
and TCM-FEMT on various symptoms and quality of life (QOL) in patients with cancer; risk of bias for the selected trials also
was assessed.
Methods. Studies were identified by searching electronic databases (MEDLINE via both PubMed and Ovid, Cochrane
Central, China National Knowledge Infrastructure, Chinese Scientific Journal Database, China Biology Medicine,
and Wanfang Database). All randomized controlled trials (RCTs) using acupuncture, Tuina, Tai Chi, Qigong, or
TCM-FEMT published before October 2, 2014, were selected, regardless of whether the article was published in Chinese
or English.
Results. We identified 67 RCTs (5465 patients) that met our inclusion criteria to perform this meta-analysis.
Analysis results showed that a significant combined effect was observed for QOL change in patients with terminal
cancer in favor of acupuncture and Tuina (Cohen’s d: 0.21e4.55, P < 0.05), whereas Tai Chi and Qigong had no
effect on QOL of breast cancer survivors (P > 0.05). The meta-analysis also demonstrated that acupuncture produced
small-to-large effects on adverse symptoms including pain, fatigue, sleep disturbance, and some gastrointestinal
discomfort; however, no significant effect was found on the frequency of hot flashes (Cohen’s d ¼ 0.02; 95%
CI ¼ 1.49 to 1.45; P ¼ 0.97; I2 ¼ 36%) and mood distress (P > 0.05). Tuina relieved gastrointestinal discomfort.
TCM-FEMT lowered depression level. Tai Chi improved vital capacity of breast cancer patients. High risk of bias was
present in 74.63% of the selected RCTs. Major sources of risk of bias were lack of blinding, allocation concealment,
and incomplete outcome data.
Conclusion. Taken together, although there are some clear limitations regarding the body of research reviewed in this
study, a tentative conclusion can be reached that acupuncture, Tuina, Tai Chi, Qigong, or TCM-FEMT represent beneficial
adjunctive therapies. Future study reporting in this field should be improved regarding both method and content of
interventions and research methods. J Pain Symptom Manage 2016;51:728e747. Ó 2016 American Academy of Hospice and
Palliative Medicine. Published by Elsevier Inc. All rights reserved.
Key Words
Acupuncture, massage, Tai Chi, Qigong, cancer, symptom management, meta-analysis
W.-W. T, H. J., X.-M. T., and L.-Y. S. contributed equally to E-mail: taoweiwei2003@163.com; or Ping Jiang, E-mail:
this work and should be regarded as co-first authors. dywx.2009@163.com
Address correspondence to: Wei-Wei Tao, MS, College of Accepted for publication: November 20, 2015.
Nursing, Dalian Medical University, 9 West Section,
Lvshun South Road, Dalian, People’s Republic of China.
Ó 2016 American Academy of Hospice and Palliative Medicine. 0885-3924/$ - see front matter
Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpainsymman.2015.11.027
Vol. 51 No. 4 April 2016 Interventions for Symptom Management and QOL 729
Introduction gentle movements and postures of the exercise are de-

Cancer is one of the most common causes of death signed to achieve a harmonious flow of energy (qi) in
worldwide. It accounted for 7.6 million deaths the body so as to improve physical fitness and overall
(around 13% of all deaths) in 2008, and deaths well-being. External qigong is usually performed by a
from cancer internationally are projected to rise to trained practitioner using their hands to direct
more than 11 million in 2030,1 according to the World emitted ‘‘qi’’ energy onto the patient’s body in the
Health Organization (WHO). Conventional treatment diagnosis and treatment of various diseases.12 Tai
for cancer, such as surgery, chemotherapy, radio- Chi includes relaxation, deep and regulated breathing
therapy, and endocrine therapy, aiming at curing the techniques, and slow movements, and it is an interven-
disease or prolonging life produces considerable tion that shares many characteristics with Qigong.13
beneficial medical outcomes. However, these treat- Music therapy has been defined as ‘‘the clinical and
ments cannot eradicate the diseases and are often evidence-based use of music interventions to accom-
accompanied by adverse effects. It is acknowledged plish individualized goals within a therapeutic re-
that most cancer patients suffer from both the disease lationship..’’14 In China, researchers usually apply
itself and symptoms induced by conventional treat- some traditional music based on the five elements of
ment, such as fatigue, anxiety, depression, nausea, TCM in their studies, which is called TCM-FEMT.15
and pain, and being at an increased risk of developing In the last decade, the use of these interventions has
secondary tumors.2e4 Numerous patients report increased dramatically. According to the WHO,
feeling abandoned and isolated after cessation of acupuncture is used in at least 78 countries.16
active treatment,5 which could exacerbate symptoms Although multiple studies applying these interven-
such as depression and anxiety. This may have an tions in cancer patients have reported positive
impact on their quality of life (QOL), which has outcomes,15,17e20 others have described mixed
been reported as being lower than other population evidence.19,21e25 Differences in study design, interven-
groups.6 Therefore, for cancer survivors, it is essential tions type, frequency and duration and the therapist
to look at ways of improving QOL and long-term in these interventions may produce varying results.
health outcomes. Therefore, a systematic review with a meta-analysis of
Nonpharmacologic interventions typically encom- the interventions is needed to more accurately gauge
pass a broad range of psychosocial, behavioral, and efficacy for cancer patients and evaluate the quality
environmental strategies that may complement con- of such studies.
ventional treatment to enhance QOL for cancer Although several systematic reviews with or without
patients.7 Traditional Chinese nonpharmacologic meta-analysis have addressed this issue,9,12,26e33 they
interventions include a series of approaches such as often have covered only a small section of evidence,
acupoint stimulation, Chinese massage (referred to restricting their scope to either a specific subpopula-
as ‘‘Tuina’’), Tai Chi, Qigong, and Traditional Chinese tion of patients with cancer or a specific type of inter-
Medicine Five-Element Music Therapy (TCM-FEMT). vention, or both, thus making broader conclusions
The use of these interventions for the management difficult. Moreover, many reviews are outdated. Several
of cancer can be traced back to the Shang Dynasty reviews also have reported the utilization of TCM
of China 3500 years ago.8 Over the centuries, various among cancer patients34e36; however, a large propor-
interventions such as acupuncture (involving the tion of the reviews discussed the use of herbal medi-
placement of solid, sterile, stainless steel needles cine, which was not involved in the interventions.
into specific points on the body), moxibustion (stimu- This study aims at closing this gap and producing an
lating skin thermally by the burning of moxa, also at overall picture of contemporary research in this field
precise locations or other specific areas),9 Chinese while exploring the hypothesis that acupuncture, Tui-
massage (a wide range of technical manipulations con- na, Tai Chi, Qigong, or TCM-FEMT may contribute to
ducted by a practitioner’s finger, hand, elbow, knee, or the QOL of cancer patients.
foot applied to muscle or soft tissue at specific body lo-
cations),10 and Qigong have been developed and used
in cancer treatment.11 Qigong, a general term for a Methods
large range of traditional Chinese energy exercises Eligibility Criteria
and therapies, is popularly practiced by a large num- Eligibility criteria are detailed following the
ber of people in Chinese communities. Basically, there Participants, Interventions, Controls, Outcomes, and
are two categories of Qigong: internal qigong vs. Studies (PICOS) framework37:
external qigong. Internal qigong or qigong exercise
is self-directed and involves the use of movements, Participants. Participants were adults aged 18 years or
meditation, and control of breathing patterns. The older who had been formally diagnosed with cancer of
730 Tao et al. Vol. 51 No. 4 April 2016
any type (solid and hematologic), any tumor stage, any disorder OR sleep disturbance OR insomnia OR dys-
kind of treatment mode, and any time since diagnosis. pnea OR shortness of breath OR peripheral neuropa-
thy OR hiccups OR hiccoughs).
Interventions. Approaches including acupoint stimu- Data extracted from the studies included year of
lation, Chinese massage (‘‘Tuina’’), Tai Chi, Qigong, publication, country of origin, number of participants
TCM-FEMT, alone or in combination, were reviewed. randomly assigned, mean age, sex, type of tumor,
Studies evaluating any types of herbal medicine, point treatment status, type of intervention, duration of
injection, Thai massage, Japanese massage, Swedish- intervention, control condition, and outcomes
style massage, reflexology, aromatherapy massage, (including assessment instruments).
and other music therapy were excluded. All articles were read by two independent reviewers
(W. -W. T. and X. -M. T.), and data from the articles
Controls. No treatment (usual care) and active (atten- were validated and extracted and entered into Excel.
tion placebo) control conditions were both consid- Discrepancies between reviewers were resolved by
ered. Nevertheless, studies without the use of a discussion.
control condition were excluded.
Assessment of Risk of Bias
Outcomes. Primary outcomes were health-related The risk of bias was assessed using the Cochrane
QOL at post-treatment. Among the studies reviewed, Handbook for Systematic Reviews of Interventions
overall QOL was measured by the Functional (RevMan version 5.1.0, The Cochrane Collaboration,
Assessment of Cancer TherapyeBreast or Functional 2011). Six components associated with the risk of
Assessment of Cancer TherapyeGeneral (FACT-B or bias were assessed: generation of the allocation
FACT-G), Karnofsky Performance Status (KPS) score, sequence, allocation concealment, masking of
European Organization for Research and Treatment outcome assessors, selective outcome reporting,
of Cancer Quality of Life QuestionnaireeCore 30, incomplete follow-up, and other potential sources of
and the Medical Outcomes Study Short Forme36 bias. Trials with a low risk for all six components
(SF-36). Secondary outcomes included cancer-related were defined as having an overall low risk of bias. Tri-
symptoms and therapy-related adverse events such as als in which one or more of the six bias components
pain, flushes, fatigue, sleep disturbance, hair loss, were unclear or had high risk of bias were defined
negative mood, diarrhea, flatulence, nausea, and to be at high risk of bias.38 Disagreements were
vomiting. resolved by discussions among the two reviewers and,
if necessary, through discussion among the authors.
Studies. Only randomized controlled trials (RCTs) However, there were no disagreements among the
were considered. All the articles included in this two reviewers.
meta-analysis are listed in the References.
Statistical Analysis
Information Sources The descriptive data for all the included studies
Studies were identified by 1) searching electronic were entered into Excel, and statistical analyses were
databases (MEDLINE via both PubMed and Ovid, completed using Review Manager 5.0, supplied gratis
Cochrane Central, China National Knowledge Infra- by the Cochrane organization (www.cochrane.org/
structure from 1911 to October 2014; Chinese Scienti- cochrane/hbook.htm) for meta-analyses. The effect
fic Journal Database from 1989 to October 2014; sizes were calculated as Cohen’s d standardized
China Biology Medicine from 1978 to October 2014; mean effects of the interventions. Individual study ef-
and Wanfang Database from 1994 to October 2014), fect sizes were synthesized to generate an overall effect
2) scanning reference lists of relevant review articles; size using a random or fixed effects model according
and 3) contacting study authors. to the heterogeneity level, weighted by the inverse of
variance. For continuous variables, mean difference
Data Collection was calculated when outcomes were measured using
The search terms in the English and Chinese litera- the same scale, and the standardized mean difference
ture were (cancer OR tumor OR tumour OR neoplas* was used when different scales were used in different
OR oncolog*), AND (acupoint OR acupuncture OR trials, with corresponding 95% CIs. If data for SDs
moxibustion OR TCM OR Chinese massage OR were missing, they were computed for the calculation
Qigong OR Tai Chi OR TCM Five Element Music of statistical pooling of effect size by calculating the
Therapy), AND (quality of life OR pain OR nausea trial data using standard error of the mean or 95%
OR hot flashes OR hot flushes OR fatigue OR xerosto- CIs.38 For dichotomous variables, the odds ratios
mia OR ileus OR anxiety OR depression OR mood were calculated when appropriate summary statistics
were reported.39 An odds ratio of more than 1 indi- stimulation, nine of Tai Chi, seven of Chinese mas-
cates an advantage for the interventions. We also sage, three of TCM-FEMT, two of Qigong, and one
completed a sensitivity analysis to identify potential of acupuncture combined with TCM-FEMT.
outliers by removing each study one by one to One-third of the studies (22/67) used an active con-
examine the individual influence of each study on trol condition, such as sham acupuncture (n ¼ 9), oral
the overall effect size. Heterogeneity was assessed or injectable drug (n ¼ 8), psychosocial therapy
using Cochrane’s Q and I2, which calculates the pro- (n ¼ 3), music exercise (n ¼ 1), or relaxation
portion of variation attributed to heterogeneity. Statis- (n ¼ 1; Table 1).
tical heterogeneity of trial results was assessed by visual Among the 67 RCTs included in the meta-analysis,
inspection of forest plots, c2 tests, and the I2 statistic.40 the top three outcome measures reported were pain,
A P-value greater than 0.10 for c2 tests and an I2 value hiccups, and fatigue, with nine studies each. Evalua-
of less than 25% were interpreted as signifying a low tion tools for overall QOL were varied, among which
level of heterogeneity.40 Primary analyses were per- KPS score was the most frequently reported (n ¼ 7),
formed with a fixed effects model; secondary confir- followed by the FACT-G (n ¼ 4), European Organiza-
matory analyses were performed with a random tion for Research and Treatment of Cancer Quality of
effects model if there was significant heterogeneity. Life QuestionnaireeCore 30 (n ¼ 3), and SF-36
An effect size of 0.8 was considered large, 0.5 was (n ¼ 2; Table 2).
considered medium, and 0.2 was considered small.41
A two-tailed P-value of less than 0.05 was considered Risk of Bias Analysis
to be significant. Each trial was evaluated in terms of its risk of bias.
Most (50/67) had high risk of bias, 12 of 67 with mod-
erate, and five with low risk of bias. Major sources of
Results risk of bias were lack of blinding, allocation conceal-
Study Selection ment, and incomplete outcome data. Of 67 trials,
The search of the English and Chinese literature only 11 studies (10 on acupoint stimulation and one
retrieved 11,166 citations, of which 10,540 records on Tai Chi) described the method of blinding and
were excluded on the basis of the abstract alone because methods of allocation concealment. Less than half of
the reports provided specified numbers and reasons
of nonrelevance or duplication. The full texts of the
for dropouts by each subject group. No more than
remaining 626 articles were retrieved for more detailed
evaluation, and 489 articles were excluded 1) for 52% provided information on the method of random-
including herbs or other medicine (n ¼ 149), 2) inter- ization used (Fig. 2).
ventions combined with other nontraditional methods
(n ¼ 79), 3) no RCT (n ¼ 151), 4) no relevant outcomes Effects of the Interventions on Clinical Outcomes for
(n ¼ 72), 5) insufficient information (n ¼ 26), and 6) Cancer Survivors
publication not available (n ¼ 12). An additional five Effects are described according to outcome mea-
articles were incorporated, gleaned from reference sures: overall QOL (20/67 of the comparisons),
lists, for a total of 142 studies included in this review. adverse symptoms including gastrointestinal distress
Of these, 67 trials (20 in English and 47 in Chinese) (21/67), pain (9/67), fatigue (9/67), depression (7/
covering 5465 patients provided enough data for 67), anxiety (4/67), sleep disturbance (4/67), flushes
statistical pooling and were used for quantitative (3/67). Other clinical outcomes involved are vital ca-
meta-analysis (Fig. 1, Table 1).15,17,19e22,24,42e70,71e100 pacity (3/67) and body mass index (BMI; 3/67). Seven-
teen studies evaluated two or more outcomes (Table 2).
Study Characteristics
The systematic search covered the period from 1980 General QOL. Eleven trials17,19,43,59,70,71,74,75,95e97
to 2013. More than 80% of the studies (51/67) were used generic scales (e.g., KPS, SF-36, World Health Or-
conducted in China, and the others were from the ganization Quality-of-Life Scale) to assess changes in
United States (11/67), Sweden (2/67), and Australia general QOL scores after acupuncture and Tai Chi.
(2/67). The trials comprise 5456 patients. Study par- Cancer-specific QOL was measured by the FACT-G,
ticipants had a median intervention duration of FACT-B, or EORTC QLQ-30 in nine
4.9 weeks (ranging from 24 hours to 24 weeks), with studies.15,20e22,24,48,51,57,92 Fig. 3 shows that a significant
four reports in Chinese not supplying the length of combined effect was observed for general QOL change
intervention. Thirty-seven percent of the trials in patients with terminal cancer in favor of acupunc-
reported mixed cancer, 27% included breast cancer ture; Chinese massage improved cancer-specific
diagnoses, followed by colorectal cancer (9%) and QOL, and Tai Chi and Qigong had no effect on QOL
liver cancer (7%). Forty-five trials were of acupoint of breast cancer survivors.
Medline-325; Cochrane Central-

457;Chinese Scientific Journal
Database-3237;WanFang-3947 ;CN
KI-1582; CBM-1618; Total
citations (n = 11,166)
Duplicates (n =6862)
Screening of titles and abstracts

(n =4304)
Excluded(n =3678)
Clearly not relevant(n =3678)
Assessment of full-text articles

(n =626)
Excluded(n =489)
Interventions including herbs or other
medicine (n =149)
Interventions combined with other
non-traditional methods (n =79)
Method (no RCT) (n =151)
No relevant outcomes (n =72)
Identified from searches of
Insufficient information (n =26) reference lists (n=5)
Publication not available (n =12)
Studies included in systematic

Excluded (n =75) review (n = 142 )
Not providing sufficient data for
further analysis (n =75)
Studies included in quantitative

meta-analysis (n =67 )
Fig. 1. PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) flow diagram of study selection.
CNKI ¼ China National Knowledge Infrastructure; CBM ¼ China Biology Medicine; RCT ¼ randomized controlled trial.
Sensitivity analyses yielded one study75 with a great and one on Qigong). Significant large post-
impact on both effects and heterogeneity for overall treatment effects were observed for acupuncture
QOL evaluated by KPS. Meta-analyses excluding this (Figs. 4b and 4c).
study showed smaller effect sizes and heterogeneity Acupuncture was found to relieve gastrointestinal
(d ¼ 2.79; 95% CI ¼ 1.12 to 6.69; P ¼ 0.16; distress including diarrhea,45,54 nausea, and vomit-
I2 ¼ 77%). ing.47,73 Acupuncture and Chinese massage also
relieved hiccups (eight studies on acupuncture, one
Adverse Symptoms. Nine studies50,64,67e69,80,81,84,85 on Chinese massage),55,62,76e78,89,98e100 and short-
including 710 patients evaluated the impact of inter- ened hours to first flatus (five on acupuncture, three
ventions on pain (eight on acupuncture and one on on Chinese massage)44,46,56,58,63,66,81,90 and hours to
Chinese massage). A small but significant effect was first bowel movement (three on acupuncture, three
found in our meta-analysis for acupuncture on Chinese massage)44,46,56,58,66,90 (Figs. 4de4j).
(d ¼ 0.30; 95% CI¼ 0.56 to 0.03; P ¼ 0.03; Four trials involving 646 patients evaluated interven-
I2 ¼ 59%; Fig. 4a). tions on sleep level using the Pittsburgh Sleep Quality
Association of interventions and fatigue was evalu- Index (two on acupuncture, one on Chinese massage,
ated in nine trials24,42,43,52,71,72,79,80,82 involving 886 and one on Qigong).24,53,60,65 Our meta-analysis
participants (seven on acupuncture, one on Tai Chi, showed a significant combined effect of acupuncture
Vol. 51 No. 4 April 2016
Table 1
Descriptive Summary of Studies Included in the Meta-Analysis
Length of
Cancer Study Sample Experimental Control Intervention Outcome Risk of
Author Year Type Country Size Intervention Intervention (Weeks) Measures Bias
Yu L 2012 Mixed China 146 Acupoint stimulation Usual care 3 Fatigue (PFS) High
Yang JL 2012 Mixed China 180 Acupoint stimulation Standard medical care 4 Fatigue (PFS) and QOL High
(KPS)
Yu HX 2012 Gastric China 68 Acupoint stimulation Standard medical care 1 Hours to first flatus/ High
bowel movement/
defecation
Cheng L 2009 Mixed China 90 Acupoint stimulation Standard medical care 1 Remission rate of High
diarrhea
Lai JH 2011 Cervical China 100 Chinese massage Standard medical care 1 Hours to first flatus/ High
bowel movement
Chen FR 2011 Mixed China 200 Acupoint stimulation Usual care 2 Nausea and vomit High
Interventions for Symptom Management and QOL

(QLQ-C30)
Yin ZL 2012 Nasopharyngeal China 43 Acupoint stimulation Standard medical care 2 QOL (QLQ-C30) High
Qiang WM 2011 Breast China 120 Tai Chi Music exercises 16 Vital capacity, BMI, High
QOL (FACT-B)
Chen LR 2009 Colorectal China 56 Chinese massage Standard medical care 2 Pain (VAS) High
Zhang YM 2012 Breast China 60 Chinese massage Standard medical care 3 Days QOL (QLSBC) High
Jiang MY 2013 Lung China 60 Tai Chi Standard medical care 4 Fatigue (BFI) and sleep High
quality
Wang HL 2012 Mixed China 399 Chinese massage Standard medical care 2 Sleep (PSQI), anxiety High
(SAS), and
depression (SDS)
Chen CF 2012 Mixed China 36 Acupoint stimulation Standard medical care 2 Remission rate of High
diarrhea
Xue WX 2011 Liver China 42 Acupoint stimulation Intramuscular 2 Response rate of High
injections hiccups
Chen YL 2008 Colorectal China 37 Chinese massage Standard medical care Not reported Hours to first flatus/ High
bowel movement
Wan SY 2011 Mixed China 134 Acupoint stimulation Standard medical care 1 QOL (QLQ-C30) High
Wang HM 2008 Colorectal China 30 Acupoint stimulation Standard medical care 1 Hours to first flatus/ High
bowel movement/
defecation, pain
(Prince Henry
method)
Lu Y 2005 Mixed China 62 Acupoint stimulation Standard medical care 6e8 QOL (KPS) High
Zhang MF 2009 Mixed China 71 Acupoint stimulation Standard medical care 4 QOL (FACT-G), anxiety High
(SAS), sleep quality
(PSQI), and pain
(self-made scale)
Pei Y 2010 Breast China 67 Acupoint stimulation Standard medical care 6 Anxiety (SAS) and High
depression (SDS,
HAMD)
Wang XP 2006 Mixed China 66 Acupoint stimulation Intramuscular 1 Response rate of High
injections hiccups
Li PX 2006 Mixed China 50 Acupoint stimulation Standard medical care 4 QOL (KPS), appetite, High
and body weight
733
(Continued)
Table 1
734
Continued
Length of
Cancer Study Sample Experimental Control Intervention Outcome Risk of
Author Year Type Country Size Intervention Intervention (Weeks) Measures Bias
Li HY 2008 Esophageal China 45 Acupoint stimulation Standard medical care Until first flatus Hours to first flatus/ High
nasogastric tube
retention
Liu H 2010 Liver China 102 Acupoint stimulation Standard medical care 2 Pain (VAS) High
Huang YN 2012 Mixed China 80 TCM-FEMT Standard medical care 12 Fatigue (PFS) and QOL High
(QLQ-C30)
Xiang CY 2006 Mixed China 92 Acupuncture combined Standard medical care 4 Depression (SDS, Moderate
with TCM-FEMT HAMD), QOL (KPS)
Feng Y 2011 Mixed China 80 Acupoint stimulation Standard medical care 4 Depression (SDS, High
HAMD), sleep quality
(PSQI)
Meng ZQ 2010 Colon China 75 Acupoint stimulation Standard medical care 6 Days or until QOL (QOLS), hours to High
the first bowel first flatus/bowel
movement, movement
whichever came
first.
Elizabeth N 2006 Breast Sweden 38 Acupoint stimulation Relaxation 12 Frequency of hot High
fiushes per 24 hours,
symptoms (KI, VAS),
psychological well-
Tao et al.
being (SCL-90),
general mood (Mood
Scale)
Randolph HL 2006 Lung China Hong 27 Acupoint stimulation Sham acupuncture 1 Pain (VAS; The Moderate
Kong cumulative dose of
patient-controlled
analgesia morphine
used on
postoperative day 2)
Joseph AR 2005 Breast USA 96 Acupoint stimulation Sham or no 1 Nausea and emesis, High
acupuncture QOL (FACT-G)
Zhu YY 2013 Pancreatic China 68 Acupoint stimulation Standard medical care 1 Pain (NRS), depression High
(HAMD)
Zhao BM 2013 Liver China 60 Acupoint stimulation Standard medical care 4 QOL (KPS), effective High
rate of ascites
Li J 2013 Breast China 168 Acupoint stimulation Standard medical care 8 QOL (KPS), fatigue High
(BFI)
Guo LY 2014 Gynecologic China 120 Acupoint stimulation Standard medical care 3 Fatigue (BFI) High
Fu LP 2014 Esophageal China 128 Acupoint stimulation Standard medical care 12 QOL (QLQ-C30), High

curative effect
Chen YH 2013 Gastric China 60 Acupoint stimulation Standard medical care 2 QOL (SF-36), anxiety High
(SAS), and
depression (SDS)
Yan QM 2014 Mixed China 150 Acupoint stimulation Standard medical care 4 QOL(KPS), effective High
rate of ascites
Liu XM 2007 Mixed China 62 Acupoint stimulation Oral drug 3 Days Response rate of High
hiccups
Xie XP 2005 Mixed China 101 Acupoint stimulation Point injection Not reported Response rate of High
hiccups
Chen SY 2007 Mixed China 36 Acupoint stimulation Standard medical care 1 Response rate of High
hiccups
Chen HT 2006 Mixed China 60 Acupoint stimulation Intramuscular 10 Days Response rate of High
injections hiccups
Luo M 2007 Liver China 48 Acupoint stimulation Intramuscular 3 Days Response rate of High
injections hiccups
Bao FF 2012 Mixed China 60 Acupoint stimulation Intramuscular 1 Response rate of High
injections hiccups
Gary D 2013 Mixed USA 74 Acupoint stimulation Sham acupuncture 6 Fatigue (BFI), Low
depression (HADS),
QOL (FACT-G)
Caroline S 2014 Breast Australia 30 Acupoint stimulation Sham acupuncture, 6 Well-being (W-BQ12), Moderate
wait list control fatigue (BFI)
Gary D 2013 Colon USA 90 Acupoint stimulation Sham acupuncture 3 Days Time that the patient Low
first tolerated solid
food/passed fiatus or
a bowel movement,
pain, and nausea
Interventions for Symptom Management and QOL

(VAS)
Simon SMN 2013 Colorectal China Hong 165 Acupoint stimulation Sham or no 4 Days Time of first passing Moderate
Kong acupuncture fiatus/tolerated a
solid diet/walk
independently, pain
(VAS)
Michael FJ 2011 Breast USA 12 Acupoint stimulation Usual care 8 Fatigue (BFI) and QOL Moderate
(FACT-Cog)
Annelie L 2012 Breast Sweden 74 Acupoint stimulation Sham acupuncture 5 Severity of hot fiushes Low
and sweating per
24 hours
David G 2010 Mixed USA 58 Acupoint stimulation Usual care 4 Composite score of Moderate
pain, function, and
activities of daily
living (Constant-
Murley instrument)
Gary D 2008 Mixed USA 106 Acupoint stimulation Sham acupuncture 4 Pain (BPI/NRS/MQS) Low
Gary D 2007 Breast USA 72 Acupoint stimulation Sham acupuncture 5 Hot fiash frequency Low
Yang JX 2013 Breast China 30 TCM-FEMT Usual care 12 Depression (SDS) High
Lei YY 2014 Mixed China 98 TCM-FEMT Usual care Not reported Anxiety (SAS) and High
depression (SDS)
Li WJ 2014 Liver China 13 Chinese massage Intramuscular Not reported Response rate of High
injections hiccups
Xiao J 2014 Gynecologic China 151 Chinese massage Usual care 24 Hours Hours to first flatus/ High
bowel movement
Byeongsang O 2012 Mixed Australia 81 Qigong Usual care 10 QOL (FACT-G), High
cognitive function
(EORTC-CF, FACT-
Cog)
Zhen C 2013 Breast China 96 Qigong Wait list 5 Depression (CES-D), High
fatigue (BFI), sleep
quality (PSQI), QOL
(FACT-G)
Liu YC 2014 Lung China 120 Tai Chi Usual care 4 Vital capacity, maximal High
inspiratory capacity,
735
total expiration time
(Continued)
on alleviating sleep disturbance (d ¼ 1.66; 95%
Moderate
mass index; FACT-B ¼ Functional Assessment of Cancer TherapyeBreast; VAS ¼ visual analog scale; QLSBC ¼ quality of life of breast cancer; BFI ¼ Brief Fatigue Inventory; PSQI ¼ Pittsburgh Sleep Quality Index; SAS ¼
Element Music Therapy; QOLS ¼ Quality of Life Status assessment tool; KI ¼ Kupperman Index; SCL-90 ¼ Symptom Checkliste90; NRS ¼ Numerical Rating Scale; SF-36 ¼ the MOS item Short Form health survey; FACT-
PFS ¼ Piper Fatigue Scale; QOL ¼ quality of life; KPS ¼ Karnofsky Performance Score; QLQ-C30 ¼ European Organization for Research and Treatment of Cancer Quality of life QuestionnaireeCore 30; BMI ¼ body
Self-rating Anxiety Scale; SDS ¼ Self-rating Depression Scale; FACT-G ¼ Functional Assessment of Cancer TherapyeGeneral; HAMD ¼ Hamilton scale for depression; TCM-FEMT ¼ Traditional Chinese Medicine Five-
Quality-of-Life Scale; FACIT-F ¼ Functional Assessment of Chronic Illness TherapyeFatigue survey; HADS ¼ hospital anxiety and depression scale; W-BQ12 ¼ The well-being questionnaire with 12 items;
Cog ¼ Functional Assessment of Cancer TherapydCognitive scale; MQS ¼ Medication Quantification Scale; CES-D ¼ Center for Epidemiologic Studies Depression Scale; WHOQOL ¼ World Health Organization
Risk of
Bias
CI ¼ 2.69 to 0.63; P ¼ 0.002; I2 ¼ 87%;
High
High
High
High
High
Figs. 4ke4l).
Our meta-analysis also revealed that interventions
significantly lowered emotional distress, including
muscular strength,
depression (four on acupuncture, two on TCM-
mass index, QOL
QOL (WHOQOL),
Vital capacity, body
aerobic capacity,
function of side
FEMT, one on Chinese massage)19,53,61,65,74,87,88 and
Body mass index

Outcome
Measures
QOL (FACIT-F),
and flexibility
QOL (FACT-B)
upper limbs
anxiety (two on acupuncture, one on TCM-FEMT,

QOL (SF-36)
(FACT-B)
one on Chinese massage),26,53,61,88 and exert a large
effect (depression: d ¼ 7.43; 95% CI ¼ 12.96 to
1.91; P ¼ 0.008; I2 ¼ 98%; anxiety: d ¼ 7.82;
95% CI ¼ 13.74 to 1.90; P ¼ 0.01; I2 ¼ 97%).
Further meta-analysis showed that TCM-FEMT
Intervention
relieved depression while acupuncture had no effect

Length of
(Weeks)
on depression or anxiety (Figs. 4me4q).

10
24
16
12
12
12
In three studies67,83,86 involving 184 breast cancer

patients, the effect of acupuncture on hot flashes
was evaluated; however, no significant effect was found
in the meta-analysis (d ¼ 0.02; 95% CI ¼ 1.49 to
1.45; P ¼ 0.97; I2 ¼ 36%; Fig. 4r).
Standard medical care
Psychosocial therapy
Intervention
Control
Other Clinical Outcomes. Three trials23,93,94 involving

249 patients assessed the effect of Tai Chi on the
Usual care
Usual care
BMI of breast cancer patients and were included in

our meta-analysis, which demonstrate a small but
significant combined effect (d ¼ 0.28; 95% CI,
0.53 to 0.03; P ¼ 0.03; I2 ¼ 0%). Another three
Continued
Table 1
trials49,93,94 evaluated the effect of Tai Chi on vital

capacity; a significant effect was observed in the
Experimental
Intervention
meta-analysis (d ¼ 339.53; 95% CI ¼ 78.50 to

600.56; P ¼ 0.01; I2 ¼ 88%; Fig. 5).
EORTC-CF ¼ European Organization for Research and Treatment of Cancer-Cognitive Function.
Tai Chi
Tai Chi
Tai Chi
Tai Chi
Tai Chi
Tai Chi
Discussion
Sample
This is the largest meta-analysis to date on the ef-

Size
22
120
110
19
21
21
fects of acupuncture, Tuina, Tai Chi, Qigong, or

TCM-FEMT on symptom management and QOL in
patients with cancer. In this report, we searched for
Country
all the reports in which the interventions were used

Study
in clinical cancer trials published before October 2,

China
China
USA
USA
USA
USA
2014, regardless of whether the article was published

in Chinese or English. This systematic review identi-
fied 66 published RCTs (covering 5465 patients)
reporting comparisons of treatment against either
Cancer
Type
standard medical care or active control. Although we

Breast
Breast
Breast
Breast
Breast
Breast
retrieved studies from many countries around the

world, more than half the studies were performed in
China. Whereas the studies covered various cancer
2007
2010
2011
2012
2011
2008
Year
sites and treatments, the evidence clearly clustered

in studies on breast cancer. According to our meta-
analyses, acupuncture produced small-to-large effects
Mustian KM
on adverse symptoms including pain, fatigue, sleep

Michelle CJ
Rausch SM
Wang YL
disturbance, and some gastrointestinal discomfort,

Lisa KS
Author
He JH
but no significant effect was found on the frequency

of hot flashes and mood distress. Tuina relieved
Table 2
Summary of Meta-Analysis Results
No. of No. of
Comparison Studies Subjects Statistical Method Effect Size
General QOL change

By KPS 7 762 4 Studies: MD (IV, random, 95% CI); 4.55 (0.75 to 8.34)
3 Studies: OR (M-H, fixed, 95% CI) 2.32 (1.26 to 4.27)
By generic QOL scales (e.g., SF-36, 4 222 SMD (IV, random, 95% CI) 3.49 (0.99 to 5.99)
WHOQOL)
Cancer-specific QOL change (e.g., FACT-G, 9 686 SMD (IV, random, 95% CI) 0.97 (0.18 to 1.76)
FACT-B, or EORTC QLQ-C30)
Symptoms improvement
Pain (by VAS or NRS) 9 710 SMD (IV, random, 95% CI) 0.35 (0.60 to 0.09)
Fatigue
Response rate by PFS 2 326 OR (M-H, fixed, 95% CI) 2.40 (1.40 to 4.13)
Means by BFI 7 560 MD (IV, random, 95% CI) 1.19 (2.19 to 0.18)
Gastrointestinal distress
By hours to first flatus 8 671 MD (IV, random, 95% CI) 11.52 (16.50 to 6.54)
By hours to first bowel movement 6 461 MD (IV, random, 95% CI) 8.22 (13.50 to 2.95)
Response rate of diarrhea 2 126 OR (M-H, fixed, 95% CI) 3.48 (1.06 to 11.47)
Nausea and vomit (by subscales of 2 328 MD (IV, fixed, 95% CI) 0.21 (0.30 to 0.11)
QLQ-C30)
Response rate of hiccups 9 488 OR (M-H, fixed, 95% CI) 5.57 (3.42 to 9.08)
Sleep disturbance (by PSQI) 4 646 MD (IV, random, 95% CI) 2.53 (4.17 to 0.89)
Depression (by SDS) 7 826 MD (IV, random, 95% CI) 7.4 3 (12.96 to 1.91)
Anxiety (by SAS) 4 624 MD (IV, random, 95% CI) 7.82 (13.74 to 1.90)
Hot flushes (by median number/24 hours) 3 184 MD (IV, random, 95% CI) 0.02 (1.49 to 1.45)
Other clinical outcomes
BMI 4 270 SMD (IV, random, 95% CI) 0.28 (2.08 to 0.61)
Vital capacity 3 350 MD (IV, random, 95% CI) 339.53 (78.5 to 600.56)
QOL ¼ quality of life; KPS ¼ Karnofsky Performance Score; OR ¼ odds ratio; MD ¼ mean difference; SF-36 ¼ the MOS item Short Form health survey;
WHOQOL ¼ World Health Organization Quality-of-Life Scale; SMD ¼ standardized mean difference; FACT-G ¼ Functional Assessment of Cancer Therapye
General; FACT-B ¼ Functional Assessment of Cancer TherapyeBreast; EORTC QLQ-C30 ¼ European Organization for Research and Treatment of Cancer Qual-
ity of Life QuestionnaireeCore 30; VAS ¼ visual analog scale; NRS ¼ Numerical Rating Scale; PFS ¼ Piper Fatigue Scale; BFI ¼ Brief Fatigue Inventory; QLQ-
C30 ¼ European Organization for Research and Treatment of Cancer Core Questionnaire; PSQI ¼ Pittsburgh Sleep Quality Index; SDS ¼ Self-rating Depression
Scale; SAS ¼ Self-rating Anxiety Scale; BMI ¼ body mass index; M-H ¼ Mantel Haenszel.
Twenty-three studies valued two or more outcomes.
gastrointestinal discomfort, TCM-FEMT lowered fatigue were found in our results than that of other
depression level, and Tai Chi improved the vital capac- meta-analyses in this field,9,29 but these two reviews
ity of breast cancer patients. were based on only a few selected trials on acupunc-
In our meta-analysis, the effect size of acupuncture ture and moxibustion and omitted other interventions
on pain, fatigue, and hiccups are consistent with the such as Chinese massage, Tai Chi, Qigong, and
reports of previous meta-analyses.27,101 A somewhat TCM-FEMT. One exception is that the significant
lower effect (d ¼ 0.76) on clinical efficacy measured effects of Tai Chi on BMI are different from
by KPS score was reported in an earlier meta- a published meta-analysis,32 which failed to include a
analysis; however, this meta-analysis only included tri- high-quality trial.94
als on acupuncture in lung cancer.102 Larger effect Strikingly, our meta-analysis demonstrated that
sizes for acupuncture on cancer-specific QOL and acupoint stimulation and Chinese massage exert
Fig. 2. Risk-of-bias analysis.

a Experimental Control Mean Difference Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
Li J 2013 80 1.4 79 75 1 79 33.7% 5.00 [4.62, 5.38]
Li PX 2006 67.78 8.01 27 63.48 4.87 23 25.8% 4.30 [0.68, 7.92]
Xiang CY 2006 59.26 16.43 46 63.52 13.61 46 17.9% -4.26 [-10.43, 1.91]
Yan QM 2014 59.23 11.23 70 48.12 17.15 80 22.6% 11.11 [6.52, 15.70]
Total (95% CI) 222 228 100.0% 4.55 [0.75, 8.34]

Heterogeneity: Tau² = 11.10; Chi² = 15.62, df = 3 (P = 0.001); I² = 81%
-20 -10 0 10 20
Test for overall effect: Z = 2.35 (P = 0.02)
Favors experimental Favors control
b Experimental Control Odds Ratio Odds Ratio

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
Lu Y 2005 25 32 22 30 35.6% 1.30 [0.41, 4.16]
Yang JL 2012 54 60 42 60 30.1% 3.86 [1.41, 10.57]
Zhao BM 2013 21 30 16 30 34.4% 2.04 [0.71, 5.89]
Total (95% CI) 122 120 100.0% 2.32 [1.26, 4.27]

Total events 100 80
Heterogeneity: Chi² = 1.99, df = 2 (P = 0.37); I² = 0%
0.05 0.2 1 5 20
c Experimental Control Std. Mean Difference Std. Mean Difference

Lisa KS 2012 104.94 6.6 9 108.96 6.06 10 29.4% -0.61 [-1.53, 0.32]
Mustian KM 2008 121.7 7 11 124.3 8 10 30.7% -0.33 [-1.20, 0.53]
Wang YL 2010 62.5 8.05 62 57.13 7.48 58 39.9% 0.69 [0.32, 1.05]
Total (95% CI) 82 78 100.0% -0.01 [-0.90, 0.89]

-10 -5 0 5 10
d Experimental Control Std. Mean Difference Std. Mean Difference

Byeongsang O 2012 8.41 1.32 23 -4.25 2.17 31 8.7% 6.71 [5.29, 8.13]
Gary D 2013 70.2 11.94 47 67.9 14.61 50 11.8% 0.17 [-0.23, 0.57]
Huang YN 2012 58.85 4.21 40 52.3 7.78 40 11.7% 1.04 [0.57, 1.51]
Joseph AR 2005 75.4 2.4 32 77.6 2.4 31 11.6% -0.91 [-1.43, -0.39]
Rausch SM 2007 2.15 5.99 13 0.75 4.72 4 9.8% 0.23 [-0.89, 1.35]
Wan SY 2011 54.53 6.68 70 43.17 5.35 64 11.8% 1.86 [1.45, 2.26]
Yin ZL 2012 86.15 6.34 22 84.85 6.11 21 11.4% 0.20 [-0.39, 0.80]
Zhang YM 2012 254.3 45.75 30 221.7 41.76 30 11.5% 0.73 [0.21, 1.26]
Zhen C 2013 82.1 15.4 49 82.8 15.9 46 11.8% -0.04 [-0.45, 0.36]
Total (95% CI) 326 317 100.0% 0.97 [0.18, 1.76]

Heterogeneity: Tau² = 1.34; Chi² = 159.80, df = 8 (P < 0.00001); I² = 95%
-10 -5 0 5 10
e Experimental Control Std. Mean Difference Std. Mean Difference

Gary D 2013 70.2 11.94 47 67.9 14.61 50 35.7% 0.17 [-0.23, 0.57]
Joseph AR 2005 75.4 2.4 32 77.6 2.4 31 33.1% -0.91 [-1.43, -0.39]
Yin ZL 2012 86.15 6.34 22 84.85 6.11 21 31.2% 0.20 [-0.39, 0.80]
Total (95% CI) 101 102 100.0% -0.17 [-0.88, 0.53]

-10 -5 0 5 10
Fig. 3. Forest plots of effect sizes for interventions on QOL. a) Acupuncture on QOL of terminal cancer survivors by means of
KPS, b) acupuncture on QOL by remission rate of KPS, c) Tai Chi on QOL of breast cancer patients by generic QOL scales, d)
interventions on QOL by cancer-specific QOL scales, e) acupuncture on QOL by cancer-specific QOL scales, f) Chinese mas-
sage on QOL by cancer-specific QOL scales, and g) Qigong on QOL by cancer-specific QOL scales. QOL ¼ quality of life;
KPS ¼ Karnofsky Performance Score; M-H ¼ Mantel Haenszel.
f Experimental Control Std. Mean Difference Std. Mean Difference

Wan SY 2011 54.53 6.68 70 43.17 5.35 64 51.1% 1.86 [1.45, 2.26]
Zhang YM 2012 254.3 45.75 30 221.7 41.76 30 48.9% 0.73 [0.21, 1.26]
Total (95% CI) 100 94 100.0% 1.31 [0.21, 2.41]

-10 -5 0 5 10
g Experimental Control Std. Mean Difference Std. Mean Difference

Byeongsang O 2012 8.41 1.32 23 -4.25 2.17 31 49.5% 6.71 [5.29, 8.13]
Zhen C 2013 82.1 15.4 49 82.8 15.9 46 50.5% -0.04 [-0.45, 0.36]
Total (95% CI) 72 77 100.0% 3.30 [-3.32, 9.92]

-50 -25 0 25 50
Fig. 3. (continued).
a significant effect on gastrointestinal discomfort of bias; only 17 (29.82%) have low or moderate risk
including easing diarrhea, hiccups, nausea and vomit- of bias. Only 16% provided information regarding
ing, and shortening time to first flatus and first bowel adequate concealment of the randomization sequence
movement (Fig. 4).The reduction in the last two out- and methods of blinding. Because of the nature of
comes mentioned previously is important for reducing acupuncture, Tuina, Tai Chi, Qigong, and TCM-
the risk of paralytic ileus by stimulating digestive pro- FEMT, it may be difficult to blind participants to inter-
pulsion, particularly after gastrointestinal surgery. We vention delivery. Few trials attempted to blind the
also found that Chinese massage and TCM-FEMT outcome assessors to minimize potential methodolog-
contribute to the combined effect on QOL in cancer ical bias. Therefore, bias may affect most of the RCTs.
patients but were not involved in any published review. Although lack of allocation concealment and blinding
Although the forest plots seem to provide consider- may be associated with exaggerated effect estimates
able evidences for effects of acupuncture, Tuina, Tai for subjective outcomes, the degree of bias seems to
Chi, Qigong, and TCM-FEMT on cancer-related symp- be rather limited.103 Incomplete follow-up data pose
toms and QOL, some effect sizes are modest. The another problem, although this may be caused by
study heterogeneity may be one of the most important the unfavorable prognoses inherent in many cancer
factor contributing to this limited effect. First, inter- diagnoses. Selective outcome reporting was present
ventions differed among studies. Intervention dura- in one-third of the studies, sometimes precluding
tion ranged from 24 hours to half a year. Different the quantitative analysis of effectsda common prob-
types and stages of disease were used among and lem in clinical trials.104 Because unreported outcomes
even within studies; the studies recruited more than were often nonsignificant, this may have inflated the
eight types of cancer patients including breast cancer, computed effect sizes.105 Low compliance with proto-
lung cancer, gastric cancer, and so forth. As well, the cols and lack of motivation to continue participating
overall QOL of cancer patients was evaluated by influence intervention impact and can be other poten-
various tools, including the KPS, SF-36, FACT-B, and tial sources of bias and also an indicator that the inter-
other tools used no more than once. Furthermore, vention is inappropriate for the recruited subjects.
peri-intervention or accompanying clinical treatment Many studies included only small samples, thus raising
varied among studies ranging from surgery, radio- the issue of insufficient power.106 However, small
therapy, various regimens of chemotherapy, and studies often produced larger effects than large
various combinations thereof. In addition, the num- studies did.
ber of subjects in some studies was generally too small; Limitations of our systematic review and any sys-
the smallest sample size was only 13. This makes the tematic review in general pertain to the potential
means and SDs of indicators vulnerable to individual incompleteness of the evidence reviewed. Although
outliers and higher or lower effects prone to be de- an attempt was made to retrieve both published and
tected because of chance. Therefore, the reliability unpublished research and to conduct a comprehen-
of the evidence presented here is clearly limited. sive search, some studies may have been missed.
Risk of bias was present in many RCTs. More than Moreover, some of the identified studies could not
80% of the studies were burdened with a high risk be retrieved. One other problem is the fact that
a Experimental Control Std. Mean Difference Std. Mean Difference

David G 2010 3.6 2.4 28 5.8 2.3 30 11.5% -0.92 [-1.47, -0.38]
Elizabeth N 2006 3.7 1.8 17 3.5 1.5 14 8.6% 0.12 [-0.59, 0.82]
Gary D 2008 2.53 1.86 62 2.96 2 57 15.8% -0.22 [-0.58, 0.14]
Gary D 2013 5 1.9 39 4.8 2.3 42 13.9% 0.09 [-0.34, 0.53]
Liu H 2010 3.18 1.65 51 4.01 1.94 51 14.9% -0.46 [-0.85, -0.06]
Randolph HL 2006 3 3.6 13 3 1.4 12 7.5% 0.00 [-0.78, 0.78]
Simon SMN 2013 2.1 1.2 55 3.4 2.2 55 15.1% -0.73 [-1.11, -0.34]
Zhu YY 2013 3.46 1.82 28 3.46 1.33 37 12.6% 0.00 [-0.49, 0.49]
Total (95% CI) 293 298 100.0% -0.30 [-0.56, -0.03]

-2 -1 0 1 2
b
Experimental Control Odds Ratio Odds Ratio
Yang JL 2012 26 60 18 60 59.5% 1.78 [0.84, 3.79]
Yu L 2012 27 73 11 73 40.5% 3.31 [1.49, 7.35]
Total (95% CI) 133 133 100.0% 2.40 [1.40, 4.13]

Total events 53 29
0.05 0.2 1 5 20
c Experimental Control Std. Mean Difference Std. Mean Difference

Caroline S 2014 3.1 2.7 9 5 1.8 10 19.3% -0.80 [-1.74, 0.14]
Gary D 2013 4.8 1.88 47 5.1 2 50 21.5% -0.15 [-0.55, 0.25]
Guo LY 2014 3.42 0.7 40 5.99 0.5 40 20.0% -4.18 [-4.98, -3.39]
Li J 2013 4.12 1.05 79 5.75 2.1 79 21.6% -0.98 [-1.31, -0.65]
Michael FJ 2011 2.13 1.23 5 4.38 2.53 7 17.6% -0.99 [-2.23, 0.26]
Total (95% CI) 180 186 100.0% -1.41 [-2.58, -0.24]

-4 -2 0 2 4
d Experimental Control Std. Mean Difference Std. Mean Difference

Li HY 2008 45.2 5.3 22 68.7 3.7 23 17.3% -5.07 [-6.31, -3.83]
Meng ZQ 2010 68.26 23.38 35 65.24 17.5 40 21.0% 0.15 [-0.31, 0.60]
Simon SMN 2013 48 21.6 55 55.2 26.4 55 21.2% -0.30 [-0.67, 0.08]
Wang HM 2008 48 25.44 15 79.2 48.48 15 19.9% -0.78 [-1.53, -0.04]
Yu HX 2012 37.5 11.2 34 58.3 11.7 34 20.6% -1.80 [-2.36, -1.23]
Total (95% CI) 161 167 100.0% -1.44 [-2.58, -0.29]

-10 -5 0 5 10
e
Experimental Control Std. Mean Difference Std. Mean Difference
Chen YL 2008 25.26 18.36 24 30.38 16.54 16 32.4% -0.28 [-0.92, 0.35]
Lai JH 2011 45.26 2.12 50 53.21 3.65 50 33.3% -2.64 [-3.18, -2.10]
Xiao J 2014 18 3.43 79 26 4.29 72 34.3% -2.06 [-2.46, -1.66]
Total (95% CI) 153 138 100.0% -1.68 [-2.88, -0.47]

-10 -5 0 5 10
Fig. 4. Forest plots of effect sizes for interventions on adverse symptoms. a) Acupuncture on pain; b) acupuncture on remis-
sion rate of fatigue by Piper Fatigue Scale; c) acupuncture on fatigue by Brief Pain Inventory; d) acupuncture on hours to first
flatus; e) Chinese massage on hours to first flatus; f) acupuncture on hours to first bowel movement; g) Chinese massage on
hours to first bowel movement; h) acupuncture on remission rate of diarrhea; i) acupuncture on nausea and vomit; j)
acupuncture on hiccups; k) acupuncture, Chinese massage, and Qigong on sleep disturbance; l) acupuncture on sleep distur-
bance; m) acupuncture, Chinese massage, and TCM-FEMT on depression; n) acupuncture on depression; o) TCM-FEMT on
depression; p) acupuncture, Chinese massage, and TCM-FEMT on anxiety; q) acupuncture on anxiety; and r) acupuncture on
hot flushes. TCM-FEMT ¼ Traditional Chinese Medicine Five-Element Music Therapy; M-H ¼ Mantel Haenszel.
f
Meng ZQ 2010 119.04 47.97 36 119.38 60.21 40 34.1% -0.01 [-0.46, 0.44]
Wang HM 2008 24 5.52 15 24 5.76 15 32.7% 0.00 [-0.72, 0.72]
Yu HX 2012 33.1 8.9 34 54.7 9.1 34 33.2% -2.37 [-3.00, -1.74]
Total (95% CI) 85 89 100.0% -0.79 [-2.31, 0.73]

-4 -2 0 2 4
g
Study or Subgroup Mean SD Total Mean SD Total Weight IV, Fixed, 95% CI IV, Fixed, 95% CI
Chen YL 2008 12.54 3.82 24 18.56 7.29 16 13.8% -1.08 [-1.76, -0.40]
Lai JH 2011 24.65 9.1 50 34.15 8.45 50 36.2% -1.07 [-1.49, -0.65]
Xiao J 2014 11 3.25 79 17 5.12 72 50.0% -1.41 [-1.76, -1.05]
Total (95% CI) 153 138 100.0% -1.24 [-1.49, -0.99]

-4 -2 0 2 4
Test for overall effect: Z = 9.62 (P < 0.00001)
h
Chen CF 2012 17 18 16 18 27.6% 2.13 [0.18, 25.78]
Cheng L 2009 42 45 35 45 72.4% 4.00 [1.02, 15.68]
Total (95% CI) 63 63 100.0% 3.48 [1.06, 11.47]

Total events 59 51
0.01 0.1 1 10 100
i
Experimental Control Mean Difference Mean Difference
Chen FR 2011 1.33 0.12 102 1.5 0.58 98 64.9% -0.17 [-0.29, -0.05]
Fu LP 2014 1.31 0.16 64 1.58 0.63 64 35.1% -0.27 [-0.43, -0.11]
Total (95% CI) 166 162 100.0% -0.21 [-0.30, -0.11]

-1 -0.5 0 0.5 1
j
Bao FF 2012 30 32 18 28 8.8% 8.33 [1.64, 42.39]
Chen HT 2006 33 40 11 20 18.7% 3.86 [1.16, 12.81]
Chen SY 2007 21 24 10 12 12.2% 1.40 [0.20, 9.75]
Liu XM 2007 29 32 20 30 14.1% 4.83 [1.18, 19.80]
Luo M 2007 18 24 8 24 14.6% 6.00 [1.71, 21.04]
Wang XP 2006 31 35 14 31 12.4% 9.41 [2.67, 33.14]
Xie XP 2005 51 54 34 47 14.7% 6.50 [1.72, 24.53]
Xue WX 2011 21 22 13 20 4.5% 11.31 [1.24, 102.72]
Total (95% CI) 263 212 100.0% 5.81 [3.54, 9.56]

Total events 234 128
0.01 0.1 1 10 100
k
Feng Y 2011 7.92 1.22 40 11.44 1.89 40 27.1% -3.52 [-4.22, -2.82]
Wang HL 2012 7.4 2.2 201 10.03 3.91 198 27.3% -2.63 [-3.25, -2.01]
Zhang MF 2009 7.88 3.42 32 12.57 4.53 39 20.9% -4.69 [-6.54, -2.84]
Zhen C 2013 10.1 3.1 49 9.6 2.9 46 24.7% 0.50 [-0.71, 1.71]
Total (95% CI) 322 323 100.0% -2.53 [-4.17, -0.89]

-20 -10 0 10 20
l Experimental Control Std. Mean Difference Std. Mean Difference

Feng Y 2011 7.92 1.22 40 11.44 1.89 40 49.3% -2.19 [-2.75, -1.63]
Zhang MF 2009 7.88 3.42 32 12.57 4.53 39 50.7% -1.14 [-1.65, -0.63]
Total (95% CI) 72 79 100.0% -1.66 [-2.69, -0.63]

-4 -2 0 2 4
m Experimental Control Mean Difference Mean Difference

Chen YH 2013 45.24 5.01 30 57.54 5.78 30 14.4% -12.30 [-15.04, -9.56]
Feng Y 2011 43.64 5.28 40 50.76 5.42 40 14.6% -7.12 [-9.46, -4.78]
Lei YY 2014 33.24 4.21 50 54.02 5.34 48 14.7% -20.78 [-22.69, -18.87]
Pei Y 2010 45.55 6.31 31 44.44 7.04 31 14.2% 1.11 [-2.22, 4.44]
Wang HL 2012 24.38 2.89 201 29.41 3.43 198 14.9% -5.03 [-5.65, -4.41]
Xiang CY 2006 40.76 14.26 46 39.41 11.6 46 13.1% 1.35 [-3.96, 6.66]
Yang JX 2013 51.87 5.67 15 60.07 4.17 15 14.1% -8.20 [-11.76, -4.64]
Total (95% CI) 413 408 100.0% -7.43 [-12.96, -1.91]

-20 -10 0 10 20
n
Chen YH 2013 45.24 5.01 30 57.54 5.78 30 24.0% -2.24 [-2.90, -1.59]
Feng Y 2011 43.64 5.28 40 50.76 5.42 40 25.2% -1.32 [-1.80, -0.83]
Pei Y 2010 37.32 5.01 31 39.11 5.77 31 25.1% -0.33 [-0.83, 0.17]
Xiang CY 2006 40.76 14.26 46 39.41 11.6 46 25.7% 0.10 [-0.31, 0.51]
Total (95% CI) 147 147 100.0% -0.93 [-1.89, 0.04]

-4 -2 0 2 4
o Experimental Control Std. Mean Difference Std. Mean Difference

Lei YY 2014 33.24 4.21 50 54.02 5.34 48 50.3% -4.30 [-5.03, -3.57]
Yang JX 2013 51.87 5.67 15 60.07 4.17 15 49.7% -1.60 [-2.44, -0.77]
Total (95% CI) 65 63 100.0% -2.96 [-5.60, -0.32]

-20 -10 0 10 20
p
Chen YH 2013 45.2 8.12 30 61.34 7.54 30 23.5% -16.14 [-20.11, -12.17]
Lei YY 2014 35.45 3.03 50 47.03 6.72 48 25.4% -11.58 [-13.66, -9.50]
Pei Y 2010 37.32 5.01 31 39.11 5.77 36 25.0% -1.79 [-4.37, 0.79]
Wang HL 2012 24.63 3.27 201 27.08 3.19 198 26.1% -2.45 [-3.08, -1.82]
Total (95% CI) 312 312 100.0% -7.82 [-13.74, -1.90]

-20 -10 0 10 20
q Experimental Control Std. Mean Difference Std. Mean Difference

Chen YH 2013 45.2 8.12 30 61.34 7.54 30 49.3% -2.03 [-2.66, -1.40]
Pei Y 2010 37.32 5.01 31 39.11 5.77 36 50.7% -0.33 [-0.81, 0.16]
Total (95% CI) 61 66 100.0% -1.17 [-2.84, 0.51]

-4 -2 0 2 4
r
Annelie L 2012 5.7 4.1 38 4.5 3.7 36 43.4% 0.30 [-0.15, 0.76]
Elizabeth N 2006 4.1 3.3 17 4.5 2.2 14 18.2% -0.14 [-0.84, 0.57]
Gary D 2007 6.2 4.2 39 7.6 5.7 28 38.4% -0.28 [-0.77, 0.20]
Total (95% CI) 94 78 100.0% -0.00 [-0.30, 0.30]

-4 -2 0 2 4
most of the included trials (51/67) were carried out Future RCTs of acupuncture, Tuina, Tai Chi,
in Chinese regions that have been shown to pro- Qigong, and TCM-FEMT for adverse symptoms
duce almost no negative studies.107 Moreover, it and QOL of cancer patients should adhere to
could be argued about the rationale for some accepted standards of trial methodology. The
meta-analyses pooling data from too few trials. studies included in this review show a number of de-
The main reasons for conducting meta-analyses fects that have been noted by other related reviews,
are to increase power, improve precision, answer for example, the frequency and duration of treat-
questions not asked by individual studies, settle con- ment, using validated primary outcome measures,
troversies arising from conflicting results, and adequate statistical tests, and heterogeneous com-
generate new hypotheses; a meta-analysis can be parison groups. Furthermore, even though it is
performed by combining two or more trials.108 difficult to blind subjects to some forms of interven-
However, the use of statistics does not guarantee tions, using assessor blinding is important for
that the results are valid. In our case, for the evalu- reducing bias. We recommend that future publica-
ation of acupuncture’s effect on diarrhea, the con- tions give detailed information of randomization
clusions from the meta-analyses are from only two methods including sequence generation and
RCTs with small sample sizes; therefore, the conclu- randomization concealment and blinding informa-
sions must remain tentative. tion, including whether blinding is used and who
a Experimental Control Std. Mean Difference Std. Mean Difference

He JH 2011 -0.7 3.6 51 -0.2 3.1 46 40.2% -0.15 [-0.55, 0.25]
Michelle CJ 2011 -0.42 0.75 9 0.29 0.61 10 6.8% -1.00 [-1.97, -0.03]
Mustian KM 2006 -1.2 5.06 11 1.4 5.76 10 8.4% -0.46 [-1.33, 0.41]
Qiang WM 2011 -0.82 2.27 56 -0.14 3.08 52 44.6% -0.25 [-0.63, 0.13]
Total (95% CI) 127 118 100.0% -0.28 [-0.53, -0.03]

-2 -1 0 1 2
b Experimental Control Mean Difference Mean Difference

He JH 2011 2,500.4 554.4 51 2,390.6 334.3 46 32.3% 109.80 [-70.43, 290.03]
Liu YC 2014 2,638.25 468.36 60 2,018.05 524.32 60 32.4% 620.20 [442.31, 798.09]
Qiang WM 2011 2,631.4 302.01 56 2,339.5 334.1 52 35.3% 291.90 [171.47, 412.33]
Total (95% CI) 167 158 100.0% 339.53 [78.50, 600.56]

-500 -250 0 250 500
Fig. 5. Forest plots of effect sizes for Tai Chi on other clinical outcomes. (a) body mass index and (b) vital capacity.
was blinded, if available. Detailed reporting accord- 10. Goats GC. Massagedthe scientific basis of an ancient
ing to the requirements presented by the CON- art: part 2. Physiological and therapeutic effects. Br J Sports
SORT (Consolidated Standards of Reporting Med 1994;28:153e156.
Trials) statement will make it easier to reach clear 11. Liu WS. Application of Chinese medicine in clinical
conclusions regarding study quality.109 oncology. Beijing: People’s Health Publishing House, 2005.
Although there are some clear limitations regarding 12. Chan CL, Wang CW, Ho RT, et al. A systematic review of
the body of research reviewed in this study, a tentative the effectiveness of qigong exercise in supportive cancer
conclusion can be reached that at least acupuncture care. Support Care Cancer 2012;20:1121e1133.
may alleviate some symptoms including pain, fatigue, 13. Lee MS, Choi TY, Ernst E. Tai chi for breast cancer
gastrointestinal distress, sleep disturbance but has no patients: a systematic review. Breast Cancer Res Treat 2010;
120:309e316.
effect on hot flashes. Tuina relieved gastrointestinal
discomfort. TCM-FEMT lowered depression level. Tai 14. American Music Therapy Association. Definitions and
Chi improved vital capacity of breast cancer patients. quotes about music therapy. Available at: http://www.
musictherapy.org/about/quotes/. Accessed August 13,
Acupuncture and Tuina improved overall QOL, 2012.
whereas Tai Chi and Qigong exert no effect on QOL
15. Huang Y, Yang X, Yang Q. Effect of traditional Chinese
of breast cancer survivors. Future study reporting
medicine five elements music on cancer-related fatigue of
should be improved regarding both research methods chemotherapy patients. Chin J Mod Nurs 2012;18:
and content of interventions. 1412e1414.
16. World Health Organization. WHO traditional medicine
strategy 2002-2005. Geneva, Switzerland: World Health Orga-
Disclosures and Acknowledgments nization, 2002.
No funding was received for this study, and the 17. Sprod LK, Janelsins MC, Palesh OG, et al. Health-
authors declare no conflicts of interest. related quality of life and biomarkers in breast cancer survi-
vors participating in tai chi chuan. J Cancer Surviv 2012;6:
The authors thank the health informatics experts 146e154.
for their assistance in developing the search strategy.
18. Wan S, Huang J, Gao J. Influence of acupoint massage
on quality of life of tumor patients accepting chemotherapy.
Chin Nurs Res 2011;25:317e318.
References 19. Xiang C, Guo Q, Liao J. Effect of therapy of traditional
1. World Health Organization. Cancer. [2008-9-23] Avail- Chinese medicine five element music and electroacupunc-
able at: http://www.whoint/mediacentre/factsheets/fs297/ ture on the depression levels of cancer patients. Chin
en/ J Nurs 2006;41:969e972.
2. Boon HS, Olatunde F, Zick SM. Trends in complemen- 20. Oh B, Butow PN, Mullan BA, et al. Effect of medical
tary/alternative medicine use by breast cancer survivors: Qigong on cognitive function, quality of life, and a
comparing survey data from 1998 and 2005. BMC Womens biomarker of inflammation in cancer patients: a randomized
Health 2007;7:4. controlled trial. Support Care Cancer 2012;20:1235e1242.
3. Sirois FM, Gick ML. An investigation of the health 21. Deng G, Chan Y, Sjoberg D, et al. Acupuncture for the
beliefs and motivations of complementary medicine clients. treatment of post-chemotherapy chronic fatigue: a random-
Soc Sci Med 2002;55:1025e1037. ized, blinded, sham-controlled trial. Support Care Cancer
4. Findley PA, Sambamoorthi U. Preventive health ser- 2013;21:1735e1741.
vices and lifestyle practices in cancer survivors: a population 22. Roscoe JA, Matteson SE, Morrow GR, et al. Acustimula-
health investigation. J Cancer Surviv 2009;3:43e58. tion wrist bands are not effective for the control of
5. Smithson J, Paterson C, Britten N, et al. Cancer chemotherapy-induced nausea in women with breast cancer.
patients’ experiences of using complementary therapies: J Pain Symptom Manage 2005;29:376e384.
polarization and integration. J Health Serv Res Policy 23. Mustian KM, Katula JA, Zhao H. A pilot study to assess
2010;15:54e61. the influence of tai chi chuan on functional capacity among
6. Monti DA, Sufian M, Peterson C. Potential role of breast cancer survivors. J Support Oncol 2006;4:139e145.
mind-body therapies in cancer survivorship. Cancer 2008; 24. Chen Z, Meng Z, Milbury K, et al. Qigong improves
112:2607e2616. quality of life in women undergoing radiotherapy for breast
7. Tong G, Geng Q, Cheng J, et al. Effects of psycho- cancer: results of a randomized controlled trial. Cancer
behavioral interventions on immune functioning in cancer 2013;119:1690e1698.
patients: a systematic review. J Cancer Res Clin Oncol 25. Yin ZL. Effect of TCM on quality of life in nasopharyn-
2014;140:15e33. geal cancer patients after radiotherapy. Chin J Phys Med
8. Zhou DH. Clinical oncology of Chinese medicine. Bei- Rehabil 2012;34:618e620.
jing: People’s Health Publishing House, 2003. 26. Chen HY, Li SG, Cho WC, et al. The role of acupoint
9. He XR, Wang Q, Li PP. Acupuncture and moxibustion stimulation as an adjunct therapy for lung cancer: a system-
for cancer-related fatigue: a systematic review and meta-anal- atic review and meta-analysis. BMC Complement Altern Med
ysis. Asian Pac J Cancer Prev 2013;14:3067e3074. 2013;13:362.
27. Choi TY, Lee MS, Ernst E. Acupuncture for cancer 45. Cheng L. Effect of moxibustion moxa at zusanli on
patients suffering from hiccups: a systematic review and cancer-related diarrhea. J Nurs Sci 2009;24:43e44.
meta-analysis. Complement Ther Med 2012;20:447e455.
46. Lai JH. Chinese massage at Yongquan on gastrointes-
28. Garcia MK, McQuade J, Haddad R, et al. Systematic tinal function in cervical cancer patients. Today Nurse
review of acupuncture in cancer care: a synthesis of the 2011;14:124e125.
evidence. J Clin Oncol 2013;31:952e960.
47. Chen FR, Wang M, Xue J. Study on influence of ginger
29. Lee MS, Choi TY, Park JE, et al. Moxibustion for cancer separated moxibustion on quality of life of elderly patients
care: a systematic review and meta-analysis. BMC Cancer with advanced cancer. Chin Nurs Res 2011;18:2951e2952.
2010;10:130.
48. Yin ZL, Meng ZX, Lin SY, et al. The intervention effect
30. Chao LF, Zhang AL, Liu HE, et al. The efficacy of acu- of chemotherapy in patients with gynecological tumor relax-
point stimulation for the management of therapy-related ation training method. Chin J Phys Med Rehabil 2012;34:
adverse events in patients with breast cancer: a systematic re- 618e620.
view. Breast Cancer Res Treat 2009;118:255e267.
49. Qiang WM, Dong FQ, Yan L, et al. Comparison of two
31. Lee MS, Chen KW, Sancier KM, et al. Qigong for can- different exercise programs in breast cancer patients after
cer treatment: a systematic review of controlled clinical trials. postoperative adjuvant chemotherapy. Chin J Nurs 2011;
Acta Oncol 2007;46:717e722. 46:537e540.
32. Yan JH, Pan L, Zhang XM, et al. Lack of efficacy of Tai
50. Chen LR, Hou XC. Effect of chiropractic and psycho-
Chi in improving quality of life in breast cancer survivors: a
logical intervention on patients with cancer pain and bad
systematic review and meta-analysis. Asian Pac J Cancer Prev
mood. Hebei J Traditional Chin Med 2009;31:1897e1898.
2014;15:3715e3720.
33. Lee MS, Pittler MH, Ernst E. Is Tai Chi an effective 51. Zhang YM. Effects of hand massage therapy on the
adjunct in cancer care? A systematic review of controlled quality of life in breast cancer patients undergoing chemo-
clinical trials. Support Care Cancer 2007;15:597e601. therapy. Chin J Nurs 2012;47:247e250.
34. Li X, Yang G, Li X, et al. Traditional Chinese medicine 52. Jiang MY. Influence of shadowboxing on improving
in cancer care: a review of controlled clinical studies pub- cancer-related fatigue and sleeping quality of patients with
lished in Chinese. PLoS One 2013;8:e60338. advanced lung cancer. Chin Nurs Res 2013;27:420e421.
35. Liu J, Li X, Liu J, et al. Traditional Chinese medicine in 53. Wang HL, Liu CC, Wen XX. The effect of a foot mas-
cancer care: a review of case reports published in Chinese sage combined with psychological intervention to improve
literature. Forschende Komplementarmedizin 2011;18: the sleep quality of patients with cancer. Zhejiang Clin
257e263. Med J 2012;14:1309e1310.
36. Yang G, Li X, Li X, et al. Traditional Chinese medicine 54. Chen CF. Treatment of moxibustion effecting on
in cancer care: a review of case series published in the Chi- cancer-associated diarrhea in 18 cases. Chin J Integr Tradit
nese literature. Evid Based Complement Alternat Med West Med Digest 2012;20:514e515.
2012;2012:751046. 55. Xue WX. Liver cancer after interventional chemo-
37. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA therapy intractable hiccups acupuncture curative effect
statement for reporting systematic reviews and meta- observation. Chin J Misdiagnostics 2011;11:3061.
analyses of studies that evaluate healthcare interventions:
56. Chen YL. Acupressure reduces abdominal distention
explanation and elaboration. BMJ 2009;339:b2700.
after laparoscopic resection in colorectal cancer patients.
38. Higgins JPT, Green S, eds. Cochrane handbook for sys- Chin J Misdiagnosis 2008;8:7627e7628.
tematic reviews of interventions version 5.1.0. The Cochrane
Collaboration, 2011. Available at: www.cochranehandbook. 57. Wan SY, Huang JP, Gao J. Influence of acupoint mas-
org. sage plus induced imagining relaxation on quality of life
of tumor patients accepting chemotherapy. Chin Nurs Res
39. Tierney JF, Stewart LA, Ghersi D, et al. Practical 2011;25:317e318.
methods for incorporating summary time-to-event data
into meta-analysis. Trials 2007;8:16. 58. Wang HM, Fan ZM. The clinical study on gastrointes-
tinal function recovery after colorectal carcinoma resection
40. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring in fast track programme using acupuncture. Master’s Thesis.
inconsistency in meta-analyses. BMJ 2003;327:557e560. 2011. Available at: http://d.g.wanfangdata.com.cn/Thesis_
41. Faul F, Erdfelder E, Lang AG, et al. G*Power 3: a flex- Y1945947.aspx.
ible statistical power analysis program for the social, 59. Lu Y, Zhang Y, Jing NC. Acupuncture improved quality
behavioral, and biomedical sciences. Behav Res Methods of life in cancer patients via clinical study. J Oncol Pract
2007;39:175e191. 2005;19:299e301.
42. Yu L. Effect of moxibustion on cancer related fatigue 60. Zhang MF, Ruan JW, Gu MF. Effects of acupuncture
Guide of China Medicine 2012;10:591e593. combined with relaxation training on pain, anxiety and
43. Yang JL. Effects of moxibustion on cancer related sleep quality in patients with advanced cancer. Chin J Phys
fatigue in elderly advanced cancer patients. Chin J Gerontol Med Rehabil 2009;31:674e677.
2012;32:4408e4410.
61. Pei Y, Yang J, Zhang J. To observe the state of depres-
44. Yu HX. Effects of moxibustion on gastrointestinal func- sion and anxiety in patients with acupuncture in the treat-
tion after surgery in elderly gastric cancer patients. Zhejiang ment of breast cancer. Chin J Inf On Tradit Chin Med
J Tradit Chin Med 2012;47:284. 2010;17:62.
62. Wang XP, Ding Y. The effect of music therapy on qual- 80. Deng G, Wong WD, Guillem J, et al. A phase II, ran-
ity of life in patients with chemotherapy for advanced cancer. domized, controlled trial of acupuncture for reduction of
J Clin Med Pract 2006;10:16. postcolectomy ileus. Ann Surg Oncol 2013;20:1164e1169.
63. Li HY, Yang DN. Acupuncture recovered gastrointes- 81. Ng SS, Leung WW, Mak TW, et al. Electroacupuncture
tinal function after surgery for esophageal cancer. Jiangsu reduces duration of postoperative ileus after laparoscopic
J Tradit Chin Med 2008;40:83e84. surgery for colorectal cancer. Gastroenterology 2013;144:
64. Liu H. The effect of acupuncture and moxibustion and 307e313.e1.
acupuncture point injection on treatment of pain of hepato- 82. Johnston MF, Hays RD, Subramanian SK, et al. Patient
cellular carcinoma. J Hebei Tradit Chin Med Pharmacol education integrated with acupuncture for relief of cancer-
2010;25:37e38. related fatigue randomized controlled feasibility study.
65. Feng Y, Wang XY, Li SD, et al. Clinical research of BMC Complement Altern Med 2011;11:49.
acupuncture on malignant tumor patients for improving 83. Liljegren A, Gunnarsson P, Landgren BM, et al.
depression and sleep quality. J Tradit Chin Med 2011;31: Reducing vasomotor symptoms with acupuncture in breast
199e202. cancer patients treated with adjuvant tamoxifen: a random-
66. Meng ZQ, Garcia MK, Chiang JS, et al. Electro- ized controlled trial. Breast Cancer Res Treat 2012;135:
acupuncture to prevent prolonged postoperative ileus: a ran- 791e798.
domized clinical trial. World J Gastroenterol 2010;16:
84. Pfister DG, Cassileth BR, Deng GE, et al. Acupuncture
104e111.
for pain and dysfunction after neck dissection: results of
67. Nedstrand E, Wyon Y, Hammar M, et al. Psychological a randomized controlled trial. J Clin Oncol 2010;28:
well-being improves in women with breast cancer after treat- 2565e2570.
ment with applied relaxation or electro-acupuncture for
vasomotor symptom. J Psychosom Obstet Gynaecol 2006; 85. Deng G, Rusch V, Vickers A, et al. Randomized
27:193e199. controlled trial of a special acupuncture technique for
pain after thoracotomy. J Thorac Cardiovasc Surg 2008;
68. Wong RH, Lee TW, Sihoe AD, et al. Analgesic effect of 136:1464e1469.
electroacupuncture in postthoracotomy pain: a prospective
randomized trial. Ann Thorac Surg 2006;81:2031e2036. 86. Deng G, Vickers A, Yeung S, et al. Randomized,
controlled trial of acupuncture for the treatment of hot
69. Zhu YY. Effect of acupuncture at specific points com- flashes in breast cancer patients. J Clin Oncol 2007;25:
bined with thumbtach needle on pain and depression in 5584e5590.
pancreatic cancer patients. Chinese General Practice 2013;
16:1923e1925. 87. Yang JX. Effect of Chinese medicine five-tone therapy
intervention on the depression state of the chemotherapy
70. Zhao BM. Effect of acupuncture on ascites in liver can- patients with breast cancer. Chin J Mod Nurs 2013;19:
cer. Health Way 2013;12:411. 4461e4463.
71. Li J. Clinical research on treating advanced breast can-
cer related fatigue by moxibustion. Clin J Chin Med 2013;5: 88. Lei YY. Effect of five elements music of traditional Chi-
1e4. nese medicine on the tumor patients under hyperthermia
nursing. J Nurses Train 2014;29:335e336.
72. Guo LY. Effect of acupuncture therapy on cancere
related fatigue of patients with gynecological malignant tu- 89. Li WJ. Effect of Chinese massage on hiccups in patients
mor after chemotherapy. J Clin Acupuncture Moxibustion with liver cancer. Yunnan J Tradit Chin Med Materia Med
2014;30:67e70. 2014;35:102.
73. Fu LP. Effect of moxibustion on quality of life in esoph- 90. Xiao J. Effect of Chinese massage combined with point
ageal cancer patients. Shaanxi J Tradit Chin Med 2014;35: pressure on ventosity in patients with gynecologic cancer.
77e79. Yiayao Qianyan 2014;16:332e333.
74. Chen YH. Effect of moxibustion on depression and 91. Liu YC. Effect of Tai Chi on respiratory function in pa-
quality of life in patients with advanced gastric cancer. tients with lung cancer. Chin J Trauma Disabil Med 2014;22:
J Nurs (China) 2013;20:63e65. 211.
75. Yan QM. Effect of moxibustion on ascites in advanced 92. Rausch SM. Evaluating psychosocial effects of two inter-
cancer patients. J Qilu Nurs 2014;20:123e124. vention, tai chi and spiritual growth groups, in women with
76. Chen HT. Clinical observation of acupuncture therapy breast cancer. MS Thesis. Richmond, VA: College of Human-
on hiccuping in liver cancer. J Clin Acupuncture Moxibus- ities and Sciences, Virginia Commonwealth University, 2007.
tion 2006;22:18e19. Available at: http://scholarscompass.vcu.edu/etd/819/.
77. Luo MS, Wu SB. Acupuncture for 70 cases of hiccup 93. He JH. Rehabilitation effect of systematic exercise in
after operation of primary liver cancer. World J Acupunc- adjuvant chemotherapy for breast cancer patients. Chin
ture-Moxibustion 2007;17:52e54. J Rehabil 2011;26:204e206.
78. Bao FF. Effect of acupuncture on hiccups in cancer 94. Janelsins MC, Davis PG, Wideman L, et al. Effects of Tai
patients after radio-chemotherapy. Zhejiang J Tradit Chin Chi Chuan on insulin and cytokine levels in a randomized
Med 2012;47:117. controlled pilot study on breast cancer survivors. Clin Breast
Cancer 2011;11:161e170.
79. Smith C, Carmady B, Thornton C, et al. The effect of
acupuncture on post-cancer fatigue and well-being for 95. Mustian KM, Palesh OG, Flecksteiner SA. Tai Chi
women recovering from breast cancer: a pilot randomised Chuan for breast cancer survivors. Med Sport Sci 2008;52:
controlled trial. Acupunct Med 2013;31:9e15. 209e217.
96. Wang YL, Sun XY, Wang YB. Different exercise on 103. Wood L, Egger M, Gluud LL, et al. Empirical evidence
breast cancer patients postoperative upper limb function of bias in treatment effect estimates in controlled trials with
and the influence of the quality of life. Chin J Phys Med different interventions and outcomes: meta-epidemiological
Rehabil 2012;34:64e65. study. BMJ 2008;336:601e605.
97. Li PX. Effect of acupuncture on anorexia in advanced 104. Smyth RM, Kirkham JJ, Jacoby A, et al. Frequency and
gastric cancer patients. N J Tradit Chin Med 2006;38:67e68. reasons for outcome reporting bias in clinical trials: inter-
98. Liu XM, Huang ZF. Acupuncture treatment for intrac- views with trialists. BMJ 2011;342:c7153.
table hiccup in patients with cancer of 32 cases. Liaoning 105. Faller H, Schuler M, Richard M, et al. Effects of psycho-
J Tradit Chin Med 2007;34:214e215. oncologic interventions on emotional distress and quality of
99. Xie XP. Observation of electroacupuncture and press- life in adult patients with cancer: systematic review and meta-
ing auricular point with pill in treating patients with tumor analysis. J Clin Oncol 2013;31:782e793.
suffering from hiccup. Cancer Res Prev Treat 2005;32: 106. Moyer A, Sohl SJ, Knapp-Oliver SK, et al. Characteris-
42e43. tics and methodological quality of 25 years of research inves-
100. Chen SY, Huang LQ. Effect of electroacupuncture tigating psychosocial interventions for cancer patients.
treatment on zusanli(st36) for intractable hiccups with ma- Cancer Treat Rev 2009;35:475e484.
lignant tumors of 24 cases. J Qilu Nurs 2007;13:43e44. 107. Vickers A, Goyal N, Harland R, et al. Do certain coun-
101. Lu L, Liao M, Zeng J, et al. Quality of reporting and its tries produce only positive results? A systematic review of
correlates among randomized controlled trials on acupunc- controlled trials. Control Clin Trials 1998;19:159e166.
ture for cancer pain: application of the CONSORT 2010 108. Higgins JPT, Green S. Cochrane handbook for system-
statement and STRICTA. Expert Rev Anticancer Ther atic reviews of interventions. West Sussex, UK: Wiley-Black-
2013;13:489e498. well, 2008.
102. Chan HY, Chui YY, Chan CW, et al. Exploring the influ- 109. Schulz KF, Altman DG, Moher D, et al. CONSORT
ence of traditional Chinese medicine on self-care among 2010 statement: updated guidelines for reporting parallel
Chinese cancer patients. Eur J Oncol Nurs 2014;18: group randomized trials. Ann Intern Med 2010;152:
445e451. 726e732.

1 s2.0 S0885392416000555 Main

Uploaded by

Copyright:

Available Formats

1 s2.0 S0885392416000555 Main

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 s2.0 S0885392416000555 Main

Uploaded by

Copyright:

Available Formats

728 Journal of Pain and Symptom Management Vol. 51 No.

Effects of Acupuncture, Tuina, Tai Chi, Qigong, and Traditional

Introduction gentle movements and postures of the exercise are de-

Medline-325; Cochrane Central-

Screening of titles and abstracts

Assessment of full-text articles

Studies included in systematic

Studies included in quantitative

Interventions for Symptom Management and QOL

Vol. 51 No. 4 April 2016

Interventions for Symptom Management and QOL

on alleviating sleep disturbance (d ¼ 1.66; 95%

Vital capacity, body

FEMT, one on Chinese massage)19,53,61,65,74,87,88 and

Body mass index

anxiety (two on acupuncture, one on TCM-FEMT,

relieved depression while acupuncture had no effect

on depression or anxiety (Figs. 4me4q).

In three studies67,83,86 involving 184 breast cancer

Other Clinical Outcomes. Three trials23,93,94 involving

BMI of breast cancer patients and were included in

trials49,93,94 evaluated the effect of Tai Chi on vital

meta-analysis (d ¼ 339.53; 95% CI ¼ 78.50 to

This is the largest meta-analysis to date on the ef-

fects of acupuncture, Tuina, Tai Chi, Qigong, or

all the reports in which the interventions were used

in clinical cancer trials published before October 2,

2014, regardless of whether the article was published

standard medical care or active control. Although we

retrieved studies from many countries around the

sites and treatments, the evidence clearly clustered

on adverse symptoms including pain, fatigue, sleep

disturbance, and some gastrointestinal discomfort,

but no significant effect was found on the frequency

General QOL change

Fig. 2. Risk-of-bias analysis.

a Experimental Control Mean Difference Mean Difference

Total (95% CI) 222 228 100.0% 4.55 [0.75, 8.34]

b Experimental Control Odds Ratio Odds Ratio

Total (95% CI) 122 120 100.0% 2.32 [1.26, 4.27]

c Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 82 78 100.0% -0.01 [-0.90, 0.89]

d Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 326 317 100.0% 0.97 [0.18, 1.76]

e Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 101 102 100.0% -0.17 [-0.88, 0.53]

f Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 100 94 100.0% 1.31 [0.21, 2.41]

g Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 72 77 100.0% 3.30 [-3.32, 9.92]

a Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 293 298 100.0% -0.30 [-0.56, -0.03]

Total (95% CI) 133 133 100.0% 2.40 [1.40, 4.13]

c Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 180 186 100.0% -1.41 [-2.58, -0.24]

d Experimental Control Std. Mean Difference Std. Mean Difference

Total (95% CI) 161 167 100.0% -1.44 [-2.58, -0.29]

Total (95% CI) 153 138 100.0% -1.68 [-2.88, -0.47]

Total (95% CI) 85 89 100.0% -0.79 [-2.31, 0.73]