Complementary Therapies in Medicine (2013) 21, 65—72

Available online at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/ctim

The role of acupuncture in emergency department

settings: A systematic review夽
Kun Hyung Kim a, Byung Ryul Lee b, Ji Ho Ryu c, Tae-Young Choi d,
Gi Young Yang b,∗

a
Department of Acupuncture & Moxibustion Medicine, Korean Medicine Hospital, Pusan National University, Yangsan, South Korea
b
Division of Clinical Medicine, School of Korean Medicine, Pusan National University, Yangsan, South Korea
c
Department of Emergency Medicine, School of Medicine, Pusan National University, Yangsan, South Korea
d
Korea Institute of Oriental Medicine, Daejeon, South Korea
Available online 29 December 2012

KEYWORDS Summary
Objectives: Patients with non-emergent and non-life threatening conditions often present to
Acupuncture;
the emergency department (ED), which hinders the efficient utilisation of healthcare resources.
Emergency
Acupuncture has frequently been used for such common conditions, although not in the ED
department;
context. This study aimed to evaluate the current evidence for acupuncture treatment in the
Systematic review;
ED and to inform future randomised controlled trials (RCTs) for acupuncture in ED settings.
Complementary and
Methods: Four English databases (MEDLINE, EMBASE, CENTRAL and AMED) were systematically
alternative medicine
searched to identify studies that tested the effects of acupuncture in ED settings using the
search terms of ‘‘acupuncture’’ and ‘‘emergency’’. Data extraction and the risk of bias assess-
ments were performed by two independent reviewers.
Results: Of the 102 screened studies, two RCTs and two uncontrolled observational studies were
deemed eligible. Sample sizes ranged from 42 to 100. The conditions treated included various
musculoskeletal and non-musculoskeletal symptoms and showed substantial clinical hetero-
geneity. Acupuncture was delivered in conjunction with standard medical care with the goal
of immediate pain alleviation; in one RCT, acupuncture resulted in a positive outcome, but
it did not in the other. The risk of bias was generally high or unclear. Uncontrolled studies
reported beneficial effects for acupuncture, although these studies were prone to bias. Two
studies reported mild and transient adverse events associated with acupuncture.
Conclusions: The current evidence is insufficient to make any recommendations concerning
the use of acupuncture in the ED. The effectiveness and safety as well as the feasibility of
acupuncture should be tested in future RCTs.
© 2012 Elsevier Ltd. All rights reserved.

夽 This study was supported by the Pusan National University Research Grant 2010. Choi T-Y was supported by Korea Institute of Oriental

Medicine (C12080 and K12130).

∗ Corresponding author at: Division of Clinical Medicine, School of Korean Medicine, Pusan National University, Yangsan 626-870, South

Korea. Tel.: +82 55 360 5963; fax: +82 55 360 5519.

E-mail address: [email protected] (G.Y. Yang).

0965-2299/$ — see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ctim.2012.12.004
66 K.H. Kim et al.

Contents

Methods ................................................................................................................... 66
Study search.......................................................................................................... 66
Quality assessment ................................................................................................... 66
Data analysis ......................................................................................................... 66
Results .................................................................................................................... 67
General characteristics of included studies ........................................................................... 67
Risk of bias ........................................................................................................... 69
Adjunctive use of acupuncture with conventional ED management vs conventional ED management alone ........... 69
Adjunctive use of acupuncture in uncontrolled case series ........................................................... 70
Safety of acupuncture ................................................................................................ 70
Discussion ................................................................................................................. 70
Strengths and limitations ............................................................................................. 71
Implications for future research ...................................................................................... 71
Implications for clinical practice ..................................................................................... 71
Conclusion ................................................................................................................ 71
Conflict of interest statement ............................................................................................. 71
References .............................................................................................................. 71

Acupuncture is increasingly used for the management of evaluate the existing evidence of the most representative
pain and other acute/chronic conditions in various clinical type of acupuncture (i.e., needle insertion into the body).
settings. When acupuncture is used as a sole or adjunctive Randomised or non-randomised controlled trials and
treatment in general practitioner (GP) practices or tertiary other uncontrolled case series were eligible, as it was
hospital settings, better health outcomes and less analgesic expected that there would be a paucity of literature for
consumption after acupuncture treatments were observed the use of acupuncture in ED settings and our aim was to
which might contribute to the reduction of potentially harm- identify the current status of evidence for this topic.
ful drug over-consumption.1—4 However, extrapolation of One author searched the database and screened poten-
these data is often challenging, as the clinical context and tially eligible studies after reading the title and abstract
environment in the emergency department (ED) setting may of identified studies. After the initial screening, which was
be significantly different from those noted in GP clinics or conducted by checking titles and abstracts, a more thorough
other secondary/tertiary hospitals.5 Given this information, investigation was performed using full-text access. This pro-
it is important to analyse acupuncture use based on clini- cess was checked by a second author who was unblinded to
cal data generated in ED settings, for which there is very author’s name and affiliation of included studies. For eli-
limited evidence. This study aimed to systematically evalu- gible RCTs, data extraction and a risk of bias assessment
ate the effectiveness and safety of acupuncture for various were conducted independently by two authors. Detailed
conditions in ED settings. These data may thus inform rele- characteristics in four domains (patients, intervention, com-
vant factors necessary for the design of future randomised parison and outcomes) were extracted. For the extraction
controlled trials (RCTs). of intervention-related information, the revised STandards
for Reporting Interventions in Clinical Trials of Acupunc-
ture (STRICTA) items were used to describe the details of
Methods acupuncture treatments used in the study context.6

Study search
Quality assessment
Four English databases (EMBASE, MEDLINE, AMED and CEN-
TRAL) were searched from their inception to July 2012. The Cochrane risk of bias tool was used to assess the internal
Search terms included ‘acupuncture’ for intervention and validity of each study.7
‘emergency’ for study setting. No terms about the study The two authors independently assessed the risk of bias,
design were set, as we expected to find only a small number and any inconsistency in assessment results were discussed
of previous studies in this field. No language restriction was with a third reviewer.
imposed.
We defined acupuncture as the insertion of a needle
at certain points of the body regardless of the underlying Data analysis
theoretical framework (i.e., traditional Asian medicine
theory or Western medical acupuncture) and the subse- A qualitative summary of results in the included studies
quent use of stimulation methods (i.e., manual or electrical was performed. Clinical heterogeneity was assessed through
stimulation). Acupuncture techniques that did not involve data analysis. Where appropriate, we tried to conduct a
needle insertion or point injection of herbal medicine were meta-analysis (quantitative analysis) with statistical hetero-
excluded, as the aim of this systematic review was to geneity assessment.
Acupuncture in emergency department settings 67

Idenficaon
English database (n =102)

Records after duplicates

removed (n = 102) Records excluded (n=80)
- Not relevant (n = 57)
Screening

-Not ED settings (n=3)

-Not clinical study (n=5)
-Not needle acupuncture (n=3)
Records screened -Case reports (n=8)
(n = 102 ) -Study protocol (n=1)
-Language not available (n=3)

Full-text articles assessed

Full-text articles excluded, with
for eligibility
Eligibility

reasons (n=18)
(n = 22 )
- Not ED settings (n=8)
-Case report (n=5)
-Not clinical study (n=3)
-Conference proceeding with
Studies included in insufficient reporting (n=1)
qualitative synthesis -Only for infants (n=1)
(n = 4)
- RCTs (n=2)
- UOSs (n=2)
Included

Studies included in
quantitative synthesis
(meta-analysis)
(n = 0 )

Figure 1 PRISMA flowchart of study selection. ED, emergency department; RCT, randomised controlled trial; UOS, uncontrolled
observational study.

Results extremities.10 In two RCTs, conventional ED management

served as the control. For three studies, concomitant treat-
Of the 102 initially searched hits, two RCTs and two uncon- ments included standard medical treatment in the ED.8,9,11
trolled observational studies (UOSs) were eligible and were In one case series, analgesic medication was not allowed
included in the analysis. Because only two RCTs with consid- during or before acupuncture treatment but other inter-
erable clinical heterogeneity were identified, quantitative ventions for acute extremity injuries, for example, ice,
synthesis of the study results (meta-analysis) was not per- elevation and splinting, were provided when appropriate.10
formed. Fig. 1 illustrates the selection process for this The details of characteristics are provided in Table 1.
review. A variety of acupuncture techniques were identified in
the review. One RCT used only auricular acupuncture and
instructed patients to retain needles for 4 to 6 days after
General characteristics of included studies administration of acupuncture.8 Another RCT employed
a mixture of auricular acupuncture, traditional Chinese
Three studies8—10 were conducted in the U.S. and one case medicine (TCM) style body acupuncture, electroacupunc-
series was conducted in Germany.11 Three studies were ture and active puncture techniques that involved an active
conducted in the ED of a medical centre8—10 and one was range of motion movements with distal-point needling.9 One
conducted in the general emergency service of the Red case series used TCM, Japanese Kiko Matsumoto, auricu-
Cross.11 Conditions for which acupuncture was employed lar and trigger point acupuncture.11 Another case series
included acute pain syndromes,8 common acute conditions did not specify which acupuncture technique was used.10
with a range of musculoskeletal and non-musculoskeletal Acupuncture practitioners involved in the treatment were
symptoms9,11 and acute, non-penetrating injuries of the reported to be licensed,10 use TCM style9 or be medical
 68
Table 1 Summary characteristics of included studies.

Study Design Patient Intervention Outcomes Summary of results

Harkin (2007)9 RCT (n = 45) (1) Age: Acupuncture (TCM style (1) Pain VAS Effects were similar in
(T/C) = (40.25 ± 14.68/ and auricular (2) Physiological parameters (respiratory both groups, showing
45.46 ± 16/18) acupuncture) plus rate, heart rate, systolic and diastolic blood 25.90 ± 17.64% and
(2) Female: conventional medical pressure) 22.18 ± 24.08% of pain
(T/C) = (42%/47%) treatments vs (3) Length of stay in ED reduction in acupuncture
(3) Condition: common conventional medical (4) Patient satisfaction with the treatment and control group,
acute conditions treatments alone Time points: immediate post-treatment respectively
including headache,
musculoskeletal pain and
other
non-musculoskeletal
symptoms
Goertz (2006)8 RCT (n = 100) (1) Age: Auricular acupuncture (1) Pain verbal NRS Significantly higher pain
(T/C) = (30.40 ± 9.67/ with standard medical (2) ED administration of medications and reduction in acupuncture
32.76 ± 7.45) care vs standard medical prescription information group (2.18 of mean
(2) Female: care alone (3) Medication consumption after discharge difference between
(T/C) = (42%/64%) Time points: groups) immediately
(3) Condition: acute pain immediate post-treatment and 24-h after after acupuncture
syndromes discharge treatment but similar at
24-h follow-up (0.37 of
mean difference
between groups)
Fleckenstein UOS (n = 60) (1) Age: 55.4 ± 23.0 Acupuncture (TCM and (1) 4-point symptom assessment scale Benefits were reported
(2011)11 (2) Female: 57% Japanese Kiko (marked, slight improvement, no change, after acupuncture
(3) Condition: various Matsumoto style, deterioration) treatment.
symptoms including auricular and trigger (2) Symptom improvement VAS (in 21 of 60
nausea, pain, dyspnea points) with emergency patients)
and other symptoms medical interventions Time points:
immediate post-treatment
Arnold (2009)10 UOS (n = 20) (1) Age: 33.0 (SD not Acupuncture (details not (1) Pain NRS Benefits were reported
reported) reported) with the (2) Patient satisfaction with acupuncture after acupuncture
(2) Female: 30% standard ED (3) Complication related to the acupuncture treatment with
(3) Condition: acute, interventions (ice, Time points: diminishing tendency of
non-penetrating injury elevation and splinting NRS measured immediately after observed effects at 72-h
of extremities, mainly when appropriate) acupuncture, 30 and 60 min follow-up.

K.H. Kim et al.

contusion and non-bony except the use of post-acupuncture
injury of a joint traditional analgesic All outcomes were measured within 72 h of
medication discharge by telephone interview
Abbreviations: RCT, randomised controlled trial; UOS, uncontrolled observational study; T/C, treatment group/control group; TCM, traditional Chinese medicine; VAS, visual analogue
scale; ED, emergency department; SD, standard deviation; VAS, visual analogue scale; NRS, numeric rating scale.
Acupuncture in emergency department settings 69

acupuncturists.8 One case series did not report the quali- Adjunctive use of acupuncture with conventional
fications of the acupuncture practitioner.11 In all included ED management vs conventional ED management
studies, patients were included after the ED physician con- alone
firmed that it was appropriate for the subject to participate
in the acupuncture treatments. In one study, acupunc- In one RCT, 100 patients with acute pain syndromes
ture treatments were allowed for 3—5 min.11 Goertz et al. presenting to the ED were randomised into either an
reported the administration of auricular acupuncture on acupuncture group or a standard medical care group.8 The
the two fixed points, a seemingly rapid procedure that mean (standard deviation) baseline pain level as measured
required notably less time when compared to traditional by a numeric rating scale (NRS) was 6.98 (1.68) and 7.78
body acupuncture, although no information on the total (1.84) in the acupuncture and standard medical care group,
treatment time in the ED department was reported.8 One respectively, suggesting a severe degree of baseline pain
case series found that the median length of stay in the ED in all participants. Immediately after a single session of
was 135 min for patients who received acupuncture, while auricular acupuncture treatment, the mean change in the
it was 90 min in patients who did not receive acupuncture pain visual analogue scale (VAS) was 2.33 (1.99) and 0.15
treatments. The authors suggested that this was an accept- (0.59) in the acupuncture and standard medical care group,
able amount of time for acupuncture treatment in the ED.10 respectively, showing statistically significant mean differ-
In the RCT conducted by Harkin and Parker9 , an acupunc- ences between the groups. However, the difference was
turist consultation, which lasted 30—40 min with 20 min of diminished when measured at a 24-h telephone follow-up
electroacupuncture stimulation, was provided. This type of (the mean change in the pain VAS was 3.41 and 3.05 in the
consultation may require considerable amounts of time in ED treatment and control groups, respectively). The authors
settings.9 The full details of the acupuncture intervention reported an assessment of medication administration and
are described in Appendix 1. prescription information in the ED as well as medication
consumption after the emergency room (ER) visit. How-
ever, those results were not fully reported, and the authors
Risk of bias stated that no statistically significant difference between
the groups was observed.
In two RCTs, the lack of patient and practitioner blinding In another RCT, 45 patients were randomised to either
was a major source of risk of bias. The RCT by Harkin and the acupuncture group (n = 32) or the conventional treat-
Parker9 generally showed high or unclear risk of bias in all ment group (n = 13).9 Immediately after a single session of
of the seven predefined domains, whereas that of Goertz acupuncture treatment, pain scores were equally decreased
et al. showed low risk of bias in the random sequence gen- from moderate to severe levels to mild levels (from 64.3%
eration, allocation concealment, outcome assessor blinding and 54.7% to 38.4% and 32.8% in the acupuncture and control
and incomplete outcome reporting.8 Details of the risk of groups, respectively). Other physiological parameters, such
bias assessment are provided in Table 2. as respiratory rate, heart rate and systolic and diastolic

Table 2 Risk of bias in included RCTs.

Domains Goertz (2006)8 Harkin (2007)9

Assessment Description Assessment Description

Random sequence generation L A computer random number U No report on

generator was used. randomisation methods
Allocation concealment L A set of sealed, numbered U No report on allocation
opaque envelops was used. concealment
Patient blinding H No blinding H No blinding
Practitioner blinding H No blinding H No blinding
Outcome assessor blinding L Outcome assessors were H No blinding
blinded.
Incomplete outcome reporting L Missing outcome data were U No information on the
balanced in numbers across number of patients who
two groups. underwent
post-treatment
assessment
Selective outcome reporting U Do not know whether U Do not know whether
research protocol is research protocol is
available available
Other sources of bias H The number of female is not L No other remarkable
balanced across two groups. sources of bias
Abbreviations: RCT, randomised controlled trial; L, low risk of bias; H, high risk of bias; U, uncertain risk of bias. Risk of bias was assessed
by the Cochrane review handbook.7
70 K.H. Kim et al.

blood pressure, were also measured with no significant in an uncontrolled case series. Two uncontrolled studies
between-group differences at post-treatment assessment. reported beneficial outcomes immediately after acupunc-
The length of ED stay and patients’ self-perceived effects of ture treatment, although the significant risk of bias due
the given treatments were similar in both groups. Willing- to their uncontrolled nature prohibited the estimation of a
ness to undergo the assigned treatments in the future was causal relationship between acupuncture treatment and the
reported in 27 of 32 patients in the acupuncture group and observed beneficial outcomes. Risk of bias was also evident
10 of 13 in the standard care group. Overall, acupuncture in two RCTs due to their unblinded design. The non-specific
showed similar effects when compared to the conventional effects of acupuncture, for example, the effects of con-
treatment group with respect to acute pain management text and the additional attention devoted to patients in the
and patient-perceived effectiveness. acupuncture group during administration of acupuncture,
may have played a role in the overall observed benefits in
the acupuncture group. Overall, the current evidence found
Adjunctive use of acupuncture in uncontrolled
in this study was insufficient to accept or repute the use of
case series acupuncture in the ED settings.
Although the evidence concerning the effectiveness of
Two uncontrolled case series reported the beneficial effects acupuncture in symptom management in the ED cannot be
of acupuncture.10,11 Outcomes were measured using the VAS determined, some of the included studies aimed to test the
for pain, pain NRS, patient-reported 4-point scale (marked feasibility of acupuncture treatment in the ED setting.9,10
improvement, slight improvement, no change or symptom At least in the context of the included studies, it seems
deterioration) and patients’ satisfaction with acupuncture feasible to test the role of acupuncture treatment in ED sett-
treatment. Both studies assessed outcomes immediately ings. However, the studies were primarily conducted in the
after acupuncture treatment. One study further measured U.S. and Germany, and the feasibility of acupuncture treat-
pain using the VAS at 30 and 60 min post-acupuncture treat- ment in the ED in different contexts or countries may not
ment and showed a reduction in pain.10 Telephone follow-up be the same as those in included studies. Because there are
within 72 h of discharge was conducted to measure pain NRS, only two RCTs, studies testing the feasibility of acupuncture
patient satisfaction and adverse events. Of the 20 patients, treatment should also be replicated in different contexts.
13 patients responded to the follow-up telephone interview A single-session acupuncture treatment was performed
and were noted to have a median NRS score of 3 out of 10. in all studies, which may not be clinically valid for other
The authors stated that this score was similar to that of the chronic conditions or in different clinical settings (e.g., pri-
historical control group (patients treated in the ED without mary care clinics or inpatient wards). This may partially
acupuncture treatment), which was noted to be 2.5 out of explain the diminished effects of acupuncture after the
10. discharge of ED. Given the limited time and resources for
acupuncture treatments in conventional ED settings, how-
Safety of acupuncture ever, it is unclear whether cumulative administration of
acupuncture is feasible in current ED settings. Whether and
Adverse events associated with acupuncture were reported how more sessions of acupuncture could be provided in ED
in one RCT and one case series.9,10 Adverse events were settings for better clinical outcomes should be explored in
observed in 6 of 32 patients in the acupuncture group in one future research.
RCT and occurred 14 times in the treatment of 20 patients in Both musculoskeletal and non-musculoskeletal condi-
one case series. All of these adverse events were minor and tions were treated in these studies, and the conditions
transient. Reported adverse events included pain and minor were characterised by heterogeneous disease characteris-
bleeding at the needled site, light-headedness, sweating, tics. Although individualised acupuncture treatments were
itching and erythema at the puncture site. used based on various theoretical frameworks (i.e., TCM,
Korean, Japanese and trigger point acupuncture) in the two
studies, conditions with different pathologies and charac-
Discussion teristics may have diluted the effect of acupuncture, which
were assessed using a single outcome (e.g., VAS). This might
In this review, we found that only two RCTs were available also partially explain the diminished effects of acupuncture
that had studied the use of acupuncture in the ED with a over time. Subgroup analysis by the treated conditions was
relatively small number of participants. No sham-controlled not available from the included studies due to the fact that
studies were found, demonstrating the lack of evidence for treatment results were not presented by condition. In future
the specific effects of acupuncture in ED settings as well trials, fewer heterogeneous inclusion criteria with respect to
as the focus of previous research on the adjunctive use disease condition may improve the interpretability of study
of acupuncture to standard medical care in the ED. The results. For example, a study protocol of multicentre RCTs
use of acupuncture in addition to standard medical care for acupuncture in low back pain, ankle sprain and migraine
in the ED showed inconsistent outcomes with respect to its through the EDs of four large tertiary teaching hospitals in
analgesic effects immediately after acupuncture treatment Australia was recently published.12 In that study, both pain
(one study showed favourable outcomes with acupuncture, ratings on the VAS as a primary outcome and condition-
while the other did not). After discharge from the ED, specific functional outcomes as secondary outcomes are to
patient-reported pain relief was similar in both groups, be measured.
suggesting the short-term effects of a single acupuncture The integration of various health-care interventions,
session in an ED setting. This tendency was also observed for example, physical therapy13,14 and allied health
Acupuncture in emergency department settings 71

interventions,15 for non-life-threatening conditions in the in the ED setting. The studies identified showed that it
ED was encouraged to improve the treatment process and was feasible to provide acupuncture treatment in the ED
clinical outcome and to enhance the cost-effectiveness of for non-emergent musculoskeletal pain and other non-
medical services in the ED. In our review, simple clinical musculoskeletal conditions, although this may not be the
outcomes, including the pain VAS or patients’ self-perceived case in different ED settings and cultures. Adverse events
improvements, were reported. How the use of acupuncture associated with acupuncture were minimal. However, more
treatment can make the triage process or the fast-track data on the safety of acupuncture practised in the ED are
management of common non-life-threatening conditions needed, as the risk of context-specific adverse events (e.g.,
more efficient was not formally addressed in the included forgotten needles in a busy ED context that may cause
studies. However, it is an important research agenda with migration of needles, which should be removed by surgical
respect to the utilisation of acupuncture in the ED. Future interventions) could not be excluded. Overall, the routine
studies should also address the process- and cost-related use of acupuncture in the ED will not be supported until more
benefits of acupuncture use in the ED. sound evidence is established by several well-conducted
RCTs that address the feasibility, clinical effectiveness and
Strengths and limitations safety of acupuncture.

To the best of our knowledge, this is the first systematic Conclusion

review of the role of acupuncture treatment in ED settings.
This review is based on two small RCTs; thus, we found insuf- There is insufficient evidence for the use of acupuncture
ficient evidence concerning the use of acupuncture in an in ED settings. The paucity of RCTs and the suboptimal
ED setting. However, this study also highlights the need for methodological qualities used in the studies are the primary
future research to address the paucity of evidence found barriers for evaluating the evidence on this topic. Future
in the review process. Although we imposed no language well-conducted RCTs are needed to determine whether
restriction on the search strategy, only English databases acupuncture can be established as an adjunctive or stand-
were searched, and some studies that reported in Russian alone treatment option for common non-life-threatening
and French were excluded due to resource limitation. Thus, conditions in the ED.
those excluded studies as well as potential unidentified eli-
gible studies indexed in local non-English databases (such as
Chinese and Korean databases) may exist. Thus, our review
Conflict of interest statement
may be susceptible to language bias. The simple search
terms used in this study may not detect potentially relevant None declared.
articles that deal with more specific topics (e.g., acupunc-
ture and acute conditions), and this should be pointed out References
as a limitation of our study. Finally, this review focussed
on the qualitative synthesis of trial results and the assess- 1. Thomas KJ, MacPherson H, Thorpe L, Brazier J, Fitter M, Camp-
ment of internal validity. Thus, the external validity and bell MJ, et al. Randomised controlled trial of a short course
generalisability of our study results are beyond the scope of traditional acupuncture compared with usual care for per-
of this review, although they are important for the replica- sistent non-specific low back pain. British Medical Journal
tion of study results or the inclusion of evidence into clinical 2006;333(7569):623.
guidelines.16 2. van den Berg I, Tan L, van Brero H, Tan KT, Janssens
AC, Hunink MG. Health-related quality of life in patients
with musculoskeletal complaints in a general acupuncture
Implications for future research practice: an observational study. Acupuncture in Medicine
2010;28(3):130—5.
Large, well-conducted RCTs should be performed to eval- 3. Wang LP, Zhang XZ, Guo J, Liu HL, Zhang Y, Liu CZ, et al.
uate the effectiveness of acupuncture in the ED setting. Efficacy of acupuncture for acute migraine attack: a multicen-
The feasibility, acceptability and outcomes associated with ter single blinded, randomized controlled trial. Pain Medicine
health-care resource utilisation should also be tested and 2012;13(5):623—30.
replicated in various ED contexts. Observational studies 4. Witt C, Brinkhaus B, Jena S, Linde K, Streng A, Wagenpfeil S,
et al. Acupuncture in patients with osteoarthritis of the knee:
regarding the safety of acupuncture and the acceptabil-
a randomised trial. Lancet 2005;366(9480):136—43.
ity of acupuncture both to ED staff and to patients may 5. Bennetts S, Campbell-Brophy E, Huckson S, Doherty S, National
be conducted in EDs that are already running acupuncture Health and Medical Research Council’s National Institute for
treatment in conjunction with conventional ED treat- Clinical Studies National Emergency Care Pain Management
ments. Study reporting should adhere to the CONSORT and Initiative. Pain management in Australian emergency depart-
acupuncture-specific CONSORT extension (i.e., STRICTA) to ments: current practice, enablers, barriers and future direc-
understand the detailed study context and procedure as well tions. Emergency Medicine Australasia 2012;24(2):136—43.
as to maximise the external validity of trial results. 6. MacPherson H, Altman DG, Hammerschlag R, Youping L, Taixiang
W, White A, et al. Revised STandards for Reporting Interven-
tions in Clinical Trials of Acupuncture (STRICTA): extending the
Implications for clinical practice CONSORT statement. PLoS Medicine 2010;7(6):e1000261.
7. Higgins JPT, Altman DG, Sterne JAC. Chapter 8: assessing risk
The current evidence is insufficient to provide any rec- of bias in included studies. In: Higgins JPT, Green S, editors.
ommendation concerning the utilisation of acupuncture Cochrane handbook for systematic reviews of interventions
72 K.H. Kim et al.

version 5.1.0 (updated March 2011). The Cochrane Collabo- and migraine in emergency departments: study protocol for a
ration; 2011. Available: http://www.cochrane-handbook.org,; randomized controlled trial. Trials 2011;12:241.
[accessed 24.10.12]. 13. Lebec MT, Jogodka CE. The physical therapist as a mus-
8. Goertz CM, Niemtzow R, Burns SM, Fritts MJ, Crawford CC, Jonas culoskeletal specialist in the emergency department. The
WB. Auricular acupuncture in the treatment of acute pain syn- Journal of Orthopaedic and Sports Physical Therapy 2009;39(3):
dromes: a pilot study. Military Medicine 2006;171(10):1010—4. 221—9.
9. Harkin C, Parker R. A prospective, randomised control 14. Fleming-McDonnell D, Czuppon S, Deusinger SS, Deusinger
trial of acupuncture for select common conditions within RH. Physical therapy in the emergency department: develop-
the Emergency Department. Journal of Chinese Medicine ment of a novel practice venue. Physical Therapy 2010;90(3):
2007;85:45—52. 420—6.
10. Arnold AA, Ross BE, Silka PA. Efficacy and feasibility of 15. Arendts G, Fitzhardinge S, Pronk K, Donaldson M, Hutton
acupuncture for patients in the ED with acute, nonpenetrating M, Nagree Y. The impact of early emergency depart-
musculoskeletal injury of the extremities. American Journal of ment allied health intervention on admission rates in older
Emergency Medicine 2009;27(3):280—4. people: a non-randomized clinical study. BMC Geriatrics
11. Fleckenstein J, Schottdorf J, Kreimeier U, Irnich D. Acupuncture 2012;12:8.
in emergency medicine: results of a case series. Der Anaesthe- 16. Glasziou P, Meats E, Heneghan C, Shepperd S. What is missing
sist 2011;60(9):854—62 [article in German]. from descriptions of treatment in trials and reviews? British
12. Cohen M, Parker S, Taylor D, Smit de V, Ben-Meir M, Cameron P, Medical Journal 2008;336(7659):1472—4.
et al. Acupuncture as analgesia for low back pain, ankle sprain