Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.
com
Support Care Cancer (2018) 26:415–425
https://doi.org/10.1007/s00520-017-3955-6
REVIEW ARTICLE
Effects of acupuncture on cancer-related fatigue: a meta-analysis
Yan Zhang 1 & Lu Lin 2 & Huiling Li 2 & Yan Hu 3 & Li Tian 1
Received: 19 July 2017 / Accepted: 30 October 2017 / Published online: 11 November 2017
# Springer-Verlag GmbH Germany, part of Springer Nature 2017
Abstract
Purpose This study was designed to critically evaluate the
effect of acupuncture on cancer-related fatigue (CRF).
Methods Seven databases (Cochrane Library, Embase,
Medline, Web of Science, CBM, Wanfang, and CNKI) were
systematically reviewed from inception to November 2016 for
randomized controlled trials (RCTs). Two reviewers critically
and independently assessed the risk of bias using Cochrane
Collaboration criteria and extracted correlated data using the
designed form. All analyses were performed with Review
Manager 5.
Results Ten RCTs, including 1327 patients (acupuncture,
733; control, 594), meeting the inclusion criteria for the
meta-analysis were identified. Acupuncture had a marked ef-
fect on fatigue in cancer patients, regardless of concurrent
anti-cancer treatment, particularly among breast cancer pa-
tients. The meta-analysis also indicated that acupuncture
could significantly mitigate CRF compared with sham acu-
puncture or usual care. Acupuncture for 20–30 min/session
three times/week for two or three weeks, twice weekly for
two weeks and weekly for six weeks, and weekly for six
weeks had substantial effects on CRF. Six RCTs reported the
occurrence of adverse events, whereas five reported none. The
The authors have full control of all primary data and agree to allow the
journal to review the data if requested.
* Li Tian
tianlisz@suda.edu.cn
1
The First Affiliated Hospital of Soochow University, P.O. Box 102,
Suzhou, China
2
School of Nursing, Soochow University, P.O. Box 203,
Suzhou, China
3
School of Nursing, Fudan University, P.O. Box 401, Shanghai, China
remaining study reported some manageable events, including
spot bleeding and bruising.
Conclusions Acupuncture is effective for CRF management
and should be recommended as a beneficial alternative thera-
py for CRF patients, particularly for breast cancer patients and
those currently undergoing anti-cancer treatment.
Keywords Acupuncture . Cancer-related fatigue .
Meta-analysis
Introduction
Among cancer patients, 52.07% endure clinically significant
cancer-related fatigue (CRF) (≥ 4) [1], the most distressing
symptom correlated with the disease process and anti-cancer
treatments [2]. CRF significantly deteriorates the quality of
life (QOL) of both cancer patients and their families, including
their confidence in conquering cancer [3]. Therefore, manage-
ment of CRF is an urgent need. Unfortunately, exact and ef-
fective pharmacological strategies for the management of
CRF are lacking. In response, an increasing number of studies
of complementary and alternative medicine (CAM) have been
performed. As an indispensable part of CAM, traditional
Chinese medicine (TCM) has been gradually acknowledged
worldwide for the management of CRF [4]. In TCM, CRF is
divided into the category of consumptive disease due to non-
restoring imbalance between Yin and Yang and insufficient
viscera’s vital qi and blood in the long term caused by cancer,
anti-cancer treatments, and other factors, such as depression
and pain [5]. Consequently, TCM holds that regulating the
viscera and tonifying qi and blood are keys to the successful
treatment of CRF. In TCM, acupuncture alone or in combina-
tion with moxibustion is often used to tonify qi and blood.
Acupuncture involves the insertion of thin needles into certain
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com
416
acupoints at specific angles in accordance with TCM theory
and the twirling, lifting, and thrusting of these needles. The
main therapeutic effects of acupuncture include dredging the
meridians (channels and collaterals), regulating Yin and Yang,
and strengthening the body’s resistance to eliminate pathogen-
ic factors. Moxibustion can warm and activate meridians via
the deep bodily penetration of radiant heat produced by burn-
ing moxa, thereby restoring the vitality of the human body.
Increasing attention has been paid to acupuncture as an
alternative strategy for CRF management. However, previous
studies of the effect of acupuncture on CRF have yielded
controversial results. In addition, additional trials with rela-
tively high methodological quality and relatively large sample
sizes have been published in recent years; therefore, this meta-
analysis was performed to critically evaluate the effectiveness
of acupuncture in CRF management and thereby reach a more
convincing conclusion with respect to acupuncture for CRF
and provide a particularly useful reference for constructing an
efficient acupuncture regimen.
Methods
The meta-analysis was performed following the PRISMA
guidelines for systematic reviews and meta-analyses [6].
Searching strategies
Seven databases, namely the Cochrane Central Register of
Controlled Trials, Embase, Medline, Web of Science, China
Biology Medicine (CBM), Wanfang, and China National
Knowledge Infrastructure (CNKI) databases, were systemati-
cally retrieved from inception through November 2016 for
correlative randomized controlled trials (RCTs) without lan-
guage restrictions. These searches were performed using the
following keywords: Bacupuncture,^ Bacupuncture therapy,^
Bfatigue or asthenia,^ Bcancer-related fatigue,^ Bcancer,^
Bcarcinoma,^ Btumour or tumor,^ and Bmalignance.^ Two
reviewers first screened the literature by scanning the titles
and abstracts and then read the full text of potentially eligible
trials to decide whether they should be included in the meta-
analysis. The search strategy for the database Medline is
shown in the Appendix as an example. Additionally, further
potentially relevant papers were searched using the reference
lists of the identified articles.
Inclusion criteria
Participants
Trials that focused on adult cancer patients (≥ 18 years) were
included.
Support Care Cancer (2018) 26:415–425
Interventions and controls
Trials that compared acupuncture (including electro-
acupuncture) with any type of control group were included,
and those that focused on laser acupuncture or electro-
acupuncture without needles were excluded.
Outcomes
Trials with CRF as an outcome were included in a systematic
review; among these trials, studies with extractable CRF
scores were included in the meta-analysis.
Studies
Only RCTs were eligible.
Data extraction
The general information of the trials, the epidemiological data
of the patients, the characteristics of the intervention and the
control groups (treatment acupoint, insertion technique, dura-
tion, frequency, and program length), the outcome measures,
and the adverse events were independently extracted by two
reviewers using a pre-designed information sheet.
Risk of bias assessment
Two reviewers independently evaluated the risk of bias in the
included RCTs using the Cochrane assessment tool, which
consists of the following seven domains: Badequate sequence
generation, allocation concealment, blinding of participants
and personnel, blinding of outcome assessment, incomplete
outcome data, selective reporting, and other bias^ [7]. Each
question can be rated as follows: yes (+), low risk of bias;
unclear (?), unclear risk of bias; no (−), high risk of bias.
Data analysis
The meta-analysis was performed using Review Manager
Software (5.0.2 version). We first assessed the clinical hetero-
geneity, and if moderate clinical diversity was identified, we
then conducted a subgroup analysis under the premise that
each stratum included at least two trials because clinical het-
erogeneity always resulted in statistical heterogeneity [7]. I2
was adopted to weigh the statistical diversity among the stud-
ies in each analysis. The selection of a fixed-effects model
(P > 0.1 and I2 < 50%) or a random-effects model (P < 0.1
and I2 ≥ P < 0) depended on the values of P and I2. Descriptive
analyses should be selected instead of a meta-analysis if
P < 0.1 and the sources of diversity are unknown.
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Support Care Cancer (2018) 26:415–425 417

Results cancer patients who had completed anti-cancer treatment (off-

treatment).
A total of 1312 studies were retrieved in the literature search.
Eleven were included in the systematic review, of which only
10 met the inclusion criteria of the meta-analysis. One trial Interventions
was excluded from further analysis due to insufficient data.
Figure 1 presents the process of screening and selection of Two trials used an electrical stimulator of the needles for some
these studies. acupoints [11, 17], and two trials integrated moxibustion with
acupuncture for some acupoints [8, 9]. Ten RCTs provided
Characteristics of the included trials details of the treatment acupoints [8–15, 17, 18], whereas
one study did not. Each trial described the insertion technique.
The characteristics of the participants, acupuncture details, The duration of the acupuncture sessions ranged from 20 to
controls, outcome measures, and adverse events are shown 30 min. The frequency of the acupuncture sessions varied
in Table 1. from one to three times weekly over a period of two to
10 weeks.
Participants
Controls
Among the 11 RCTs, five were conducted in breast cancer
patients, one in breast cancer and endometrium cancer pa-
Five trials used sham acupuncture as a control intervention
tients, one in patients with gynecologic cancer, and the other
[10, 11, 13, 17, 18], and only one trial used the same treatment
four in various types of malignancy. The mean age of all
acupoint as the true acupuncture protocol [17]. Eight trials
included patients ranged from 51.0 to 64.9 years. The ethnic-
compared acupuncture and typical care [8–12, 14–16]. One
ity was reported in seven trials. Two trials were conducted
trial used American ginseng as a control intervention [9].
with Chinese populations, and three trials were performed in
white populations. In the four studies that reported cancer
stage, most were stages I to III. Only four studies focused on

Fig. 1 Flow chart diagram of trial Records identified through Additional records identified
identification and selection database searching through other sources
(n= 1309) (n= 3)

duplicates removed
(n=691)

Citations screened by title and

abstract
(n=621)

Articles excluded
(n=578)

Full text articles assessed

for eligibility
(n= 43)

Full-text articles excluded (n=32)

Allocation not randomized: 2
Treatment arm not appropriate: 20
Control arm not appropriate: 6
Fatigue not quantified: 4
Studies included in qualitative
synthesis
(n=11)
Articles excluded (n=1)
Not providing sufficient data
for further analysis:1
Studies included in quantitative
synthesis (meta-analysis)
(n= 10)
 Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Table 1 Studies included in the systematic review and meta-analysis

418

Author, Sample Cancer type Mean age Ethnicity Caner stage Current anti- Intervention: treatment acupoint, Comparison Fatigue Adverse
year (± SD) tumor treatment insertion technique, duration, scale events
Gender frequency, program length

Lian et al. 2015 AG 22 Various AG 53.5 ± Asian Not Off-treatment DU20, EX-HN3, EX-HN5, UC BFI None
[8] CG 23 14.9 reported GB34, KI3, LI4, LR3, PC6,
Female 14, SP6, SP9, and ST36; lift,
male 8 thrust, and twirl the needle
CG 54.5 ± until BDe Qi^ (e.g., sensation
14.8 of soreness, tingling),
Female 13, moxibustion on RN4, RN6,
male 10 RN12, and KI1
simultaneously; 30 min; 3
times per week; 2 weeks
Guo et al. 2014 AG 40 AG: cervical 17, AG 51.0 ± Asian I to IV Chemotherapy BL18, BL23, RN4, RN6, RN12, CG1: American ginseng, BFI Not reported
[9] CG1 40 ovarian 20, 5.5 PC6, SP6, and ST36; lift, 2000 mg/d P.O., 3
CG2 40 endometrial 7 CG1 52.0 ± thrust, and twirl the needle times per day, 3 weeks
CG1: cervical 3.7 until BDe Qi^ (e.g., sensation CG2: UC
14, ovarian 12, CG2 53.0 ± of soreness, tingling), then
endometrial 14 6.2 needle warming moxibustion
CG2: cervical Female on ST36, RN4, and RN6;
11, ovarian 18, 20 min; 3 times per week;
endometrial 7 3 weeks
Mao et al. 2014 AG 22 Breast cancer AG 57.5 ± % white I to III Hormonal therapy SP6 and ST36; lift, thrust, and CG 1: sham acupuncture BFI Not reported
[10] CG1 22 10.1 AG 59% twirl the needle until BDe Qi^ (non-penetrating
CG2 23 CG1 60.9 ± CG1 77% (e.g., sensation of soreness, needles at
6.5 CG2 78% tingling); 30 min; twice non-acupuncture,
CG2 60.6 ± weekly for 2 weeks and avoid eliciting the BDe
8.2 weekly for 6 weeks; 8 weeks Qi^ sensations;
Female remainder was
the same as AG)
CG 2: UC
Smith et al. AG 10 Breast cancer AG 55.0 ± % Caucasian Not Off-treatment KI3, KI27, RN4, RN6, SP6, and CG 1: sham acupuncture BFI Not reported
2013 [11] CG1 10 8.8 AG 100% reported ST 36; using a Park device; (non-invasive sham
CG2 10 CG1 53.0 ± CG1 90% 20 min; twice weekly for needle using the Park
12.5 CG2 90% 3 weeks and weekly for device on the fixed
CG2 58.0 ± 3 weeks; 6 weeks sham points, the rest
7.5 was same with AG)
Female CG 2: UC
Molassiotis AG 56 Breast cancer Not reported % Caucasian I to III Surgery, LI4, SP6, and ST36; puncture CG: UC MFI None
et al. 2013 CG 49 Female AG 94% chemotherapy, perpendicularly, with a
[12] CG 91% radiotherapy depth of 0.5 to 1 in.,
depending on patients’ size,
sensitivity, and health state;
20 min; weekly, 10 weeks
AG 47 Various Not reported Not reported Off-treatment BFI None
Support Care Cancer (2018) 26:415–425
 Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Table 1 (continued)

Author, Sample Cancer type Mean age Ethnicity Caner stage Current anti- Intervention: treatment acupoint, Comparison Fatigue Adverse
year (± SD) tumor treatment insertion technique, duration, scale events
Gender frequency, program length

Deng et al. CG 50 AG 54(46, HT6, KI3, LI11, RN4, RN6, CG: sham acupuncture
2013 [13] 58) SP6, and ST36; insert the (performed exactly as
Female 39, needle until BDe Qi^ (a the true acupuncture,
male 8 sense of warmth, heaviness, or except for the use of
CG 53(45, aching); 20 min; weekly; 6 weeks blunt-tipped needles
59) and points that were a
Female 41, few millimeters off the
Support Care Cancer (2018) 26:415–425

male 9 meridians and away

from the points used
in the true
acupuncture)
Molassiotis AG 227 Breast cancer AG 52 (30, % white Not reported Surgery, LI4, SP6, and ST36; puncture CG: UC MFI Not reported
et al. CG 75 75) AG 93% chemotherapy, perpendicularly, with a
2012 [14] CG 53 (25, CG 95% radiotherapy depth of 0.5 to 1 in.,
80) depending on patient size,
Female sensitivity, and health state;
20 min; weekly, 6 weeks
Johnston et al. AG 5 Breast cancer AG 55.0 ± % white Not reported Chemotherapy, KI 3, LI4, SP6, and ST36; lift, CG: UC BFI None
2011 [15] CG 7 6.4 AG 80% hormonal thrust, and twirl the needle
CG 53.0 ± CG 57% therapy, until BDe Qi^ sensation;
7.2 radiotherapy 30 min; weekly; 8 weeks
Female additionally involving
self-care training alongside
acupuncture in the first four
sessions
Lim et al. 2011 AG 10 Various AG 55.0 ± Not IV Off-treatment Acupuncture points were CG: UC ESAS None
[16] CG 8 11.1 reported chosen based on the
Female 8, symptoms experienced;
male 2 insert the needle until BDe
CG 64.9 ± Qi^ (sensation of heaviness
8.7 and fullness), then connect
Female 7, to an electrical stimulator
male 1 (0.3-ms duration, 4-Hz
alternating current); 20 min;
weekly; 4 weeks
Balk et al. 2009 AG 16 Breast and AG 54.0 ± Not Not r Surgery, chemotherapy, KI3, LI4, RN6, SP6, and CG: sham acupuncture FACIT-F None
[17] CG 11 endometrium 9.1 reported eported radiotherapy ST36; insert the needle to (the telescoping
cancer CG 53.7 ± elicit a Bneedle grab^ blunt-edged needle
9.0 sensation, the needles at was used at the
Female KI3 and ST36 were same acupuncture
connected to points as the true
electro-acupuncture device acupuncture protocol)
419
 Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Table 1 (continued)
420

Author, Sample Cancer type Mean age Ethnicity Caner stage Current anti- Intervention: treatment acupoint, Comparison Fatigue Adverse
year (± SD) tumor treatment insertion technique, duration, scale events
Gender frequency, program length

(1 Hz), a heat lamp was

placed over RN6 point on
the lower abdomen;
30 min; 1–2 times per
week; 4–6 weeks
Molassiotis AG 15 Various AG: not Not Not Off-treatment LI4, SP6, and ST36; puncture CG: sham acupressure MFI Spot bleeding
et al. 2007 CG 16 reported reported reported perpendicularly, with a (self-pressure in three (n = 2),
[18] Female 9, depth of 0.5 to 1 in., points, i.e., LI12, bruising
male 6 depending on patients’ size, GB33, BL61); 1 min (n = 1),
CG: not sensitivity, and health state, for each point; daily; feeling of
reported to elicit BDe Qi^; 20 min; 3 2 weeks discomfort
Female 11, times per week; 2 weeks (n = 1),
male 5 nausea
(n = 1)

AG acupuncture group, CG control group, DU20 Baihui, EX-HN3 Yintang, EX-HN5 Taiyang, RN12 Zhongwan, PC6 Neiguan, ST36 Zusanli, SP6 Sanyinjiao, RN6 Qihai, RN4 Guanyuan, BL23 Shenshu,
BL18 Ganshu, LI4 Hegu, SP9 Yinlingquan, GB34 Yanglingquan, LR3 Taichong, KI3 Taixi, KI1Yongquan, KI27 Shufu, HT6 Yinxi, LI11 Quchi, LI12 Erjian, GB33 Xiyangguan, BL61 Pucan, BFI Brief
Fatigue Inventory, MFI Multidimensional Fatigue Inventory, ESAS Edmonton Symptom Assessment System, FACIT-F Functional Assessment of Chronic Illness Therapy–Fatigue subscale
Support Care Cancer (2018) 26:415–425
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Support Care Cancer (2018) 26:415–425 421

Blinding? (Blinding of participants and personnel(perfomance bias))

Risk of bias in individual trials

Blinding? (Blinding of outcome assessment (detection bias))

The overall risk of bias is presented in Fig. 2. The major
sources of risk of bias seemed to correlate with allocation
concealment, blinding of participants and personnel, and
blinding of outcome assessment. The individual risk of bias
for each trial is presented in Fig. 3. Four trials [12, 14, 15, 17]
had a high risk of bias (mainly due to the allocation conceal-

Incomplete outcome data addressed?

ment, the blinding of participants, and the blinding of outcome

Adequate sequence generation?

assessment), and seven trials had a relatively low risk of bias.

Free of selective reporting?

One trial [17] had a high risk of bias regarding incomplete

Allocation concealment?
outcome data.

Free of other bias?

Analysis of overall effects

The meta-analysis of CRF change scores in the 10 RCTs in-

dicated that acupuncture has a marked beneficial effect [stan-
dardized mean difference (SMD) = − 1.26, 95%CI (− 1.80, Balk 2009 + + + + – + ?
− 0.71), P < 0.01] (Fig. 4), supporting acupuncture as an
alternative therapy for CRF. The funnel plot indicated a mild Deng 2013 + + + + + + ?

publication bias, and the sensitivity analysis revealed that the Guo 2014 + ? ? ? + + ?
analysis model was relatively stable.
Johanston 2011 + – – – + + ?

Subgroup and sensitivity analysis Lian 2015 + + ? + + + +

Lim 2011 + + ? + + + ?
Treatment status
Mao 2014 + + ? + + + ?
The meta-analyses of the four trials with only off-treatment + + ? + + + –
Molassiotis 2007
patients [SMD = − 1.38, 95%CI (− 2.16, − 0.61), P < 0.01]
and of the six trials with patients undergoing anti-cancer treat- Molassiotis 2012 + – – – + + ?

ment [SMD = − 1.16, 95%CI (− 1.91, − 0.42), P < 0.01] both Molassiotis 2013 + – – – + + ?
demonstrated that acupuncture had a large effect on CRF com-
pared with the control group. Smith 2013 + + + + ? + ?

Fig. 3 Risk of bias assessment by individual trials

Malignancy type
(Table 2), and the model was verified to be relatively stable
Acupuncture significantly mitigated CRF in breast cancer pa- through a sensitivity analysis. A meta-analysis of four studies
tients [SMD = − 1.20, 95%CI (− 1.82, − 0.72), P < 0.01]

Fig. 2 Overall risk of bias assessment using the Cochrane tool

Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

422 Support Care Cancer (2018) 26:415–425

acupuncture Control Std. Mean Difference Std. Mean Difference

Study or Subgroup Mean SD Total Mean SD Total Weight IV, Random, 95% CI IV, Random, 95% CI
Balk 2009 -4.95 1.54 16 -3.9 1.39 11 4.2% -0.69 [-1.48, 0.11]
Deng 2013 -1.2 1.68 34 -1.2 1.82 40 4.4% 0.00 [-0.46, 0.46]
Guo 2014 -0.9 0.62 40 -0.93 0.79 40 4.5% 0.04 [-0.40, 0.48]
Guo 2014 2.66 0.56 40 3.17 0.36 40 4.4% -1.07 [-1.54, -0.60]
Guo 2014 2.66 0.56 40 1.51 0.72 40 4.4% 1.77 [1.25, 2.29]
Guo 2014 -0.9 0.62 40 1.27 0.7 40 4.3% -3.25 [-3.93, -2.57]
Johanston 2011 -4.2 1.32 5 -1.62 2.2 7 3.6% -1.26 [-2.56, 0.05]
Lian 2015 -3 0.26 22 -0.6 0.51 23 3.5% -5.79 [-7.17, -4.40]
Mao 2014 -1.4 0.66 21 0.5 0.38 22 4.0% -3.48 [-4.46, -2.51]
Mao 2014 -0.4 0.59 22 -0.1 0.36 23 4.4% -0.61 [-1.21, -0.01]
Mao 2014 -0.4 0.59 22 -0.5 0.61 22 4.4% 0.16 [-0.43, 0.76]
Mao 2014 -1.4 0.66 19 0.2 0.51 21 4.1% -2.68 [-3.55, -1.80]
Mao 2014 -1.4 0.66 21 -0.6 0.56 20 4.3% -1.28 [-1.96, -0.60]
Mao 2014 -1.4 0.66 19 -0.7 0.43 19 4.3% -1.23 [-1.93, -0.53]
Molassiotis 2007 -3.6 2.88 13 -0.9 2.78 13 4.2% -0.92 [-1.74, -0.11]
Molassiotis 2007 -5.9 2.75 13 -0.1 2.55 13 4.0% -2.12 [-3.11, -1.13]
Molassiotis 2012 -3.72 1.22 227 -0.62 1.24 75 4.5% -2.52 [-2.85, -2.19]
Molassiotis 2013 -0.85 1.12 65 -0.35 0.95 65 4.5% -0.48 [-0.83, -0.13]
Smith 2013 -3.2 2.56 9 -1.1 1.8 10 4.0% -0.92 [-1.87, 0.04]
Smith 2013 -3.2 2.56 9 -0.8 2.1 10 4.0% -0.98 [-1.95, -0.02]
Smith 2013 -2.5 2.14 9 -0.3 1.61 10 4.0% -1.12 [-2.10, -0.13]
Smith 2013 -3.1 2.4 9 -0.6 1.9 10 4.0% -1.11 [-2.09, -0.13]
Smith 2013 -3.1 2.4 9 -1.5 1.91 10 4.0% -0.71 [-1.64, 0.23]
Smith 2013 -2.5 2.14 9 -0.5 1.65 10 4.0% -1.01 [-1.98, -0.04]

Total (95% CI) 733 594 100.0% -1.26 [-1.80, -0.71]

Heterogeneity: Tau² = 1.67; Chi² = 390.12, df = 23 (P < 0.00001); I² = 94%
-4 -2 0 2 4
Test for overall effect: Z = 4.53 (P < 0.00001)
Favours experimental Favours control

Fig. 4 Overall effect of acupuncture on cancer-related fatigue

that included mixed tumor types (e.g., lung, liver, rectal, na-
sopharyngeal, gastric, ovarian, ampullary, cervical, endome-
trial, and pancreatic cancers) showed that acupuncture also
had a large effect on CRF management in these patients
[SMD = − 1.80, 95%CI (− 3.29, − 0.32), P = 0.02] (Table 2).
P = 0.02], and weekly for six weeks [SMD = − 1.66, 95%CI
(− 1.93, − 1.39), P < 0.01] had a large effect on CRF. Ten of
the 11 RCTs reported the detailed treatment acupoints and
insertion technique, and each trial included SP6 (Sanyinjiao)
and ST36 (Zusanli) among the treatment acupoints.

Acupuncture regimen Adverse events

Treatment with acupuncture only significantly mitigated the
CRF of cancer patients compared with sham acupuncture or
the usual treatment protocol [SMD = − 1.30, 95%CI (− 1.82,
− 0.78), P < 0.01] (Table 2). Acupuncture combined with
moxibustion had no overall effect on CRF changes
[SMD = − 1.58, 95%CI (− 3.44, − 0.28), P = 0.10]; however,
it showed a significant and large effect on CRF changes com-
pared with typical care at the end of the intervention
[SMD = − 4.44, 95%CI (− 6.92, − 1.96), P < 0.01].
Acupuncture for 20 min/session [SMD = − 0.95, 95%CI
(− 1.64, − 0.26), P < 0.01] or 30 min/session
[SMD = − 1.79, 95%CI (− 2.73, − 0.85), P < 0.01], three
times/week for two [SMD = − 2.16, 95%CI (− 2.73, − 1.59),
P < 0.01] or three weeks [SMD = − 0.93, 95%CI (− 1.30,
− 0.56), P < 0.01], twice weekly for two weeks and weekly
for six weeks [SMD = − 1.80, 95%CI (− 3.29, − 0.32),
Seven studies reported the occurrence of adverse events [8,
12, 13, 15–18], whereas four trials did not report any adverse
events. Only one study reported five adverse events [18];
however, no serious events were directly correlated with acu-
puncture (Table 1).
Discussion
This meta-analysis, which included 10 RCTs that provided a
total of 24 effect sizes, showed that acupuncture had a marked
beneficial effect [SMD = − 1.26, 95%CI (− 1.80, − 0.71),
P < 0.01] on reducing CRF. In addition, this meta-analysis
revealed that acupuncture significantly alleviated CRF in pa-
tients who were not receiving treatment and in patients who
were undergoing anti-cancer treatment (Table 2); this broad
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com

Support Care Cancer (2018) 26:415–425 423

Table 2 Subgroup analyses of

aerobic exercise on CRF changes Outcome type k ES Sample size SMD 95%CI P

AG CG L U

Treatment status
Off-treatment 4 10 136 149 − 1.38 − 2.16 − 0.61 < 0.01
Anti-cancer treatment 6 14 597 445 − 1.16 − 1.91 − 0.42 < 0.01
Malignancy type
Breast cancer 5 15 475 334 − 1.2 − 1.82 − 0.72 < 0.01
Mixed 4 5 98 100 − 1.80 − 3.29 − 0.32 0.02
Acupuncture type
Acupuncture only 5 16 496 353 − 1.30 − 1.82 − 0.78 < 0.01
Acupuncture combined with moxibustion 2 5 182 183 − 1.58 − 3.44 0.28 0.10
Control group
Sham acupuncture 5 10 165 168 − 0.83 − 1.27 − 0.39 < 0.01
Usual care 7 12 488 345 − 1.74 − 2.74 − 0.74 < 0.01
Number of total sessions
6 sessions 4 4 296 151 − 2.52 − 4.35 − 0.69 < 0.01
9 sessions 2 4 98 98 − 1.26 − 2.90 0.39 0.14
10 sessions 2 3 107 107 − 1.69 − 3.19 − 0.18 0.03
Acupuncture duration
20 min 6 15 566 426 − 0.95 − 1.64 − 0.26 < 0.01
30 min 4 9 167 168 − 1.79 − 2.73 − 0.85 < 0.01
Program length
2 weeks 2 2 35 36 − 3.92 − 7.51 − 0.33 0.03
6 weeks 3 4 279 135 − 1.10 − 2.59 0.40 0.15
8 weeks 2 3 47 49 − 2.01 − 3.49 − 0.53 < 0.01
Frequency of acupuncture (per week)
1 times 4 4 331 187 − 1.06 − 2.40 0.18 0.12
3 times 3 5 128 129 − 2.35 − 4.14 − 0.55 0.01

k number of studies, ES number of effect size, AG acupuncture group, CG control group, SMD standardized mean
difference effect size, L lower, U upper

effect of acupuncture is valuable for cancer patients. Most
patients with CRF become too weak to undergo other alterna-
tive therapies for this condition, such as exercise that requires
physical strength; however, acupuncture does not require ex-
tra autonomic activity. The implementation of acupuncture
therapy includes two basic elements: acupoint selection and
needle manipulation technique.
Of 11 trials, 10 described the treatment acupoints, and all
studies included SP6 (convergent acupoint of the liver, spleen,
and kidney Yin channels) and ST36 (stomach meridian), the
energy-associated points [19]. However, few studies other
than the two aforementioned investigations have described
the basis underlying the selection of acupoints. Four studies
[10, 12, 14, 15] selected only two to four energy-associated
points (LI4, SP6, ST36, and KI3) and showed a large effect on
CRF patients with breast cancer. In TCM, moxibustion is of-
ten combined with acupuncture; however, among the included
trials, only two studies [8, 9] examined the simultaneous use
of acupuncture and moxibustion. These studies involved
moxibustion on the energy-associated points RN4, RN6, and
RN12, which are located in the abdomen and are inappropriate
for deep puncture, particularly for patients with abdominal
tumors. Three trials [11, 13, 17] used the same insertion tech-
nique for these points that was utilized for non-abdominal
points. Based on our meta-analysis, the effect of acupuncture
integrated with moxibustion on changes in CRF was statisti-
cally significant compared with typical care and slightly dif-
fered from the effect of American ginseng. KI1 located in the
planta pedis, the Jing point of the kidney, is effective for
chronic fatigue syndrome [20] and is recommended for CRF
management. Only one trial [8] selected KI1 integrated with
other energy-associated points. Moreover, other trials did not
use similar acupoints compared with this trial except KI1;
therefore, we cannot determine the exact effect of KI1.
Seven trials [8–10, 13, 15, 16, 18] described the insertion
technique as Bthe needle was inserted until ‘De Qi’, a sensa-
tion of soreness, tingling, etc.,^ which is the classic puncture
standard for acupuncture in TCM. Two trials [12, 14]
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com
424
Bpunctured perpendicularly with a depth of 0.5 to 1 inch.^
One trial [17] used a Bneedle grab^ sensation as the insertion
criterion. The remaining trial [11] used a Park device, and
Bneedles were inserted to depth according to standard texts^
without any interpretation.
In addition, the duration, frequency, and program length of
acupuncture were also important. However, the meta-analysis
indicated that the alleviation effect on CRF depended on only
the number of acupuncture sessions and not duration, frequen-
cy, or program length. Four trials [8, 13, 14, 18] implementing
six sessions of acupuncture on regimen A (20 or 30 min, three
times per week for two weeks) or regimen B (20 min, weekly
for six weeks) showed that both acupuncture regimens with
six sessions mitigated CRF. The two trials [9, 11]
implementing nine sessions of acupuncture on regimen A
(20 min, three times per week for three weeks) or regimen B
(20 min, twice per week for three weeks and weekly for three
weeks) indicated that both acupuncture regimens with nine
sessions mitigated CRF. Two trials [10, 12] implementing 10
sessions of acupuncture on regimen A (30 min, twice per
week for two weeks and weekly for six weeks) or regimen
B (20 min, weekly for 10 weeks) showed that both acupunc-
ture regimens with 10 sessions mitigated CRF. This finding
may suggest that clinical professionals can formulate an acu-
puncture plan based on the specific situation and convenience
of patients.
Limitations of the current study
Although a comprehensive review of literature related to acu-
puncture for CRF was performed, regardless of cancer type
and patients’ treatment statuses, there remain certain limita-
tions that must be considered when interpreting the results of
this study. First, among the 10 trials included in the meta-
analysis, three trials had a sample of less than 30 subjects.
Second, it is known that various symptoms, such as pain, sleep
disturbance, nutrition, and activity level, among others, can
influence CRF level; however, the included RCTs did not
report variations in these factors that contribute to CRF, which
may have significantly influenced study outcomes. Therefore,
further studies with high-quality methodology, larger sample
sizes, and a multi-center design are needed.
Relative to existing meta-analyses of acupuncture for CRF
[21, 22], this meta-analysis included 3 additional newly pub-
lished RCTs with relatively large sample sizes and relatively
high methodological quality; these RCTs provided 11 addi-
tional effect sizes. Therefore, compared with prior researchers
who addressed this topic, we performed more detailed sub-
group analyses in accordance with the clinical heterogeneity
determined based on the number of included effect sizes; thus,
our conclusions are more convincing and can provide a more
meaningful reference for clinical professionals to use when
formulating acupuncture regimens to alleviate CRF.
Support Care Cancer (2018) 26:415–425
Conclusions
Acupuncture is effective for CRF management and should be
recommended as a beneficial alternative therapy for CRF pa-
tients, particularly for breast cancer patients and those current-
ly undergoing anti-cancer treatment. The causes of CRF
should be identified before the formulation of the acupuncture
regimen. Energy-associated points are essential for the
acupoint selection, and other points corresponding to the spe-
cific viscera or symptoms should be included. Due to the
location of the latter acupoints, i.e., RN4, RN6, RN12, and
KI1 (bilateral), moxibustion is more suitable than acupunc-
ture. When formulating the acupuncture regimen, the patient’s
specific situation and convenience should be prioritized over
the duration, frequency, and program length of acupuncture.
Author’s contributions Zhang Y and Tian L performed the meta-
analysis and wrote the first draft of the manuscript. Li HL and Hu Y
supervised the work. Tian L and Lin L revised the final manuscript. All
authors read and approved the final manuscript.
Compliance with ethical standards
Ethical approval For this type of study, formal consent is not required.
Conflict of interest The authors declare that they have no conflicts of
interest.
Grant This study was supported by the Suzhou Science and
Technology Development Project (SYS 201526).
Appendix
Table 3 A detailed search strategy for Medline
#1: Acupunct*OR acupuncture therapy
#2: fatigue .mp OR asthenia. mp
#3: cancer*.mp OR carcino*.mp OR tumour*.mp OR tumor*.mp OR
malignan*.mp
#4: Randomized Controlled Trial[Ptyp]
#5: animals[noexp] NOT humans
#6: #1 AND#2 AND#3 AND#4NOT#5
References
1. Tian L, Lin L, Li HL, Chen KJ, Zhang XJ, Qian SJ, Hu Y (2016)
Prevalence and associated factors of cancer-related fatigue among
cancer patients in eastern China. Oncologist. https://doi.org/10.
1634/the oncologist. 2015-0537
2. Hofman M, Ryan JL, Figueroa-Moseley CD, Jean-Pierre P,
Morrow GR (2007) Cancer-related fatigue: the scale of the
Licenciado para - Rodrigo Aquilini de Barros - 44351324864 - Protegido por Eduzz.com
Support Care Cancer (2018) 26:415–425
problem. Oncologist 12:4–10. https://doi.org/10.1634/
theoncologist.12-S1-4
3. Scott JA, Lasch KE, Barsevick AM, Piault-Louis E (2011) Patients’
experiences with cancer-related fatigue: a review and synthesis of
qualitative research. Oncol Nurs Forum 38:E191–E203. https://doi.
org/10.1188/11.onf.e191-e203
4. Lehmann H (2013) Acupuncture in ancient China: how important
was it really? J Integr Med 11:45–53. https://doi.org/10.3736/
jintegrmed2013008
5. Ma Z (2010) Analysis of syndrome and associated factors of CRF
in breast cancer patients. Beijing University of Chinese Medicine,
Beijing
6. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group
(2009) Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. Ann Intern Med 151:264–269.
https://doi.org/10.7326/0003-4819-151-4-200908180-00135
7. Furlan AD, Pennick V, Bombardier C, van Tulder M, Editorial
Board, Cochrane Back Review Group (2009) 2009 updated method
guidelines for systematic reviews in the Cochrane Back Review
Group. Spine (Phila Pa 1976) 34:1929–1941. https://doi.org/10.
1097/BRS.0b013e3181b1c99f
8. Lian WL (2015) Research on TCM differentiation and acupuncture
therapy on cancer-related fatigue. Guangzhou University of
Chinese Medicine, Guangzhou
9. Guo LY, Liu SL, Peng Y, Yu H, Li Q, Wang DP (2014) Effect of
acupuncture therapy on cancer-related fatigue of patients with gy-
necologic malignant tumor after chemotherapy. Clin J Tradit Chin
Med 30:67–70
10. Mao JJ, Farrar JT, Bruner D, Zee J, Bowman M, Seluzicki C,
DeMichele A, Xie SX (2014) Electroacupuncture for fatigue, sleep,
and psychological distress in breast cancer patients with aromatase
inhibitor-related arthralgia: a randomized trial. Cancer 120:3744–
3751. https://doi.org/10.1002/cncr.28917
11. Smith C, Carmady B, Thornton C, Perz J, Ussher JM (2013) The
effect of acupuncture on post-cancer fatigue and well-being for
women recovering from breast cancer: a pilot randomised con-
trolled trial. Acupunct Med 31:9–15. https://doi.org/10.1136/
acupmed-2012-010228
12. Molassiotis A, Bardy J, Finnegan-John J, Mackereth P, Ryder WD,
Filshie J, Ream E, Eaton D, Richardson A (2013) A randomized,
controlled trial of acupuncture self-needling as maintenance therapy
425
for cancer-related fatigue after therapist-delivered acupuncture. Ann
Oncol 24:1645–1652. https://doi.org/10.1093/annonc/mdt034
13. Deng G, Chan Y, Sjoberg D, Vickers A, Yeung KS, Kris M, Straus
D, Cassileth B (2013) Acupuncture for the treatment of post-
chemotherapy chronic fatigue: a randomized, blinded, sham-
controlled trial. Support Care Cancer 21:1735–1741. https://doi.
org/10.1007/s00520-013-1720-z
14. Molassiotis A, Bardy J, Finnegan-John J, Mackereth P, Ryder DW,
Filshie J, Ream E, Richardson A (2012) Acupuncture for cancer-
related fatigue in patients with breast cancer: a pragmatic random-
ized controlled trial. J Clin Oncol 30:4470–4476. https://doi.org/10.
1200/jco.2012.41.6222
15. Johnston MF, Hays RD, Subramanian SK, Elashoff RM, Axe EK,
Li JJ, Kim I, Vargas RB, Lee J, Yang L, Hui KK (2011) Patient
education integrated with acupuncture for relief of cancer-related
fatigue randomized controlled feasibility study. BMC Complement
Altern Med 11:49. https://doi.org/10.1186/1472-6882-11-49
16. Lim JTW, Wong ET, Aung SKH (2011) Is there a role for acupunc-
ture in the symptom management of patients receiving palliative
care for cancer? A pilot study of 20 patients comparing acupuncture
with nurse-led supportive care. Acupunct Med 29:173–179. https://
doi.org/10.1136/aim.2011.004044
17. Balk J, Day R, Rosenzweig M, Beriwal S (2009) Pilot, randomized,
modified, double-blind, placebo-controlled trial of acupuncture for
cancer-related fatigue. J Soc Integr Oncol 7:4–11
18. Molassiotis A, Sylt P, Diggins H (2007) The management of
cancer-related fatigue after chemotherapy with acupuncture and
acupressure: a randomised controlled trial. Complement Ther
Med 15:228–237. https://doi.org/10.1016/j.ctim.2006.09.009
19. Cheng XN (2009) Chinese acupuncture and moxibustion, 3rd edn.
Foreign Language Press, Beijing
20. Chen XQ, Jiang ZX, Xu ZP (2013) Effect of Yongquan moxibus-
tion on 38 patients with chronic fatigue syndrome and sleep disor-
ders. Chin Acupunct Mox 33:450
21. Zeng Y, Luo T, Finnegan-John J, Cheng AS (2014) Meta-analysis
of randomized controlled trials of acupuncture for cancer-related
fatigue. Integr Cancer Ther 13:193–200. https://doi.org/10.1177/
1534735413510024
22. He XR, Wang Q, Li PP (2013) Acupuncture and moxibustion for
cancer-related fatigue: a systematic review and meta-analysis.
Asian Pac J Cancer Prev 14:3067–3074. https://doi.org/10.7314/
APJCP.2013.14.5.3067