Study Protocol Systematic Review Medicine ®

OPEN

Acupuncture for treating whiplash-associated

disorder
A systematic review and meta-analysis protocol
∗
Seunghoon Lee, KMD, PhDa, , Dae-Hyun Jo, KMD, MSb, Kun Hyung Kim, KMD, PhDc

Abstract
Background: This review aims to evaluate the effectiveness and safety of acupuncture treatment for patients with whiplash-
associated disorder (WAD).
Methods: We will search the following databases from their inception to October 2018: MEDLINE, Embase, the Cochrane Central
Register of Controlled Trials, the Cumulative Index to Nursing and Allied Health Literature, the Allied and Complementary Medicine
Database, 1 Chinese database (China National Knowledge Infrastructure), 1 Japanese database (Japan Science and Technology
Information Aggregator, Electronic), and 5 Korean databases (KoreaMed, Research Information Service System, Korean Studies
Information Service System, Database Periodical Information Academic, and Oriental Medicine Advanced Searching Integrated
System). All randomized controlled trials of acupuncture for WAD will be considered for inclusion without language restrictions. The
risk of bias will be assessed using the Cochrane risk of bias tool. The mean difference or standard mean difference for continuous
data and risk ratio for dichotomous data will be calculated with 95% confidence intervals.
Dissemination: The results of this review will be disseminated through peer-reviewed journal articles or conference presentations,
and may provide important guidance for clinicians and patients regarding the use of acupuncture treatment for treating WAD.
Trial registration number: PROSPERO 2018: CRD42018106964.
Abbreviations: CI = confidence interval, MD = mean difference, NDI = neck disability index, NRS = numerical rating scale, QoL =
quality of life, RCT = randomized controlled trial, ROM = range of motion, RR = risk ratio, SMD = standardized mean difference, VAS
= visual analog scale, WAD = whiplash-associated disorder.
Keywords: acupuncture, protocol, systematic review, traffic accident, whiplash injury, whiplash-associated disorder

1. Introduction symptoms of WAD include neck pain and stiffness, and other pain
in the back, shoulder, and temporomandibular joints. Apart from
1.1. Description of the condition pain, WAD symptoms include dizziness, visual disturbance,
The term of whiplash associated disorder (WAD) was introduced fatigue, sleep disturbance, anxiety, depression, memory difficulties,
by the Quebec Task Force (QTF) in 1995 to reflect the spectrum and psychological distress.[2–4] The incidence of WAD is estimated
of clinical symptoms following bony or soft-tissue injuries to be 300 per 100,000 inhabitants, although these rates differ
resulting largely from motor vehicle collisions.[1] The predominant between countries.[2] The associated annual costs after road
accidents in 2016 is estimated to be more than £35 billion in the
United Kingdom[5] and more than US $21 billion in the South
Funding: This study was supported by the Convergence of Conventional
Medicine and Traditional Korean Medicine R&D program funded by the Ministry
Korea, which accounts for approximately 1.4% of the gross
of Health & Welfare through the Korea Health Industry Development Institute domestic product of South Korea;[6] these estimates are consistent
(KHIDI) (HI16C2365). even when the method of calculation is altered to include output,
The authors have no conflicts of interest to disclose. medical care, damage to property, and police cost. Several
Supplemental Digital Content is available for this article. prospective studies have report that most recovery takes place
a
Department of Acupuncture & Moxibustion Medicine, b Department of up to 3 months following the initial injury,[7,8] and, therefore,
Acupuncture & Moxibustion Medicine, Graduate School, Kyung Hee University, proper treatment management in the acute and subacute stages is
Seoul, c Department of Acupuncture & Moxibustion, School of Korean Medicine, important to prevent the development of chronic conditions.[2]
Pusan National University, Yangsan, South Korea.
∗
Correspondence: Seunghoon Lee, Department of Acupuncture and
Moxibustion Medicine Kyung Hee University Korean Medicine Hospital, 23 1.2. Description of the intervention
Kyunghee dae-ro, Dongdaemun-gu, Seoul 02447, South Korea
(e-mail: [email protected]).
Acupuncture is defined as an intervention that stimulates specific
points (e.g., traditional acupuncture points, myofascial trigger
Copyright © 2018 the Author(s). Published by Wolters Kluwer Health, Inc.
This is an open access article distributed under the Creative Commons points, or tender points) using needles with various manipu-
Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and lations (e.g., manual or electrical stimulation). It has mainly been
reproduction in any medium, provided the original work is properly cited. used for the treatment of musculoskeletal diseases such as neck,
Medicine (2018) 97:41(e12654) back, or knee pain, and its effectiveness and safety have been
Received: 10 September 2018 / Accepted: 12 September 2018 supported by many rigorous clinical trials.[9] Recently, the
http://dx.doi.org/10.1097/MD.0000000000012654 application of acupuncture therapy has been extended to include

1
Lee et al. Medicine (2018) 97:41 Medicine

treatment for psychological symptoms related to posttraumatic 2.2.2. Types of participants. All patients who suffered from the
stress disorder, as well as comorbidities for chronic pain, such as any symptoms related to WAD, such as musculoskeletal pain,
insomnia, depression, or anxiety.[10,11] sensorimotor control disturbances, or psychological problems
will be included. The diagnosis criteria and classification of WAD
1.3. How the intervention might work will not be limited.

The mechanism of acupuncture treatment for pain and 2.2.3. Types of interventions. Acupuncture treatment using
psychological symptoms remains unclear. Some studies have needling with various types of stimulation (e.g., manual or
proposed mechanisms of acupuncture analgesia according to: electrical) on specific points (e.g., traditional acupuncture points,
local effect mediated by adenosine A1 receptors[12,13] or myofascial trigger points, or tender points) will be included.
myofascial trigger point inactivation;[14] segmental effect based However, trials involving non-penetrating stimulation on specific
on the gate-control theory of pain;[15] and a general effect points (e.g., acupressure, magnets, moxibustion, transcutaneous
through descending inhibitory pain control by serotonin and electrical nerve stimulation, or laser therapy) and penetrating
noradrenaline.[16] Moreover, the mechanism of acupuncture for stimulation with the insertion of medical materials (e.g., herbal
psychological symptoms in chronic pain or mental illness is acupuncture or thread embedding acupuncture) will not be
thought to involve modulation of opioid peptides and mono- included in the review.
amines, such as noradrenaline, serotonin, or dopamine in the The control intervention will be considered as no treatment/
brain.[17] waiting list, sham acupuncture, and active treatment (e.g.,
medication or physiotherapy).[22] However, trials in which
acupuncture was compared with other forms of acupuncture or
1.4. Why it is important to perform this review herbal medication will be excluded. When the acupuncture group
The value of acupuncture treatment for treating WAD is received acupuncture and other active treatment simultaneously,
controversial based on the limited systematic reviews and only trials in which the same active treatment was administered to
clinical practice guidelines available. One systematic review the both groups will be included.
reported that the evidence of acupuncture for WAD was 2.2.4. Types of outcome measures. The time frame of
inconclusive due to limited data.[18] Updated evidence suggests outcome measurements will be determined as a short-term (up
that acupuncture may not be effective in treating neck pain to 12 weeks after injury) and a long-term outcome (more than 12
induced by WAD because acupuncture treatment has not been weeks after injury).
shown to reduce neck pain at a level that was clinically
significant. However, the trials included in this review only
2.2.4.1. Primary outcomes.
compared acupuncture therapy with sham acupuncture, which is
known to have more than a placebo effect and not be valid as an 1. Severity of pain: the measurement of relevant pain using any
inert placebo control.[19,20] Moreover, they did not conduct a scale (e.g., visual analog scale (VAS) (0–100 mm or 0–10 cm)
systematic search, and they only evaluated literature published in or numerical rating scale [NRS]) will be analyzed.
English.[20,21] Therefore, it is worth conducting a systematic 2. Function: relevant overall function and disability using any
review that includes comparative effectiveness trials, as well as scale or range (e.g., neck disability index [NDI])
sham-controlled trials, with an up-to-date systematic search to
determine whether acupuncture is an effective treatment option
for WAD. 2.2.4.2. Secondary outcomes.
1. Quality of life (QoL): assessed using a validated scale (e.g.,
1.5. Objective 36-item Short-Form [SF-36] or Euro-QoL)
2. Range of movement (ROM) of the neck
The objective of this systematic review is to evaluate the benefits 3. Psychological measurements
and harms of acupuncture therapy for patients with WAD in 4. Clinical global improvement in symptoms
comparison to those who received with no treatment, sham 5. Adverse events related to acupuncture treatment
acupuncture, routine/usual care, conventional medicine, or other
active treatments.
2.3. Search methods for identification of studies
2.3.1. Electronics searches. The following 12 databases will be
2. Methods
searched from inception to October 2018: MEDLINE (1946 to
2.1. Study registration October Week 4 2018), Embase (1980 to October 4, 2018), the
Cochrane Central Register of Controlled Trials (The Cochrane
The protocol of review methods has been registered, prospec- Library, 2018 Issue 10), the Cumulative Index to Nursing and
tively (CRD42018106964; http://www.crd.york.ac.uk/ PROS- Allied Health Literature (CINAHL, 1982 to October 2018), the
PERO). Allied and Complementary Medicine Database (AMED, 1985 to
October 2018), 1 Chinese database (China National Knowledge
2.2. Criteria for including studies in this review Infrastructure (CNKI)), 1 Japanese database (Japan Science and
2.2.1. Types of studies. Prospective randomized controlled Technology Information Aggregator Electronic (J-STAGE)), and
trials (RCTs) of acupuncture treatment for WAD will be 5 Korean databases (KoreaMed, Research Information Service
included in the review. Nonrandomized controlled trials, System (RISS), Korean Studies Information Service System
observational studies, qualitative studies, and laboratory studies (KISS), Database Periodical Information Academic (DBpia),
will be excluded. Language will not be restricted for study and Oriental Medicine Advanced Searching Integrated System
eligibility. (OASIS)).

2
Lee et al. Medicine (2018) 97:41 www.md-journal.com

2.3.2. Search for other resources. The WHO International 2.4.5. Unit of analysis issues. When unit of analysis issues arise
Clinical Trials Registry Platform will be also searched for ongoing in the studies that assessed outcome variable repeatedly (at more
and recently completed studies. Bibliographic references in relevant than one time point), we will categorized the assessments into 3
publications will be manually searched to avoid missing eligible different measurement time frames after the traffic accident: acute
trials. stage (until 4 weeks), subacute stage (until 12 weeks), and chronic
stage (over 12 weeks). If more than 2 assessments are reported in
2.3.3. Search strategy. The search terms will consist of 2 parts: the same time frame, only the last assessment in the time frame
WAD (e.g., whiplash, traffic, or neck injury) and acupuncture (e.g., will be chosen for analysis.
acupuncture, electroacupuncture, or dry needling). The detailed
search strategies for MEDLINE are presented in the online 2.4.6. Dealing with missing data. For missing or incomplete
supplementary appendix 1, http://links.lww.com/MD/C530. data, we will attempt to contact the original study authors to
request the missing data. If the additional data cannot be obtained,
only the available data will be analyzed, and the potential impact
2.4. Data collection and analysis
2.4.1. Selection of studies. Two review authors (SL and KHK) of the missing data will be addressed in the discussion.
will independently screen the titles and abstracts for potentially 2.4.7. Assessment of heterogeneity. Heterogeneity will be
eligible studies identified by the searches. The authors will assessed preferentially by visual inspection of the forest plot, and
independently select and record their decisions according to a x2 test with a significance level of P < .10 will define the
predefined criteria on a standard eligibility form. If a disagreement presentation of heterogeneity. Additionally, I2 statistic will be
between 2 reviewers for study selection cannot be resolved through assessed to quantify the inconsistencies among the studies, with a
discussion, a third reviewer will resolve the disagreement. The flow value of more than 50% indicating a meaningful heterogeneity. I2
process of filtration will be summarized in a Preferred Reporting of 0% to 40% may be unimportant, 30% to 60% may be
Items for Systematic Reviews and Meta-Analysis (PRISMA)- moderate, 50% to 90% may be substantial, and 75% to 100%
compliant flow chart (http://www.prisma-statement.org). may be considerable heterogeneity.[24]
2.4.2. Data extraction and management. Two review authors
2.4.8. Assessment of reporting biases. Funnel plot will be
(SL and KHK) will independently extract data from the articles
used to detect reporting bias when more than 10 studies are
using a standard data extraction form (e.g., author, year of
available,[24] and Egger’s regression method will be used to
publication, country, study design, sample size, participants,
determine funnel plot asymmetry.[25]
condition, acupuncture intervention, control intervention, out-
come measures, main results, and adverse events) after reading the 2.4.9. Data synthesis. The meta-analysis using Review Manag-
full text of each article. Details of the acupuncture treatment and er software (RevMan, version 5.3 for Windows; the Nordic
control interventions will be extracted based on the revised Cochrane Centre, Copenhagen, Denmark) will be used. A
STandards for Reporting Interventions in Clinical Trials of random effects model will be used to calculate the pooled effect
Acupuncture (STRICTA) guidelines.[23] Any disagreement regard- estimates, because substantial clinical heterogeneity is expected
ing the extracted data will be resolved through discussion or across the included studies in this review. If considerable
consultation among the reviewers. When the data are insufficient or heterogeneity (I2 ≥ 75%) cannot be explained by the clinical
unclear, we will contact the first author or the corresponding author and methodological diversity, the data will not be pooled.[24]
through e-mail or telephone to request additional information. When a study has more than 2 acupuncture groups with different
stimulation styles (e.g., manual or electrical stimulation) or points
2.4.3. Assessment of risk of bias. Two review authors (SL and
(e.g., local or distal points), meta-analysis will be conducted in
KHK) will independently perform the quality assessment using the
careful consideration of whether the data of the different
tool for assessing risk of bias based on the Cochrane Handbook for
acupuncture groups will be combined into one merged
Systematic Reviews of Interventions.[24] The following domains
acupuncture group.[26]
will be assessed: random sequence generation; allocation conceal-
ment; blinding of participants/personnel; blinding of outcome 2.4.10. Subgroup analysis and investigation of heterogene-
assessors; incomplete outcome data; selective outcome reporting; ity. When sufficient numbers of studies are available, subgroup
and other sources of bias (including factors that are likely to analysis will be performed to identify the heterogeneity among
influence the results, such as extreme baseline imbalance of age, studies according to the following:
comorbidities, disease onset, or physical conditions). The risk of
bias will be categorized into 3 levels: low, high, and unclear risk 1. Type of acupuncture stimulation (e.g., manual versus electrical
of bias. Any disagreement will be resolved through discussion or needle stimulation)
consultation among the reviewers. 2. Type of control (e.g., no treatment/waitlist, sham acupuncture,
routine/usual care, conventional medicine, or other active
2.4.4. Measures of treatment effect. For continuous outcomes, treatments)
the mean difference (MD) with 95% confidence intervals (CIs) 3. Severity of signs and symptoms assessed by the QTF grading
will be presented. If the methods or scales measuring the system (e.g., grade I, II or III, IV).
treatment effect among the studies in the analysis are not the
same, the standardized mean difference (SMD) will be used. For
2.4.11. Sensitivity analysis. Where appropriate, sensitivity
dichotomous outcomes, the risk ratio (RR) will be used to
analysis will be conducted to evaluate whether the results are
measure the treatment effect with 95% CIs. Ordinal data will be
robust in the review according to the following:
converted to dichotomous data when the data needs to be pooled.
For example, global assessments which were graded as 1. Methodological qualities (e.g., whether random sequence
“recovery,” “markedly effective,” “effective,” and “ineffective” generation, allocation concealment, and assessor blinding are
will be dichotomized into “improved” or “not improved.” adequately conducted or not)

3
Lee et al. Medicine (2018) 97:41 Medicine

2. Statistical method (random-effects model vs fixed-effects References

model)
[1] Spitzer WO, Skovron ML, Salmi LR, et al. Scientific monograph of the
Quebec Task Force on Whiplash-Associated Disorders: redefining
2.4.12. Summary of evidence. The results of the main “whiplash” and its management. Spine 1995;20:1S–73S.
outcomes (primary outcomes and adverse events) will be [2] Sterling M, Kenardy J. Whiplash: Evidence Base for Clinical Base for
summarized in the “Summary of findings” tables. The quality Clinical Practice. Elsevier Australia, Chatswood, NSW:2011.
[3] Treleaven J. Dizziness, unsteadiness, visual disturbances, and postural
of evidence in the main outcomes will be evaluated using the
control: implications for the transition to chronic symptoms after a
Grading of Recommendations Assessment, Development whiplash trauma. Spine 2011;36:S211–217.
and Evaluation (GRADE) software with respect to the [4] Kasch H, Turk DC, Jensen TS. Whiplash injury: Perspectives on the
following aspects: limitations in the design and implementa- Development of Chronic Pain. 2016;Wolters Kluwer Health, Philadel-
tion; indirectness of evidence; unexplained heterogeneity or phia:Available at: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&CSC=
Y&NEWS=N&PAGE=booktext&D=books3&AN=01906635/1st_Edi
inconsistency of results; imprecision of results; and high tion.
probability of publication bias. The quality of evidence will [5] Department for Transport. Guidance documents—RAS 60003 Total
be categorized into 4 levels: high, moderate, low, and very low value of prevention of reported accidents by severity and cost element:
quality.[24] GB 2016. Available at: https://www.gov.uk/government/statistical-data-
sets/ras60-average-value-of-preventing-road-accidents. Accessed August
22, 2018.
[6] Korea Road Traffic Authority. Estimation and assessment of traffic
2.5. Ethics and dissemination
accidentincurred costs in 2016 [in Korean]. Available at: http://taas.
Ethical approval is not necessary as this study is a systematic koroad.or.kr/web/bdm/srs/selectStaticalReportsList.do?menuId=
WEB_KMP_IDA_SRS_RTE. Accessed August 22, 2018.
review. The results of this review will be disseminated through
[7] Kamper SJ, Rebbeck TJ, Maher CG, et al. Course and prognostic factors
peer-reviewed journal articles and conference presentations. of whiplash: a systematic review and meta-analysis. Pain 2008;138:
617–29.
[8] Sterling M, Hendrikz J, Kenardy J. Compensation claim lodgement and
3. Discussion health outcome developmental trajectories following whiplash injury: a
prospective study. Pain 2010;150:22–8.
This is the study protocol of a systematic review and meta- [9] Vickers AJ, Cronin AM, Maschino AC, et al. Acupuncture for chronic
analysis for the use of acupuncture therapy to treat patients with pain: individual patient data meta-analysis. Arch Intern Med 2012;172:
WAD. A recent systematic review[20] and a current clinical 1444–53.
guideline[21] suggested that there was not sufficient data to [10] Kim YD, Heo I, Shin BC, et al. Acupuncture for posttraumatic stress
disorder: a systematic review of randomized controlled trials and
determine the effectiveness of needle acupuncture for WAD; prospective clinical trials. Evid Based Complement Alternat Med
additionally, they reported that electroacupuncture was not likely 2013;2013:615857.
to be effective and it should not be considered as a form of [11] Grant S, Colaiaco B, Motala A, et al. Acupuncture for the treatment of
treatment for neck pain. However, these reviews are based on adults with posttraumatic stress disorder: a systematic review and meta-
studies that included only trials with sham needling, which may analysis. J Trauma Dissociation 2018;19:39–58.
[12] Goldman N, Chen M, Fujita T, et al. Adenosine A1 receptors mediate
be as powerful as verum acupuncture although the value of local anti-nociceptive effects of acupuncture. Nat Neurosci 2010;13:
acupuncture could be evaluated comparing with standard 883–8.
treatment as in a comparative effectiveness trials.[19] Moreover, [13] Takano T, Chen X, Luo F, et al. Traditional acupuncture triggers a local
it is known that language restrictions can impact the estimates of increase in adenosine in human subjects. J Pain 2012;13:1215–23.
[14] Chou L-W, Kao M-J, Lin J-G. Probable mechanisms of needling
effectiveness in complementary and alternative medicine inter- therapies for myofascial pain control. Evid Based Complement Alternat
ventions.[27] However, these studies only searched English Med 2012;2012:705327.
databases, and did not include studies in the Korean or Chinese [15] White A, Cummings M, Filshie J. An Introduction to Western Medical
literature. A recent review emphasized that Korean studies should Acupuncture. Elsevier Health Sciences, UK, Saintt Louis:2008.
be considered in a systematic review for acupuncture treatment to [16] Silva JRT, Silva ML, Prado WA. Analgesia induced by 2- or 100-Hz
electroacupuncture in the rat tail-flick test depends on the activation of
avoid potential language bias.[28] different descending pain inhibitory mechanisms. J Pain 2011;12:51–60.
Therefore, we will search all relevant literature, without any [17] Filshie J, White A, Cummings M. Medical Acupuncture: A Western
language restrictions, in the Korean, Chinese, and Japanese Scientific Approach. 2016;Elsevier,
databases to include any relevant trials of acupuncture for [18] Moon T-W, Posadzki P, Choi T-Y, et al. Acupuncture for treating
whiplash associated disorder: a systematic review of randomised clinical
treating patients with WAD. The results of this systematic review trials. Evid Based Complement Alternat Med 2014;2014:870271.
will provide a summary of the current evidence on the [19] Lundeberg T, Lund I, Sing A, et al. Is placebo acupuncture what it is
effectiveness and safety of acupuncture for WAD. This evidence intended to be? Evid Based Complement Alternat Med 2011;2011:
will be a useful resource to patients, practitioners, and health 932407.
policy-makers who want to consider acupuncture for WAD as a [20] Wong JJ, Shearer HM, Mior S, et al. Are manual therapies, passive
physical modalities, or acupuncture effective for the management of
primary form of treatment or an adjuvant therapy to patients with whiplash-associated disorders or neck pain and associated
conventional treatments. disorders? An update of the Bone and Joint Decade Task Force on Neck
Pain and Its Associated Disorders by the OPTIMa collaboration. Spine J
2016;16:1598–630.
Author contributions [21] Côté P, Wong JJ, Sutton D, et al. Management of neck pain and
associated disorders: a clinical practice guideline from the Ontario
Conceptualization: Seunghoon Lee, Kun Hyung Kim Protocol for Traffic Injury Management (OPTIMa) Collaboration. Eur
Funding acquisition: Seunghoon Lee Spine J 2016;25:2000–22.
Methodology and project administration: Seunghoon Lee, Kun [22] American Society of Anesthesiologists Task Force on Acute Pain
ManagementPractice guidelines for acute pain management in the
Hyung Kim, Dae-Hyun Jo
perioperative setting: an updated report by the American Society of
Writing – original draft: Seunghoon Lee Anesthesiologists Task Force on Acute Pain Management. Anesthesiol-
Writing – review & editing: Kun Hyung Kim, Dae-Hyun Jo ogy 2012;116:248–73.

4
Lee et al. Medicine (2018) 97:41 www.md-journal.com

[23] MacPherson H, Altman DG, Hammerschlag R, et al. Revised STandards [26] Littell JH, Corcoran J, Vijayan P. Systematic Reviews and Meta-
for Reporting Interventions in Clinical Trials of Acupuncture Analysis. Oxford University Press, Oxford, New York:2008.
(STRICTA): extending the CONSORT statement. J Evid Based Med [27] Pham B, Klassen TP, Lawson ML, et al. Language of publication
2010;3:140–55. restrictions in systematic reviews gave different results depending on
[24] Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of whether the intervention was conventional or complementary. J Clin
Interventions Version 5.1.0 [updated March 2011]. The Cochrane Epidemiol 2005;58:769–76.
Collaboration, 2011. Available at: http://www.cochrane-handbook.org. [28] Kim KH, Kong JC, Choi J-Y, et al. Impact of including Korean
[25] Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis randomized controlled trials in Cochrane reviews of acupuncture. PLoS
detected by a simple, graphical test. BMJ 1997;315:629–34. One 2012;7:e47619.