THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE

Volume 23, Number 3, 2017, pp. 164–179

ª Mary Ann Liebert, Inc.
DOI: 10.1089/acm.2016.0155

REVIEW ARTICLE

Acupuncture for the Treatment of Peripheral Neuropathy:

A Systematic Review and Meta-Analysis
Alexandra Dimitrova, MD, Charles Murchison, MS, and Barry Oken, MD, PhD

Abstract
Objectives: Neuropathy and its associated pain pose great therapeutic challenges. While there has been a
recent surge in acupuncture use and research, little remains known about its effects on nerve function. This
review aims to assess the efficacy of acupuncture in the treatment of neuropathy of various etiologies.
Methods: The Medline, AMED, Cochrane, Scopus, CINAHL, and clintrials.gov databases were systemati-
cally searched from inception to July 2015. Randomized controlled trials (RCTs) assessing acupuncture’s
efficacy for poly- and mononeuropathy were reviewed. Parallel and crossover RCTs focused on acupuncture’s
efficacy were reviewed and screened for eligibility. The Scale for Assessing Scientific Quality of Investigations
in Complementary and Alternative Medicine was used to assess RCT quality. RCTs with score of >9 and
active control treatments such as sham acupuncture or medical therapy were included.
Results: Fifteen studies were included: 13 original RCTs, a long-term follow-up, and a re-analysis of a prior
RCT. The selected RCTs studied acupuncture for neuropathy caused by diabetes, Bell’s palsy, carpal tunnel
syndrome, human immunodeficiency virus (HIV), and idiopathic conditions. Acupuncture regimens, control
conditions, and outcome measures differed among studies, and various methodological issues were identified.
Still, the majority of RCTs showed benefit for acupuncture over control in the treatment of diabetic neuropathy,
Bell’s palsy, and carpal tunnel syndrome. Acupuncture is probably effective in the treatment of HIV-related
neuropathy, and there is insufficient evidence for its benefits in idiopathic neuropathy. Acupuncture appears to
improve nerve conduction study parameters in both sensory and motor nerves. Meta-analyses were conducted
on all diabetic neuropathy and Bell’s palsy individual subject data (six RCTs; a total of 680 subjects) using a
summary estimate random effects model, which showed combined odds ratio of 4.23 (95% confidence interval
2.3–7.8; p < 0.001) favoring acupuncture over control for neuropathic symptoms.
Conclusions: Acupuncture is beneficial in some peripheral neuropathies, but more rigorously designed
studies using sham-acupuncture control are needed to characterize its effect and optimal use better.

Keywords: neuropathic pain, neuropathy, acupuncture, integrative medicine

Introduction Peripheral neuropathy (PN) is broadly defined as damage to

the peripheral nervous system caused by a primary lesion or
dysfunction,10 with polyneuropathy involving multiple nerves
A cupuncture has recently emerged as an impor-
tant integrative medicine treatment, in both the hospital
and clinic settings.1,2 In 1998, an National Institutes of Health
and mononeuropathy involving a single nerve. The most
common causes of polyneuropathy are diabetes, thyroid dis-
Consensus Development Panel concluded that acupuncture is orders, vitamin B12 deficiency, alcohol abuse, chemotherapy,
efficacious for postoperative and chemotherapy-induced nau- and human immunodeficiency virus (HIV) infection.10,11 In
sea and vomiting, and helpful in other conditions, including spite of exhaustive workups, no cause is identified in one third
stroke rehabilitation.3 Newer evidence-based reviews have of PN cases, and they are considered idiopathic.11,12 The most
shown the therapeutic benefits of acupuncture for chronic low- common mononeuropathies are carpal tunnel syndrome (CTS),
back pain,4,5 migraine and tension headache,6,7 chemotherapy- which is caused by demyelination of the median nerve at the
induced nausea and vomiting,8 and other pain conditions.9 wrist,10 and Bell’s palsy, which involves the facial nerve and

Department of Neurology, Oregon Health and Science University, Portland, OR.

164
ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 165

causes facial weakness.13 PN is often accompanied by painful Outcome measures

paresthesias, which present a great therapeutic challenge.10,14 One of the following outcome measures was required
Integrative treatments such as acupuncture are used in pain for inclusion: nerve conduction studies (NCS); validated
clinics with increasing popularity and are gaining acceptance questionnaires such as Symptom Severity Score, the Vi-
among the U.S. public and academic community.15,16 This sual Analogue Scale, Global Symptom Score, or McGill
acceptance of acupuncture has prompted growing research into Pain Questionnaire score; or clearly defined study-specific
integrative therapies for the challenging treatment of neuro- objective criteria distinguishing responders from non-
pathic pain. To the authors’ knowledge, there has been no responders.
critical, systematic appraisal of the effects of acupuncture
across all types of neuropathy, including both poly- and
mononeuropathy. Data extraction
The review team consisted of two neurologists with
Objectives expertise in acupuncture, neuromuscular disorders, neuro-
The objective of this review is to examine critically the physiology, and statistical methodology. Initially, an ab-
evidence in the published literature for the safety and effi- stract review was performed to determine study eligibility.
cacy of acupuncture in the treatment of PN and associated Studies were excluded if they were not RCTs or did not
symptoms compared to sham acupuncture or conventional have acupuncture or PN as their focus. Other reasons for
medical therapy. Furthermore, acupuncture’s effectiveness exclusion were the lack of a non-acupuncture control
across various neuropathic conditions, as evidence permits, condition, such as comparing acupuncture with manual
is also assessed. manipulation to acupuncture without manual manipulation
or comparing acupuncture plus moxibustion to acupunc-
ture alone. The remaining studies were reviewed in detail.
Methods
The Scale for Assessing Scientific Quality of Investigations
Search strategy in Complementary and Alternative Medicine Version (SAS-
QI CAM), an instrument specifically designed to evaluate
The following electronic databases were searched from in-
RCTs focused on integrative therapies,17 was utilized. It as-
ception to July 2015: Medline, Cochrane Library, Scopus,
sesses methodological problems and sources of bias com-
AMED, CINAHL, and Clintrials.gov. The key search terms
monly encountered in integrative medicine research, such as
used were ([‘‘acupuncture’’ OR ‘‘acup*’’ OR ‘‘laser acupunc-
allocation concealment, investigator and subject blinding,
ture’’ OR ‘‘moxibustion’’ OR ‘‘electroacupuncture’’] AND
blinded outcome assessment, reproducibility of experimental
[‘‘neuropathy’’ OR ‘‘peripheral neuropathy’’ OR ‘‘neuropathic
procedures, and sound statistical analysis. The two reviewers
pain’’ OR ‘‘neuralgia’’ OR ‘‘carpal tunnel syndrome’’ OR
assigned each paper a SAS-QI CAM score independently and
‘‘Bell’s palsy’’]). Review papers were used to identify addi-
in a blinded fashion. As the two reviewers were in agreement,
tional references. Language restrictions were set to English and
there was no need to use a third reviewed as tiebreaker. Based
subject restrictions to humans. Gray literature in clintrials.gov
on a recent study by D’Silva et al.,18 a SAS-QI CAM score of
was reviewed for unpublished relevant data. Parallel and
>9 (out of 21) served as the cutoff for inclusion based on RCT
crossover RCTs were selected if their focus was acupuncture as
overall scientific quality.
an intervention for the treatment of PN, regardless of blinding.
RCTs were selected if they included at least one group receiving
acupuncture and one control group receiving sham acupuncture Data synthesis plan
or another active treatment (e.g., medication).
All included RCTs on acupuncture for diabetic peripheral
neuropathy (DPN) and Bell’s palsy used composite improve-
Subject characteristics ment scales such as ‘‘markedly relieved/improved/failed’’19 or
Only adults (age >18 years) were included in the search ‘‘marked effectiveness/effectiveness/failure,’’20 reported at the
parameters, as neuropathy, with the exception of Bell’s palsy level of the individual subject. These were converted to di-
and hereditary neuropathies, is rare in children. The type of chotomous ‘‘no improvement’’ and ‘‘improvement’’ variables
PN studied was further stratified by etiology as diabetic, and included in meta-analysis using standard pooling proce-
chemotherapy-induced, HIV-induced, post-herpetic, idio- dures based on odds ratios. Contingency tables were developed
pathic, or mononeuropathy such as CTS or Bell’s palsy. with acupuncture as the treatment marginal and improvement
from baseline as the outcome marginal. Fisher’s exact test was
used to calculate the odds ratios as the primary endpoints.
Intervention
Weights for the studies were calculated based on reciprocal of
Acupuncture was defined as the stimulation of acupuncture the variance and normalized based on the total sum of the
points by needles that pierce the skin, with or without electric weights used in the meta-analyses. In addition to a standard
stimulation, with or without moxibustion (heating with the fixed-effects Mantel–Haenszel model, a random effects meta-
mugwort herb). Methods of stimulating acupuncture points analysis was done due to the expected increase in heterogeneity
without needle insertion (laser, acupressure) were also sear- resulting from combining Bell’s palsy and DPN studies. Be-
ched. Controls needed to be active and included sham acu- yond the basal evaluation, a separate meta-analysis was con-
puncture or relevant medical treatments such as drugs or ducted exclusively on the DPN subjects. All statistical analyses
injections. Co-interventions were allowed only if they were were conducted using R v3.0.221 with additional utility from
given to both the intervention and the control group. the ‘‘rmeta’’ package.22
166 DIMITROVA ET AL.

Results for the treatment of neuropathy caused by Bell’s palsy,45,59

Included studies
CTS,55,56,60–62 diabetes,19,20,63,64 HIV,57,58,65 and a combi-
nation of diabetes and idiopathic causes.66 Of the included
Initial searches generated 1008 records, and 772 re- RCTs, nine studied manual acupuncture with or without
mained after duplicate removal (Fig. 1). Of these, 726 moxibustion, three studied electroacupuncture, and one stud-
studies were excluded after abstract review, leaving 46 to ied a combination of manual and electroacupuncture. Other
be reviewed in detail for eligibility. Initial searches iden- active interventions included per os (P.O.) medications, in-
tified studies with a focus on acupuncture for the treatment tramuscular (i.m.)/intravenous (i.v.) injections, or splinting. As
of DPN, chemotherapy-induced neuropathy, HIV-related planned, all selected studies had an active control arm.
neuropathy, idiopathic neuropathy, zoster neuralgia, trau- All included RCTs studied adult male and female out-
matic peroneal neuropathy, and various cranial neuropa- patients, with the exception of Yu et al. who studied dia-
thies such as Bell’s palsy and trigeminal neuralgia. Of the betic neuropathy in the hospital setting19 (Table 2). While
46 studies that underwent full review, 15 were excluded most studies originated in China,19,20,45,56,59,63,64 others
because they were not RCTs23–37: three for an improper involved outpatients in Iran,60 Thailand,61 and the United
control condition, including verum acupuncture38–40; three States.58,62,65,66 Across studies, sample sizes ranged from
because they did not focus on acupuncture/involved multi- 16 to 480 subjects, treatment courses ranged from 20 days
ple interventions41–43; one44 because it was identified as a to 3 months, and acupuncture frequency ranged from daily
duplicate to an included paper45; one because it was de- to weekly. There was significant variability in how neu-
scriptive only46; and eight because of a low SAS-QI CAM ropathy was diagnosed, including neurologic exam, symp-
score (<10).47–54 tom scales, NCS, and serologic testing. The same applied to
Following a detailed review, 15 papers met the final in- outcome measures, which ranged from subjects’ symptom
clusion criteria (Table 1): 13 RCTs, a 13-month follow-up55 ratings to validated clinical scales (Table 3), NCS (Table 4),
of an original RCT,56 and a re-analysis57 of a previously serologic testing, and study-specific graded improvement
published paper.58 The selected studies focus on acupuncture scales.

FIG. 1. Database search flow diagram.

ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 167

Table 1. Study Quality Ratings Using the Scale for Assessing Scientific
Quality of Investigations in Complementary and Alternative Medicine
Study Condition SASQI-CAM Intervention Control
Li et al., 200445 Bell’s palsy 13/21 Acupuncture, moxibustion 1. Acupuncture, moxibustion,
meds P.O. and i.m.
2. Meds P.O. and i.m.a
Tong et al., 200959 Bell’s palsy 14/21 Acupuncture 1. Prednisolone P.O. +
pepcidine P.O.a
2. Conservative Tx
(non-medical)
Khosrawi et al., 201260 CTS 19/21 Acupuncture, splinting Sham acupuncture, splinting,
Vitamins B1 and B6 P.O.
Kumnerddee et al., 201061 CTS 14/21 Electroacupuncture Splinting
Yao et al., 201262 CTS 17/21 Acupuncture, splinting Sham acupuncture, splinting
Yang et al., 2009,56 201155 CTS 18/21 Acupuncture Prednisolone P.O.
Yu et al., 200119 DM 10/21 Electroacupuncture Meds P.O. and i.m.
Zhang et al., 201063 DM 11/21 Acupuncture Inositol P.O.
Zheng et al., 200464 DM 10/21 Acupuncture, electroacupuncture, Mecobalamin P.O.
snow lotus i.m. injection
Zuo et al., 201020 DM 12/21 Acupuncture, vitamin B12 i.v. Vitamin B12 i.v.
Anastazi et al., 201365 HIV 18/21 Acupuncture, moxibustion Sham acupuncture, placebo
moxibustion
Schiflett et al., 201157 HIV 15/21 Acupuncture Control (non-acupuncture)
points
Shlay et al., 199858 HIV 19/21 1. Acupuncture 1. Control (non-acupuncture)
2. Acupuncture, amitriptyline points
3. Acupuncture, placebo P.O. 2. Control points,
amitriptyline
3. Control points, placebo
P.O.
4. Amitriptyline
5. Placebo P.O.
Penza et al., 201166 DM/IPN 12/21 Electroacupuncture Sham electroacupuncture
a
Control condition included in meta-analysis.
CTS, carpal tunnel syndrome; DM, diabetes mellitus; i.m., intramuscular; P.O., per os (oral); HIV, human immunodeficiency virus; i.v.,
intravenous; IPN, idiopathic peripheral neuropathy; Tx, treatment; SASQI-CAM, scale for assessing scientific quality of investigations in
complementary and alternative medicine.

DPN ( p = 0.03). However, this was not true after adjustment for
Four RCTs met the eligibility criteria.19,20,63,64
These studies multiple comparisons. Shiflett et al.57 reanalyzed data from
were conducted in China and used manual acupuncture20,63,64 or this trial using baseline pain as a covariate in a repeated-
electroacupuncture19,64 in combination with snow lotus injec- measures analysis of covariance and found that baseline pain
tion in acupuncture points64 and i.v. vitamin B12.20 All studies intensity was significant ( p < 0.001) and so was the interaction
used P.O., i.m., or i.v. medication control (Table 2). Outcome of acupuncture and amitriptyline over time ( p < 0.017).
measures included study-specific graded composite im- Amitriptyline alone or acupuncture alone showed significant
provement scales based on symptoms,19,20,63,64 clinical reduction in Gracely Pain Scale score at 6 weeks. However,
signs,63 exam,19,63 labs,19 and NCS.19,20 The included studies only acupuncture sustained this effect at 14 weeks (Tables 2
reported significant improvement in acupuncture groups com- and 3). When amitriptyline and acupuncture were combined,
pared with control in neurologic exam,19,20,64 neuropathic they did not differ from placebo.
symptoms,19,63,64 composite improvement scales,19,20,63,64 Anastasi et al.65 also found significant reduction in Gracely
NCS,19,20 and blood rheology.19 Pain Scale following a 6-week course of manual acupuncture/
moxibustion compared to sham acupuncture/placebo mox-
ibustion control (Table 2). Additionally, the acupuncture group
HIV-related neuropathy had significant improvement on the Subjective Peripheral
Neuropathy Screen compared to its pretreatment baseline.
One included study of manual acupuncture for HIV-related
neuropathy from the early-mid 1990s58 had enrollment diffi-
Idiopathic neuropathy
culties and a 20–35% dropout rate due to significant HIV
morbidity. This led to a complicated ‘‘factorial option’’ design One RCT on electroacupuncture for idiopathic neuropathy
with eight trial arms, which made a direct comparison of was included.66 Four subjects with diabetic neuropathy and 12
amitriptyline to acupuncture impossible. The study found with idiopathic neuropathy were enrolled in a double-blind
significantly higher pain relief in the acupuncture group crossover study comparing electroacupuncture to sham-
 Table 2. Study Design, Population, Outcome Measures, and Results Summary
Acupuncture
Reference Study design Study population intervention Acupuncture course Control intervention Outcome measures Reported findings
45
Li 2004 3-arm RCT n = 480 Acupuncture, Treatment course: 1. Acupuncture, House–Brackmann Scale, Active intervention group vs. Control 2:
Bell’s palsy moxibustion 5· per week for moxibustion, FDIP, FDIS, Improvement Improvement Scale significantly better in active
e
China 4 weeks (20 total) prednisone, vitamin Scale intervention ( p < 0.05), ‘‘cured’’ ( p = 0.013),
Outpatients B1, dibasole P.O., ‘‘Obviously Improved’’ ( p = 0.024); House–
vitamin B12 i.m. Brackmann ( p = 0.005), FDIP ( p = 0.009)
2. Prednisone, vitamin Acupuncture Control 1 vs. Control 2
B1, dibasole P.O., ‘‘Obviously Improved’’ ( p = 0.014), House–
vitamin B12 i.m. Brackmann ( p = 0.005), FDIP ( p = 0.009)
Tong 200959 3-arm RCT n = 119 Acupuncture, eye Treatment course: 1. Prednisolone + House–Brackmann Scale Acupuncture vs. Control 1 vs. Control 2:
Bell’s palsy care, education 3· per week until pepcidine P.O., eye No significant difference in rate of recovery
China recovery or up to care, education among the three groups
Outpatients 3 months 2. Eye care, home facial Overall improvement to Grade 3 or better:
exercises, education 96.4% acupuncture vs. 86.9% Control 1 vs. 89.5%
Control 2
Khosrawi 2-arm RCT n = 64 Acupuncture, nightly Treatment course: Sham acupuncture, NCS including DML, DSL, Post vs. pre intervention:
201260 CTS wrist splinting 2· per week for vitamins B1 and SNCV, GSS Acupuncture group GSS improvement ( p < 0.001)
Iran 4 weeks (8 total) B6 P.O. Control group GSS unchanged ( p = 0.17)
Outpatients Acupuncture group SNCV faster ( p = 0.02)
Acupuncture vs. control at 4-week follow-up:
GSS improvement ( p < 0.001)
DSL improvement ( p = 0.07)
SNCV improvement ( p = 0.02)
Kumnerddee 2-arm RCT n = 61 Electroacupuncture Treatment course: Nightly wrist splinting BCTS including Symptom Post vs. pre intervention:
201061 CTS 2· per week for Severity Scale (SSS) and Acupuncture group reduction in SSS, FSS, and VAS

168
Thailand 5 weeks (10 total) Functional Status Scale score ( p < 0.05)
Outpatients (FSS), VAS Control group reduction in SSS ( p = 0.008), no
change in FSS and VAS
Acupuncture vs. control:
Greater VAS reduction in the acupuncture group
( p = 0.028)
Yao 201262 2-arm RCT n = 41 Acupuncture, nightly Treatment course: six Sham acupuncture six NCS including: DML, CMAP Post vs. pre intervention CTSAQ:
CTS wrist splinting weekly sessions weekly sessions, amplitude, CSI, CTSAQ Acupuncture group ( p < 0.05)
USA nightly wrist splinting Scale, key- and tip-pinch Control group ( p < 0.05)
Outpatients assessment 3-month follow-up CTSAQ improvement:
Acupuncture group ( p = 0.17)
Control group ( p = 0.02)
3-month tip-pinch, key-pinch, CSI:
No difference from baseline for either group
Acupuncture vs. control:
No difference in CTSAQ post intervention
Yang 200956 2-arm RCT n = 77 Acupuncture Treatment course: Prednisolone P.O. daily NCS including: DML, CMAP Post vs. pre intervention:
CTS 2· per week for for 4 weeks amplitude, MNCV, DSL, GSS acupuncture group ( p < 0.01)
China 4 weeks (8 total) SNAP amplitude, W-P GSS control group ( p < 0.01)
Outpatients SNCV, GSS NCS acupuncture group: decrease in DML, DSL,
increase in W-P SNCV, SNAP amplitudes ( p < 0.05)
NCS control group: decrease in DML, DSL,
Increase in W-P SNCV, SNAP amplitudes, APB
CMAP amplitude ( p < 0.05)
Acupuncture vs. control at 4-week follow-up:
GSS improvement post treatment ( p = 0.15)
Nocturnal awakening post treatment ( p = 0.03)
DML improvement ( p = 0.012)
(continued)
 Table 2. (Continued)
Acupuncture
Reference Study design Study population intervention Acupuncture course Control intervention Outcome measures Reported findings
Yang 2011 Same Same Same Same Same Same Post vs. pre intervention:
(LTF)55 13-month acupuncture group NCS all different from
baseline except for MNCV
13-month control group NCS no different from
baseline except for MNCV, W-P CNCV
( p < 0.01), DSL ( p < 0.05)
Acupuncture vs. control:
Improved 7- and 13-month GSS ( p < 0.01)
Improved DML and MNCV at 13 months ( p < 0.01)
Yu 200119 2-arm RCT n = 78 Electroacupuncture Treatment course: 10 Thiamine P.O., vitamin NCV, blood rheology, Acupuncture vs. control:
DM daily sessions · 3 B12 i.m.—same course neurologic exam, Overall improvement ( p < 0.01)
China Break: 3 days symptomatic improvement, NCV pre treatment ( p > 0.05)
a
Inpatients Improvement Scale NCV post treatment ( p < 0.01)
Post- vs. pre-intervention rheology:
Acupuncture group ( p < 0.01)
Control group ( p > 0.05)
Zhang 201063 2-arm RCT n = 65 Acupuncture Treatment course: 14 Inositol P.O. daily for Neurologic exam, Acupuncture vs. control:
DM daily sessions · 5 3 months symptomatic improvement, Overall improvement ( p < 0.05)
b
China Break: 5 days Improvement Scale
Outpatients

169
Zheng 200464 2-arm RCT n = 104 Acupuncture, Treatment course: 10 Mecobalamin P.O. Symptom score, clinical sign Post vs. pre intervention:
DM electroacupuncture, every other day daily · 2 months score, Improvement Scalec Acupuncture group overall improvement ( p < 0.01)
China snow lotus i.v. sessions · 2 Control group overall improvement ( p < 0.01)
Outpatients Acupuncture vs. control:
Overall improvement ( p < 0.001)
Symptom score ( p < 0.05)
Clinical sign score ( p < 0.05)
Zuo 201020 2-arm RCT n = 75 Acupuncture, Treatment course: daily Methylcobalamin i.v. NCV (MNCV, SNCV), Acupuncture vs. control:
DM methylcobalamin sessions for 4 weeks daily for 4 weeks Neurologic Exam Score, Symptom score ( p < 0.05)
China i.v. Improvement Scaled Neurologic Exam Score ( p < 0.001)
Outpatients Post vs. pre intervention:
Acupuncture group Neurologic Exam Score
( p < 0.001)
Acupuncture group median, peroneal MNCV
and SNCV ( p < 0.001)
Control group median, peroneal SNCV ( p < 0.001)
Anastasi 2-arm RCT n = 50 Acupuncture, Treatment course: Sham acupuncture, GPS, Subjective Peripheral Acupuncture vs. control:
201365 HIV moxibustion 2· per week for placebo moxibustion Neuropathy Screen (SPNS) GPS significantly better in acupuncture at 2-week
United States 6 weeks (12 total) follow-up ( p < 0.05)
Outpatients Post vs. pre intervention:
SPNS average score significantly improved in
acupuncture at 2-week follow-up ( p < 0.05)
(continued)
 Table 2. (Continued)
Acupuncture
Reference Study design Study population intervention Acupuncture course Control intervention Outcome measures Reported findings
58
Shlay 1998 8 groups n = 250 Acupuncture Treatment course: 14 Sham acupuncture, GPS, Global Pain Relief Acupuncture vs. sham acupuncture:
HIV weeks: 2· per week amitriptyline, placebo Score,f QOL questionnaire, No difference in pain or quality of life at 6 or 14
United States for 6 weeks, followed pill Neurologic Summary weeks
Outpatients by weekly for 8 Score More patients in the acupuncture group reported
weeks (20 total) moderate or more pain relief ( p = 0.03), not
significant after adjustment for multiple
comparisons
Amitriptyline vs. placebo:
No difference in pain or quality of life at 6 or 14
weeks
Schiflett 2011 2 · 2 table n = 125 Same Same Same Same Covariate analysis using baseline pain with repeated
(reanalysis HIV measures at weeks 6 and 14:
of Shlay United States Baseline pain significant ( p < 0.001), acupuncture +
1998)57 Outpatients amitriptyline interaction significant (p < 0.017)
ANCOVA with baseline pain as covariate:
Acupuncture associated with significant global pain
relief at weeks 6 and 14

170
Penza 201166 2-arm RCT n = 16 Electroacupuncture Treatment course: six Sham electroacupuncture VAS, BDI, SF-36 QOL, Acupuncture vs. control:
Crossover DM/IPN sessions every Patient’s Global No significant difference after treatment
United States 5–7 days, 12-week Impression of Change Post vs. pre intervention:
Outpatients break, crossover (PGIC) No significant difference compared to baseline
in either group
Crossover:
No significant difference based on order
of treatment
a
Markedly relieved/improved/failed based on exam and symptoms.
b
Remarkable effectiveness/effectiveness/ineffectiveness based on symptom and clinical sign score.
c
Marked/some/none based on symptoms, exam, labs, NCVs.
d
Marked effectiveness/effectiveness/failure based on symptoms.
e
Cured/obviously improved/improved/no improvement based on House–Brackmann Scale and Facial Disability Index.
f
Global Pain Relief Score—complete/a lot/moderate/slight/none/worse.
BCTS, Boston Carpal Tunnel Score; BDI, Beck Depression Inventory; CMAP, Compound Muscle Action Potential; CSI, Combined Sensory Index; CTSAQ, Carpal Tunnel Self-Assessment Questionnaire; DML, distal motor latency, DSL, distal sensory
latency; FDI, Facial Disability Index; FDIP, FDI Physical; FDIS, FDI Social; GPS, Gracely Pain Score; GSS, Global Symptom Scale; MNCV, motor nerve conduction velocities; NCS, nerve conduction studies; NCV, nerve conduction velocity; PN,
peripheral neuropathy; QOL, quality of life; SNAP, sensory nerve action potential; SF-36 QOL, San Francisco-36 Quality of Life inventory; SNCV, sensory nerve conduction velocities; W-P SNCV, wrist–palmar sensory nerve conduction velocity; VAS,
Visual Analog Scale.
 Table 3. Overview of Clinical Scales Used as Outcome Measures
Neuropathy-
related Post-acupuncture changes
Validated scale used condition Study/condition compared to baseline
BCTS including Symptom Severity CTS Kumnerddee 201061 Improvement in SSS, FSS at 5 weeks
( p < 0.05)
Scale (SSS) and Functional Status
Scale (FSS)
Carpal Tunnel Self-Assessment CTS Yao 201262 Improved post acupuncture ( p < 0.05)
Questionnaire (CTSAQ) No difference from baseline at 3 months
56
Global Symptom Score (GSS) CTS Yang 2009 Improved at 4 weeks ( p < 0.01)
Yang 201155 Improved at 7 and 13 months ( p < 0.01)
Khosrawi 201260 Improved at 4 weeks ( p < 0.001)
Facial Disability Index (FDIP, FDIS) Bell’s palsy Li 200445 Improved FDIP compared to control
at 4 weeks ( p = 0.009)
House–Brackmann Scale Bell’s palsy Li 2004 Improved over medication control
( p = 0.005)
Tong 200959 96.4% improvement to grade 3 or better
Gracely Pain Scale (GPS) HIV Anastasi 201365 Improved at 2-week follow-up ( p < 0.05)
HIV Schiflett 201157 N/A
HIV Shlay 199858 Greater pain relief at 14 weeks ( p = 0.03)
Karnofsky Score HIV Schiflett 2011 No difference at 6 or 14 weeks
Shlay 1998 No difference at 6 or 14 weeks
Subjective Peripheral Neuropathy HIV Anastasi 2013 Average score improved at 2 weeks
( p < 0.05)
Screen (SPNS)
Visual Analog Scale (VAS) HIV, DM, Kumnerddee 2010 Improved at 5 weeks ( p < 0.05)
IPN Penza 201166
Beck Depression Inventory DM/IPN Penza 2011 No difference after six treatments
SF-36 QOL DM/IPN Penza 2011 No difference after six treatments

Table 4. Nerve Conduction Study Changes in Acupuncture and Control Groups

Change from baseline Change from baseline Change in acupuncture
Nerve conduction study parameter in acupuncture group in control group vs. control
Motor NCS
Compound muscle action potential (CMAP) amplitude
Median nerve 5 Yang 200956 [ Yang 2009 5 Yang 2009
\\ Yang 201155 5 Yang 2011 5 Yang 2011
Distal motor latency (DML)
Median nerve 5 Khosrawi 201260 5 Khosrawi 2012 5 Khosrawi 2012
Y Yang 2009 Y Yang 2009 Y Yang 2009
ZZ Yang 2011 5 Yang 2011 ZZ Yang 2011
Motor nerve conduction velocity (MNCV)
Ulnar, tibial nerve \\ Yu 200119 5 Yu 2001 \\ Yu 2001
Median nerve 5 Yang 2009 5 Yang 2009 5 Yang 2009
5 Yang 2011 \\ Yang 2011 \\ Yang 2011
\\ Yu 2001 \\ Yu 2001 \\ Yu 2001
\\ Zuo 201020 5 Zuo 2010 [ Zuo 2010
Peroneal nerve \\ Zuo 2010 5 Zuo 2010 [ Zuo 2010
Sensory NCS
Sensory nerve action potential (SNAP) amplitude
Median nerve [ Yang 2009 [ Yang 2009 5 Yang 2009
\\ Yang 2011 \\ Yang 2011 5 Yang 2011
Distal sensory latency (DSL)
Median nerve 5 Khosrawi 2012 5 Khosrawi 2012 5 (p = 0.07) Khosrawi 2012
Y Yang 2009 Y Yang 2009 5 Yang 2009
ZZ Yang 2011 Y Yang 2011 5 Yang 2011
Sensory nerve conduction velocity (SNCV)
Median nerve [ Khosrawi 2012 5 Khosrawi 2012 [ Khosrawi 2012
[ Yang 2009 [ Yang 2009 5 Yang 2009
\\ Yang 2011 5 Yang 2011 5 Yang 2011
\\ Zuo 2010 \\ Zuo 2010 5 Zuo 2010
Peroneal nerve \\ Zuo 2010 \\ Zuo 2010 5 Zuo 2010
\\, increased ( p < 0.01); [,increased ( p < 0.05); 5, no significant change ( p > 0.05); Y, decreased ( p < 0.05); ZZ, decreased ( p < 0.01).

171
172 DIMITROVA ET AL.

electroacupuncture (eight subjects per group). There was no the acupuncture plus medical therapy ‘‘Control 1’’ over the
difference between acupuncture and control in any of the medical therapy ‘‘Control 2.’’ In contrast, Tong et al.59
outcome measures used: Visual Analogue Scale, Patient’s found no significant difference on the House–Brackmann
Global Impression of Change, Beck Depression Inventory, Scale, as all three groups had high improvement rates.
and the SF-36 Quality of Life Instrument (Table 3). Both
groups showed no difference from baseline. The authors did Acupuncture-induced changes in nerve
not analyze data from the diabetic and idiopathic neuropathy conduction parameters
subjects separately (Table 2).
NCS were performed in included trials on DPN and CTS
(Table 4). With respect to motor nerve function, acupuncture
CTS
produced significant effects on median nerve compound muscle
All four RCTs56,60,61,62 used validated CTS-specific scales action potential (CMAP) amplitude, median nerve distal motor
(Table 2) and reported significant improvement with manual latency (DML), and motor nerve conduction velocity (NCV) of
acupuncture56,60,62 and electroacupuncture61 compared with the median, ulnar, and peroneal nerves. Compared with medi-
baseline in the Global Symptom Score,56,60 Boston Carpal cation control, acupuncture caused significantly greater im-
Tunnel Score,61 Carpal Tunnel Self-assessment Ques- provement in DML and in median, ulnar, and motor NCV.
tionnaire,62 and Visual Analogue Scale61 (Table 3). Three of Sensory NCS revealed that acupuncture caused an increase in
the four RCTs included showed significantly greater effec- sensory nerve action potential (SNAP) amplitude in the median
tiveness of acupuncture compared with night splinting,61 nerve, lowered median nerve distal sensory latency (DSL), and
sham acupuncture,60 oral vitamin B1 and B6,60 and oral increased median and peroneal nerve NCV. However, compared
prednisolone.56 In contrast, Yao et al.62 revealed no differ- to medication control, acupuncture showed significant im-
ence in CTS symptoms between acupuncture plus night provement only in median nerve sensory NCV60 (Table 4).
splinting compared to sham acupuncture plus night splinting, With respect to longer-term NCS changes, acupuncture-
with both groups improved from baseline. Most long-term treated groups had faster sensory NCV, a trend ( p = 0.07)
follow-ups favored acupuncture at 4 weeks,56,60 5 weeks,61 7 toward shorter DSL60 at 4 weeks, and shorter DML and
months, and 13 months,55 with only a single study showing motor NCV at 13 months.55
no difference between acupuncture and control at 3 months.62
Meta-analysis on acupuncture for Bell’s palsy
Bell’s palsy and diabetic neuropathy
Two RCTs with 48045 and 11959 subjects met the inclu- Outcomes from all included trials on DPN and Bell’s
sion criteria (Table 2). Li et al.45 included an intermediate palsy were reported on the individual subject level and in-
treatment group, ‘‘Control 1,’’ which received both acu- cluded in this meta-analysis. All trials used medication
puncture and medical therapy, and a medical therapy control. Data from Li et al.’s45 ‘‘Control 1’’ were excluded
‘‘Control 2’’ group. Tong et al.59 included an oral steroid because acupuncture was administered, and from Tong
group ‘‘Control 1’’ and a home exercises ‘‘Control 2’’ group et al.’s59 ‘‘Control 2’’ because it was not an active control.
(Table 2). Li et al.45 found significant improvement in the The third arms of both trials were included in the data syn-
House–Brackmann Scale and Facial Disability Index in the thesis. A forest plot for acupuncture against control treatment
acupuncture group compared with medical therapy and of was constructed using contingency tables of acupuncture

FIG. 2. Meta-analysis results for diabetes mellitus and Bell’s palsy.

ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 173

against the control condition (Fig. 2). Meta-analysis on the fied multiple methodological problems and suggested that
DPN subjects showed an overall increased effect of acu- their findings were inconclusive. Neither study included the
puncture over control, which was still present ( p < 0.001) four English language trials on DPN identified in the present
when correcting for study heterogeneity (mean odds ratio study.19,20,63,64
[OR] = 4.85; 95% confidence interval [CI] 2.2–11). Meta- Similar conclusions were reached in this study to those of
analysis on the Bell’s palsy subjects also showed an in- Chen et al.70 who found acupuncture to be superior to oral
creased effect of acupuncture over control ( p = 0.042) when thiamine plus i.m. vitamin B12,19 i.v. vitamin B12,20 oral
correcting for heterogeneity (mean OR = 2.78; 95% CI 1– mecobalamin,64 and oral inositol.63 Similar to the above re-
7.4). Similarly, the pooled meta-analysis of all trials studying views, multiple methodological problems were encountered.
diabetic neuropathy and Bell’s palsy, when adjusting for
study heterogeneity, found a mean OR of 4.23 in favor of
Acupuncture for Bell’s palsy
acupuncture over control (95% CI 2.3–7.8, p < 0.001).
The same six Chinese RCTs71–76 were included in two
Cochrane reviews on Bell’s palsy77,78 and in a systematic
Acupuncture safety, adverse events
review79 published by the same group. Collectively, these
The included studies varied greatly in how adverse events six RCTs suggested that acupuncture leads to improved
were defined and monitored. Overall acupuncture treatment facial nerve recovery. However, the authors warned that the
appeared safe, without serious adverse events, except in the conclusions cannot be trusted due to multiple methodolog-
early study on HIV-related neuropathy58 where both groups ical flaws. Tong et al.59 was not referenced. Li et al.45 was
experienced significant morbidity. There was a trend toward excluded from the Cochrane reviews due to use of mox-
lower rates of life-threatening events in the acupuncture ibustion. Moxibustion is the stimulation caused by burning
group (3/58 subjects) compared with the control group (10/ the herb Artemisia vulgaris over an acupuncture point and is
56; p = 0.06), with a high combined 2-year mortality rate widely regarded as an acupuncture modality. Trials using
(28.8%). The remaining studies reported no acupuncture- moxibustion were included in the present review, as long as
related serious adverse events. Minor adverse events in- acupuncture was also used as active intervention.
cluded minor bruising with acupuncture,56,59,65 which in one In 2012, Kim et al.80 reviewed eight RCTs, among them
study was as common as 20% of subjects61; local discomfort Tong et al.,59 two studies included in the Cochrane re-
with needle insertion; pain following session; and uncom- views,72,74 and additional five Chinese language RCTs,81–85
fortable paresthesias during acupuncture in 5% of subjects.56 all published prior to 2010, which were not included in the
Cochrane 2010 review for unclear reasons. The authors
excluded moxibustion, and this may be why Li et al.45 was
Discussion
not selected. The authors found that both acupuncture plus
To date, no critical appraisal of acupuncture’s effect on medication and acupuncture alone were more effective than
PN of various etiologies has been conducted. Prior sys- medication alone. The authors concluded that acupuncture
tematic reviews were published in the areas of acupuncture may be a useful adjunct or an alternative to drug therapy.
for DPN, CTS, and Bell’s palsy. The evidence is summa- However, they identified multiple methodological flaws.
rized below, along with the authors’ perspective. The latest review on the subject included 14 RCTs86—all
six RCTs from the Cochrane reviews, all RCTs reviewed by
Kim et al., plus two additional Chinese language RCTs.87,88
Acupuncture for DPN
The authors did not critically assess RCT quality except for
The included RCTs involved manual acupuncture20,63 or rating them ‘‘high,’’ ‘‘low,’’ and ‘‘unclear.’’ Meta-analysis
electroacupuncture,19 which were superior to control con- pooled data from 1541 individual subjects, using ‘‘effective
ditions in DPN, as was the combination of both therapies.64 rate’’ as the outcome. Acupuncture was found to be superior
Further studies are needed to compare acupuncture to to control interventions (risk ratio [RR] = 1.14; 95% CI
electroacupuncture in DPN. 1.04–1.25). The authors felt that their results were incon-
By far the largest number of RCTs identified in prelimi- clusive due to various methodological flaws.
nary searches dealt with acupuncture for DPN. However, As the present search was limited to the English language,
only four RCTs met the inclusion criteria due to poor overall only two trials on manual acupuncture for Bell’s palsy were
quality of the reviewed studies. In 2012, Bo et al.67 at- included, with 599 subjects combined, which had different
tempted a systematic review and identified 75 RCTs (73 in results. Li et al.45 found that acupuncture was more effec-
Chinese) on acupuncture for DPN. They concluded that the tive than a combination of acupuncture and medical ther-
majority of those were of low-moderate methodological apy or medical therapy alone. Conversely, Tong et al.59
quality, while none of the trials met all of the CONSORT68 found no difference between acupuncture and the two
and STRICTA69 criteria. A systematic review was thus not control groups—oral steroids and home facial exercises—
conducted. as all three groups had high rates of improvement from
In 2013, Chen et al.70 reviewed 25 Chinese RCTs baseline. Studying episodic, self-limiting conditions such
studying manual acupuncture for DPN. None of the included as Bell’s palsy presents a challenge because the remission
RCTs met all the CONSORT and STRICTA criteria, and rate is 85% within the first 3 weeks and close to 100%
there were no clear inclusion criteria. Meta-analysis on 23/ within 6 months.89
25 RCTs suggested that acupuncture is more effective on a Much has been written about the bias of acupuncture
‘‘global symptom improvement scale’’ than mecobalamin, research coming from China. As early as 1998, Vickers
vitamin B1 and B12, and no treatment. The authors identi- et al.90 cautioned about publishing bias and the lack of
174 DIMITROVA ET AL.

negative trials. While this study is almost 20 years old, there acupuncture to be more effective than control in five RCTs.
are no negative RCTs in any of the selected reviews, which The authors cautioned that the selected RCTs had multiple
raises questions about continued publication bias. methodological flaws and were underpowered.

Acupuncture for CTS Acupuncture point selection

56,60
The present review found that manual acupuncture In addition to varied acupuncture treatment courses, the
and electroacupuncture61 were more effective than control included RCTs used a wide variety of acupuncture points
in three of the four included RCTs on CTS.56,60,61 Both (Table 5). Point selection rationale was not justified in any
acupuncture and sham-acupuncture groups improved sig- RCTs, except for one.63
nificantly in the fourth included RCT.62 A possible expla- There was some consistency in point selection in the
nation may be that CTS has a relatively high remission rate, treatment of CTS and Bell’s palsy. All included CTS trials
ranging from 33% at 6 months91 to >50% at 1 year.92 involved points on the ipsilateral pericardium meridian:
All four included studies on CTS involved needle PC6,56,60,62 PC7,56,60–62 and PC8.61 Pericardium meridian
placement near the median nerve. Yang et al. found this points are commonly used in CTS treatment due to their
effect to be sustained at 13 months.55 Further studies are close association with the median nerve, which is affected
needed on acupuncture’s long-term effects. Similar to the in CTS. Similarly, both of the included Bell’s palsy trials
present findings, Sim et al.93 reviewed six RCTs and found used ipsilateral points on the stomach meridian—ST2,59

Table 5. Acupuncture Point Selection

Acupuncture Moxibustion point
Study Condition modality Acupuncture point selection selection (if any)
Li 200445 Bell’s palsy Acupuncture, Ipsilateral: ST4, ST6, ST7, LI4, Hanging
moxibustion GB14, SI17 moxibustion –
Bilateral: LI4 5 min at each point
Tong 200959 Bell’s palsy Acupuncture Ipsilateral: ST2, ST4, ST6, EX-HN7, N/A
GB14, TW17
Bilateral: LI4
Khosrawi 201260 CTS Acupuncture Ipsilateral: PC6, PC7 N/A
Kumnerddee 201061 CTS Electroacupuncture Ipsilateral: LI4, LI11, PC7, PC8, N/A
2 Ba-Xie points digits 2/3 and 3/4
Yao 201262 CTS Acupuncture Ipsilateral: PC6, PC7, SP6 Contralateral: N/A
TW5, LI4, LI11, GB34
Yang 200956 CTS Acupuncture Ipsilateral: PC6, PC7 N/A
Yu 200119 DM Electroacupuncture Bilateral: LI4, LI11, LI15, TW5, GB30, N/A
GB34, ST36, ST41, ST44
Zhang 201063 DM Acupuncture Bilateral: BL18, BL20, BL23, BL58, N/A
ST36, SP6, SP3, Yishu, ST40, GB 34;
Also CV6, CV4
Custom points:
Bilateral LI4, LI10, LI11, LI15, ST31,
ST32, ST34, ST43, ST44, SP10,
BL17—for blood stasis
Bilateral SP8, SP9—for phlegm
Bilateral Bafeng and Baxie—for severe
numbness of the hands and feet
Zheng 200464 DM Acupuncture, Manual acupuncture: bilateral T7Jiaji, N/A
electroacupuncture, BL23, GB30; Also CV4, CV6
snow lotus i.m. Electroacupuncture: bilateral SP6
injection Injection of snow lotus herb bilaterally in
T7 Jiaji and SP6
Zuo 201020 DM Acupuncture Bilateral ST36, SP6, KI3, LI4, LI11, N/A
TW5, CV4, and CV6
Anastazi 201365 HIV Acupuncture, Not provided Not provided
Moxibustion
Shlay 199858 HIV Acupuncture Bilateral SP6, SP7, SP9, Ba Feng, N/A
KI2, KI3
Penza 201166 DM/IPN Electroacupuncture Bilateral ST36, SP6, LR3, BL60 N/A
ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 175

ST4,45,59 ST6,45,59 ST7,45 and GB1445,59—which are in close tervention—moxibustion, medication or herbs, or splinting.
proximity to branches of the facial nerve. Both trials also This is further complicated by the use of various acupuncture
included needling of bilateral LI4, which is one of the most modalities—manual acupuncture,20,45,56,58–60,62,63,65 electro-
commonly used points for facial pain. acupuncture,19,61,66 or a combination of both.64 Manual and
The RCTs focused on acupuncture for diabetic, HIV- electroacupuncture may have different mechanisms of action
related, and idiopathic neuropathy used a greater variety of and different effects on neuropathy and neuropathic pain.
acupuncture points with little overlap. This may be because Unfortunately, not enough electroacupuncture trials were
these conditions cause polyneuropathy with more complex available to draw meaningful comparisons with manual acu-
symptoms compared with mononeuropathies. puncture for each of the reviewed neuropathic conditions.

Overview of statistical bias Methodological problems with sample size calculations

Statistical bias within the larger meta-analysis was eval- Only two of the included RCTs58,65 contained a sample
uated using assessments of study heterogeneity (I2) both size calculation. Therefore, it is possible that the majority of
within each of the disease subsets and in the pooled cohort. the included trials are underpowered (Table 6). In particular
This combined approach was utilized to evaluate any dis- Penza et al.66 only enrolled eight subjects per group, and
crepancy among the studies both within and across disease only 12 subjects had idiopathic neuropathy. There was no
states. In all cases, the heterogeneity is not significant statistical rationalization for this sample size and this uneven
(Bell’s: I2 = 6%, p = 0.55; diabetes mellitus [DM]: I2 = 47%, split of patients with idiopathic and diabetic neuropathy.
p = 0.12; combined: I2 = 28%, p = 0.22), indicating the effect
of acupuncture intervention is appropriately similar among Methodological problems with improper control
the studies. Even the comparatively high I2 value for DM is and blinding
an indication of ‘‘moderate’’ heterogeneity according to the In Traditional Chinese Medicine, there is a belief that a
Cochrane criteria.94 The analysis of heterogeneity suggests sensation called de qi is crucial to acupuncture’s therapeutic
that even in the absence of negative results, the effect of benefit. De qi has been described as aching, soreness,
acupuncture treatment is moderately consistent within and heaviness, warmth, coolness, tingling, numbness at the site
across neuropathic conditions. of the needle,96 or as radiating paresthesias.97 Because of
this belief, many RCTs from China are designed to compare
Suggested mechanism of action de qi elicited by needle manipulation to manipulation-free
While the selected trials employed varied acupuncture needling of an acupuncture point.39,40 Any studies that in-
regimens, they all involved acupuncture points located near volved passive needle insertion in acupuncture point as
peripheral nerves, such as points close to the median nerve control were excluded.
for CTS and close to the facial nerve for Bell’s palsy Lack of blinding of both subjects and investigators was a
(Table 5). It is possible that acupuncture needles exert direct common methodological flaw of the selected studies. None
effect on an underlying nerve and peri-neural tissues, either of the studies originating in China19,20,45,56,59,63,64 used
through manual manipulation or electric current. This could sham acupuncture, and the subjects were not blinded
explain why acupuncture appears to have an effect not only (Table 6). In 7 of the 13 included studies,19,20,45,56,59,63,64 the
on neuropathic symptoms, but also on NCS parameters outcome assessors were not blinded to treatment assignment,
(Table 4). These effects were sustained up to 13 months55 in which is source of investigator bias.
CTS and were significantly different from the control con-
dition. Further studies are needed before any definitive Methodological problems with placebo and expectancy
conclusions can be drawn about acupuncture’s effect on Most of the included RCTs carried out a greater number
NCS parameters in neuropathic conditions. of visits or interventions in the acupuncture groups
(Table 2), which may have affected subject expectancy. In
Methodological problems with standardization many cases, the authors alluded to positive subject expec-
tations from acupuncture, including higher dropout rates in
One common problem in acupuncture research is the lack
the control condition and subjects seeking acupuncture in-
of standardization of point selection, number of needles
dependently.56,59 Subject expectations and the role of pla-
used, needle retention time, needling depth, needle manip-
cebo in general were not addressed. It is well known that
ulation, use of moxibustion, and electroacupuncture. The 13
positive expectation amplifies acupuncture-induced analge-
RCTs reviewed varied greatly in acupuncture point selection
sia in both subjective pain ratings and objective fMRI signal
and in number of sessions: from six weekly sessions for
changes.98
DPN62 to 36 sessions in Bell’s palsy.59 Consideration was
given to calculating the amount of acupuncture delivered (in
Methodological problems with outcome measures
minutes or hours). However, there is no clear concept of an
appropriate acupuncture dose and how much treatment is The 13 included RCTs contained a wide array of outcome
needed for a given condition.95 The relationship between measures, as should be expected with trials on various
number of needles and acupuncture effect is also unclear neuropathic conditions, involving both poly- and mono-
and probably not linear. neuropathy. Overall, most trials assessed subject-reported
Lastly, isolating acupuncture’s effect would be simpler in measures of improvement, which are intrinsically subjec-
RCTs that study acupuncture alone as the intervention. Half of tive. Only four of the 13 RCTs used NCS for baseline and
the included RCTs combined acupuncture with another in- outcome measures (Table 4). More studies of neuropathy
 Table 6. Summary of Methodological Quality
Sham Validated Non-validated Outcome Statistical plan
Additional active acupuncture scales outcome Subjects assessors Sample size included
Reference interventions? as control? used? measures used? NCS used? blinded? blinded? calculation? in methods?
Li 200445 Moxibustion No Yes Improvement Scale,e — — — — —
symptomatic
improvement
Tong 200959 Eye care, education No Yes — — — — — Yes
Khosrawi 201260 Nightly wrist splinting YES plus P.O. meds Yes — Yes Yes Yes — Yes
Kumnerddee 201061 — NO Yes — — — — — Yes
Yao 201262 Nightly wrist splinting YES plus nightly Yes — — Yes Yes — Yes
wrist splinting
Yang 200956 — No Yes — Yes — Yes — Yes
Yu 200119 — No Yes Improvement Scale,a Yes — — — —
symptomatic
improvement
Zhang 201063 — No No Improvement Scale,b — — — — —
symptomatic
improvement,
neurologic exam

176
Zheng 200464 Snow lotus injections No No Symptom score, — — — — —
clinical sign score,
Improvement
Scalec
Zuo 201020 Methylcobalamin i.v. No Yes Neurologic Exam Yes — — — Yes
Score,
Improvement
Scaled
Anastasi 201365 Moxibustion YES plus placebo Yes — — Yes Yes Yes Yes
moxibustion
Shlay 199858 — YES plus P.O. meds, Yes Global Pain Relief — Yes Yes Yes Yes
P.O. placebo Score,f Neurologic
Summary Score
Penza 201166 — YES Yes — — Yes Yes — Yes
a
Markedly relieved/improved/failed based on exam and symptoms.
b
Remarkable effectiveness/effectiveness/ineffectiveness based on symptom and clinical sign score.
c
Marked/some/none based on symptoms, exam, labs, NCVs.
d
Marked effectiveness/effectiveness/failure based on symptoms.
e
Cured/obviously improved/improved/no improvement based on House–Brackmann Scale and Facial Disability Index.
f
Global Pain Relief Score—complete/a lot/moderate/slight/none/worse.
—, no/none.
ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 177

should use NCS or other objective measures of improve- 8. Ezzo JM, Richardson MA, Vickers A, et al. Acupuncture-
ment such as somatosensory evoked potentials. point stimulation for chemotherapy-induced nausea or vo-
Several included studies used arbitrary, investigator- miting. Cochrane Database Syst Rev 2006;CD002285.
created scales of improvement (Table 6), such as ‘‘complete/ 9. Vickers AJ, Cronin AM, Maschino AC, et al. Acupuncture
a lot/moderate/slight/none/worse,’’58 based on a combination for chronic pain: Individual patient data meta-analysis. Arch
of symptoms, neurologic exam, and improvement on disease- Intern Med 2012;172:1444–1453.
specific, validated scales. These could be a significant source 10. Amato AA, Russell JA. Neuromuscular Disorders. New
of bias, particularly if the outcome assessors are not blinded York: McGraw-Hill Medical, 2008.
to treatment assignment. With the exception of Zhang et al.,63 11. Herskovitz S, Scelsa SN, Schaumburg HH. Peripheral
all RCTs using investigator-created scales of improvement Neuropathies in Clinical Practice. New York: Oxford Uni-
versity Press, 2010.
also used validated outcome measures.
12. Gordon Smith A, Robinson Singleton J. Idiopathic neu-
ropathy, prediabetes and the metabolic syndrome. J Neurol
Conclusions Sci 2006;242:9–14.
This systematic review suggests that acupuncture is ef- 13. Martyn CN, Hughes RA. Epidemiology of peripheral neu-
ropathy. J Neurol Neurosurg Psychiatry 1997;62:310–318.
fective in diabetic neuropathy, Bell’s palsy, and CTS, un-
14. Chong MS, Bajwa ZH. Diagnosis and treatment of neuro-
der the conditions of the included studies. Acupuncture
pathic pain. J Pain Symptom Manage 2003;25:S4–S11.
may be effective in HIV-related neuropathy. However, 15. Vincent A, Kruk KM, Cha SS, et al. Utilisation of acu-
further studies are needed to explore its potential thera- puncture at an academic medical centre. Acupunct Med
peutic role in this population. There is a need for future 2010;28:189–190.
studies that address point selection, acupuncture modality, 16. Barnes PM, Bloom B, Nahin RL; National Center for
and frequency. Further studies should also aim to minimize Health Statistics (U.S.). Complementary and Alternative
methodological flaws by incorporating proper subject and Medicine Use Among Adults and Children: United States,
investigator blinding, standardization of acupuncture in- 2007. Hyattsville, MD: Department of Health and Human
tervention, use of validated, objective outcome assess- Services, Centers for Disease Control and Prevention, Na-
ments, a priori hypotheses with proper sample and effect tional Center for Health Statistics, 2008.
size calculations, and appropriate statistical analyses. 17. Jeste DV, Dunn LB, Folsom DP, et al. Multimedia edu-
cational aids for improving consumer knowledge about
Acknowledgments illness management and treatment decisions: A review of
randomized controlled trials. J Psychiatr Res 2008;42:1–21.
This study is funded by the following grants: NIH K23 18. D’Silva S, Poscablo C, Habousha R, et al. Mind–body
AT008405 (A.D.), NIH 5T32 AT002688 (B.O.). The au- medicine therapies for a range of depression severity: A
thors wish to thank Prof. Marian McDonagh, PharmD, from systematic review. Psychosomatics 2012;53:407–423.
OHSU’s department of Medical Informatics and Clinical 19. Yu J, Cui Z. Clinical study of diabetic peripheral neurop-
Epidemiology for her contribution to this manuscript. athy treated by acupuncture. Int J Clin Acupunct 2001;12:
315–318.
20. Zuo L, Zhang L. Study on the effect of acupuncture plus
Author Disclosure Statement
methylcobalamin in treating diabetic peripheral neuropa-
No competing financial interests exist. thy. J Acu Tuina Sci 2010;8:249–252.
21. R-Core Team. R: A language and environment for statis-
tical computing. Vienna, Austria: R Foundation for Statis-
References
tical Computing,; 2013. Online document at: www
1. Diehl DL, Kaplan G, Coulter I, et al. Use of acupuncture by .R-project.org/, accessed November 3, 2016.
American physicians. J Altern Complement Med 1997;3:119– 22. Lumley T. rmeta: Meta-analysis. R package version 2.16.
126. Online document at: http://cran.r-project.org/web/packages/
2. Eisenberg DM, Kessler RC, Foster C, et al. Unconventional rmeta/rmeta.pdf, accessed November 3, 2016.
medicine in the United States. Prevalence, costs, and pat- 23. Li Z. Mingmu dihuang tang combined with electro-
terns of use. N Engl J Med 1993;328:246–252. acupuncture for treatment of diabetic oculomotor paralysis
3. NIH Consensus Conference. Acupuncture. JAMA 1998; in 52 cases. J Tradit Chin Med 2007;27:37–38.
280:1518–1524. 24. Man PL. Acupuncture analgesia for the treatment of tri-
4. Manheimer E, White A, Berman B, et al. Meta-analysis: geminal neuralgias: A series of forty-one cases. J Natl Med
Acupuncture for low back pain. Ann Intern Med 2005;142: Assoc 1975;67:115–117.
651–663. 25. Hu J. Acupuncture treatment of herpes zoster. J Tradit Chin
5. Furlan AD, van Tulder M, Cherkin D, et al. Acupuncture Med 2001;21:78–80.
and dry-needling for low back pain: An updated systematic 26. Abuaisha BB, Costanzi JB, Boulton AJ. Acupuncture for the
review within the framework of the cochrane collaboration. treatment of chronic painful peripheral diabetic neuropathy: A
Spine (Phila Pa 1976) 2005;30:944–963. long-term study. Diabetes Res Clin Pract 1998;39:115–121.
6. Linde K, Allais G, Brinkhaus B, et al. Acupuncture for 27. Green J, McClennon J. Acupuncture: An effective treat-
migraine prophylaxis. Cochrane Database Syst Rev 2009; ment for painful diabetic neuropathy. Diabet Foot J 2006;
CD001218. 9(4):182–191.
7. Linde K, Allais G, Brinkhaus B, et al. Acupuncture for 28. Donald GK, Tobin I, Stringer J. Evaluation of acupuncture
tension-type headache. Cochrane Database Syst Rev 2009; in the management of chemotherapy-induced peripheral
CD007587. neuropathy. Acupunct Med 2011;29:230–233.
178 DIMITROVA ET AL.

29. Lee S, Kim JH, Shin KM, et al. Electroacupuncture to treat 48. Lewith GT, Field J, Machin D. Acupuncture compared with
painful diabetic neuropathy: Study protocol for a three- placebo in post-herpetic pain. Pain 1983;17:361–368.
armed, randomized, controlled pilot trial. Trials 2013;14:225. 49. Cai DF. Warm-needling plus Tuina relaxing for the treat-
30. Galantino ML, Eke-Okoro ST, Findley TW, et al. Use of ment of carpal tunnel syndrome. J Tradit Chin Med 2010;
noninvasive electroacupuncture for the treatment of HIV- 30:23–24.
related peripheral neuropathy: A pilot study. J Altern Com- 50. Danciu A, Danciu E. The preference of acupuncture treat-
plement Med 1999;5:135–142. ment in autonomic diabetic neuropathy. Am J Acupunct
31. Yongping J, Stefanovic J. The acupuncture treatment of 1985;13:247–252.
peripheral neuropathy in HIV/AIDS. J Chin Med 2002; 51. Jiang H, Shi K, Li X, et al. Clinical study on the wrist-ankle
27–29. acupuncture treatment for 30 cases of diabetic peripheral
32. Lu Z, Chen Z. Electroacupuncture for treatment of 12 cases neuritis. J Tradit Chin Med 2006;26:8–12.
of infantile peroneal nerve injury. J Tradit Chin Med 2000; 52. Naeser MA, Hahn KA, Lieberman BE, et al. Carpal tunnel
20:130–131. syndrome pain treated with low-level laser and microam-
33. Branco K, Naeser MA. Carpal tunnel syndrome: Clinical peres transcutaneous electric nerve stimulation: A con-
outcome after low-level laser acupuncture, microamps trans- trolled study. Arch Phys Med Rehabil 2002;83:978–988.
cutaneous electrical nerve stimulation, and other alternative 53. Qian W, Qian H, Wu T, et al. Clinical research on acu-
therapies—an open protocol study. J Altern Complement Med puncture treatment of diabetic peripheral neuropathy. J Acu
1999;5:5–26. Tuina Sci 2004;2:12–14.
34. Zhou Y, Garcia MK, Chang DZ, et al. Multiple myeloma, 54. Shao L. Observations on the curative effect of acupuncture
painful neuropathy, acupuncture? Am J Clin Oncol 2009; on cardiac vegetative neuropathy of diabetes. J Acu Tuina
32:319–325. Sci 2004;2:18–20.
35. Phillips KD, Skelton WD, Hand GA. Effect of acupuncture 55. Yang CP, Wang NH, Li TC, et al. A randomized clinical
administered in a group setting on pain and subjective pe- trial of acupuncture versus oral steroids for carpal tunnel
ripheral neuropathy in persons with human immunodeficiency syndrome: A long-term follow-up. J Pain 2011;12:272–
virus disease. J Altern Complement Med 2004;10:449–455. 279.
36. Schroder S, Liepert J, Remppis A, et al. Acupuncture 56. Yang CP, Hsieh CL, Wang NH, et al. Acupuncture in pa-
treatment improves nerve conduction in peripheral neu- tients with carpal tunnel syndrome: A randomized con-
ropathy. Eur J Neurol 2007;14:276–281. trolled trial. Clin J Pain 2009;25:327–333.
37. Schroeder S, Meyer-Hamme G, Epplee S. Acupuncture for 57. Shiflett SC, Schwartz GE. Effects of acupuncture in re-
chemotherapy-induced peripheral neuropathy (CIPN): A ducing attrition and mortality in HIV-infected men with
pilot study using neurography. Acupunct Med 2012;30:4–7. peripheral neuropathy. Explore (NY) 2011;7:148–154.
38. Xie Z. 51 cases of occipital neuralgia treated with acu- 58. Shlay JC, Chaloner K, Max MB, et al. Acupuncture and
puncture. J Tradit Chin Med 1992;12:180–181. amitriptyline for pain due to HIV-related peripheral neu-
39. Tong Y, Guo H, Han B. Fifteen-day acupuncture treatment ropathy: A randomized controlled trial. Terry Beirn Com-
relieves diabetic peripheral neuropathy. J Acupunct Mer- munity Programs for Clinical Research on AIDS. JAMA
idian Stud 2010;3:95–103. 1998;280:1590–1595.
40. Xu SB, Huang B, Zhang CY, et al. Effectiveness of 59. Tong FM, Chow SK, Chan PY, et al. A prospective ran-
strengthened stimulation during acupuncture for the treat- domised controlled study on efficacies of acupuncture and
ment of Bell palsy: A randomized controlled trial. CMAJ steroid in treatment of idiopathic peripheral facial paralysis.
2013;185:473–479. Acupunct Med 2009;27:169–173.
41. Napadow V, Liu J, Li M, et al. Somatosensory cortical 60. Khosrawi S, Moghtaderi A, Haghighat S. Acupuncture in
plasticity in carpal tunnel syndrome treated by acupuncture. treatment of carpal tunnel syndrome: A randomized con-
Hum Brain Mapp 2007;28:159–171. trolled trial study. J Res Med Sci 2012;17:1–7.
42. Lun X. Treatment of diabetic peripheral neuropathy using 61. Kumnerddee W, Kaewtong A. Efficacy of acupuncture
Chinese herbs and acupuncture. Int J Clin Acupunct 1995; versus night splinting for carpal tunnel syndrome: A ran-
6:271–273. domized clinical trial. J Med Assoc Thai 2010;93:1463–
43. Hui F, Boyle E, Vayda E, et al. A randomized controlled 1469.
trial of a multifaceted integrated complementary-alternative 62. Yao E, Gerritz PK, Henricson E, et al. Randomized con-
therapy for chronic herpes zoster-related pain. Altern Med trolled trial comparing acupuncture with placebo acu-
Rev 2012;17:57–68. puncture for the treatment of carpal tunnel syndrome. PM R
44. Liang F, Li Y, Yu S, et al. A multicentral randomized 2012;4:367–373.
control study on clinical acupuncture treatment of Bell’s 63. Zhang C, Ma YX, Yan Y. Clinical effects of acupuncture
palsy. J Tradit Chin Med 2006;26:307. for diabetic peripheral neuropathy. J Tradit Chin Med 2010;
45. Li Y, Liang FR, Yu SG, et al. Efficacy of acupuncture and 30:13–14.
moxibustion in treating Bell’s palsy: A multicenter ran- 64. Zheng HT, Li YF, Yuan SX. Observations on 52 patients
domized controlled trial in China. Chin Med J (Engl) 2004; with diabetic peripheral neuropathy treated by needling
117:1502–1506. combined with drug. J Acu Tuina Sci 2004;2:24–26.
46. Ahn AC, Bennani T, Freeman R, et al. Two styles of acu- 65. Anastasi JK, Capili B, McMahon DJ, et al. Acu/Moxa for
puncture for treating painful diabetic neuropathy—a pilot distal sensory peripheral neuropathy in HIV: A randomized
randomised control trial. Acupunct Med 2007;25:11–17. control pilot study. J Assoc Nurses AIDS Care 2013;24:
47. Ursini T, Tontodonati M, Manzoli L, et al. Acupuncture for 268–275.
the treatment of severe acute pain in herpes zoster: Results 66. Penza P, Bricchi M, Scola A, et al. Electroacupuncture is
of a nested, open-label, randomized trial in the VZV Pain not effective in chronic painful neuropathies. Pain Med
Study. BMC Complement Altern Med 2011;11:46. 2011;12:1819–1823.
ACUPUNCTURE FOR PERIPHERAL NEUROPATHY REVIEW 179

67. Bo C, Xue Z, Yi G, et al. Assessing the quality of reports 85. Zhu L. Observations on the efficacy of combined acu-
about randomized controlled trials of acupuncture treat- puncture and medication for treating the acute stage of
ment on diabetic peripheral neuropathy. PLoS One 2012; peripheral facial paralysis. Shanghai J Acupunct Moxibust
7:e38461. 2006;25:17–18.
68. Schulz KF, Altman DG, Moher D. CONSORT 2010 86. Li P, Qiu T, Qin C. Efficacy of acupuncture for Bell’s
statement: Updated guidelines for reporting parallel group palsy: A systematic review and meta-analysis of random-
randomised trials. BMJ 2010;340:c332. ized controlled trials. PLoS One 2015;10:e0121880.
69. MacPherson H, Altman D, Hammerschlag R, et al. Revised 87. Zhao Y, He L, Zhang Q. Effectiveness of three different
STandards for Reporting Interventions in Clinical Trials treatments for peripheral facial paralysis. Chin J Clin Rehab
of Acupuncture (STRICTA): Extending the CONSORT 2005;29 41–43.
statement. PLoS Med 2010;7:e1000261. 88. Zhu H, Jiang J, Feng L, et al. Intractable facial paralysis
70. Chen W, Yang GY, Liu B, et al. Manual acupuncture for treated with stellate ganglion block plus electric acupunc-
treatment of diabetic peripheral neuropathy: A systematic ture. Chin J Pain Med 2004;5:263.
review of randomized controlled trials. PLOS ONE 2013; 89. Peitersen E. The natural history of Bell’s palsy. Am J Otol
8:e73764. 1982;4:107–111.
71. Li J. Comparison the efficacy between acupuncture and 90. Vickers A, Goyal N, Harland R, et al. Do certain countries
manipulation for Bell’s palsy. Chin Clin Med Res 2005;11: produce only positive results? A systematic review of
1715–1716. controlled trials. Controlled Clin Trials 1998;19:159–166.
72. Liu M. Comparison of acupuncture and drug treatment 91. Futami T, Kobayashi A, Ukita T, et al. Carpal tunnel
for 130 patients with facial palsy. J Clin Acupunct 1996; syndrome; its natural history. Hand Surg 1997;2:129–130.
12:56. 92. Silverstein BA, Fan ZJ, Bonauto DK, et al. The natural
73. Ma Z. Clinical Observations on acupuncture and mox- course of carpal tunnel syndrome in a working population.
ibustion treatment of HIV positive peripheral facial paral- Scand J Work Environ Health 2010;36:384–393.
ysis. Shanghai J of Acupunct Moxib 2004;23:19–20. 93. Sim H, Shin BC, Lee MS, et al. Acupuncture for carpal
74. Shao S. Acupuncture and western medicine for 58 pa- tunnel syndrome: A systematic review of randomized
tients with peripheral facial palsy. New Chin Med 1999; controlled trials. J Pain 2011;12:307–314.
30:14. 94. Higgins JPT, Green S, eds. Cochrane Handbook for Sys-
75. Yang G. Comparison of the efficacy between acupuncture tematic Reviews of Interventions. Version 5.1.0 [updated
and therapy apparatus for Bell’s palsy. J Clin Acupunct March 2011]. The Cochrane Collaboration, 2011. Online
Moxib 2001;17:28–29. document at: www.handbook.cochrane.org, accessed No-
76. Yu Y. Analysis of acupuncture for peripheral facial palsy. vember 3, 2016.
Shanghai J Acupunct Moxib 1999;18:26. 95. White A, Cummings M, Barlas P, et al. Defining an ad-
77. He L, Zhou MK, Zhou D, et al. Acupuncture for Bell’s equate dose of acupuncture using a neurophysiological
palsy. Cochrane Database Syst Rev 2007;CD002914. approach—A narrative review of the literature. Acupunct
78. Chen N, Zhou M, He L, et al. Acupuncture for Bell’s palsy. Med 2008;26:111–120.
Cochrane Database Syst Rev 2010;CD002914. 96. Hui KK, Nixon EE, Vangel MG, et al. Characterization of
79. Zheng H, Li Y, Chen M. Evidence based acupuncture the ‘‘deqi’’ response in acupuncture. BMC Complement
practice recommendations for peripheral facial paralysis. Altern Med 2007;7:33.
Am J Chin Med 2009;37:35–43. 97. Kou W, Gareus I, Bell JD, et al. Quantification of DeQi
80. Kim JI, Lee MS, Choi TY, et al. Acupuncture for Bell’s sensation by visual analog scales in healthy humans after
palsy: A systematic review and meta-analysis. Chin J Integr immunostimulating acupuncture treatment. Am J Chin Med
Med 2012;18:48–55. 2007;35:753–765.
81. Dai F, Zhang Y. Acupuncture point-penetrating method 98. Kong J, Kaptchuk TJ, Polich G, et al. An fMRI study on
combined with glucocorticoids for 36 cases of Bell’s palsy. the interaction and dissociation between expectation of
Zhejiang J Tradit Chin Med 2009;44. pain relief and acupuncture treatment. Neuroimage 2009;
82. Wang L. Clinical observation on acupuncture combined 47:1066–1076.
medicine treatment for acute idiopathic facial paralysis.
J Sichuan Tradit Chin Med 2007;25:109–110. Address correspondence to:
83. Xuan L, Wang L, Hou J, et al. Study on shendao point of Alexandra Dimitrova, MD
wild-horizontal needling for functional recovery of facial Department of Neurology
paralysis of facial muscles. Chin J Tradit Med Sci Technol Oregon Health and Science University
2007;14:6–7. 3181 SW Sam Jackson Park Road
84. Yang C, Bao J, Zhang Z, et al. Observations on the efficacy Mail Code CR120
of combined acupuncture and medication for treating in Portland, OR 97239
320 cases of facial paralysis. Sci Tech Info Gansu 2006;
35:240–241. E-mail: [email protected]