A Critical Look into the 2016 NICE Guidelines: Acupuncture for Low-Back Pain and Sciatica

Abstract

Background:

In November of 2016, the National Institute for Health and Care Excellence (NICE) Guidelines for low-back pain (LBP) and sciatica were published. According to the NICE Guidelines Development Group (GDG), acupuncture is no longer a recommended treatment for LBP and sciatica, while other therapies including nonsteroidal anti-inflammatory drugs, exercise, epidurals, and manual therapy are recommended as treatments.

Objective:

The aim of this article is to discuss how the GDG decision-making process behind the recommendations against acupuncture—while supporting common conventional treatments for LBP and sciatica—is inconsistent and lacks sufficient evidence-based justification.

Methods:

The evidence used to develop the 2016 NICE Guidelines for LBP and sciatica were critically appraised using the Grading of Recommendations, Assessment, Development, and Evaluation framework, and examined for their limitations.

Results:

There is predominantly moderate-quality evidence favoring acupuncture over sham, suggesting that the GDG's conclusion that acupuncture works through nonspecific effects is inconsistent with the NICE evidence. The NICE evidence comparing acupuncture to usual care (or wait-list) also demonstrates acupuncture's effectiveness. The GDG's analyses excluded non-English language studies, and evaluated acupuncture by different standards, compared to other recommendations.

Conclusions:

Acupuncture demonstrates efficacy and effectiveness in the treatment of LBP and sciatica. Each of the GDG's recommendations for treatment of LBP and sciatica should be reevaluated as consistently as possible by the same standards to mitigate any inconsistencies. Analyses of acupuncture should include studies without language restrictions and factor in acupuncture dose and types of sham devices to reduce potential bias in conclusions drawn.

Introduction

In November of 2016, the National Institute for Health and Care Excellence (NICE) published its guidelines for the treatment of low-back pain (LBP) and sciatica.¹ This impressive project amounted to more than 3500 pages and is the result of considerable deliberation and informed evidence-based decision making.¹ Despite this, the justification of some recommendations remains controversial.

According to the Guidelines Development Group (GDG), acupuncture is no longer a recommended treatment for LBP, because no consistent clinically important effects were available when verum acupuncture was compared with sham acupuncture.¹ This suggests that acupuncture works by nonspecific contextual effects. Acupuncture, therefore, was deemed to be not efficacious, although it was deemed effective.¹ Dissonance arises, as acupuncture was excluded based on its efficacy, while this criteria was not used to exclude recommendations of conventional therapies, such as nonsteroidal anti-inflammatory drugs (NSAIDs), exercise, epidurals, and manual therapy.¹ In addition, the quality of evidence provided to justify decisions made was insufficient, thus calling each recommendation into question.

While some of the inconsistencies in the 2016 NICE Draft Guidelines have been previously explored, the completed 2016 NICE Guidelines had yet to be assessed until now.² The purpose of this article is to demonstrate evidence that the decision-making process behind the GDG's recommendations against acupuncture—while simultaneously supporting common conventional treatments for LBP and sciatica—is inconsistent and lacks sufficient evidence-based justification.

Methods

The respective evidence provided in the 2016 NICE Guidelines for LBP and sciatica to develop treatment recommendations were critically appraised using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) framework (Table 1).^1,3 Also, limitations of the evidence the GDG used to develop these guidelines were examined.

Table 1.

Four Levels of GRADE Evidence

Quality level	Definition
High Quality Evidence	We are very confident that the true effect lies close to that of the estimate of the effect.
Moderate Quality Evidence	We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low Quality Evidence	Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect.
Very Low Quality Evidence	We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.

GRADE, Grading of Recommendations, Assessment, Development, and Evaluation.

Results

Verum Acupuncture Versus Sham

The comparison between verum acupuncture versus sham acupuncture for treating LBP and sciatica showed that of 32 outcomes, 9 were clinically and statistically significant in favor of acupuncture, 20 were statistically significant in favor of acupuncture, and 3 were statistically significant in favor of sham.¹

Of the 9 outcomes showing clinically and statistically significant improvements in favor of acupuncture, 1 was of low quality showing improved quality of life (QoL) physical domain at ≤4 months.¹ A second outcome was of high quality showing reduced psychologic distress at ≤4 months.¹ The remaining 7 outcomes were of moderate quality showing improvements in QoL physical domain at >4 months; QoL general health domain at ≤4 months; QoL function domain at ≤4 months; QoL physical role limitation domain at ≤4 months; QoL vitality domain at ≤4 months; QoL social function domain at ≤4 months; and QoL emotional role limitation domain at ≤4 months.¹ Given predominantly moderate quality evidence in favor of acupuncture, the GDG's conclusion that acupuncture works through nonspecific effects is inconsistent with the evidence.

The GDG documents in Appendix C.9 of the 2016 NICE Guidelines that any non-English study was excluded, which suggests that the evidence used to support the NICE recommendations was insufficient due to a language bias.¹ For instance, a study was conducted to explore the impact of searching Chinese-language databases in addition to English-language databases on the results of systematic reviews on acupuncture published up until 2009.⁴ It was found that 68.6% of reviews that did not search Chinese-language databases were inconclusive, while this was only true for 30.8% of the reviews that did search Chinese-language databases.⁴ In Appendix L.9 of the 2016 NICE Guidelines, the GDG excluded 8 studies because they were not written in English, thus adding potential bias to the final analysis.¹ The current authors identified 5 studies not written in English examining verum acupuncture versus sham for LBP that were not included in the 2016 NICE guidelines analysis but were included in the latest published Cochrane review for LBP.^5–10 There was no reason provided for exclusion of these studies in appendix L.9 of the 2016 NICE Guidelines.¹ Given that 3 of 5 of the studies were statistically in favor of acupuncture and that they were not included in the final analysis, it is possible that the analyses used by the GDG could have been skewed to not favor acupuncture.^8–10

Past literature suggests that no modality of sham acupuncture may be regarded as inert due to evidence depicting C tactile afferent activation, which is known to alleviate unpleasantness and increase one's sense of well-being.^11,12 This suggests that the true effect of acupuncture may be underestimated when compared to sham. Moreover, the effects of different sham modalities are unclear.¹³ The analyses in the 2016 NICE Guidelines pooled studies that used different sham acupuncture modalities, including varying insertion points and puncturing methods.¹ The lack of sham acupuncture stratification in the analyses may have added bias to the true effect of acupuncture so as to potentially affect the conclusions drawn.

Acupuncture Versus Usual Care (or Wait-List)

Even if acupuncture is assumed to not be efficacious, the NICE data comparing acupuncture versus usual care (or wait-list) demonstrated acupuncture's effectiveness. Of 20 outcomes, 9 showed clinically and statistically significant findings in favor of acupuncture, with the remaining 11 outcomes showing statistically significant findings in favor of acupuncture.¹

Of the 9 outcomes showing clinically and statistically significant improvement of acupuncture compared to usual care (or wait-list), 1 was of very low quality showing reduced pain severity at ≤4 months.¹ A second outcome was of high quality showing improved QoL physical domain at ≤4 months.¹ The remaining 7 outcomes were of moderate quality showing improvements in QoL bodily pain domain at ≤4 months; function (Roland & Morris Disability Questionnaire, Hannover Functional Ability Questionnaire, and Pain Disability index) at ≤4 months; function (Hannover Functional Ability Questionnaire) at >4 months; days with analgesics at ≤4 months; and 50% responder criteria.¹

Despite having potentially serious limitations, a cost–utility analysis favored acupuncture plus usual care over usual care alone for LBP with or without sciatica (incremental cost-effectiveness ratio: $6,480 USD per quality-adjusted life-year gained), supporting acupuncture's use further.¹

Acupuncture Versus NSAIDs

The GDG compared acupuncture with NSAIDs and found low-quality evidence suggesting no clinically significant difference for all but 1 outcome.¹ NSAIDs were favored for pain severity ≤4 months, although the outcome was powered by a single study (n_treatment = 29, n_control = 29), which only used a single session of acupuncture.^1,14 This suggests that acupuncture might be comparable to NSAIDs, yet NSAIDs are recommended over acupuncture. There is low-to-moderate–quality evidence to support NSAIDs as clinically superior to placebo in terms of pain, QoL, and function.¹ However, there is predominantly moderate-to-high–quality evidence to support verum acupuncture as clinically superior to sham in terms of QoL and function.¹ Additionally, the clinically superior result from the study for pain outcomes in favor of NSAIDs might not be relevant to the public as the result was determined by comparing etoricoxib, a drug that is not approved for use in North America, with placebo.¹⁵ Although the current authors' argument is subject to indirectness, there is higher-quality evidence for acupuncture than for NSAIDs in each therapy's comparison with sham/placebo. This highlights another inconsistency in the GDG's decision-making process.

Conventional Therapies Versus Sham

While exercise, epidural injections, and manual therapy are recommended treatments, there is insufficient evidence to justify their efficacy. Only single trials were available that compared individual exercise with sham and group exercise with sham, respectively. The first trial (n_treatment = 86, n_control = 95) used an active treatment sham, while the latter trial (n_treatment = 14, n_control = 12) lacked validity pertaining to the sham and suffered from imprecision.^16,17 Moreover, evidence for outcomes in favor of epidural injections of local anesthetics and steroids is supported by only single studies (n_treatment = 28, n_control = 37; n_treatment = 80, n_control = 80) comparing the intervention to sham.^18,19

Finally, manipulation, mobilization, or soft-tissue techniques were recommended as part of multimodal treatment packages.¹ There was questionable evidence for manual manipulation versus sham and soft-tissue therapy versus sham. Of 14 outcomes, only 1 outcome was clinically significant in favor of manual manipulation (moderate evidence: QoL physical domain improvement at ≤4 months), although powered by a single study (n_treatment = 97, n_control = 95).²⁰ Of 3 outcomes, only 1 was clinically significant in favor of soft-tissue therapy (very low quality: reduction of pain severity at ≤4 months).¹ Meanwhile, all outcomes for manual therapy and soft-tissue therapy in multimodal treatment packages compared to sham or single-modality massage were powered by single studies (n_treatment = 20, n_control = 20; n_treatment = 31, n_control = 29).^21,22

Discussion

The NICE evidence comparing verum acupuncture to sham for treating LBP and sciatica, when examined by the GRADE framework, demonstrated predominantly moderate-quality evidence in favor of acupuncture.¹ This evidence for acupuncture efficacy might have been even greater if non-English studies (5 of which were included in the latest Cochrane review of acupuncture for LBP) were included in the NICE analyses.^5–10 Similarly, the NICE evidence showed predominantly moderate-quality evidence demonstrating the effectiveness of acupuncture compared to usual care (or wait-list) treatments.¹

These positive results for acupuncture for treating LBP and sciatica were present despite the GDG pooling studies that used different types of sham acupuncture interventions, such as minimal acupuncture or nonpuncture sham devices.¹ The GDG also pooled studies that administered the treatment once^14,23,24 with studies that administered the treatment up to twenty times.²⁵ Notably, a systematic review on acupuncture for chronic pain conducted a stratified analysis exploring intervention frequency.²⁶ It was found that 6 or more treatments were associated with positive outcomes, which emphasizes the precedence for the analyses to account for various treatment frequencies.²⁶

There is better (moderate-to-high quality) evidence supporting verum acupuncture over sham in terms of QoL and function than evidence for NSAIDs (low-to-moderate quality).¹ The GDG concluded on the basis of a single study comparing acupuncture (1 session) to NSAIDs for pain relief that NSAIDs produced superior pain relief, which, as noted above, is based on an inadequate dosage of acupuncture.¹⁴

Similarly, while the GDG endorsed exercise, epidural injections, and manual therapy as recommended treatments for LBP and sciatica, there is insufficient evidence powered by 1 or 2 studies that these recommendations were based on.^16–22

Conclusions

To make a recommendation against acupuncture for treating LBP and sciatica based on its efficacy, there is a need to ensure that the respective analysis includes studies without language restrictions, distinguishes among types of sham (such as minimal acupuncture and nonpuncture sham devices), and considers the dose of acupuncture administered in trials examined. Another comparison can be made to prove the efficacy of acupuncture while avoiding the issue of using sham devices, which are suggested to be active controls.^11,12 Specifically, future analyses should compare acupuncture to medications that have been proven in multiple low-risk randomized controlled trials. Next, there is a need for more high-quality trials on the efficacy of NSAIDs and comparing NSAIDs with acupuncture to clarify further the inconsistency in recommending NSAIDs as well as NSAIDs over acupuncture for LBP and sciatica. Issues with safety of NSAIDs, compared to acupuncture and other conventional care, should also be considered as part of any analysis of interventions of LBP and sciatica. Finally, the current authors hope that each of the GDG's recommendations, such as for conventional therapies, are reconsidered and reevaluated as consistently as possible to the same standards to mitigate any inconsistencies.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

National Institute of Health and Clinical Excellence (NICE). Low Back Pain and Sciatica in Over 16s: Assessment and Management (NG59). London: NICE; 2016.

Cummings

. Exercise Not Acupuncture Recommended by NICE for Low Back Pain—Balanced Assessment, Bias or Error? Acupuncture in Medicine Blog, BMJ Blogs. Online document at: blogs.bmj.com/aim/2016/03/31/nice-exercise-not-acupuncture/ Accessed November 14, 2016.

Balshem

, Helfand

, Schünemann

, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011; 64(4):401–406.

Lui

, Smith

, Terplan

. Heterogeneity in search strategies among Cochrane acupuncture reviews: Is there room for improvement?. Acupunct Med. 2010; 28(3):149–153.

Furlan

, van Tulder

, Cherkin

, et al. Acupuncture and dry-needling for low back pain. Cochrane Database Syst Rev. 2005; 1:CD001351.

Araki

, Kawamura

, Mataka

, et al. Randomized controlled trial comparing the effect of manual acupuncture with sham acupuncture for acute low back pain [in Japanese]. J Jpn Soc Acupunct Moxibustion. 2001; 51(3):382.

Inoue

, Kitakouji

, Ikeuchi

, et al. Randomized controlled pilot study comparing acupuncture with sham acupuncture for lumbago [in Japanese]. J Jpn Soc Acupunct Moxibustion. 2000; 50(2):356.

Inoue

, Kitakouji

, Ikeuchi

, et al. Randomized controlled pilot study comparing manual acupuncture with sham acupuncture for lumbago (2nd report) [in Japanese]. J Jpn Soc Acupunct Moxibustion. 2001; 51(3):412.

Lopacz

, Gralewski

. A trial of assessment of the results of acupuncture or suggestion in the treatment of low back pain. Neur Neurochir Pol. 1979; 8:405–409.

10.

Von Mencke

, Wieden

, Hoppe

, Porschke

, Hoffmann

, Herget

. Acupuncture of shoulder pain and low back pain: Two prospective double-blind studies [in German]. Akupunktur. 1988; 4:204–215.

11.

Lund

, Lundeberg

. Are minimal, superficial or sham acupuncture procedures acceptable as inert placebo controls?. Acupunct Med. 2006; 24(1):13–15.

12.

Mohr

, Binkofski

, Erdmann

, Buchel

, Helmchen

. The anterior cingulate cortex contains distinct areas dissociating external from self-administered painful stimulation: A parametric fMRI study. Pain. 2005; 114(3):347–357.

13.

Dincer

, Linde

. Sham interventions in randomized clinical trials of acupuncture—a review. Complement Ther Med. 2003; 11(4):235–242.

14.

Shin

, Ha

, Lee

, et al. Effects of motion style acupuncture treatment in acute low back pain patients with severe disability: A multicenter, randomized, controlled, comparative effectiveness trial. Pain. 2013; 154(7):1030–1037.

15.

U.S. Food and Drug Administration—Center for Drug Evaluation and Research. Briefing Information. Joint Meeting With the Arthritis Advisory Committee and the Drug Safety and Risk Management Advisory Committee. U.S. Food and Drug Administration, 2005. Online document at: www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4090b1.htm Accessed November 14, 2016.

16.

Albert

, Manniche

. The efficacy of systematic active conservative treatment for patients with severe sciatica: A single-blind, randomized, clinical, controlled trial. Spine (Phila Pa 1976). 2012; 37(7):531–542.

17.

Smith

, Kolt

, McConville

. The effect of the Feldenkrais method on pain and anxiety in people experiencing chronic low back pain. N Z J Physiother. 2001; 29(1):6–14.

18.

Ghahreman

, Ferch

, Bogduk

. The efficacy of transforaminal injection of steroids for the treatment of lumbar radicular pain. Pain Med. 2010; 11(8):1149–1168.

19.

Karppinen

, Malmivaara

, Kurunlahti

, et al. Periradicular infiltration for sciatica—a randomized controlled trial. Spine (Phila Pa 1976). 2001; 26(9):1059–1067.

20.

Haas

, Vavrek

, Peterson

, Polissar

, Neradilek

. Dose-response and efficacy of spinal manipulation for care of chronic low back pain: A randomized controlled trial. Spine J. 2014; 14(7):1106–1116.

21.

Senna

, Machaly

. Does maintained spinal manipulation therapy for chronic nonspecific low back pain result in better long-term outcome?. Spine (Phila Pa 1976). 2011; 36(18):1427–1437.

22.

Schimmel

, de Kleuver

, Horsting

, Spruit

, Jacobs

WCH

, van Limbeek

. No effect of traction in patients with low back pain: A single centre, single blind, randomized controlled trial of Intervertebral Differential Dynamics Therapy. Eur Spine J. 2009; 18(12):1843–1850.

23.

Inoue

, Kitakoji

, Ishizaki

, et al. Relief of low back pain immediately after acupuncture treatment—a randomised, placebo controlled trial. Acupunct Med. 2006; 24(3):103–108.

24.

Marignan

. Auriculotherapy treatment protocol for low-back pain: A randomized trial. Med Acupunct. 2014; 26(3):154–160.

25.

Leibing

, Leonhardt

, Köster

, et al. Acupuncture treatment of chronic low-back pain—a randomized, blinded, placebo-controlled trial with 9-month follow-up. Pain. 2002; 96(1–2):189–196.

26.

Ezzo

, Berman

, Hadhazy

, Jadad

, Lao

, Singh

. Is acupuncture effective for the treatment of chronic pain? A systematic review. Pain. 2000; 86(3):217–225.