Acupuncture for chronic nonspecific low back pain: An overview of systematic reviews

Available online at www.sciencedirect.com

ScienceDirect European Journal of Integrative Medicine 7 (2015) 94–107

Review article

Acupuncture for chronic nonspecific low back pain: An overview of systematic reviews Yingchun Zeng a , Joanne Wai-yee Chung b,∗ a

Department of Obstetrics and Gynecology, The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong Province, China b Department of Health and Physical Education, The Hong Kong Institute of Education, Hong Kong, China Received 1 July 2014; received in revised form 31 October 2014; accepted 2 November 2014

Abstract Introduction: Chronic nonspecific low back pain (cnLBP), which cannot be attributed to a specific pathology is very common. As a result acupuncture is frequently used by patients as a treatment option. This overview aimed to summarize and evaluate the available systematic reviews on the clinical effectiveness and cost-effectiveness of acupuncture for the management of cnLBP, and to identify the safety of acupuncture for the management of cnLBP. Methods: Systematic reviews of acupuncture and cnLBP were sourced from five databases. Publications between January 2003 and May 2014 were included for analysis. Quality appraisal of included systematic reviews was assessed by the Overview Quality Assessment Questionnaire. Results: Seventeen systematic reviews were included. Five found that acupuncture was more effective when compared with a no treatment/waiting list control, as there were eight systematic reviews and meta-analysis providing positive and consistent findings. Seven systematic reviews providing positive findings of the comparison of acupuncture to sham acupuncture/passive modality treatment. Three systematic reviews of multiple RCTs also indicated positive and consistent findings of the comparison of acupuncture plus an intervention vs an intervention alone. Overall, findings on the effectiveness of acupuncture for cnLBP were consistent. Conclusions: As there is a range of diverse acupuncture styles used for patients with cnLBP, future trials are needed to establish the standardization of acupuncture in terms of the length of treatment sessions, frequency of sessions, number of needles needed per treatment, placement of needle insertion, depth of needle insertion, and whether needle stimulation achieves De Qi. © 2014 Elsevier GmbH. All rights reserved. Keywords: Acupuncture; Chronic nonspecific low back pain; Overview

Introduction Back pain is regarded as one of the most common musculoskeletal complaints, and the second most common condition for patients seek primary care consultation as many people experiencing back pain during their lifetime [1,2]. The term nonspecific back pain is used to describe back pain that is not attributed to a specified pathology or symptom pattern [2]. A recent systematic review indicated that a global lifetime prevalence of low back pain is up to 63.2% [3]. Krismer and van

∗

Corresponding author. Tel.: +852 29488479; fax: +852 29486000. E-mail addresses: [email protected] (Y. Zeng), [email protected] (J.W.-y. Chung). http://dx.doi.org/10.1016/j.eujim.2014.11.001 1876-3820/© 2014 Elsevier GmbH. All rights reserved.

Tulder [4] reported an even higher lifetime prevalence of low back pain, which is 60–85% in Europe. The reported prevalence varies significantly depending on the definitions of low back pain used [5]. Low back pain may refer to pain, muscle tension or stiffness occurring between the costal margin and gluteal folds [6]. Rozenberg et al. [7] indicated that more than 90% of patients are categorized into nonspecific LBP cases. One of the main characteristics of nonspecific low back pain is its recurrent nature, which is an essential factor for predisposing the individuals to chronic illness [2,8]. While the diagnosis and treatment of low back pain has improved, disability arising from chronic nonspecific low back pain (cnLBP) appears to be increasing [9]. As cnLBP is significantly prevalent in the working population, it also results in obvious financial burden to society due to loss of working hours and ability to work. For

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

healthcare systems, cnLBP is one of the most costly medical conditions [10]. The direct healthcare cost of low back pain in United States is estimated as $25 billion annually, possibly over $100 million if direct and indirect healthcare expenses are taken together [11]. Choosing an efficacious, safe, and cost-effective treatment for low back pain patients can be challenging due to many stakeholders involving patients, health providers, policy makers, and third-party payers [12]. While pharmacologic treatment seems to be one of the most effective treatment options for cnLBP, its side effects arises safety concerns on drug therapies for cnLBP. Cassileth et al. [13] suggested that complementary therapies can be incorporated as an adjunct to pain management by enhancing or decreasing the need for pharmacologic treatment. In recent decades, complementary therapies are used widely for treating cnLBP. Acupuncture is the most common complementary therapies for cnLBP [14]. It is a safe and cost-effective treatment compared to medication, injection, and surgical therapies for cnLBP [15]. Aims The overview was aimed to summarize and evaluate the available systematic reviews on the clinical effectiveness and cost-effectiveness of acupuncture for the management of cnLBP, and to identify the safety of acupuncture for the management of cnLBP. Methods Data sources and searches Systematic reviews of acupuncture and chronic nonspecific low back pain were sourced from five databases: Medline, The Cochrane Library, Allied and Complementary Medicine Database (AMED), Scopus, and CAJ (Chinese Academic Journal) Full-text Database. The publications between January 2003 and May 2014 were included for analysis. The search was conducted on May 15, 2014 and articles published in English and Chinese were included. The search terms of “acupuncture”, “acupuncture therapy”; “low back pain”, “chronic”, “review”, “review literature”, “meta-analysis”, and “systematic” were included. The selection of studies was shown in Fig. 1. Inclusion and exclusion criteria Systematic reviews were included; the study population had chronic (duration of symptoms >12 weeks) nonspecific low back pain (pain no known underlying pathology or disease or related to pregnancy). Exclusion criteria were; those reviews which included populations with acute (<6 weeks duration) or sub-acute (3–12 weeks duration) low back pain, or specific low back pain caused by specific pathological entities such as infection, inflammatory disorders, systemic diseases or metastatic diseases.

95

Fig. 1. PRISMA flow diagram of included systematic reviews.

Interventions Acupuncture was defined as “a process involving that needles were required to be inserted into the skin (without an injection) at classical meridian points, extra points or Ah-shi points (painful points), accompanying a definite feeling of “De Qi” [16,17]. De Qi (arrival of energy) was “a sensation of numbness or distention sometimes generated by stimulating acupuncture needles by hand or with an electrical current” [18]. According to acupuncture theory, activation of De Qi may be indicated that acupuncture is reaching the correct placement and exerting potentially beneficial effects. Sham acupuncture was defined as “any intervention designed to make patients believing that he/she is receiving acupuncture by either puncturing a location near the acupoint with tingling only but not De Qi, or stimulated acupuncture technique using a toothpick or other needle-like object in the needle guidetube” [19]. Acupuncture that did not involve needle insertion such as laser acupuncture, or electro-acupuncture without needles, was excluded. Outcome measures The primary outcome was the effectiveness of acupuncture for cnLBP. The outcome measures were; pain intensity, patient global assessment of pain, and specific functional status related to cnLBP. The secondary outcome was the cost-effectiveness and the safety of acupuncture for cnLBP. Relevant outcome measures include the presence and frequency of adverse effects (e.g., abdominal pain) and withdrawals due to adverse events of acupuncture interventions. Quality assessment of the systematic reviews selected It is essential to include existing systematic reviews that adhere to high methodological standards [20]. According to Pieper et al. [21], the most commonly used assessment tool for overviews of systematic reviews were the Overview Quality Assessment Questionnaire (OQAQ). Two independent

96

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

reviewers assessed the methodological quality of each included review using the OQAQ list (YZ & JC). Any disagreement was resolved by discussion and referral to the papers for achieving consensus. Results Seventeen systematic reviews [16–18,22–35] were included in this review (Fig. 1). The characteristics of these included systematic reviews are summarized in Table 1. Description of the characteristics of included reviews Majority (n = 12) [16–18,22–27,29,33,35] aimed to investigate the effectiveness of acupuncture therapy for cLBP. The primary objective of five systematic reviews [28,30–32,34] was to summarize the evidence on the cost-effectiveness of acupuncture. There were two systematic reviews [25,33] examining the safety of acupuncture for cLBP. Some (n = 7) compared acupuncture with other CAM therapies [16–18,23,25,26,33]. Due to heterogeneity of the primary trials, only four of the systematic reviews [17,22,23,25] were combined the findings of the primary trials using meta-analysis. One systematic review [28] included a variety of studies with different designs such as randomized controlled trials (RCTs), case studies and observations studies. This systematic review included patients and acupuncture specialists as study subjects. Other systematic reviews only included one type of primary studies of RCTs. Outcome comparisons include acupuncture vs no treatment or inactive treatment (sham acupuncture, placebo medication) or active treatment (massage, spinal manipulation, medication, transcutaneous electrical nerve stimulation (TENS)). Other outcome comparisons included acupuncture plus an intervention vs an intervention alone. Outcome measures include pain intensity, functioning, adverse events and cost-effectiveness. Four meta-analysis [17,22,23,25] presented pain outcomes using a standardized mean difference or weighted mean difference. Other reviews presented pain outcomes using qualitative synthesis. Effectiveness of acupuncture on pain reductions for cnLBP Acupuncture vs no treatment or waiting list control. Xu et al. [17] conducted a meta-analysis of 13 RCTs on acupuncture for cLBP in long-term follow-up, and reported that the overall effect using a random effect model of acupuncture on pain reduction was −0.64 (95% CI: −1.13 to −0.14). Vickers et al. [22] reported that the overall effect of acupuncture on pain reduction for cnLBP was 0.49 (95% CI: 0.33–0.64). In Manheimer et al.’s [18] review, acupuncture is more effective than no treatment and the overall effect using a random effect model was 0.69 (95% CI: 0.40–0.98) for patients with cLBP. Rubinstein et al. [23] reviewed the effects of acupuncture, spinal manipulation therapy (SMT), and herbal medicine for cnLBP, and found that the mean weighted difference (MWD) of pain relief and disability with comparison to no treatment or waiting list control were −24.1 (95% CI: −31.52 to −16.88), and −0.61 (95% CI: −0.90 to −0.33) respectively. A systematic review

reported that the effects of acupuncture on the outcomes of pain and functional disability were achieving moderate to large treatment effects in the comparison with no treatment [24]. Three additional systematic reviews reported similar positive findings that acupuncture is more effective than no treatment for cnLBP [16,25,26]. From these eight systematic reviews and meta-analysis [16–18,22–26], there is consistent evidence that acupuncture was more effective than a no treatment or a waiting list control, and the overall effect ranged from 0.49 to 0.69. Acupuncture vs no treatment, placebo medication or passive modality treatment (e.g. sham acupuncture, sham TENS). One systematic review indicated that sham acupuncture was as effective as acupuncture [26]. On the contrary, Xu et al. [17] found that the overall effect of acupuncture compared with sham treatment was −0.26 (95% CI: −0.56 to −0.05). Vickers et al. [22] also found that acupuncture is more effective than sham acupuncture, and the overall effect using a random effect model was 0.20 (95% CI: 0.09–0.32). In Manheimer et al.’s [18] systematic review, the overall effect was larger than Xu et al. and Vickers et al.’s systematic reviews, and was 0.54 (95% CI: 0.35–0.73). Within a short-term follow-up (up to 3 months), Rubinstein et al. [23] reported that the MWDs of acupuncture comparison to sham, placebo medication or passive modalities were −5.88 (95% CI: −11.20 to −0.55), and −0.18 (95% CI: −0.32 to −0.04) respectively. Yuan et al. [24] indicated that in comparisons of sham acupuncture, and placebo TENS the pain outcomes generally achieved small or moderate effect sizes. Two additional systematic reviews reported similar positive findings that acupuncture is more effective in pain reduction than sham treatment for cnLBP at short-term follow-ups (<3 months) (effect size ranged from −0.21 to −1.11) [16,25]. There is relatively consistent evidence for supporting that acupuncture is more effective than sham acupuncture as seven systematic reviews demonstrated positive findings, and the overall effect ranged from 0.26 to 0.54. Acupuncture plus an intervention (e.g. physiotherapy, standard medical care, or exercise) vs intervention alone. For the outcome of pain and functional disability, Yuan et al. [24] found that moderate to large treatment effects have been achieved in the comparison of acupuncture plus conventional treatment (e.g. physiotherapy, standard medical care, or exercise) vs conventional treatment. Within a short-term follow-up (up to 3 months), Rubinstein et al. [23] reported that the MWDs of pain relief and disability were −9.80 (95% CI: −14.93 to −4.67), and −1.04 (95% CI: −1.46 to −0.61) respectively. A cochrane review by Furlan et al. [16] also indicated that acupuncture in addition to conventional therapies for cLBP was more effective than these therapies alone. Although there were no clear effect size reported for acupuncture as an effective adjutant therapy for cnLBP, all these three reviews [16,23,24] concluded that acupuncture added into conventional therapy were better than conventional therapy alone. Acupuncture vs other active treatments (e.g. massage, SMTspinal manipulative therapy). Two systematic reviews [17,24] reported relatively positive findings: Xu et al. [17] conducted

Table 1 Summary of 17 included reviews. Review types

Databases, year, languages

Primary aim

Study sample

Study interventions

Outcome measures

Main findings and conclusion

Xu et al. (2013) [17]

Meta-analysis of RCTs

Medline, AMED, EMBASE, CENTRAL, ISRCTN, mRCT, CNKI Up to Jan 2012 Chinese, English

To determine the effectiveness of acupuncture therapy

2678 patients from 13 RCTs

Acupuncture vs no treatment Acupuncture vs sham acupuncture Acupuncture vs other treatment

Pain intensity Disability Spinal flexion QOL

The overall effect sizes of acupuncture on pain relief, disability, spinal flexion, and QOL were −0.43 (95% CI, −0.64 to −0.21), −0.43 (95% CI, −0.66 to −0.21), −1.04 (95% CI, −1.56 to −0.51), and 0.47 (95% CI, 0.15–0.78) respectively. Acupuncture is effective in providing long-term pain relief of CLBP, but this effect may be due to non-specific effects arising from skin manipulation.

Ambrosio et al. (2012) [30]

Systematic review of RCTs

AMED, MEDLINE, EMBASE, and specific health economic databases: HEED, NHS EED Up to January 2010 English

To synthesize data from economic evaluations to determine whether acupuncture for the treatment of chronic pain conditions is good value for money

Total 8 economic evaluation studies included with cLBP, neck pain, dysmenorrhoea, migraine and headache, and osteoarthritis

Acupuncture

Cost-effectiveness analysis (CEA), cost-utility analysis (CUA), cost per quality adjusted life year (QALY)

From seven CUAs, acupuncture was found to be clinically effective but cost more. The cost per QALY gained ranged from 2527 to 14,976 pounds per QALY, which was below typical thresholds of willingness to pay. Overall, this review suggested that acupuncture interventions are cost-effective when compared with routine care.

Furlan et al. (2012) [25]

Systematic review and meta-analysis of RCTs

Medline, EMBASE, The Cochrane Library, CINAHL, AMED, etc. Up to February 2010 English, Chinese, Japanese

To systematically review and compare the efficacy, cost-effectiveness, and safety of acupuncture, manipulation, massage and mobilization in adults with neck and low-back pain

A total of 147 RCTs included CAM therapies for neck and low back pain, 33 RCTs of acupuncture for LBP patients

Acupuncture vs inactive treatment (no treatment, placebo medication) Acupuncture vs active treatment (manipulation, massage, a combination of physical modalities such as the light, electricity, heat)

Efficacy: pain intensity, pain disability, Cost-effectiveness Safety: frequency of adverse events

Compared with inactive treatment, acupuncture reduced pain significantly at short-term follow-ups (<3 months); compared with active treatment, manipulation and massage were significantly better than acupuncture in reducing pain intensity and disability; Acupuncture was cost-effective compared to usual care and no treatment; Acupuncture was less often reported adverse events including soreness/pain at the site of needling, minor bleeding, dizziness, or headache. No difference between acupuncture group of subjects compared with usual care or TENS for the proportion of subjects with any adverse events.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Author and year

97

98

Table 1 (Continued) Review types

Databases, year, languages

Primary aim

Study sample

Study interventions

Outcome measures

Main findings and conclusion

Hutchinson et al. (2012) [26]

Systematic review of RCTs

Medline 1950–2011 English

To assess the effectiveness of acupuncture in the treatment of cnLBP

13,874 adults with cnLBP from 7 RCTs

Acupuncture vs sham Pain intensity acupuncture Pain disability Acupuncture vs Spinal flexion conventional therapy QOL (including physical Cost-effectiveness therapy, massage, general exercise, and drugs) (Individualized/standardized/simulated) Acupuncture vs usual care Acupuncture vs placebo TENS

Kim et al. 2012 [31]

Systematic review of RCTs

Medline, EMBASE, The Cochrane Library, Scopus, NHS CRD databases, EBSCOhost, CNKI, KoreaMed, KISS, RISS Up to March 2011 English, Chinese, Korean

To summarize the evidence on the cost-effectiveness of acupuncture

17 RCTs of acupuncture interventions, 3 RCTs included LBP patients

Acupuncture

Cost-effectiveness analysis (CEA), cost-utility analysis (CUA), and cost-benefit analysis (CBA)

All CUAs showed that acupuncture with or without usual care was cost-effective compared with waiting list control or usual care alone. In the CEAs, acupuncture was beneficial at a relatively low cost in Six European and Asian countries.

Vickers et al. (2012) [22]

Meta-analysis of RCTs

Medline, Cochrane Reviews, clinicians.gov Up to December 2010 English

To determine the effective size of acupuncture for chronic pain conditions: back and neck, osteoarthritis, chronic headache, and shoulder pain

Including five types of chronic pain, 10 RCTs for CNLBP

Acupuncture vs no treatment Acupuncture vs sham acupuncture

Pain intensity

For pain intensity reduction of CNLBP, effective size for acupuncture compared to no treatment was 0.49, and for acupuncture compared to sham acupuncture was 0.20. Acupuncture is effective for the treatment of chronic pain and is therefore a reasonable referral option.

“. . .acupuncture is more effective than no treatment but no conclusions can be drawn about its effectiveness over other treatment modalities. Sham acupuncture may be as effective as acupuncture which challenges the importance of needling along a meridian, the depth the needles need to be inserted and whether stimulation of the needles influences the effectiveness of treatment. In practice, acupuncture is often used as an adjunct to other therapy modalities. . .”

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Author and year

Systematic review of RCTs

Medline, EMBASE, CINAHL, CENTRAL, PEDro Up to December 2008 English, Dutch, German

To assess the effects of acupuncture, SMT and herbal medicine for cnLBP

20 RCTs of acupuncture for treatment of 5590 CNLBP patients

Acupuncture vs no treatment or waiting list control Acupuncture vs sham, placebo or passive modalities Acupuncture plus an intervention vs intervention alone Acupuncture vs any other intervention Acupuncture vs other types of acupuncture

Trigkiliday (2010) [29]

Systematic review of RCTs

PubMed Till July 2009 English

To evaluate a treatment guideline (a course of acupuncture of up to 10 sessions over 12 weeks) for the efficacy of acupuncture

4 RCTs cLBP patients

Acupuncture vs sham Pain intensity acupuncture Pain disability Acupuncture vs conventional therapy Acupuncture vs usual care (including physiotherapy, manipulation, exercise, and medications) (Individualized/standardized/simulated) Acupuncture vs usual care

Acupuncture can be effective in managing patients with LBP, especially the patients have positive expectations about acupuncture, suggesting a strong psychological element. A course of acupuncture of up to 10 sessions over 12 weeks was justified.

Johnston et al. (2008) [35]

Systematic review of RCTs

Medline, EMBASE, Cochrane, AMED, PEDro Up to December 2005 English

To assess current use of expertise-based randomization in trials of acupuncture or spinal manipulation for LBP

4 RCTs LBP patients

Acupuncture or spinal manipulation

Differential expertise bias may exist if a majority of clinicians participating in a RCT in which the skill of the providing clinician may influence results have greater expertise in 1 of the 2 interventions under evaluation. Random allocation of participants to clinicians with expertise in the specific interventions under investigation can overcome differential expertise bias. Investigators designing acupuncture or spinal manipulation trials in which 2 or more active therapies are compared do not make use of expertise-based randomization.

Pain Disability

Comparing acupuncture or competing spinal manipulation therapies

Within/at 3 month follow-up: Of the 1st comparison, the effect sizes for pain relief and disability were −24.1 (95% CI: −31.52 to −16.88), and −0.61 (95% CI: −0.90 to −0.33); Of the 2nd comparison, −5.88 (95% CI: −11.20 to −0.55), and −0.18 (95% CI: −0.32 to −0.04); Of the 3rd comparison, −9.80 (95% CI: −14.93 to −4.67), and −1.04 (95% CI: −1.46 to −0.61); Of the 4th comparison, −8.50 (95% CI: −11.04 to −5.96), and −0.53 (95% CI: −0.67 to −0.38); Of the 5th comparison, inconsistent evidence

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Rubinstein et al. (2010) [23]

99

100

Table 1 (Continued) Review types

Databases, year, languages

Primary aim

Study sample

Study interventions

Outcome measures

Main findings and conclusion

Yuan et al. (2008a) [24]

Systematic review of RCTs

Medline, EMBASE, AMED, ISI Web of Science Till 2008 English

To explore the effectiveness of acupuncture for nLBP

23 RCTs including 6359 nLBP patients

Acupuncture vs no treatment Acupuncture vs sham acupuncture Acupuncture vs conventional therapy Acupuncture vs placebo TENS Acupuncture and conventional therapy vs conventional therapy

Pain Functional disability Cost-effectiveness

For outcome of pain and functional disability, moderate to large treatment effect sizes have been achieved in the comparison of # 1 & # 5, whereas other groups of comparisons generally achieved small or moderate effect sizes. Twenty-four percent of comparisons achieved the minimal clinically importance difference (−20% or more) on pain reduction. Acupuncture as a cost-effective adjunct to standard care is clear cut.

Yuan et al. (2008b) [28]

Systematic review of RCTs, case studies, surveys, textbooks, and Chinese expert opinions

Medline, PubMed, EMBASE, AMED, ProQuest, CINAHL, ISI Web of Science, Cochrane Controlled Trials Register Chinese medical databases Till 2006 English Chinese

To investigate the difference in acupuncture treatment regimens for LBP among textbooks, clinical studies and clinical practice, and explore reasons for such differences

Patients and experts

Main intervention: acupuncture Co-interventions: Moxibustion and cupping reported by textbooks and clinical practice. various co-interventions in surveys including medication, physical therapy etc. Electroacupuncture advocated by Chinese textbooks, surveys and Clinical trials

Common points for needle, number of needles, the needle retention time, frequency and total number of session per treatment

BL23, BL25 and BL40 were generally reported by all sources as common acupuncture points. Other common points include K13, Jiaji points, Ashi and trigger points. De Qi sensation was reported by all resources as the common needle sensation, 10 needle points reported in surveys, 11 in clinical studies, Chinese experts tended to use less points, median as 5. The duration of needle retention was around 20min, Chinese experts retain the needle for longer as 25.5 min. One or two times per week or five to six per week as treatment frequency. For treatment of cnLBP, the most outstanding differences were in treatment frequency, number of points needled and co-interventions

Keller et al. (2007) [27]

Systematic review of RCTs

The Cochrane Library, Medline, CINHAL, AMED Till December 2005 English

To estimate the effect sizes of common LBP treatments comparing the interventions to placebo/sham or no treatment comparison groups

47 RCTs of LBP patients

Acupuncture, behavioral therapy, exercise therapy, TENS, manipulation, and medication therapy

Pain intensity Physical functioning

For cLBP, acupuncture had the largest effect sizes (SMD: 0.61), TENS had the smallest effect size (SMD: 0.22)

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Author and year

Systematic review of prospective, controlled, cost-effectiveness studies

Medline, EMBASE, CINAHL, AMED, The Cochrane Library, NHS Economic Evaluation Database, Health Technology Assessments up to October 2005 English

To summarize and assess all prospective, controlled, cost-effectiveness studies of complementary therapies carried out in the UK

6 cost-effectiveness studies included LBP and chronic headache patients

Acupuncture, spinal manipulation

Health benefits, cost

Main outcome measures of effectiveness favored the complementary therapies, but effective size were small and of uncertain clinical relevance. A cost-utility analyses the incremental cost per QALY was less than 10,000 pounds, and estimates of the incremental cost of achieving improvements in quality of life. But the estimates obtained may represent the cost-effectiveness nonspecific effects associated with acupuncture.

Furlan et al. (2005) [16]

Systematic review of RCTs

CENTRAL, MEDLINE, EMBASE Till February 2003 English, Chinese, Japanese, Norwegian, Polish, German

To determine the effectiveness of CAM therapies compared to placebo, no intervention, or other interventions

Total 35 RCTs included, 10 RCTs of cnLBP patients

Acupuncture vs no treatment Acupuncture vs sham therapy Acupuncture vs other interventions Addition of acupuncture to other interventions

Pain Function

Acupuncture is more effective than no treatment or sham treatment for CNLBP but there are no differences in effectiveness compared with other conventional therapies in short-term (up to 3 months). Acupuncture is not more effective than other conventional and “alternative” treatments. Acupuncture in addition to conventional therapies for CLBP is more effective than these therapies alone. Dry-needling appears to be a useful adjunct to other therapies for CLBP.

Manheimer et al. (2005) [18]

Meta-analysis of RCTs

Medline, Cochrane Central, EMBASE, AMED, CINAHL, and 2 specialized European databases Till August 2004 English

To assess acupuncture’s effectiveness for treating LBP

33 RCTs of LBP patients

Needle acupuncture vs. sham acupuncture or sham TENS Acupuncture vs no treatment Acupuncture vs other active treatments (e.g. medication, massage, TENS, spinal manipulation)

Pain, function, global improvement, fit for work

Acupuncture is more effective than sham [effective size: 0.54 (95% CI, 0.35–0.73)] or no treatment [effective size: 0.69 (95% CI, 0.40–0.98)] for patients with cLBP. Evidence about acupuncture’s effectiveness compared with other active treatment is inconclusive.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Canter et al. (2006) [32]

101

102

Table 1 (Continued) Review types

Databases, year, languages

Primary aim

Study sample

Study interventions

Outcome measures

Main findings and conclusion

van der Roer et al. (2005) [34]

Systematic review of RCTs

Pubmed, EMBASE, The Cochrane Library Till July 2004 English

To discover the most cost-effective treatment for LBP

Total 17 RCTs included, five on CLBP patients

Acupuncture, McKenzie exercises, chiropractic treatment, exercise therapy, multidisciplinary rehabilitation intervention, ergonomics interventions, neuroreflextherapy, bed rest and worksite visit

CEA, CUA, cost-benefit analysis

Among 17 included economic evaluations of LBP interventions, a single conclusion on the most cost-effective treatment for LBP could not be made because the study population, interventions and outcomes were very diverse.

Cherkin et al. (2003) [33]

Systematic review of RCTs

Medline, EMBASE, the Cochrane Controlled Trials Register, till April 2003 English

To summarize evidence about the effectiveness, safety, and costs of the most popular CAM therapies for back pain

20 RCTs of acupuncture included, 3 RCTs of massage, and 26 RCTs of spinal manipulation included

Acupuncture, massage, spinal manipulation

Pain Adverse events Cost

The effectiveness of acupuncture remains unclear. All of these treatments seem to be relatively safe. No evidence of savings for acupuncture therapy was identified. Massage may reduce the costs of care after an initial course of therapy.

Abbreviations: AT, acupuncture therapy; AMED, Allied and Complementary Medicine Database; CAM, Complementary and Alternative Medicine; CBA, cost-benefit analysis; CEA, cost-effectiveness analysis; CENTRAL, Cochrane Central Register of Controlled Trials; CLBP, chronic low back pain; CNKI, China National Knowledge Infrastructure; CNLBP, chronic nonspecific low back pain; CUA, cost-utility analysis; DARE (Database of Abstracts and Reviews of Effectiveness); HEED, Health Economics Evaluation Database; ISRCTN, International Standard Randomized Controlled Trial Number; KISS, Korean Studies Information Service System; LBP: low back pain; NHS EED, NHS Economic Evaluation Database; RCT: Randomized Controlled Trial; RISS, Research Information Sharing Service; SMT, spinal manipulation therapy; TENS, transcutaneous electrical nerve stimulation.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Author and year

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

a meta-analysis of 13 RCTs on acupuncture for cLBP in longterm follow-up and reported the overall effect of acupuncture comparison to other treatment subgroups was −0.49 (95% CI: −0.90 to −0.09). Rubinstein et al. [24] compared acupuncture with other active treatment modalities and reported that the MWDs were −8.50 (95% CI: −11.04 to −5.96), and −0.53 (95% CI: −0.67 to −0.38) respectively. In Keller et al.’s systematic review [27], there were five types of active treatment for cLBP, acupuncture had the largest treatment effect of 0.61 ((95% CI: 0.41–0.81), and TENS had the smallest treatment effect of 0.19 (95% CI: −0.13 to 0.51) [27]. But this systematic review calculated the effect size for each type of treatment, rather than acupuncture compared with other active treatment directly. However, two systematic reviews [16,25] reported negative findings: Furlan et al. [16] pointed out that acupuncture is not more effective than other conventional and “alternative” treatments. In a late systematic review, Furlan et al. [25] found that manipulation was significantly better than acupuncture in reducing pain intensity and disability at short-term follow-ups (<3 months) (−10.0, 95% CI: −14.0, −4.0 vs −6.0, 95% CI: −16.0, 2.0). There were three other reviews [18,24,26] which indicated inconclusive findings. Thus, there is conflicting evidence whether acupuncture is more effective than other treatment modalities. Acupuncture vs other types of acupuncture. Only two studies [16,23] examined the effects of different acupuncture techniques for cnLBP. Both reviews concluded that there was inconclusive evidence and no clear recommendations could be made about the most effective acupuncture technique. Overall, acupuncture is more effective when compared with; no treatment, a waiting list control, sham treatment, and when acting as adjunct treatment modalities. For the comparisons with other active treatment or comparing with other types of acupuncture, no recommendations could be made. Only commonly used acupuncture techniques could be summarized: one systematic review recommended common acupuncture points (e.g. BL23, BL25, BL40, K13, Jiaji points, Ashi and trigger points), number of points needled (10–11 needles), duration of needling (20–25.5 min), frequency of treatment per weeks (1–2 or 5–6) and total number of treatment times/sessions (around 10 sessions) [28]. In the UK, a National Institute for Health and Clinical Excellence (NICE) guideline [36] recommended that total treatment could consist of up to 10 sessions, and Trigkilidas concluded that a course of acupuncture of up to 10 sessions over 12 weeks was justified [29,36]. Cost-effectiveness and safety of acupuncture There were seven systematic reviews [24,25,30–32,34,35] that investigated cost-effectiveness of acupuncture for LBP. Five reviews [24,25,30–32] reported positive findings: acupuncture was cost-effective compared with usual care and no treatment. Other systematic reviews performed an incremental analysis of costs and effects. Kim et al. [31] found that all cost-utility analyses (CUAs) showed that acupuncture with or without usual care was cost-effective compared with waiting list control or usual

103

care alone. In the cost-effectiveness analysis (CEAs), acupuncture was beneficial at a relatively low cost in six European and Asian countries. Canter et al. [32] found that a cost-utility analysis the incremental cost per quality-adjusted life year (QALY) was less than 10,000 pounds, and estimates of the incremental cost of achieving improvements in quality of life. But the estimates obtained may represent the cost-effectiveness nonspecific effects associated with acupuncture. Yuan et al. [24] concluded that acupuncture as a cost-effective adjunct to standard care is clear cut. However, two reviews reported inconclusive findings: Cherkin et al. [33] indicated that no evidence of savings for acupuncture therapy was identified. van der Roer et al. [34] identified a variety of treatment interventions such as acupuncture, exercise therapy, massage, manipulative treatment and multidisciplinary treatment. This review included 17 economic evaluation studies on LBP interventions. The most cost-effective treatment for LBP could not be made because these included studies were diverse in terms of study population, interventions and outcomes. When considering the safety of acupuncture treatment, two systematic reviews [25,33] found that reporting adverse events by study participants was less often for those having acupuncture compared to other treatment such as manipulation, Any adverse events were only minor including soreness/pain at the site of needling, minor bleeding, dizziness, or headache. Therefore, acupuncture seems be to relatively safe.

Methodological issues of included reviews Each systematic review had a formal assessment of methodological quality by using the OQAQ. The assessment results of these 17 reviews were presented in Table 2. The overall score of methodological quality was from 1 to 9. Seven reviews [16–18,24,25,30,34] had very high methodological standard. Six reviews [22,23,26,27,31,35] had high methodological quality but with minor or minimal flaws (with OQAQ score higher than 6). Four reviews [28,29,32,33] had low methodological quality, and two of them [28,29] had major flaws. The most common methodological weaknesses were lacking of assessment tool to evaluate the validity of included primary studies among these systematic reviews, and methods used to combine the findings of trials were not reported or unclear. One systematic review by Johnston et al. [35] emphasized the research designs issues and reported that the quality of research designs was varied. Johnston et al. [35] indicated that the quality of the acupuncture treatment delivered in the trials were largely dependent on the variation in acupuncturists’ expertise. In other words, variation in acupuncturists’ expertise would influence the quality of acupuncture treatment for LBP. Therefore, Johnston et al. [35] suggest that random allocation of participants to clinicians with expertise in the specific interventions under investigation may be possible strategy to overcome differential expertise bias. From this overview of systematic reviews, there were no trials included in these systematic reviews which stated the use of expertise-based randomization designs to overcome the expertise bias.

104

Reviews (author and year)

Search methods used to find evidence on the primary aim stated

The search for evidence reasonably comprehensive

Inclusion criteria of selection studies reported

Bias in the selection of studies avoided

Assessment tool used for the validity of included studies

Validity of all the studies referred to the text assessed using appropriate criteria

Methods used to combine the findings of the relevant (to reach a conclusion) reported

Findings of the relevant studies combined relevant to the primary aim

Conclusions of the review supported by the data and/or analysis reported in the review

Total score (9) (‘+’ as 1; ‘−’or ‘?’ as 0)

Xu et al. (2013) Ambrosio et al. (2012) Furlan et al. (2012) Hutchinson et al. (2012) Kim et al. (2012) Vickers et al. (2012) Rubinstein et al. 2010) Trigkiliday (2010) Johnston et al. (2008) Yuan et al. (2008a) Yuan et al. (2008b) Keller et al. (2007) Canter et al. (2006) Furlan et al. (2005) Manheimer et al. (2005) van der Roer et al. (2005) Cherkin et al. (2003)

+ + + + + + + + + + + + + + + + +

+ + + + + + + – + + + + + + + + +

+ + + + + – + – + + + + + + + + ?

+ + + + ? ? + – + + ? + ? + + + ?

+ + + – + ? + – – + – – – + + + –

+ + + ? ? + + – ? + ? ? ? + + + ?

+ + + – + + + ? + + – + ? + + + –

+ + + + + + + + + + + + + + + + +

+ + + + + + + ? + + + + + + + + +

9 9 9 6 7 6 6 2 7 9 2 7 5 9 9 9 4

OQAQ: Overview Quality Assessment Questionnaire; ‘+’: yes; ‘−’: no; ‘?’: unclear. The overall score is from 1 to 9. OQAQ  4: having extensive or major flaw; 6: having minor or minimal flaw.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Table 2 Methodological quality of included reviews by OQAQ.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Discussion This overview identified a large amount of evidence on the effectiveness of acupuncture therapies for management of cnLBP. Among the included systematic reviews published in the last decade, there is consistent evidence that acupuncture was more effective than no treatment, waiting list control, or sham acupuncture. There is conflicting evidence whether acupuncture is more effective than other treatment modalities such as massage, or SMT. According to the levels of evidence described by the Cochrane Back Review Group, there are four levels of evidence based on the quality and number of trials of systematic reviews, they are strong (consistent findings provided by a systematic review of multiple high-quality RCTs); moderate (consistent findings provided by a systematic review of multiple low-quality RCTs-at least four or at least two highquality RCTs); limited (inconsistent findings from a systematic review of multiple RCTs-at least four); and no evidence (no RCTs) [37]. Based on this guideline, there was strong evidence when acupuncture compared with no treatment or passive treatment modalities, as there were eight systematic reviews and meta-analysis providing positive and consistent findings. For the comparison of acupuncture plus an intervention vs an intervention alone, three systematic reviews of multiple RCTs also indicated positive and consistent findings. Two comparisons showed inconsistent findings from systematic reviews, and there was limited evidence when acupuncture compared with other active treatment modalities or other types of acupuncture. More trials are needed to test the effectiveness of acupuncture with other active treatment modalities, in particular comparing the effectiveness of different acupuncture techniques. Some studies adopted traditional Chinese medicine (TCM) styles of acupuncture, and there were studies used Western styles of acupuncture techniques. Different styles of acupuncture techniques have different treatment protocols in terms of total number of treatment sessions, and frequency of treatment. One survey study explored these practice characteristics of acupuncture in China and Europe and found that differences were significant perceived by Chinese and European acupuncture practitioners [38]. A systematic review by Yuan et al. [28] had indicated these differences between TCM styles and other acupuncture styles. Yuan et al. [28] found that Chinese experts tend to use less acupoints, but retain the needle for longer. Other outstanding differences were treatment frequency and total number of treatment sessions among different acupuncturists. White et al. commented that the ‘core’ components of ‘adequate’ acupuncture protocol consist of acupoints selections, number of points needled, depth of insertion, responses elicited, needle stimulation-method and strength, needle retention time, needle types and the experience of the acupuncturist [39]. As there is a range of diverse acupuncture styles used for patients with cnLBP, future trials are needed to establish the standardization of acupuncture in terms of the length of treatment sessions, the frequency of sessions; the number of needles needed per treatment, the placement of needle insertion, the depth of needle insertion, and the needle stimulation to achieve De Qi.

105

Establishing the standardization of the total number of acupuncture session and the duration of treatment sessions can refer to UK NICE guideline in the review by Trigkilidas (10 session within 12 weeks) [29,36]. The lack of agreement on what is standard acupuncture protocol is an obstacle to good patient care, and to research into acupuncture [39]. Hence, to establish standard protocol by testing the effectiveness of different acupuncture styles for cnLBP is urgently needed, and to identify the most effective styles for cnLBP could be done by direct comparison between different treatment protocols in homogeneous groups of patients. Reviews on the cost-effectiveness of acupuncture, suggested that there were generally consistent findings from five systematic reviews, but two systematic reviews reported inconclusive findings. Lin et al. conducted a systematic review of guidelineendorsed treatments for low back pain, and concluded that acupuncture is one of the cost-effective treatments for chronic low back pain [40]. Acupuncture for low back pain also reduces the utilization of health services, as patients with low back pain were less likely to visit physicians after acupuncture treatment [41]. According to the Cochrane Back Review Group’s guideline and the findings of this overview, there is moderate evidence for the cost-effectiveness of acupuncture when comparing with no treatment, usual care and wait-list controls. Further high-quality trials testing cost-effectiveness will provide an even stronger evidence base. Once more sound evidence on cost-savings resulting from acupuncture for cnLBP is accumulated, it could provide a strong reasons for policy change and improve access within current healthcare systems [42]. There were some limitations to be considered when interpreting the findings of this overview. Although the search strategies seemed thorough, this overview may not have located all relevant systematic reviews or meta-analyses. In the searching stage, this overview conducted extended searching and did not use of the terms of ‘non-specific’. But coming to the stage of data synthesis, this overview only included the sub-group of cnLBP patients. This overview focused on evaluating systematic reviews or metaanalytic reviews, essential details of those primary trials had not been included. While this overview used a quality assessment tool to assess those included systematic and meta-analysis reviews to guarantee the data quality, the quality of primary data from individual trials could not be necessarily high. Collectively, these limitations may limit the conclusions of the study findings. Conclusion In this overview, there appears to be strong evidence for the effectiveness and moderate evidence for cost-effectiveness for the use of acupuncture for treating cnLBP. Acupuncture in comparison with no treatment/wait-list control or other passive treatment modalities is statistically effective for cnLBP which is currently supported by good evidence from systematic reviews and meta-analysis, but there is limited evidence that acupuncture was superior to other active interventions. This means that, according to the evidence summarized above, acupuncture should be considered as one of active treatment options for cnLBP.

106

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107

Author contributions Yingchun Zeng helped in the conduct of study, searched literatures and data extractions and preparation of the manuscript; Joanne Wai-yee Chung helped in design and conduct of the study, searched literature and data extraction and in the revision of the manuscript. Funding None. Conflicts of interest All authors stated that we have no conflicts of interest. References [1] Quinn F, Hughes C, Baxter D. Complementary and alternative medicine in the treatment of low back pain: a systematic review. Phys Ther Rev 2006;11:107–16. [2] Walker J. Back pain: pathogenesis, diagnosis and management. Nurs Stand 2012;27:49–56. [3] Hoy D, Bain C, Williams G, March L, Brooks P, Blyth F, et al. A systematic review of the global prevalence of low back pain. Arthritis Rheum 2012;24:181–91. [4] Krismer M, van Tulder M. The Low Back Pain Group of the Bone and Joint Health Strategies for Europe Project, low back pain (non-specific). Best Prac Res Clin Rheumatol 2007;21:77–91. [5] Institute of Medicine (US) Committee on Advancing Pain Research, Care, and Education. Relieving pain in America: a blueprint for transforming prevention, care, education, and research. Washington, DC: National Academies Press; 2011. Available from: http://www.ncbi.nlm.nih.gov/ books/NBK91497/ [6] Chou R. Pharmacological management of low back pain. Drugs 2010;70:387–402. [7] Rozenberg S, Foltz V, Fautrel B. Treatment strategy for chronic low back pain. Joint Bone Spine 2012;79:555–9. [8] Balagué F, Mannion AF, Pellisé F, Cedraschi C. Non-specific low back pain. Lancet 2012;379:4–10. [9] Murphy AY, van Teijlingen ER, Gobbi MO. Inconsistent grading of evidence across countries: a review of low back pain guidelines. J Manipulative Physiol Ther 2006;29:576–81. [10] Frost H, Lamb SE, Doll HA, Carver PT, Stewart-Brown S. Randomized controlled trial of physical therapy compared with advice for LBP. BMJ 2004;329:708. [11] Lewis K, Abdi S. Acupuncture for low back pain: a review. Clin J Pain 2010;26:60–9. [12] Ammendolia c, Furlan AD, Imamura M, Irvin E, van Tulder M. Evidenceinformed management of chronic low back pain with needle acupuncture. Spine J 2008;8:160–72. [13] Cassileth B, Trevisan C, Gubili J. Complementary therapies for cancer pain. Curr Pain Headache Rep 2007;114:265–9. [14] Lin JG, Chen WL. Review: acupuncture analgesia in clinical trials. Am J Chin Med 2009;37:1–18. [15] Dorsher PT. Acupuncture for chronic pain. Tech Reg Anesth Pain Manag 2011;15:55–63. [16] Furlan AD, Van Tulder M, Cherkin D, Tsukayama H, Lao L, Koes B, et al. Acupuncture and dry-needling for low back pain: an updated systematic review within the framework of the cochrane collaboration. Spine 2005;30(8):944–63. [17] Xu M, Yan S, Yin X, Li X, Gao S, Han R, et al. Acupuncture for chronic low back pain in long-term follow-up: a meta-analysis of 13 randomized controlled trials. Am J Chin Med 2013;41:1–19.

[18] Manheimer E, White A, Berman B, Forys K, Ernst E. Meta-analysis. Acupuncture for low back pain. Ann Intern Med 2005;142(8):651–63. [19] Cherkin DC, Sherman KJ, Kahn J, Erro JH, Deyo RA, Haneuse SJ. Effectiveness of focused structural massage and relaxation massage for chronic low back pain: protocol for a randomized controlled trial. Trials 2009;10:96. [20] Whitlock EP, Lin JS, Chou R, Shekelle P, Robinson KA. Using existing systematic reviews in complex systematic reviews. Ann Intern Med 2008;148:776–82. [21] Pieper D, Buechter R, Jerinic P, Eikermann M. Overviews of reviews often have limited rigor: a systematic review. J Clin Epidemiol 2012;65: 1267–73. [22] Vickers AJ, Cronin AM, Maschino AC, Lewith G, MacPherson H, Foster NE, et al. Acupuncture for chronic pain: individual patient data metaanalysis. Arch Intern Med 2012;172(19):1444–53. [23] Rubinstein SM, van Middelkoop M, Kuijpers T, Ostelo R, Verhagen AP, de Boer MR, et al. A systematic review on the effectiveness of complementary and alternative medicine for chronic non-specific low-back pain. Eur Spine J 2010;19:1213–28. [24] Yuan J, Purepong N, Kerr DP, Park J, Bradbury I, McDonough S. Effectiveness of acupuncture for low back pain: a systematic review. Spine 2008;33:E887–900. [25] Furlan AD, Yazdi F, Tsertsvadze A, Gross A, Van Tulder M, Santaguida L, et al. A systematic review and meta-analysis of efficacy, cost-effectiveness, and safety of selected complementary and alternative medicine for neck and low-back pain. Evid Based Complement Altern Med 2012;2012:1–61. [26] Hutchinson AJP, Ball S, Andrews JCH, Jones G. The effectiveness of acupuncture in treating chronic non-specific low back pain: a systematic review of the literature. J Orthopaedic Surgery Res 2012;7:36. [27] Keller A, Hayden J, Bombardier C, van Tulder M. Effect size of non-surgical treatments for non-specific low-back pain. Eur Spine J 2007;16:1776–88. [28] Yuan J, Kerr D, Park J, Liu XH, McDonough S. Treatment regimens of acupuncture for low back pain – a systematic review. Complement Ther Med 2008;16(5):295–304. [29] Trigkilidas D. Acupuncture therapy for chronic lower back pain: a systematic review. Ann R Coll Surg Engl 2010;92(7):595–8. [30] Ambrósio EMM, Bloor K, MacPherson H. Costs and consequences of acupuncture as a treatment for chronic pain: a systematic review of economic evaluations conducted alongside randomised controlled trials. Complement Ther Med 2012;20(5):364–74. [31] Kim SY, Lee H, Chae Y, Park HJ, Lee H. A systematic review of cost-effectiveness analyses alongside randomized controlled trials of acupuncture. Accupunct Med 2012;30:273–85. [32] Canter PH, Coon JT, Ernst E. Cost-effectiveness of complementary therapies in the United Kingdom – a systematic review. Ecam 2006;3:425– 32. [33] Cherkin DC, Sherman KJ, Deyo RA, Shekelle PG. A review of the evidence for the effectiveness, safety, and cost of acupuncture, massage therapy, and spinal manipulation for back pain. Ann Intern Med 2003;138:898– 906. [34] van der Roer N, Goossens M, Evers S, van Tulder MW. What is the most cost-effective treatment for patients with low back pain? A systematic review. Best Prac Res Clin Rheumatol 2005;19:671–84. [35] Johnston BC, Da Costa BR, Devereaux PJ, Akl EA, Busse JW. The use of expertise-based randomized controlled trials to assess spinal manipulation and acupuncture for low back pain: a systematic review. Spine 2008;33(8):914–8. [36] National Institute for Health and Clinical Excellence. Low back pain. Early management of persistent nonspecific low back pain. NICE Clinical Guideline 88; 2009. http://www.nice.org.uk/CG88fullguideline [37] van Tulder M, Furlan A, Bombardier C, Bouter L. Editorial Board of the Cochrane Collaboration Back Review Group. Updated method guidelines for systematic reviews in the Cochrane collaboration back review group. Spine 2003;28:1290–9. [38] Robinson N, Lorenc A, Ding W, Jun J, Bovey M, Wang X. Exploring practice characteristics and research priorities of practitioners of traditional acupuncture in China and the EU – a survey. J Ethnopharmacol 2012;140:604–13.

Y. Zeng, J.W.-y. Chung / European Journal of Integrative Medicine 7 (2015) 94–107 [39] White A, Cummings M, Barlas P, Cardini F, Filshie J, Foster NE, et al. Defining an adequate dose of acupuncture using a neurophysiological approach – a narrative review of the literature. Acupunct Med 2008;26:111–20. [40] Lin CWC, Haas M, Maher CG, Machado LAC, van Tulder MW. Cost-effectiveness of guideline-endorsed treatment for low back pain: a systematic review. Eur Spine J 2011;20:1024–38.

107

[41] Moritz S, Liu MF, Rickhi B, Xu TJ, Paccagnan P, Quan H. Reduced health resource use after acupuncture for low-back pain. J Altern Complement Med 2011;17:1015–9. [42] Molassiotis A. Managing cancer-related fatigue with acupuncture: is it all good news for patients. Acupunct Med 2013;31:3–4.