acupuncture for chronic low back pain in long-term follow ... · acupuncture versus no treatment)...
TRANSCRIPT
Acupuncture for Chronic Low Back Painin Long-Term Follow-Up: A Meta-Analysis
of 13 Randomized Controlled Trials
Mai Xu,* Shi Yan,† Xu Yin,§ Xiuyang Li,‡ Shuguang Gao,* Rui Han,¶
Licheng Wei,* Wei Luo* and Guanghua Lei*
*Department of Orthopaedics, Xiangya Hospital
Central South University, Changsha, Hunan 410008, China
†The Institute of Documents on Traditional Chinese Medicine
‡Department of Acupuncture and Moxibustion
Shandong University of Traditional Chinese Medicine
Jinan, Shandong 250000, China
§Department of Pathology in Basic Medicine
Beijing University of Traditional Chinese Medicine
Chao Yang, Beijing 100029, China¶Department of Orthopaedics, Qingdao Haici Hospital
Qingdao, Shandong 266000, China
Abstract: Chronic low back pain is one of the most common reasons that people seek medicaltreatment, and the consequent disability creates a great financial burden on individuals andsociety. The etiology of chronic low back pain is not clear, which means it is often refractoryto treatment. Acupuncture has been reported to be effective in providing symptomatic reliefof chronic low back pain. However, it is not known whether the effects of acupuncture aredue to the needling itself or nonspecific effects arising from the manipulation. To determinethe effectiveness of acupuncture therapy, a meta-analysis was performed to compare acu-puncture with sham acupuncture and other treatments. Overall, 2678 patients were identifiedfrom thirteen randomized controlled trials. The meta-analysis was performed by a randommodel (Cohen’s test), using the I-square test for heterogeneity and Begg’s test to assess forpublication bias. Clinical outcomes were evaluated by pain intensity, disability, spinalflexion, and quality of life. Compared with no treatment, acupuncture achieved better out-comes in terms of pain relief, disability recovery and better quality of life, but these effectswere not observed when compared to sham acupuncture. Acupuncture achieved betteroutcomes when compared with other treatments. No publication bias was detected.
Correspondence to: Dr. Guanghua Lei, Department of Orthopaedics, Xiangya Hospital, Central South University.No. 87 XiangYa Road, Changsha, Hunan 410008, China. Tel: (þ86) 731-8432-7326, Fax: (þ86) 731-8432-7354,
E-mail: [email protected]
The American Journal of Chinese Medicine, Vol. 41, No. 1, 1–19© 2013 World Scientific Publishing Company
Institute for Advanced Research in Asian Science and MedicineDOI: 10.1142/S0192415X13500018
1
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Acupuncture is an effective treatment for chronic low back pain, but this effect is likely to beproduced by the nonspecific effects of manipulation.
Keywords: Meta-Analysis; Chronic Low Back Pain; Efficacy; Disability; Acupuncture.
Introduction
Low back pain, which carries a 70–85% lifetime risk in the general population (Andersson,1999), is been one of the most common reasons for people to seek medical treatment (Hartet al., 1995; Deyo et al., 2006). It also creates a great financial burden as a result of theconsequent disability and the utilization of health services (Frymoyer and Cats-Baril, 1991;Frymoyer, 1992; Waddell, 1996; 2010, Apeldoorn et al., 2010; Wieser et al., 2010;Guidelines for lower back pain treatment ignored, 2010). Chronic low back pain is a poorlydefined condition as the underlying pathological causes are often unclear (Mayer et al.,2010). Some physiotherapists and researchers have argued that the chronic disability in lowback pain should be primarily considered as a psychosocial dysfunction (Cai et al., 2007;Astfalck et al., 2010). Conventional medication is often ineffective in its treatment,whereas alternative therapies have sometimes been shown to achieve excellent outcomes interms of both pain control and rehabilitation (Kumar et al., 2009; Witt et al., 2009;Sherman et al., 2010).
Acupuncture is an ancient therapy that originates from traditional Chinese philosophy,where the disease is considered to occur as a consequence of the imbalance between the Yinand Yang forces in the body. Inserting needles into the specific acupoints on the meridianlines can restore the balance of the body. Many studies have confirmed the effect of acu-puncture on pain (Kaptchuk, 2002; Thicke et al., 2011). However, most of these findingscannot be replicated due to insufficient sample sizes or limitations in the methodology, andthe utility of acupuncture in modern medicine remains controversial. Meta-analyses havetherefore been applied to explore the effects of pain relief. Various systematic reviewsregarding the use of acupuncture for back pain have been published (Ernst and White, 1998;Van Tulder et al., 1999; Furlan et al., 2005a, 2005b; Manheimer et al., 2005; Yuan et al.,2008; Kong et al., 2009; Chou, 2010; Linde et al., 2010; Rubinstein et al., 2010; Furlan etal., 2012). Some have confirmed that acupuncture has a marginally greater effect in relievingpain compared to sham acupuncture and no additional treatment (Manheimer et al., 2005).Others showed that acupuncture could only improve pain relief and functional improvementimmediately after the sessions and within a short-term follow-up period, compared to notreatment or sham therapy (Furlan et al., 2005b). Although the local and endocrine controlmechanisms of acupuncture have been clearly explained (Luu and Boureau, 1989; Chu,2002; Cai et al., 2009; Wang et al., 2010), its impact on chronic low back pain has inevitablybeen attributed to non-specific effects (Linde et al., 2010).
We raise the null hypothesis that acupuncture is equally as effective as blank treatments,sham acupuncture, or conventional care, as well as other alternative therapies, such astranscutaneous electrical nerve stimulation and massage. We attempted to confirm this
2 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
hypothesis with a meta-analysis of the literature, which demonstrated that acupuncture iseffective for chronic low back pain due to the non-specific effects that arise from themanipulation.
Materials and Methods
Review Objective
This meta-analysis is reported according to updated methodological guidelines for sys-tematic reviews published by the Cochrane back review group (2009) (Chou, 2010).The following comparisons were investigated: (1) acupuncture compared with blanktreatment in a relevant comparison (i.e., a comparison of acupuncture with another treat-ment versus the same other treatment alone) or an irrelevant comparison (a comparison ofacupuncture versus no treatment) to test whether acupuncture is useful for the treatment ofchronic low back pain; (2) acupuncture compared with sham acupuncture to test thenonspecific effects that arise from manipulation; and (3) acupuncture compared with otherconventional or alternative treatments to test whether acupuncture is superior to these othertreatments.
Search Strategy
Two reviewers independently performed a computer-aided search on Pubmed/Medline,AMED, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), ISIProceedings for Conference Abstracts, International Standard Randomized Controlled TrialNumber (ISRCTN) Register and the meta-Register for Randomized Controlled Trials(mRCT), China National Knowledge Infrastructure (CNKI) and the Wan Fang database upto 25 January 2012. A search of specific databases was also performed, included theComplementary and Alternative Medicine Specialist Library (from the National Library ofHealth, UK), the Physiotherapy Evidence Database (PEDro), and the Cumulative Index ofNursing and Allied Health (CINAHL).
In order to find as many studies as possible, we used a loose search strategy. Searchterms included “acupuncture”, “low back pain”, “needle”, “back pain” with the limits setfor randomized controlled trials (RCTs). The terms were searched both as free terms andMedical Subject Headings (MeSH) terms and connected by the Boolean operators “AND”and “OR”. Titles and abstracts of the potentially involved articles were reviewed first, todecide whether the studies contained any relevant information. If the trial was deemed to beeligible by either author, the full text would be obtained for further review. Keywords ofthe references of all relevant publications were traced, including systematic reviews, meta-analyses and identified RCTs, in order to retrieve citations omitted by the electronic search.We then sent a copy of the selected studies to all authors of the studies involved to ask ifthey were acknowledged in any other published trials. The search was repeated near thecompletion of the data analysis (15 February 2012) to ensure an integrated investigation.Figure 1 details the trial search flow of our meta-analysis.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 3
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Inclusion Criteria
We included therapeutic RCTs, and pain intensity was established as the primary outcome,while the secondary outcomes were the indexes related to the clinical outcome (e.g.,disability, quality of life and spinal flexion). After the literature review, we found that thedefinition of the term “chronic” in terms of low back pain was often confusing. Thus, weprimarily included articles in the meta-analysis that defined “chronic patients” as those whohad suffered from pain for at least the last three weeks (Rubinstein et al., 2010), and thensecondarily those that had suffered pain for 12 weeks (Chou, 2010). “Acupuncture” wasdefined as a process wherein needles are penetrated into the skin without an injection,accompanying a definite feeling of “De Qi” (an energetic feedback that the practitionerfeels when reaching the correct placement of an acupuncture needle). “Sham” acupuncturewas defined as either puncturing a location near the acupoint with tingling only but not “DeQi”, or simulated acupuncture technique using a toothpick or other needle-like object in theneedle guidetube (Cherkin et al., 2009). This has been found to be effective in acupuncture-naïve patients with back pain (Sherman et al., 2002). Long-term follow-up refers to out-comes that were measured between four weeks and one year (Chou, 2010).
When a single same trial was described in more than one publication, we only includedthe latest report. We excluded studies with non-clinical outcome measures (e.g., a study onthe mechanism), comparisons between the various acupuncture methods (e.g., moxibus-tion,electroacupuncture, warm needles, with or without Thermal Design Power), three-dimensional finite element analysis, animal studies, case reports, comparative studieswithout randomization, expert opinions, editorials, letters, practice guidelines and reviews.
Figure 1. Selection process of trials for the meta-analysis to compare acupuncture with blank, sham acupuncture,
or other treatments for chronic low back pain. RCT¼ randomized controlled trial.
4 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Risk of Bias and Studies Quality Evaluation
The risk of bias was appraised in accordance with the Cochrane Back Review GroupCriteria List for Methodologic Quality Assessment of Randomized, Controlled Trials (vanTulder et al., 2003). The report quality of the acupuncture intervention assessment wasmade based upon revised Standards for Reporting Interventions in Clinical Trials ofAcupuncture (STRICTA). Disagreements or inconsistencies were resolved through dis-cussion in cooperation with a third author (XY L).
Data Extraction
Two authors (SY and XY) independently extracted data from each included study induplicate with the use of a standardized form. Data for general information (author,publication year), mean follow-up duration, minimal or mean pain duration, mean patientage, sex distribution and the number of patients were listed. The sample size was halved ifthe acupuncture group was compared with two treatments. Again, disagreements weresettled by discussion between the review authors and if necessary with help from a thirdauthor.
Statistical Methods
We entered eligible trials into Stata/SE 11.0 software. The I-square score was used toquantify heterogeneity between trials, and > 50% was considered as significantly hetero-geneous. The trial results were factitiously divided into primary outcome measures (painintensity) and secondary outcome measures (pain disability, spine flexion and quality oflife). When an interval data was the only available data to describe the discrete trend, weused approximations to estimate standard deviation (SD) as 95% confidence interval (CI)divided by “4” or (75–25% CI) divided by “2”. We employed a random model to meta-analyze the end points when a high percentage was obtained for I2 (> 50%). The standardmean difference (SMD) was obtained as the effect size by Cohen’s test and associated 95%CIs. The effect size was defined as 0.20 for small, 0.50 for medium, and 0.80 for largeeffects (Cohen, 1992). Stratified analysis by treatment was used to determine whether theoutcomes were stable. For pain intensity, pain disability and spinal flexion, negative effectsizes indicated a beneficial effect of acupuncture, while quality of life was antithetical.Begg’s rank test was used to assess publication bias. Consistency in direction was definedas when 75% or more of the studies showed either a benefit or no benefit (Chou, 2010).
Results
Study Characteristics of Eligible Studies and Quality Assessment
We considered 132 potentially relevant references. Sixty-six studies were excluded afterthe titles and abstracts were reviewed. Twenty-three studies were excluded because of
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 5
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
duplication and 34 more were excluded as a result of an evaluation index of no interest(e.g., adverse events and psychological distress). A small amount of the index could not bemerged (Coan et al., 1980) or had an inappropriate study design, such as analgesic intakeand activity improvement (Mendelson et al., 1983; Giles and Muller, 2003). An additionalmanual search of reference lists generated nine further citations that were not obtained fromthe electronic search, of which five were excluded because of the language (Korean). Afterthe application of our inclusion and exclusion criteria, 2678 patients were identified from13 articles (Table 1) (Grant et al., 1999; Leibing et al., 2002; Molsberger et al., 2002; Kerret al., 2003; Meng et al., 2003; Yeung et al., 2003; Brinkhaus et al., 2006; Thomas et al.,2006; Haake et al., 2007; Szczurko et al., 2007; Cherkin et al., 2009; Itoh et al., 2009;Zaringhalam et al., 2010). Of the thirteen trials, one was from Northern Ireland, one fromHong Kong, one from Japan, one from Canada, one from Iran, two from the UnitedKingdom, two from the USA, and four from Germany.
The demographic data are shown in Table 1. The acupuncture invention assessment andrisk of bias assessment are listed in Tables 2 and 3, respectively. Most of the involvedstudies had flaws in the reporting style for acupuncture, the depth of needle insertion,needle retention times, the setting and context of treatment, and the background of thepractitioner. The main biases that affected the results were performance bias and detectionbias. Begg’s rank test showed that there was no publication bias in terms of pain intensityand pain disability, with p values (continuity corrected) of 0.544 and 0.537, respectively.Figures 2A and 2B show the Begg’s funnel plots that visualize the tests for publication biasregarding pain intensity and disability as a result of pain.
Table 1. Details of Trials of Acupuncture versus Sham Acupuncture, Blank and Other Treatments forChronic Low Back Pain
Author Year Mean ofFollow-Up
Duration of Pain No. of Patients Mean Age Sex Ratio(M:F)
Leibing E 2002 9m NA* 131 48.1y 55:76Molsberger AF 2002 3m Mean 9.9 yr 186 50y 97:89Meng CF 2003 4w >12 weeks 47 71y 18:29
Kerr DP 2003 6m NA* 60 41y 28:32Brinkhaus B 2006 52w 14.7 yr 298 58.8y 96:202Thomas KJ 2006 24m NA* 239 NA* 94:145Cherkin DC 2009 52w 3–12mo 638 47y 242:396
Yeung CKN 2003 3m NA* 52 53y 9:43Grant DJ 1999 3m >6mo 60 74y 6:54Kazunori I 2009 12w >6mo 26 > 65y 9:17
Szczurko O 2007 12w >6 weeks 75 46.6y NA*Haake M 2007 6m 8 years 1162 50y 470:692Zaringhalam J 2010 10w NA* 80 50–60y all men
Note: *NA¼ data not available.
6 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Table 2. Acupuncture Intervention Assessment According to the Revised Standards for Reporting Inter-ventions in Clinical Trials of Acupuncture (STRICTA)
1 2 3 4 5 6 TotalScore
1a 1b 1c 2a 2b 2c 2d 2e 2f 2g 3a 3b 4a 4b 5 6a 6b
Leibing E 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 17Molsberger AF 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 15Meng CF 0 1 1 1 1 0 1 1 1 1 1 1 1 0 0 1 1 13Kerr DP 0 1 0 1 0 0 1 1 0 1 1 1 1 1 0 1 1 11
Brinkhaus B 0 0 1 1 1 0 1 0 0 1 1 1 1 1 1 1 1 12Thomas KJ 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 1 1 5Cherkin DC 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 17
Yeung CKN 0 1 0 1 1 1 1 1 0 1 1 1 1 1 0 1 1 13Grant DJ 1 1 1 1 0 0 0 0 1 1 1 1 1 0 0 1 1 11Kazunori I 0 0 1 1 0 0 0 1 1 1 1 1 0 0 0 1 1 9
Szczurko O 0 0 1 1 1 1 1 0 1 1 1 1 1 0 0 1 1 12Haake M 0 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 14Zaringhalam J 0 1 0 1 1 1 1 1 1 1 1 1 1 0 0 1 1 13
1: ACUPUNCTURE RATIONALE1a: Style of acupuncture (e.g., Traditional Chinese Medicine (TCM), Japanese, Korean, Western medical,
Five Element, ear acupuncture, etc).1b: Reasoning for treatment provided, based on historical context, literature sources and/or consensus
methods, with references where appropriate.1c: Extent to which treatment was varied.
2: DETAILS OF NEEDLING2a: Number of needle insertions per subject per session (mean and range where relevant).2b: Names (or location if no standard name) of points used (uni-/bilateral).2c: Depth of insertion, based on a specified unit of measurement or on a particular tissue level.2d: Response sought (e.g., De Qi or muscle twitch response).2e: Needle stimulation (e.g., manual, electrical).2f: Needle retention time.2g: Needle type (diameter, length and manufacturer or material).
3: TREATMENT REGIMEN3a: Number of treatment sessions.3b: Frequency and duration of treatment sessions.
4: OTHER COMPONENTS OF TREATMENT4a: Details of other interventions administered to the acupuncture group (e.g., moxibustion, cupping, herbs,
exercises, lifestyle advice).4b: Setting and context of treatment, including instructions to practitioners, and information and explanations
to patients.
5: PRACTITIONER BACKGROUND5: Description of participating acupuncturists (qualification or professional affiliation, years in acupuncture
practice, other relevant experience).
6: CONTROL OR COMPARATOR INTERVENTIONS6a: Rationale for the control or comparator in the context of the research question, with sources that justify the
choice(s).6b: Precise description of the control or comparator. If sham acupuncture or any other type of acupuncture-
like control is used, provide details as for items 1–3 above.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 7
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Primary Meta-Analysis
Our primary method was to compare the outcome of acupuncture versus blank treatments,sham acupuncture and other treatments as reported in all 13 involved studies using meta-analysis.
Pain intensity. Figure 3 shows the results of pain intensity changes. Pain intensity wasreported in 18 studies from 13 articles, in which eight compared acupuncture with shamacupuncture, five with a group blank and five with other treatments. Visual analog scales(VAS) were employed in nine articles; the studies by Cherkin et al. (2009) and Yeung et al.(2003) used the numerical rating scale, Haake et al. (2007) used the Von Korff ChronicPain Grade Scale and Thomas et al. (2006) used the McGill present pain index, all of
Table 3. Risk of Bias Assessment of Involved Randomized Controlled Trials*
Single/Multicenter
PowerAnalysis
A B C D E F G H I J K
Leibing E Single center þ þ þ þ þ � þ ? þ � þ �Molsberger AF Single center � þ þ þ þ � � þ þ � þ þMeng CF Multicenter � þ þ þ � � � � þ � þ þKerr DP Single center þ þ þ þ þ � þ þ þ � þ �Brinkhaus B Multicenter þ þ þ þ þ � þ ? þ þ þ þThomas KJ Multicenter þ þ þ þ � � þ ? þ þ þ �Cherkin DC Multicenter þ þ þ ? þ þ þ þ ? þ þ þYeung CKN Single center þ ? þ þ � � þ þ þ þ þ þGrant DJ Single center � ? þ � � � þ ? þ þ þ �Kazunori I Single center � þ ? þ þ � þ þ ? � þ �Szczurko O Single center þ þ þ � � � þ þ þ þ þ þHaake M Multicenter � þ þ þ þ � þ þ þ þ þ þZaringhala MJ Single center � þ þ þ ? ? ? þ þ þ þ �*Risk of bias is assessed according to Cochrane Back Review Group Criteria List for Methodological
Quality Assessment of Randomized, Controlled Trials
A. Was the method of randomization adequate?
B. Was the treatment allocation concealed?
C. Were the groups similar at baseline regarding the most important prognostic indicators?D. Was the patient blinded to the intervention?
E. Was the care provider blinded to the intervention?
F. Was the outcome assessor blinded to the intervention?
G. Were co-interventions avoided or similar?
H. Was the compliance acceptable in all groups?
I. Was the dropout rate described and acceptable?
J. Was the timing of the outcome assessment in all groups similar?
K. Did the analysis include an intention-to-treat analysis?
þ indicates yes; − indicates no; ? indicates don’t know.
Internal validity criteria refer to characteristics of the study that might be related to selection bias (criteria A, B,H), performance bias (criteria C, D, I, J), attrition bias (criteria F, G), and detection bias (criteria E, K).
8 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
which are very similar scoring systems to the VAS. The heterogeneity across the subgroupstudies was high. The overall SMD was �0.43 (95% CI, �0.64 to �0.21); the blanktreatment and other treatment subgroups were significantly different, with a SMD of �0.64(95% CI, �1.13 to �0.14) and �0.49 (95% CI, �0.90 to �0.09), respectively. Althoughthe effect size showed that acupuncture could relieve pain more effectively than shamtreatment (SMD, �0.26; 95% CI, �0.56 to 0.05), this finding should not be consideredstatistically significant as the 95% CI crossed 0. The effect size was small when acu-puncture was compared with other treatments and moderate when compared with the blanktreatment. The subgroup effect sizes are shown in Fig. 4A.
(A) Begg’s funnel plot with pseudo 95% confidence limits for pain intensity
(B) Begg’s funnel plot with pseudo 95% confidence limits for disability as a result of pain
Figure 2. Begg’s funnel plot with pseudo 95% confidence limits for pain intensity (A) and disability as a result ofpain (B). The diameter of circle indicates the sample size of the individual study. *The point representing
Zaringhalam et al. (2010) is covered by the point representing Thomas KJ (2006), making the number of spots 17.SMD¼ standard mean difference; s.e. of: SMD¼ standard error of standard mean difference.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 9
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Disability as a result of pain. Figure 5 shows the results regarding disability as a resultof pain. Data regarding this parameter were reported in 12 studies from nine articles, inwhich four compared acupuncture with sham acupuncture, four compare acupuncture withthe blank treatment group and four with other treatments. The Roland disability ques-tionnaire was employed in the studies by Meng et al. (2003), Zaringhalam et al. (2010) andCherkin et al. (2009), while the pain disability index was used by Leibing et al. (2002) andBrinkhaus et al. (2006), Kazunori et al. (2009) Zaringhalam et al. (2010) and Thomas et al.(2006) used the Oswestry pain disability index, and the Aberdeen low back pain scale wasused by in Yeung et al. (2003). All the evaluation indexes were converted into a hundredpercentage point system.
The heterogeneity across the subgroup studies was low, with the exception of the othertreatment subgroup. The SMD of the overall cohort was �0.43 (95% CI, �0.66 to �0.21);Acupuncture in both the blank and other treatment subgroups gained better outcomes as theSMD was �0.58 (95% CI, �0.82 to �0.34) and �0.71 (95% CI, �1.29 to �0.21). Theeffect size of acupuncture was moderate compared with the other treatment and the blanktreatment groups. However, compared with the sham treatment group, acupuncture was notstatistically different. The subgroup effect sizes are shown in Fig. 4B.
Figure 3. Meta-analysis for pain intensity. The diamond represents the summary treatment effect over all analyzedstudies. A negative SMD represents that the acupuncture is superior to control intervention in relieving pain.
10 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Spinal flexion. Figure 6 shows the results regarding spinal flexion. Data were reportedin five studies, of which three compared acupuncture with sham acupuncture, one withblank treatment and one with other treatments. The fingertip-to-floor distance is employedin three studies while the range of movement of spinal flexion was assessed in two studies.Heterogeneity was low in the sham treatment subgroup (I-square¼ 0.0%) and high overall(I-square¼ 77.4%). The other treatments subgroup was the only one that significantlyfavored acupuncture, with an SMD of �1.04 (95% CI, �1.56 to �0.51). The effect size ofthis finding was high. The subgroup effect sizes are shown in Fig. 4C.
Quality of life. Quality of life was measured by physical and mental functioningscoring systems. Relevant data was available in five studies, of which three comparedacupuncture with sham acupuncture, one with blank treatment and one with othertreatments. The evaluation indexes used were SF-12 and SF-36. Figure 7 shows thecontribution of each study to the meta-analysis. Heterogeneity was low in the shamsubgroup (I2 ¼ 17:0%), but the overall effect was significantly heterogeneous (I2 ¼ 85:1%).When acupuncture was compared with sham acupuncture, the SMD was significantlyin favor of acupuncture (SMD, 0.22; 95% CI, 0.03 to 0.40) but the effect size wasstill smaller than the overall (SMD, 0.47; 95% CI, 0.15 to 0.78), blank treatment
Figure 4. Meta-analysis for pain intensity, disability, spinal flexion and quality of life in a random model (Cohen’stest). Stratification is analyzed by different interventions. (A) Pain intensity of patients who suffered from pain forat least the last 3 weeks. (B) Disability of patients who suffered from pain for at least the last 3 weeks. (C) Spinal
flexion of patients who suffered from pain for at least the last 3 weeks. (D) Quality of life of patients who sufferedfrom pain for at least the last 3 weeks. (E) Pain intensity of patients who suffered from pain for at least the last 12weeks. (F) Disability of patients who suffered from pain for at least the last 12 weeks. (G) Spinal flexion of
patients who suffered from pain for at least the last 12 weeks. (H) Quality of life of patients who suffered from painfor at least the last 12 weeks.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 11
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Figure 5. Meta-analysis for disability. The diamond represents a summary of the treatment effect across allanalyzed studies. A negative SMD demonstrates that acupuncture is superior to the control intervention forrecovering from disability.
Figure 6. Meta-analysis for spinal flexion. The diamond represents a summary of the treatment effect across all
analyzed studies. The overall SMD is not statistically significant as the 95% confidence interval includes 0. Thiscorresponds to a p value > 0.05.
12 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
(SMD, 0.87; 95% CI, 0.54 to 1.21) and other treatments (SMD, 0.71; 95% CI, 0.53 to0.89) (Fig. 4D).
Secondary Meta-Analysis
We secondarily meta-analyzed the outcomes of acupuncture versus blank, sham acu-puncture and other treatments as reported in the studies that clearly defined chronic pain aspain which lasted longer than 12 weeks (Grant et al., 1999; Molsberger et al., 2002; Menget al., 2003; Brinkhaus et al., 2006; Haake et al., 2007; Cherkin et al., 2009; Itoh et al.,2009). The results are listed in Figs 4E–4H. Compared with sham acupuncture, acu-puncture showed no superior benefit in the treatment of chronic low back pain in all fourevaluation indexes (pain intensity, disability, spinal flexion and quality of life). Further-more, acupuncture only showed a better outcome in terms of disability and quality of lifewhen compared with the blank and other treatments only in pain disability.
Consistency in Direction
The consistency in direction of pain intensity and quality of life changes was poor, as only55.5% (10/18) and 60% (3/5) of the studies showed a trend that was consistent withsubgroup effects, respectively. However, the consistency was good for disability and spinalflexion, as 75% (9/12) and 100% (5/5) of the studies showed a trend that was consistentwith subgroup effects, respectively.
Figure 7. Meta-analysis for quality of life. The diamond represents a summary of the treatment effect across allanalyzed studies. A positive SMD demonstrates that the patients treated with acupuncture achieve a higher qualityof life when compared withthe controls.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 13
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Consistent results were obtained in 72% (8/11 and 13/18) of the studies in the blankand sham acupuncture groups, while consistency was poor in the other treatment group(6/11, 55%).
Discussion
Numerous systemic reviews and meta-analyses over the last two decades have investi-gated the effectiveness of acupuncture for the treatment of chronic low back pain. Whenacupuncture has been compared with no additional treatment or blank treatments, theresults have consistently shown that acupuncture is significantly more effective (Ernstand White, 1998; Furlan et al., 2005a, 2005b; 2012; Manheimer et al., 2005; Yuan et al.,2008; Linde et al., 2010; Rubinstein et al., 2010). However, the results are relativelyinconsistent when acupuncture is compared with sham acupuncture (Ernst and White,1998; Van Tulder et al., 1999; Furlan et al., 2005a, 2005b; 2012; Manheimer et al.,2005; Yuan et al., 2008; Kong et al., 2009; Linde et al., 2010; Rubinstein et al., 2010).Some studies found there is moderate evidence that acupuncture is more effective thansham acupuncture (Van Tulder et al., 1999; Furlan et al., 2005a, 2005b; Manheimeret al., 2005; Linde et al., 2010; Rubinstein et al., 2010), while other studies havepublished strong evidence of no significant difference between acupuncture and shamacupuncture (Ernst and White, 1998; Yuan et al., 2008; Kong et al., 2009; Furlan et al.,2012). Given the clinical heterogeneity of other treatments for chronic low back pain, itis not surprising that a consistent conclusion could not be made when comparing acu-puncture with other treatments: some are positive (Furlan et al., 2005a; Yuan et al.,2008; Rubinstein et al., 2010; Furlan et al., 2012) and some are negative (Van Tulderet al., 1999). Drawing a generalized conclusion is complicated by the fact that differentsystemic reviews and meta-analyses have used different inclusion and exclusion criteria,different definitions of “chronic pain” and “acupuncture”, and evaluated different typesof outcomes.
Acupuncture is Effective in Relieving Chronic Low Back Pain Comparedto Blank or no Treatment
Compared with blank or no treatment, this meta-analysis found that acupunctureachieves moderately better outcomes in the treatment of chronic low back pain. Thiscorresponds to the experience of acupuncturists and the mechanism of pain relief.An acutely painful stimulus can help to relieve ongoing pain, which is known as“counter irritation” (Manchikanti et al., 2009). This phenomenon explains why practi-cing acupuncture without an analgesic earns more benefits than with analgesia. Fur-thermore, the tangible analgesia induced by acupuncture is caused by a centralneurohumoral mechanism with delayed-onset characteristics (Cao, 2002). This indicatesthat acupuncture has direct and indirect effects on the area from which the nociceptiveinput originated.
14 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Acupuncture Effects are Produced by Nonspecific Effects Arising from Manipulation
Sham acupuncture provides the same sting as acupuncture but without the sense of “DeQi”. According to the theory of traditional Chinese medicine, acupuncture outside Mer-idian lines cannot disturb or harmonize the Yin and Yang (Guan and Liu, 2009). In otherwords, theoretically sham acupuncture should be of no analgesic or therapeutic benefit. Inour meta-analysis, acupuncture does not achieve better outcomes in terms of pain relief,disability rehabilitation or spinal flexion improvement compared with sham acupuncture.This indicates that the partial analgesic effect of acupuncture does not derive from thestimulation of the meridians, namely “De Qi”, but from the skin manipulation. Consideringthe effect sizes determined by the meta-analysis of these two subgroups in terms of painintensity, disability and spinal flexion, we can conclude that a majority of the analgesiceffect of acupuncture arises from other factors than “De Qi” itself. Consequently, it is notdifficult to understand why better qualified and experienced acupuncturists are not moresuccessful in treating patients with chronic low back pain (Witt et al., 2010).
Acupuncture Should Be Used in Combination with Other Treatments
Although the results of our meta-analysis show the superiority of acupuncture over othertreatments, we draw a very cautious conclusion that the therapeutic effect of acupuncturemay change along with different interventions of the other treatment subgroup. The mainindication of acupuncture is pain, and acupuncture, as an alternative treatment, has provenitself to be useful for treating pain. Even so, acupuncture is not as effective as some othertreatments, such as massage (Chou, 2010) and baclofen (Zaringhalam et al., 2010). Wetherefore recommend the use of acupuncture in combination with other treatments.
Chronic Low Back Pain for More than Three Weeks or More than 12 Weeks
We found that the effect of acupuncture in treating chronic low back pain was significantlylimited by the duration of the disease, with a much smaller effect observable when thedefinition of chronic low back pain encompassed those with pain for more than 12 weeks.This indicates that the effect of acupuncture declines with prolonged disease duration, andexplains the phenomenon that acupuncture often obtains a better outcome when treatingacute pain than chronic pain (van Tulder et al., 1999).
Limitations and Strengths of Our Meta-Analysis
Our study had several limitations. The principal limitation was the relatively finite outcomevariables of the involved RCTs, as we chose only four indexes to merge and ignored othermeaningful indexes, such as adverse events and back-specific functional status, as outlinedin Deyo et al. (1998). Another important limitation was the types of other treatmentsincluded in the analysis. In the stratified analysis, the other treatments subgroup was highlyheterogeneous as much clinical heterogeneity exists in the different interventions forchronic low back pain. We factitiously merged these, which undoubtedly resulted in a
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 15
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
synthetic effect of acupuncture compared with other treatments. As a result of resourcelimitations, we did not search other Asia databases besides Chinese publications; this mayhave reduced our estimates of the effect sizes because acupuncture RCTs published in Asiahave been shown to be positive in most cases (Vickers et al., 1998).
A major strength of this review was that we used a methodological guideline (Chou,2010) for systematic reviews of trials for treatments of neck and back pain to report ourmeta-analysis. Ideally, we would have liked to report step-by-step following this guideline;however, the included studies did not provide enough valid data to complete each part ofthe analysis required in the guidelines. Furthermore, the intervention of acupuncture waschecked and evaluated by STRICTA (MacPherson et al., 2010). Considering that thereporting of acupuncture trials is often disordered, we recommend the use of STRICTA asa checklist for authors in this field.
Conclusions
The current data shows that acupuncture is effective in providing long-term relief ofchronic low back pain, but this effect is produced by non-specific effects that arise fromskin manipulation. Further research is needed to evaluate which the effects of acupuncturecompared with other treatments. Moreover, acupuncture should be used in combinationwith other treatments in the treatment of chronic low back pain. In view of the oftenrefractory nature of chronic low back pain, comprehensive treatment plans, which mayinvolve acupuncture, are urgently needed.
References
Andersson, G.B. Epidemiological features of chronic low-back pain. Lancet 354: 581–585, 1999.Apeldoorn, A.T., R.W. Ostelo, H. van Helvoirt, J.M. Fritz, H.C. de Vet and M.W. van Tulder. The
cost-effectiveness of a treatment-based classification system for low back pain: design of arandomised controlled trial and economic evaluation. BMC Musculoskelet. Disord. 11: 58,2010.
Astfalck, R.G., P.B. O’Sullivan, L.M. Straker and A.J. Smith. A detailed characterisation of pain,disability, physical and psychological features of a small group of adolescents with non-specific chronic low back pain. Man. Ther. 15: 240–247, 2010.
Brinkhaus, B., C.M. Witt, S. Jena, K. Linde, A. Streng, S. Wagenpfeil, D. Irnich, H.U. Walther,D. Melchart and S.N. Willich. Acupuncture in patients with chronic low back pain: a ran-domized controlled trial. Arch. Intern. Med. 166: 450–457, 2006.
Cai, C., Y.H. Pua and K.C. Lim. Correlates of self-reported disability in patients with low back pain:the role of fear-avoidance beliefs. Ann. Acad. Med. Singap. 36: 1013–1020, 2007.
Cai, Y.Y., Z.S. Liu, S. Wang, Q.F. Wang, X.M. Cui and L. Qu. Influence of electroacupuncture ofmeridian acupoints on the related hormones of the hypothalamus-pituitary-adrenal axis in ratswith cerebral ischemia reperfusion injury [in Chinese]. Zhen Ci Yan Jiu 34: 297–303, 2009.
Cao, X. Scientific bases of acupuncture analgesia. Acupunct. Electrother. Res. 27: 1–14, 2002.Cherkin, D.C., K.J. Sherman, A.L. Avins, J.H. Erro, L. Ichikawa, W.E. Barlow, K. Delaney,
R. Hawkes, L. Hamilton, A. Pressman, P.S. Khalsa and R.A. Deyo. A randomized trial
16 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
comparing acupuncture, simulated acupuncture, and usual care for chronic low back pain.Arch. Intern. Med. 169: 858–866, 2009.
Chou, R. Low back pain (chronic). Clin. Evid. 2010: 1116, 2010.Chu, J. The local mechanism of acupuncture. Zhonghua Yi Xue Za Zhi 65: 299–302, 2002.Coan, R.M., G. Wong, S.L. Ku, Y.C. Chan, L. Wang, F.T. Ozer and P.L. Coan. The acupuncture
treatment of low back pain: a randomized controlled study. Am. J. Chin. Med. 8: 181–189,1980.
Cohen, J. A power primer. Psychol. Bull. 112: 155–159, 1992.Deyo, R.A., M. Battie, A.J. Beurskens, C. Bombardier, P. Croft, B. Koes, A. Malmivaara, M. Roland,
M. Von Korff and G. Waddell. Outcome measures for low back pain research. A proposal forstandardized use. Spine 23: 2003–2013, 1998.
Deyo, R.A., S.K. Mirza and B.I. Martin. Back pain prevalence and visit rates: estimates from U.S.national surveys, 2002. Spine 31: 2724–2727, 2006.
Ernst, E. and A.R. White. Acupuncture for back pain: a meta-analysis of randomized controlled trials.Arch. Intern. Med. 158: 2235–2241, 1998.
Frymoyer, J.W. Can low back pain disability be prevented? Baillieres Clin. Rheumatol. 6: 595–606,1992.
Frymoyer, J.W. and W.L. Cats-Baril. An overview of the incidences and costs of low back pain.Orthop. Clin. North. Am. 22: 263–271, 1991.
Furlan, A.D., M. van Tulder, D. Cherkin, H. Tsukayama, L. Lao, B. Koes and B. Berman. Acu-puncture and dry-needling for low back pain: an updated systematic review within the fra-mework of the cochrane collaboration. Spine 30: 944–963, 2005a.
Furlan, A.D., M.W. van Tulder, D.C. Cherkin, H. Tsukayama, L. Lao, B.W. Koes and B.M. Berman.Acupuncture and dry-needling for low back pain. Cochrane Database Syst. Rev. CD001351,2005b.
Furlan, A.D., F. Yazdi, A. Tsertsvadze, A. Gross, M. van Tulder, L. Santaguida, J. Gagnier,C. Ammendolia, T. Dryden, S. Doucette, B. Skidmore, R. Daniel, T. Ostermann andS. Tsouros. A systematic review and meta-analysis of efficacy, cost-effectiveness, and safety ofselected complementary and alternative medicine for neck and low-back pain. Evid. Based.Complement. Alternat. Med. 2012: 953139, 2012.
Giles, L.G. and R. Muller. Chronic spinal pain: a randomized clinical trial comparing medication,acupuncture, and spinal manipulation. Spine 28: 1490–1502; discussion 1502–1493, 2003.
Grant, D.J., J. Bishop-Miller, D.M. Winchester, M. Anderson and S. Faulkner. A randomizedcomparative trial of acupuncture versus transcutaneous electrical nerve stimulation for chronicback pain in the elderly. Pain 82: 9–13, 1999.
Guan, L.P. and C.Z. Liu. Progresses of study on correlativity of qi-arrival of needling with clinicaltherapeutic effect and its action mechanism [in Chinese]. Zhongguo Zhen Jiu 29: 945–948,2009.
Guidelines for lower back pain treatment ignored. For patients with low back pain, the result is higherrisk for adverse results and higher costs. Duke Med. Health News 16: 3, 2010.
Haake, M., H.H. Muller, C. Schade-Brittinger, H.D. Basler, H. Schafer, C. Maier, H.G. Endres, H.J.Trampisch and A. Molsberger. German Acupuncture Trials (GERAC) for chronic low backpain: randomized, multicenter, blinded, parallel-group trial with 3 groups. Arch. Intern. Med.167: 1892–1898, 2007.
Hart, L.G., R.A. Deyo and D.C. Cherkin. Physician office visits for low back pain. Frequency,clinical evaluation, and treatment patterns from a U.S. national survey. Spine 20: 11–19, 1995.
Kaptchuk, T.J. Acupuncture: theory, efficacy, and practice. Ann. Intern. Med. 136: 374–383, 2002.Kazunori, I., S. Itoh, Y. Katsumi and H. Kitakoji. A pilot study on using acupuncture and transcu-
taneous electrical nerve stimulation to treat chronic non-specific low back pain. Complement.Ther. Clin. Pract. 15: 22–25, 2009.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 17
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
Kerr, D.P., D.M. Walsh and D. Baxter. Acupuncture in the management of chronic low back pain: ablinded randomized controlled trial. Clin. J. Pain 19: 364–370, 2003.
Kong, J.C., M.S. Lee and B.C. Shin. Randomized clinical trials on acupuncture in Korean literature: asystematic review. Evid. Based. Complement. Alternat. Med. 6: 41–48, 2009.
Kumar, S., M.P. Negi, V.P. Sharma, R. Shukla, R. Dev and U.K. Mishra. Efficacy of two multimodaltreatments on physical strength of occupationally subgrouped male with low back pain. J. BackMusculoskelet. Rehabil. 22: 179–188, 2009.
Leibing, E., U. Leonhardt, G. Koster, A. Goerlitz, J.A. Rosenfeldt, R. Hilgers and G. Ramadori.Acupuncture treatment of chronic low-back pain — a randomized, blinded, placebo-controlledtrial with 9-month follow-up. Pain 96: 189–196, 2002.
Linde, K., K. Niemann, A. Schneider and K. Meissner. How large are the nonspecific effects ofacupuncture? A meta-analysis of randomized controlled trials. BMC Med. 8: 75, 2010.
Luu, M. and F. Boureau. Acupuncture in pain therapy: current concepts [in Chinese]. Thre Umsch.46: 518–525, 1989.
MacPherson, H., D.G. Altman, R. Hammerschlag, L. Youping, W. Taixiang, A. White and D. Moher.Revised STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA):extending the CONSORT statement. J. Altern. Complement. Med. 16: ST1-14, 2010.
Manchikanti, L., S.E. Glaser, L. Wolfer, R. Derby and S.P. Cohen. Systematic review of lumbardiscography as a diagnostic test for chronic low back pain. Pain Physician 12: 541–559, 2009.
Manheimer, E., A. White, B. Berman, K. Forys and E. Ernst. Meta-analysis: acupuncture for lowback pain. Ann. Intern. Med. 142: 651–663, 2005.
Mayer, J.M., S. Haldeman, A.C. Tricco and S. Dagenais. Management of chronic low back pain inactive individuals. Curr. Sports Med. Rep. 9: 60–66, 2010.
Mendelson, G., T.S. Selwood, H. Kranz, T.S. Loh, M.A. Kidson and D.S. Scott. Acupuncturetreatment of chronic back pain. A double-blind placebo-controlled trial. Am. J. Med. 74: 49–55, 1983.
Meng, C.F., D. Wang, J. Ngeow, L. Lao, M. Peterson and S. Paget. Acupuncture for chronic lowback pain in older patients: a randomized, controlled trial. Rheumatology 42: 1508–1517,2003.
Molsberger, A.F., J. Mau, D.B. Pawelec and J. Winkler. Does acupuncture improve the orthopedicmanagement of chronic low back pain— a randomized, blinded, controlled trial with 3 monthsfollow up. Pain 99: 579–587, 2002.
Rubinstein, S.M., M. van Middelkoop, T. Kuijpers, R. Ostelo, A. P. Verhagen, M.R. de Boer, B.W.Koes and M.W. van Tulder. A systematic review on the effectiveness of complementary andalternative medicine for chronic non-specific low-back pain. Eur. Spine J. 19: 1213–1228,2010.
Sherman, K.J., C.J. Hogeboom, D.C. Cherkin and R.A. Deyo. Description and validation of anoninvasive placebo acupuncture procedure. J. Altern. Complement. Med. 8: 11–19, 2002.
Sherman, K.J., D.C. Cherkin, A.J. Cook, R.J. Hawkes, R.A. Deyo, R. Wellman and P.S. Khalsa.Comparison of yoga versus stretching for chronic low back pain: protocol for the YogaExercise Self-care (YES) trial. Trials 11: 36, 2010.
Szczurko, O., K. Cooley, J.W. Busse, D. Seely, B. Bernhardt, G.H. Guyatt, Q. Zhou and E.J.Mills. Naturopathic care for chronic low back pain: a randomized trial. PloS. One 2: e919,2007.
Thicke, L.A., J.K. Hazelton, B.A. Bauer, C.W. Chan, E.A. Huntoon, P.J. Novotny, J.A. Sloan and D.L. Wahner-Roedler. Acupuncture for treatment of noncyclic breast pain: a pilot study. Am. J.Chin. Med. 39: 1117–1129, 2011.
Thomas, K.J., H. MacPherson, L. Thorpe, J. Brazier, M. Fitter, M.J. Campbell, M. Roman, S.J.Walters and J. Nicholl. Randomised controlled trial of a short course of traditional acupuncturecompared with usual care for persistent non-specific low back pain. BMJ 333: 623, 2006.
18 M. XU et al.
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.
van Tulder, M., A. Furlan, C. Bombardier and L. Bouter. Updated method guidelines for systematicreviews in the cochrane collaboration back review group. Spine 28: 1290–1299, 2003.
van Tulder, M.W., D.C. Cherkin, B. Berman, L. Lao and B.W. Koes. The effectiveness of acu-puncture in the management of acute and chronic low back pain. A systematic review withinthe framework of the Cochrane Collaboration Back Review Group. Spine 24: 1113–1123,1999.
Vickers, A., N. Goyal, R. Harland and R. Rees. Do certain countries produce only positive results?A systematic review of controlled trials. Control. Clin. Trials 19: 159–166, 1998.
Waddell, G. Low back pain: a twentieth century health care enigma. Spine 21: 2820–2825, 1996.Wang, S.J., B. Zhu, X.X. Ren and L.H. Tan. Experimental study on acupuncture activating the
gonadotropin-releasing hormone neurons in hypothalamus. J. Tradit. Chin. Med. 30: 30–39,2010.
Wieser, S., B. Horisberger, S. Schmidhauser, C. Eisenring, U. Brugger, A. Ruckstuhl, J. Dietrich,A.F. Mannion, A. Elfering, O. Tamcan and U. Muller. Cost of low back pain in Switzerland in2005. Eur. J. Health. Econ. 2010.
Witt, C.M., R. Ludtke, R. Baur and S.N. Willich. Homeopathic treatment of patients with chronic lowback pain: a prospective observational study with 2 years’ follow-up. Clin. J. Pain 25: 334–339, 2009.
Witt, C.M., R. Ludtke, K. Wegscheider and S.N. Willich. Physician characteristics and variation intreatment outcomes: are better qualified and experienced physicians more successful in treatingpatients with chronic pain with acupuncture? J. Pain 11: 431–435, 2010.
Yeung, C.K., M.C. Leung and D.H. Chow. The use of electro-acupuncture in conjunction withexercise for the treatment of chronic low-back pain. J. Altern. Complement. Med. 9: 479–490,2003.
Yuan, J., N. Purepong, D.P. Kerr, J. Park, I. Bradbury and S. McDonough. Effectiveness of acu-puncture for low back pain: a systematic review. Spine 33: E887–900, 2008.
Zaringhalam, J., H. Manaheji, A. Rastqar and M. Zaringhalam. Reduction of chronic non-specific lowback pain: a randomised controlled clinical trial on acupuncture and baclofen. Chin. Med. 5:15, 2010.
NONSPECIFIC EFFECTS OF ACUPUNCTURE TREATING CHRONIC LOW BACK PAIN 19
Am
. J. C
hin.
Med
. 201
3.41
:1-1
9. D
ownl
oade
d fr
om w
ww
.wor
ldsc
ient
ific
.com
by 8
3.93
.167
.56
on 0
2/24
/13.
For
per
sona
l use
onl
y.