View
253
Download
31
Category
Preview:
Citation preview
Bioelectromagnetics 29:245^256 (2008)
Review
Electrical Properties of Acupuncture Pointsand Meridians: ASystematic Review
Andrew C. Ahn,1,2,3* Agatha P. Colbert,4 Belinda J. Anderson,5 /rjan G. Martinsen,6,7
Richard Hammerschlag,8 Steve Cina,9 Peter M.Wayne,1 and Helene M. Langevin10,11
1Division for Research and Education in Complementary andIntegrativeMedical Therapies, HarvardMedical School, Boston, Massachusetts
2Division of GeneralMedicine andPrimary Care, Beth Israel DeaconessMedical Center,Boston, Massachusetts
3Center for Biomedical Engineering, Massachusetts Institute of Technology,Cambridge, Massachusetts
4Helfgott Research Institute, National College of NaturalMedicine, Portland, Oregon5Pacific College of OrientalMedicine, NewYork, NewYork6Department of Physics, University of Oslo, Oslo, Norway
7Department of Clinical andBiomedical Engineering, Rikshospitalet, Oslo, Norway8Oregon College of OrientalMedicine, Portland, Oregon
9NewEngland School ofAcupuncture,Watertown, Massachusetts10Department of Neurology, University of Vermont, Burlington,Vermont11Department of Orthopaedics andRehabilitation, University of Vermont,
Burlington,Vermont
According to conventional wisdom within the acupuncture community, acupuncture points andmeridians are special conduits for electrical signals. This view gained popularity after anecdotalreports and clinical studies asserted that these anatomical structures are characterized by lowerelectrical impedance compared to adjacent controls. To ascertain whether evidence exists to support orrefute this claim, we conducted a systematic review of studies directly evaluating the electricalcharacteristics of acupuncture structures and appropriate controls. We searched seven electronicdatabases until August 2007, hand-searched references, and consulted technical experts. We limitedthe review to primary data human studies published in English. A quality scoring system wascreated and employed for this review. A total of 16 articles representing 18 studies met inclusioncriteria: 9 examining acupuncture points and 9 examining meridians. Five out of 9 point studiesshowed positive association between acupuncture points and lower electrical resistance andimpedance, while 7 out of 9 meridian studies showed positive association between acupuncturemeridians and lower electrical impedance and higher capacitance. The studies were generally poorin quality and limited by small sample size and multiple confounders. Based on this review, theevidence does not conclusively support the claim that acupuncture points or meridians areelectrically distinguishable. However, the preliminary findings are suggestive and offerfuture directions for research based on in-depth interpretation of the data. Bioelectromagnetics,29:245256, 2008. 2008 Wiley-Liss, Inc.
Key words: impedance; resistance; capacitance; skin; electrodermal
62008Wiley-Liss, Inc.
Grant sponsor: National Center for Complementary AlternativeMedicine (NCCAM); Grant number: K23-AT003238. The con-tents of this research are solely the responsibility of the authorsand do not necessarily represent the official views of NCCAM,National Institutes of Health.
*Correspondence to: Andrew C. Ahn, Harvard Medical School,Division for Research and Education in Complementary andIntegrative Medical Therapies, 401 Park Drive Suite 22A-West,Boston, MA 02215. E-mail: aahn@hms.harvard.edu
Received for review 12 October 2007; Final revision received 28November 2007
DOI 10.1002/bem.20403Published online 31 January 2008 in Wiley InterScience(www.interscience.wiley.com).
BACKGROUND
Despite hundreds of mechanistic and clinical studiesin acupuncture, the fundamental tenets of acupuncturethe point and meridianare poorly understood. ClassicChinese medical theory recognizes about 361 points, saidto be located on 14 main meridians connecting the body.The 14 main meridians are associated with specific organsand serve as pathways for the flow of vital energy or qi[Birch and Felt, 1999; Kaptchuk, 2002]. Within theacupuncture community, it is a commonly held opinionthat these acupuncture structures are special conduitsfor electrical signals [Brewitt, 1995]. This view gainedpopularity as early as the 1950s when anecdotal reportsand clinical studies from China, Japan, France, Germany,and the United States asserted that acupuncture points andmeridians are distinguishable by their lower electricalimpedance compared to adjacent controls [Royal andRoyal, 1991; Brewitt, 1995]. As this view gained traction,devices such as acupuncture point locators and electro-diagnostic devices became commercially available toscientifically assist the clinicians in diagnosis andtreatment delivery.
The fundamental premise by which these deviceswere developed, however, has always been viewed withconsiderable skepticism from the conventional scien-tific community. Confounding factors, such as skinmoisture, electrode pressure, stratum corneum thick-ness, electrode polarization, and other factors have ledmany to assert that the reportedly distinct electricalcharacteristics are attributable to external factors and/orartifacts and not to the acupuncture point or meridian.To this day, no systematic evaluation of these studieshas been conducted.
This article aims to systematically review andsummarize the studies evaluating the electrical propertiesof acupuncture points and meridians. This review assesseswhether sufficient evidence can support or refute the claimthat acupuncture structures are electrically distinct. Tomaintain focus, we limited the review to human studies.This review does not include studies that assess theclinical significance of these electrical measures, focus onthe temporal behavior of electrical measurements atacupuncture structures alone, or offer theoretical explana-tion for this reported association without empirical data.To make this review as objective as possible, we createdand employed a scoring system to characterize the qualityof the studies that met review criteria.
EVIDENCE ACQUISITION
Search Strategy
We sought studies that evaluated the electricalproperties of acupuncture points and meridians com-
pared to appropriate controls. We searched sevenelectronic databases, including PubMed, CochraneLibrary Database, AltHealthWatch (EBSCO Host),Allied and Complementary Database (AMED), Biosis,EMBASE, and Scientific Citation Index Expanded fordates ranging from the databases inception to August2007 for all seven databases. We also searched theConference Paper Index and Dissertation Citationbut found no useful citations or references in thesetwo databases. Searches were limited to the Englishlanguage due to translation challenges.
Titles, abstracts, and keyword lists of the sevenelectronic databases were searched using the followingMeSH terms: acupuncture, acupoint, and meridiancrossed with the terms: impedance, resistance, conduc-tivity, conductance, admittance, susceptance, potential,capacitive, capacitance, electropermeable, electrophy-siologic, electrodermal, electrodiagnostic, and electro-diagnosis. The MeSH terms were exploded to maximizeyield and to capture all possible relevant studies. Welimited the search to Human or English languagestudies, if such search options were available. Wherepossible, related articles function was used to identifyadditional references. We referred to popular acupunc-ture texts to identify other potential studies, hand-searched all relevant references from papers and texts,and contacted experts in the field to identify anyadditional references.
Study Selection
For each database citation, we read the fullabstract to determine whether acupuncture pointsor meridians were measured electrically. Whenthe abstract lacked sufficient detail or was simplyabsent, the full text article was acquired for citationscontaining titles or abstracts indicating involvementof electrical stimulation or device. Full text articlescontaining primary data on electrical measures atpoints and/or meridians as well as controls weremaintained in the review and further evaluatedfor methodological quality. Controls were definedas (1) areas not containing an acupuncture point ormeridian or (2) areas within the vicinity of a pointor meridian. Based on the second definition, studiescontaining only topographic profiles of electricalmeasures around an acupuncture point were maintainedin the review. Non-English publications and studieswith animals were excluded. We did not includestudies that evaluated the clinical significance ofelectrical measures, reported electrical measurementswithout appropriate controls, or speculated on themechanistic reasons for these electrical characteristicsat acupuncture points or meridians without empiricaldata.
246 Ahn et al.
Bioelectromagnetics
Full-text articles were further screened with apredetermined set of scoring criteria and eliminated ifthey scored less than three out of the possible tenpoints. This strategy was employed based on theassumption that poor quality studies would unlikelycontribute meaningfully to our understanding ofacupuncture points and meridians.
Development and Application of theScoring Instrument
To assess study quality in a manner that minimizessubjectivity, a set of scoring criteria was created forthis review. The criteria were developed by a multi-disciplinary panel of five experts, representing therelevant disciplines of acupuncture, biomedicine,electrophysiology, biophysics, electrical engineering,and survey methodology. Each panelist first produceda list of items he/she considered important forevaluating electrical characteristics at acupuncturesites. The items were combined into one large listand resent to all panelists for reevaluation. Items withlow face validity were eliminated by consensus, withthe understanding that the instrument should becomprehensive yet flexible enough to account forstudy diversity. The items were presented in a yes/no/partial format to minimize subjectivity.
Based on panelists feedback, a preliminaryscoring instrument was created and subsequentlytested on selected articles by three academic clini-cianstwo of whom were not on the panel of experts.Problems with clarity and/or applicability of eachitem were identified, and wording was modified toaddress these concerns. The panelists categorized theitems into ten domains: subjects and settings, siteselection, controls, electrode, confounding variables,instruments, electrical input, measurements, blinding,and result/statistical analysis. Each domain has a valueof one point, yielding a total of ten points for the scoringsystem. The final instrument was resent to panelistsfor final approval. (The final scoring instrument can beviewed at http://osher. hms.harvard.edu/pu_general_pub.asp.)
Two physicians (AA and AC) with backgrounds inacupuncture and electrophysiology independentlyreviewed and scored the qualifying studies. Thesereviewers were not blinded to the study title or authorsnames. Each item within the scoring criterion wasdiscussed until a final agreement was reached. For thoseitems where an agreement could not be reached, a thirdparty evaluator was involved to achieve consensus.Relevant data were extracted from the studies andaggregated in tabular form.
EVIDENCE SYNTHESIS
Selection Results
From 1663 citations, we identified 320 withpotential relevance to this review (see Fig. 1). Afterobtaining and reading the full texts, we identified50 articles that actively measured electrical parametersat the skin. Of these articles, 10 were excluded due tolack of a control [Nakatani, 1956, 1986; Nakatani andYamashita, 1977; Motoyama et al., 1984; Motoyama,1986; Ionescutirgoviste et al., 1987; Cho and Chun,1994; Comunetti et al., 1995; Cheng et al., 2000; Fanget al., 2004]; 18 were excluded due to poor quality asdefined by a quality score of less than three out ofpossible ten points [Bergsmann and Woolley-Hart,1973; Noordergraaf and Silage, 1973; Brown et al.,1974; Ghaznavi, 1974; Roppel and Mitchell, 1975;Becker, 1976; Shenberger, 1977; Eory, 1984; Zhu et al.,1984a; Zhu et al., 1984b; Croley, 1986; Nansel andJansen, 1988; Zhu et al., 1988; Kitakoji et al., 1994;Chen, 1996; Eickhorn and Schimmel, 1999; Lu et al.,1999; Halek, 2000]; and 6 were excluded for usingelectrical parameters to initially define acupuncturepoints or meridians [Wulfson and Warren, 1976;Prokhorov et al., 2000a,b, 2006; Lee et al., 2003; Zhanget al., 2004]. Sixteen articles representing 18 studieswere maintained in the review and evaluated indetail. Nine studies evaluated the acupuncture point[Reichmanis et al., 1975, 1976; Hyvarinen andKarlsson, 1977; McCarroll and Rowley, 1979; Poonet al., 1980; Falk et al., 2000; Pearson et al., 2007], whilethe other nine evaluated the acupuncture meridian[Reichmanis et al., 1977a,b, 1979; Hu et al., 1992;Zhang et al., 1999; Martinsen et al., 2001; Johng et al.,2002; Lee et al., 2003; Ahn et al., 2005].
Acupuncture Points: Data Summary
Most studies evaluating the acupuncture pointswere of poor quality and published in journals withlow citation index (see Table 1). All studies testedthe hypothesis that acupuncture points are areas oflower resistance or impedance. Overall, the samplesizes were small (mean N 13), and study participantswere mostly healthy. DC or low frequency currentswere commonly used and thus electrical resistanceor conductance was the parameter most frequentlyevaluated. In four cases, the choice of controls wasinadequately justified and could have easily influencedthe outcome [Reichmanis et al., 1975, 1976; Falk et al.,2000; Pearson et al., 2007]. All studies employed amonopolar electrode arrangement with a smaller activeelectrode over the point(s) of interest and a largerreference electrode frequently held in the hand by the
Properties of Acupuncture Points andMeridians 247
Bioelectromagnetics
study participant. In addition, the studies tended to limittheir evaluation to the upper extremities. Only threestudies evaluated acupoints not located on the armsand evaluated points on the ear and forehead aswell [Hyvarinen and Karlsson, 1977; Falk et al., 2000;Pearson et al., 2007].
While five out of nine studies showed a positiveassociation between acupuncture points and decreasedelectrical impedance/resistance, the higher qualitystudies showed little to no association. Pearson et al.sstudy [2007], the highest rated study, used two devicesone with a DC current and the other with a 100 Hz ACcurrentto determine the resistance/impedance at an(1) acupuncture point, (2) non-acupuncture point locatedon a meridian, and (3) a non-acupuncture point not on ameridian. These three points were identified on severalbody sites that included the palm (PC8), forehead(GB14), and fourth finger (TW1). Only acupoint GB14showed any significant differencewith the non-meridian/non-acupuncture control and, furthermore, this differ-ence was seen only with the 100 Hz device. The meridianpoints, on the other hand, demonstrated significant (ornear significant) decrease in electrical impedance com-pared to the non-meridian control in all three body siteswith the exception of the DC resistance at acupoint PC8.
In evaluating acupuncture points, two generalstrategies were adopted. One strategy was to firstidentify the acupuncture point/control and then meas-
ure the electrical resistance or conductance [McCarrolland Rowley, 1979; Pearson et al., 2007]. The otherstrategy was to scan or survey the skin to identifylow resistance areas and subsequently see if theareas coincided with acupuncture points [Hyvarinenand Karlsson, 1977]. The studies that resorted tothe latter method tended to have more positiveresults than those taking the former approach. Thisimplies that the inappropriate or inaccurate identifica-tion of points might have contributed to the discrepantfindings.
The techniques applied in these studies wereremarkably diverse, particularly with respect to probetype. Reichmanis in 1975 wheeled a circular steelelectrode 0.2 cm in diameter along the meridian andcontrol line [Reichmanis et al., 1975], whereas Reich-manis in 1976 used a 6 6 array of 2 mm diameter steelelectrodes to produce a topographic representation ofelectrical conductance [Reichmanis et al., 1976]. Incontrast, Hyvarinen used two concentric electrodesa0.5 mm diameter inner electrode and a 2.5 mm diameterouter ring electrode that served as a control [Hyvarinenand Karlsson, 1977], while Poon et al. [1980] used anadhesive, conducting tape with high voltage intensities.This diversity in techniques accounts for why the resultsare neither readily comparable between studies norgeneralizable to all low-impedance points and controlson the human body.
Fig. 1. Study flow.
248 Ahn et al.
Bioelectromagnetics
TA
BL
E1
.S
tud
ies
Ele
ctri
cal
Ch
ara
cter
iza
tio
no
fA
cup
un
ctu
reP
oin
ts
Ref
eren
ces
Des
ign
Sam
ple
Ele
ctri
cal
cur-
rent
Ele
ctro
de
Sit
eo
fte
stin
gC
on
tro
lR
esu
lts
Ev
iden
ced
irec
tio
n
Qu
alit
ysc
ore
(ou
to
f1
0)
Rei
chm
anis
etal
.[1
975
]Id
enti
fied
max
imal
con
du
ctan
cep
oin
tsw
ith
in1
cm2
area
atac
up
oin
tan
dco
ntr
ol
and
com
par
ed
frac
tio
nal
incr
ease
in
con
du
ctan
ce
7D
C,
1mA
0.2
cmw
ide
stai
nle
ssst
eel
wh
eel
17
acu
po
ints
on
PC
and
LI
mer
idia
nb
/wm
etac
arpophal
angea
lan
del
bow
join
t
N
earb
yan
atom
ical
lysi
mil
arlo
cati
on
2
ou
to
f6
PC
po
ints
and
3o
ut
of
11
LI
po
ints
had
sig
nifi
can
tfr
acti
on
alin
crea
sein
con
du
ctan
cein
all
7su
bje
cts
4
Rei
chm
anis
etal
.[1
976
]O
bta
ined
top
og
rap
hic
con
du
ctio
np
rofi
les
atac
up
oin
tan
dco
ntr
ol,
and
com
par
edco
nd
uct
ance
var
iance
10
DC
,2
V0
.2cm
36
stai
nle
ssst
eel
rod
sin
gri
dp
atte
rn3
.5
2.5
cm
10
acu
po
ints
TB
4
9,
Lu
5
8(o
nfo
rear
m)
2.5
2.5
cmn
earb
yp
oin
t
ascl
ose
asfe
asib
le
76
/10
0ac
up
oin
tlo
cish
ow
edin
crea
sed
con
du
ctan
cevar
iance
com
par
edto
con
tro
l>
5su
bje
cts
wit
hin
crea
sed
con
du
ctan
cevar
ian
ceat
all
10
acu
po
ints
4
Hy
var
inen
and
Kar
lsso
n[1
97
7]
Iden
tifi
edlo
wre
sist
ance
po
ints
on
fore
arm
and
han
du
sin
gco
nce
ntr
icel
ectr
od
e
5h
ealt
hy
mal
em
edic
alst
ud
ents
DC
,6
.7V
0.5
mm
dia
met
erci
rcu
lar
elec
tro
de
wit
h2
.5m
mco
nce
ntr
icre
fere
nce
elec
tro
de
Fore
arm
,h
and
s,ea
rsS
urr
ou
nd
ing
skin
(2.5
mm
rad
ius)
Rep
rod
uci
ble
low
resi
stan
cep
oin
tsw
ere
iden
tifi
edo
nh
and
and
ears
.P
oin
tsre
sem
ble
acu
po
ints
inlo
cati
on
3
Hy
var
inen
and
Kar
lsso
n[1
97
7]
Mea
sure
dim
ped
ance
of
low
resi
stan
cep
oin
tsan
dth
ear
eaw
ith
inth
eir
vic
init
y
5h
ealt
hy
mal
em
edic
alst
ud
ents
DC
,A
C;
50
0,
10
00
Hz
0.5
mm
silv
erw
ire
Fore
arm
,ea
rsS
urr
ou
nd
ing
skin
(2.5
mm
rad
ius)
Imp
edan
cep
eru
nit
area
was
50
-to
10
0-f
old
low
erat
low
resi
stan
cep
oin
tsco
mp
ared
toad
jace
nt
skin
1
0
10
0kO
vs.
2
3MO
,re
spec
tivel
y
3
McC
arro
llan
dR
ow
ley
[19
79
]
Det
erm
ined
wh
eth
erim
ped
ance
min
imu
mex
iste
din
a5
5g
rid
aro
un
dac
up
oin
t
8A
C,
1k
Hz,
4.5
mA1
mm
silv
erw
ire
rod
,ch
lori
nat
ed7
acu
po
ints
on
han
d/a
rm
4o
ut
of
32
coll
ecte
dse
tssh
ow
edd
ecre
ased
skin
imp
edan
ceu
po
nre
pli
cati
on
Dat
ash
ow
edte
mp
ora
lvar
iab
ilit
y
4
Po
on
etal
.[1
98
0]
Iden
tifi
ed
elec
troper
mea
ble
p
oin
tsu
sin
gta
pe
elec
tro
de
9A
C,
0.5
Hz,
0
18
V3
1in
chep
idu
ctiv
eta
pe
elec
tro
de
Fore
arm
al
on
gL
u,
PC
mer
idia
nN
/AC
on
du
ctiv
e
spo
ts
on
fore
arm
wer
eid
enti
fied
.S
po
tsli
eal
on
gm
erid
ian
,bu
tnot
clea
rly
asso
ciat
edw
ith
acu
po
ints
.N
og
ener
alp
atte
rnac
ross
sub
ject
s
3
Fal
ket
al.
[20
00
]M
easu
red
2
2g
rid
imp
edan
ces
at
acti
ve
v
s.co
ntr
ol
zones
and
com
par
edm
inim
um
imp
edan
ces
b/w
zones
34
h/o
coca
ine
abu
seo
nm
eth
adon
e;1
7M
,1
7F
AC
,5
Hz,
2.5
mA1
mm
Ag
/Ag
Cl
pro
be
4ac
tiv
eea
rzo
nes
(sh
enm
en,
sym
pat
het
ic,
liver
,lu
ng
)
4co
ntr
ol
zon
eso
nh
elix
of
ear
Low
est
imp
edan
ceval
ues
fou
nd
wit
hin
acti
ve
zon
esw
ere
sig
nifi
can
tly
low
erth
anth
ose
wit
hin
contr
ol
zon
es.
Act
ive
1.3
6v
s.C
on
t1
.97
MO
(P
Recommended