American Journal of Emergency Medicine (2009) 27, 280284

www.elsevier.com/locate/ajem

Original Contribution

Efficacy and feasibility of acupuncture for patients in the

ED with acute, nonpenetrating musculoskeletal injury of
the extremities
Alissa A. Arnold BS, B. Evan Ross DOM, LAc , Paul A. Silka MD
Ruth and Harry Roman Emergency Department, Cedars-Sinai Medical Center, Cedars-Sinai Research Institute, Los Angeles,
CA 90048, USA
Received 16 August 2007; revised 13 February 2008; accepted 14 February 2008

Abstract
Objectives: We conducted a pilot study to assess the efficacy of acupuncture as an analgesic
intervention for patients presenting to the emergency department (ED) after minor acute trauma to the
extremities. In addition, we sought to assess the feasibility of performing acupuncture in this setting.
Methods: Acupuncture was used as primary analgesia for a convenience sample of ED patients with
acute, nonpenetrating extremity injury. Efficacy was measured using a visual analog scale before
treatment, immediately after acupuncture (time 0), and every 30 minutes thereafter. A telephone call was
made to patients within 72 hours to ascertain pain levels using a 0 to 10 numerical rating scale. Markers
of feasibility included average time patients spent in the fast track area of the ED vs average time in the
department (TID) for all fast track patients with similar injury.
Results: Of 47 patients approached, 20 (43%) consented to participate. The mean age of those who
consented was 33 years, and 70% (n = 14) were male. Median change in visual analog scale score for
preacupuncture vs time 0 was 16 mm, with range of 0 to 60 mm. Median numerical rating scale score at
time of discharge and at follow-up was 3. Median TID was 135 minutes, with a range of 55 to 255 minutes.
Patients with extremity injury who did not receive acupuncture had a median TID of 90 minutes.
Conclusions: This study suggests that acupuncture can be an effective analgesic intervention for patients
with acute injury to the extremities. Acupuncture did not increase patients' TID. Minor complications
were reported.
2009 Elsevier Inc. All rights reserved.

1. Introduction

Abstract presented at the 9th Annual SAEM Western Regional

Research Forum, Redondo Beach, CA, March 2006, and CAL/ACEP 34th
Annual Scientific Assembly, Palm Springs, CA, June 2005.

Deceased.
Corresponding author. Department of Emergency Medicine, CedarsSinai Medical Center, Los Angeles, CA 90048, USA. Tel.: +1 310 423 7945.
E-mail addresses: alissa.arnold@cshs.org, alissaarnold@hotmail.com
(A.A. Arnold).
0735-6757/$ see front matter 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.ajem.2008.02.018

Acupuncture and emergency medicine are not commonly

associated entities. However, because it is one of the primary
duties of the emergency physician to provide pain relief to
acutely injured or ill patients in a timely manner, it is
reasonable to explore all analgesic options in the emergency
department (ED). Furthermore, as some patients and
physicians may avoid traditional analgesics because of
potential allergic reactions, patient intolerance, fears of

Acupuncture for patients in the ED

masking disease, and concern of addiction, an investigation
of the effectiveness and feasibility of acupuncture as an
alternative method of pain management is valid.
Acupuncture has been used by millions of Americans and
is performed by thousands of physicians, dentists, and
acupuncturists for relief of pain and for a broad range of
medical conditions [1,2].
A National Institutes of Health (NIH) consensus statement has noted that previous investigations have found
acupuncture to be efficacious for treating a myriad of
conditions causing pain and discomfort [1-3]. These authors
have suggested that acupuncture is well tolerated and has low
incidence of deleterious adverse effects. The NIH has further
reported that there is sufficient evidence of acupuncture's
value to expand its use into conventional medicine and to
encourage further studies [1,2].
We conducted a pilot study to assess the efficacy of acupuncture as an analgesic intervention for patients presenting
to the ED after minor acute trauma to the extremities. In
addition, we sought to assess the feasibility of performing
acupuncture in this setting.

2. Methods
A pilot study using acupuncture as the sole method of
analgesia was performed in our ED. Our ED is a level 1
trauma center with 76 000 annual patient visits. This study
was performed in the 7-bed fast track area of the ED.
Ambulatory patients with injury or illness that is not believed
to cause significant respiratory or cardiovascular compromise are triaged to the fast track area by a certified triage
nurse. All patients are treated by a board-prepared or boardcertified emergency physician in the fast track. Institutional
review board approval was obtained for this study.
Over a period of 2 1/2 years, from August 2004 to
January 2007, a convenience sample of 20 healthy patients
between 18 and 60 years of age presenting to the ED with
acute, nonpenetrating injury of the extremities were
consented to participate in this study. One of the
investigators reviewed EmSTAT (Allscripts), our computerized patient-tracking and nurse documentation system, to
identify potential study candidates based on the patient's
chief complaint. Patients identified as having the appropriate injuries were then interviewed to determine if the
additional following criteria were met:
o presents with acute musculoskeletal extremity pain due
to nonpenetrating injury
o aged between 18 and 60 years
o had vital signs within normal limits at time of triage:
normal blood pressure with systolic blood pressure of
100 to 140 mm Hg and diastolic blood pressure less
than 100 mm Hg; heart rate of 60 to 100 beats/min;
respiratory rate not greater than 22 breaths/min and not

281

o
o
o
o
o

less 12 breaths/min; and temperature greater than

96.5F and not greater than 100.4F.
presented with no evidence of open fracture or laceration
presented with no evidence of neurovascular
compromise
was not on blood thinners
was free of chronic disease
had not undergone surgery or experienced chronic pain
in the injured extremity region within the past
6 months

For the purpose of this study, an acute injury was one

that had occurred within 36 hours of presentation.
Patients were consented in the ED by one of the authors
(A.A.A.) after this screening. For those patients who
consented to participate, the treating ED physician was
asked not to administer traditional analgesics before or during
acupuncture. If analgesic medication was requested by the
patient and administered in the ED before acupuncture,
patients were withdrawn from the study. Of our patient
sample (n = 20), all received the standard ED interventions of
ice, elevation, and splinting when appropriate.
Patients were evaluated by a licensed acupuncturist, and
appropriate therapy was performed based on their oriental
medicine working diagnosis. Three licensed acupuncturists
performed the acupuncture therapy throughout the study.
Acupuncture efficacy was measured using a visual analog
scale (VAS) to assess pain level before acupuncture therapy,
immediately after the treatment (time 0), and for every
30 minutes for the duration of the patients' stay in the ED
(time 1, time 2, etc) [4-6]. A 0 to 10 numerical rating scale
(NRS) was performed in addition to the last visual analog
score obtained during the ED visit.
Acupuncture was provided in conjunction with conventional diagnostic evaluation in the ED by the attending
emergency physician. In the event that the patient did not
experience clinically significant pain relief from the acupuncture, traditional pain medication was offered at the
discretion of the treating emergency physician and the
patient. If additional pain medication was requested and
administered in the ED after acupuncture, the patient was
withdrawn from further analysis.
Telephonic follow-up was conducted within 72 hours of
discharge to inquire about complications and obtain an NRS
score and a satisfaction with acupuncture score. Discharge
NRS scores were compared with NRS scores at the time of
telephonic follow-up. Patients' satisfaction with acupuncture
was determined by a 1 to 5 Likert scale, with 1 being not at
all satisfied and 5 being satisfied beyond expectations.
The NRS scores that are routinely obtained by a registered
nurse (RN) at discharge from all fast track patients were used
as a historical control.
The markers for feasibility of acupuncture being
performed in the ED are length of time subjects spent in
the ED, or time in the department (TID), the incidence of
complications, and overall patient satisfaction with

282
Table 1

A.A. Arnold et al.

Primary diagnoses of patients receiving acupuncture

Wrist contusion (n = 2)
Forearm contusion (n = 3)
Hand contusion (n = 1)
Leg contusion (n = 2)
Knee injury (n = 2)
Wrist injury (n = 1)
Ankle injury (n = 1)
Foot injury (n = 1)
Sprained ankle (n = 2)
Sprained foot (n = 1)
Tendonitis (n = 1)
Closed wrist fracture (n =1)
Closed ankle fracture (n = 1)
Closed hand fracture (n =1)
Closed metacarpal fracture (n =1)

Fig. 1

acupuncture. Included in TID is the time of initial

assessment by the triage nurse to the time of discharge
from the ED.
Discharge NRS scores and TID for nonacupuncture
patients were compared with discharge NRS scores and
TID for patients receiving acupuncture. All control data were
obtained from an EmSTAT database query of 20 ageappropriate fast track patients with acute extremity injury.
Complications were determined by direct observation of
the treatment and patient self-reporting.
The Wilcoxon signed rank test, median, and ranges were
used to determine statistical significance.

3. Results
Of 47 patients approached, 20 (43%) consented to participate. The mean age of those who consented was 33 years,
and 70% (n = 14) were male.
Table 1 indicates the diagnoses of patients receiving
acupuncture. Several patients had more than one diagnosis.
The most frequent primary diagnoses were contusion and
non-bony injury of a joint.
The median preacupuncture (n = 17) VAS score was
57 mm, with range of 10 to 100 mm. At time 0 (n = 16), the
median VAS score decreased to 36.5 mm, with a range
of 0 to 94 mm (Table 2) (Fig. 1). Median change in
Table 2

VAS score from preacupuncture to time 0 (n = 16)

was 16 mm, with a range of 0 to 60 mm (P b .0001).
From preacupuncture to time 1 (n = 8), median change
was 19 mm, with a range of 9 of 66 mm (P = .008)
(Table 2) (Fig. 1). Further significant change was not
reported with subsequent VAS scores during the patients'
ED course of treatment.
Three patients did not complete the protocol and were not
included in the above analysis. Each of these patients
requested and was administered additional analgesic medication in the ED immediately after acupuncture. These patients
had a median VAS score of 73 mm during preacupuncture. At
time 0, before they received the analgesic medication, it
decreased to 41 mm.
A fourth patient with the primary diagnosis of wrist
contusion was administered additional medication in the ED.
This patient had a baseline VAS score of 67 mm, which
decreased at time 0 to 47 mm immediately after acupuncture.
Approximately 60 minutes after acupuncture, the patient
developed a cephalgia and was administered 600 mg ibuprofen
in the ED for this pain. This individual was included in the VAS
analysis because of time of medication administered in relation
to acupuncture received.
Follow-up information was obtained from 13 patients
because 4 patients could not be reached. The median NRS
score at discharge and at follow-up was 3, with a range of
0 to 7 at discharge and a 0 to 6 at follow-up. The median NRS
score at discharge for the control group (n = 20) was 2.5, with
a range of 0 to 9, suggesting no difference between the
2 groups (P = .62).
The reported median satisfaction score at follow-up for
patients receiving acupuncture (n = 12) was 5, with a range
of 3 to 5.

Change in pain scores during acupuncture treatment in the ED

Median VAS scores (mm)

VAS pre
VAS T0
VAS T1
VAS T2

Median VAS score (mm) vs time.

Change in median and range VAS scores (mm)

Median

Min

Max

17
16
8
3

57
36.5
28.5
27

10
0
5
16

100
94
91
38

VAS preVAS T0
VAS preVAS T1
VAS T0VAS T1

Pre indicates preacupuncture; T0, time 0; T1, time 1; T2, time 2.

Median

Range

Interquartile range

P value

16
8
7

16
19
3

60 (0-60)
57 (9-66)
30

20.5
38.5
24

b.0001
.008
.58

Acupuncture for patients in the ED

Table 3

283

Complications of acupuncture

Complications

No. occurrences

Local pain in puncture region

Lightheadedness
Itching in puncture region
Erythema in puncture region
Perspiration
Minor bleeding in puncture region

2
4
1
1
2
4

Complications reported and observed during acupuncture

therapy are listed in Table 3. The bleeding experienced
because of acupuncture was estimated to be less than 2 mL
(or approximately 1 drop) and stopped after brief, direct
pressure. One patient reported swelling at the site of needle
insertion at the 72-hour follow-up. However, this patient did
not feel that there was a need for reevaluation of the swelling.
No other complications were observed or reported.
There was no difference in TID for patients receiving
acupuncture vs the control group. The median TID was
135 minutes, with a range of 55 to 255 minutes, for acupuncture patients, whereas median TID was 90 minutes, with
a range of 52 to 270 minutes, for the control group (P = .07).

4. Limitations
We compared patients receiving acupuncture to a
historical control group of patients who did not receive
acupuncture and presented with similar injuries in an attempt
to understand if there was a difference in the NRS obtained at
time of discharge between these groups. We consider this
methodology to have limitations.
We acknowledge that there may be variation in NRS
scores from nonacupuncture patients who had their scores
obtained by an RN unaffiliated with the study. Although
RNs routinely collect this information for all ED patients
at discharge, the obtained values may vary based on the
instructions and frame in which the NRS pain score was
obtained. All of the pain scores obtained in the
acupuncture group were obtained by a single investigator
in a similar fashion.
Further limiting comparison between the 2 groups was
our inability to obtain VAS scores from the nonacupuncture
group. Without the serial VAS scores from this group, we
were unable to understand the progression of pain during the
ED stay. In future work, we would recommend serial VAS
scores for the control group.
This pilot study was conducted as a test of concept.
Future, larger prospective studies comparing control groups
to those patients receiving acupuncture may provide further
understanding of the efficacy of this treatment in the
acute setting.
Control groups that may be considered for larger studies
with patients presenting with acute, nonpenetrating injuries

of the extremities could include patients treated with sham

acupuncture or patients treated with only physical measures
of cold therapy and splinting.
Our study also had 3 different acupuncturists performing therapy throughout the study. This may have led to
variation in therapeutic approaches despite similar injuries.
Competing orientations that may lead to different therapeutic approaches exist (ie, the use of different acupuncture points) [2]. Future studies may consider standardizing
therapeutic approaches or using a single acupuncturist.

5. Discussion
This pilot study studied the use of acupuncture as
primary analgesic treatment for patients presenting to the
ED with acute, nonpenetrating injuries to the extremities.
We found that acupuncture provided analgesia without
incidence of significant complications. Other authors have
found similar analgesic efficacy when using acupuncture,
but to our knowledge, no other study has been performed in
the ED [1-3]. In addition, we found acupuncture to be a
treatment that was feasible in our ED setting.
We observed a significant decrease in pain scores after
acupuncture. Median VAS score decreased by 20.5 mm
(Table 2) when comparing preacupuncture scores to those
obtained at time 0. This change in VAS score represents not
only a statistically significant improvement but also a
clinically significant improvement. Todd et al [5] determined
that a 13-mm decrease in VAS score represents a clinically
significant reduction in a patient's perceived pain, even in the
absence of statistical significance. Serial VAS scores suggest
that analgesic relief was sustained for the patients' duration
in the ED (Fig. 1).
Four patients requested and were administered additional
analgesic medication in the ED after acupuncture. Of these
4 patients, 3 had a median VAS score of 73 mm during
preacupuncture. At time 0, their median VAS score
decreased to 41 mm. Their median change in VAS score
was 44 mm during the immediate postacupuncture period
(pre-acupuncture to time 0), before the requested analgesic
medication was administered, suggesting that acupuncture
provided some degree of analgesic efficacy [5]. Despite
improvement in their pain scores, they were considered
protocol violations, and no further analysis of their results
was performed.
At the 72-hour follow-up, we found no significant change
in the median NRS score compared with that at ED
discharge. We did not anticipate that a single acupuncture
treatment would provide extended analgesia. We feel that
this measure suggests that our cohort did not have worsening
pain after discharge from the ED after this therapy.
In addition, we found no difference in discharge pain
scores between patients who received acupuncture and those
who did not (P = .62). This suggests that for patients with

284
acute extremity injury, acupuncture may be as effective as
traditional analgesic therapy for the duration of the patients'
stay in the ED. We recommend further study.
Several minor complications were observed or reported
during acupuncture therapy (Table 3). None of these
complications necessitated terminating the acupuncture or
required treatment. The minor complications we observed
seem to be in alignment with the NIH consensus statement
that notes that an advantage of acupuncture is the lower
incidence of deleterious adverse effects compared with that
of many drugs or other accepted medical procedures used for
the same conditions [2]. We feel that these complications are
felt to be minor, not impacting the ability to perform
acupuncture in the ED.
At the time of the 72-hour telephonic follow-up, one
patient reported swelling at the site of needle insertion. This
patient's NRS score at follow-up was 1, lower than the
median follow-up NRS score of 3 for all patients contacted
(n = 13). This individual did not request further evaluation of
this complication. We concluded that this complication was
mild, as the patient declined further care. No further
complications were reported at the time of follow-up.
We found that acupuncture in the ED did not lead to
significant delays. Median TID of patients receiving
acupuncture was 135 minutes, with a range of 55 to
255 minutes, whereas the control group had a median TID
of 90 minutes, with a range of 52 to 270 minutes. We found
no significant difference in TID between those who received
acupuncture and those who did not (P = .07). These data
suggest that acupuncture would be feasible in a fast-paced
ED setting.
Acupuncture patients (n = 12) also reported a high median
satisfaction score (5 of 5), with a range of 3 to 5.
We advocate that future studies incorporate a randomized, controlled sample of patients to receive sham
acupuncture, which has been used successfully as placebo
in past studies [2,3,7,8]. Greater acupuncturist availability
may minimize the introduced selection bias, as our
convenience sample was obtained intermittently on weekday
afternoons. In addition, we did not study individual

A.A. Arnold et al.

approaches used by the acupuncturists but would recommend that future studies follow the NIH suggested
methodology of incorporating fixed acupuncture points for
like injury or disease processes [2].

6. Conclusion
We found that acupuncture may be an efficacious, safe,
and feasible analgesic alternative for patients presenting to
the ED with minor acute injury to the extremities.

Dedication
This manuscript is dedicated in loving memory to
B. Evan Ross.

References
[1] Acupuncture. NIH consensus statement online. 1997 Nov 3-5; 15(5):134. National Library of Medicine Web site. Available at: http:www.ncbi.
nlm.nih.gov/books/bv.fcgi?rid=hstat4.chapter.18129. Accessed December 18, 2006.
[2] NIH Consensus Development Panel on Acupuncture. Acupuncture:
NIH consensus conference. JAMA 1998;280:1518-24.
[3] Marwick C. Acceptance of some acupuncture applications. JAMA
1997;278:1725-7.
[4] Todd KH. Pain assessment instruments for use in the emergency
department. Emerg Med Clin North Am 2005;23:285-95.
[5] Todd KH, Funk KG, Funk JP, et al. Clinical significance of reported
changes in pain severity. Ann Emerg Med 1996;27:485-9.
[6] Bijur PE, Silver W, Gallagher EJ. Reliability of the visual analog
scale for measurement of acute pain. Acad Emerg Med 2001;8:
1153-7.
[7] Pariente J, White P, Frackowiak RSJ, et al. Expectancy and belief
modulate the neuronal substrates of pain treated by acupuncture.
Neruorimage 2005;25:1161-7.
[8] Kaptchuk TJ. Acupuncture: theory, efficacy, and practice. Ann Intern
Med 2002;136:374-83.