Complementary Therapies in Medicine (2014) 22, 21—25

Available online at www.sciencedirect.com

ScienceDirect

jo ur nal home p ag e: www.elsev ierhea l th .com/ journa ls /c t im

The effects of auricular electroacupunctureon obesity in female patients — Aprospective randomized placebo-controlledpilot study

Regina Patricia Schukroa,∗, Christoph Heisererb,Andrea Michalek-Sauberera, Andreas Gleissc,Sabine Sator-Katzenschlagera

a Department of Special Anaesthesia and Pain Therapy, Medical University of Vienna, Austriab Department of Surgery, General Hospital Steyr, Steyr, Austriac Center for Medical Statistics, Informatics, and Intelligent Systems, Medical University Vienna, AustriaAvailable online 31 October 2013

KEYWORDSAuricularacupuncture;Electroacupuncture;Obesity

SummaryBackground: Obesity is a chronic condition related to serious morbidity and mortality of increas-ing incidence and prevalence. Several studies show a significantly higher weight loss withacupuncture treatment. This is the first prospective, randomized, double-blinded study, testingthe effects of auricular electroacupuncture on weight loss in obese female patients.Methods: 56 female obese patients (age > 18, Body Mass Index, BMI > 25) were randomized toreceive either auricular acupuncture with electrical stimulation with a P-Stim® device (verumgroup, n = 28) or placebo treatment with a P-Stim® dummy (n = 28) for four days. Three auricularacupuncture points were chosen (hunger 18, stomach 87 and colon 91). The treatment wasperformed once a week for a period of six weeks. A follow-up visit was performed after 4weeks. At each visit body weight, BMI (Body Mass Index) and body fat were measured.Results: Relative reduction of body weight was significantly greater in the verum group (−3.73%;CI = −4.70 to −2.76) than in the placebo group (−0.70%; CI = −1.57 to +0.16; p < 0.001) In addi-tion we also observed a significant reduction of BMI (p < 0.001) in the verum group (−3.62%;

CI = −4.39 to −2.84) compared to placebo (−0.82%; CI = −1.55 to −0.10; p < 0.001). No patientreported side effects related to acupuncture.Conclusion: In conclusion electrical auricular acupuncture could be a safe, additive, non-pharmacological treatment in obese patients.© 2013 Elsevier Ltd. All rights re

∗ Corresponding author at: Department of Special Anaesthesia and Pain MGuertel 18-20, A 1090 Vienna, Austria. Tel.: +43 1 40400 4144; fax: +43 1

E-mail address: regina.schukro@meduniwien.ac.at (R.P. Schukro).

0965-2299/$ — see front matter © 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.ctim.2013.10.002

served.

anagement, Medical University of Vienna, AKH Vienna, Waehringer 40400 6422.

2

I

Oophegt>aoa

osmatsaesaht

emcwtt

M

S

Tta(ctwlaii

cpa(uGacssl

cut

Tg

smp

p6bw

sSF

D

AtMaa

S

TcswpAdartf

ostaf

ccuatmrwa

2

ntroduction

besity is a chronic condition that is related to seri-us morbidity and mortality of increasing incidence andrevalence. Excess bodyweight is associated with elevatedealth risks, such as cardiovascular and cerebrovascular dis-ases as well as diabetes.1 The Body Mass Index (BMI) isenerally accepted as classification of obesity. Accordingo WHO (World Health Organization) guidelines, a BMI of25.00 kg/m2 is considered as overweight, 25—29.99 kg m−2

s preobese and over 30.00 kg−2 as obese. As the incidencef obesity increases new therapies to improve weight man-gement are looked for.

Especially the effects of complementary medicine onbesity are under closer investigation. Recent studieshowed a positive effect of acupuncture on appetite,etabolism, intestinal motility and emotional factors such

s stress. In addition, it can increase neural activity inhe ventromedial nuclei of hypothalamus, the tone of themooth muscle of the stomach and levels of encephalinnd serotonin in the plasma and brain tissue.2,3 The clinicalffects of acupuncture on obesity have been summarized byeveral meta-analyses in4,5 suggesting a positive effect ofcupuncture on weight loss in obese patients. The authorsowever criticized a lack of randomized placebo-controlledrials on this topic.

The aim of this study was to investigate if auricularlectroacupuncture combined with moderate diet recom-endations based on Traditional Chinese Medicine (TCM)

an result in bodyweight reduction in obese female patientshen compared to a placebo acupuncture treatment. It is

he first study that uses a P-Stim® auricular electroacupunc-ure device in the indication of weight loss.

aterials and methods

tudy design and patient selection criteria

his prospective randomized placebo-controlled clinicalrial was conducted at the Department of Special Anesthesiand Pain Management at the Medical University of ViennaAKH Vienna). The study was approved by the local ethicsommittee according to the declaration of Helsinki. All par-icipants were healthy, obese women (age > 18, BMI > 25)ho had no prior experience with acupuncture or auricu-

ar acupuncture. Further exclusion criteria were pregnancy,llergy against material (metal of needles or patch), bleed-ng disorders, anticoagulation, implanted pacemaker ormplantable cardioverter defibrillators.

Fifty six patients were randomized in two groups by aomputer-generated randomization table. Randomization ofatients was stratified with respect to age (<50 or ≥50 years)nd BMI (<30 or ≥30 kg/m2). Patients in the verum groupn = 28) received auricular acupuncture with electrical stim-lation using a P-Stim®-electroacupuncture device (BieglermbH. Allhangstrasse 18a, 3001 Mauerbach, Austria) at theuricular acupuncture points hunger 18, stomach 87 and

olon 91. The P-Stim®-device consists of a battery-poweredtimulator worn behind the ear, which was connected toterile, single use permanent needles (titanium, 27 gauge,ength 3 mm). Electrical stimulation consisted of biphasic

aaca

R.P. Schukro et al.

onstant current (2 mA) pulses of 1 Hz every 3 s (3 h of stim-lation followed by a pause of 3 h to avoid development ofolerance).

In the placebo group (n = 28) a P-Stim® dummy was used.he dummy-device had no power supply and had beenrinded to leave only metal plates.

An electrical conductance device meter (multipointelection penTM, Biegler GmbH, Mauerbach, Austria) whicheasures skin resistance, was used to identify acupunctureoints in both groups.

Patients received auricular electro acupuncture orlacebo for 4 days (24 h) per week during a period of

weeks. All patients received their acupuncture needlesetween 07:00 and 11:00 a.m. A follow-up was performed 4eeks after the last treatment.

At each visit, body weight, BMI and body fat were mea-ured. The Body Impedance Analysis (BIA formulas) fromchindler and Ludvik6 were used for the calculation of theFM (fat free mass) and BCM (body cell mass).

iet according to TCM guidelines

t the first visit all patients received a dietary consulta-ion following guidelines according to Traditional Chineseedicine (TCM). Patients should eat regularly (e.g. 3 times

day), avoid cold and raw food, white sugar and fast foodnd should reduce intake of dairy products.

tatistical analysis

he study was planned to detect a group difference of 4 per-ent points with respect to relative weight change betweentart and end of treatment. Since the variability of relativeeight change was unknown an a priori sample size of 60atients was targeted, a number seen in previous studies.fter the data of 40 patients were available the standardeviation was estimated (without performing an interimnalysis of the group comparison) in order to obtain a moreeliable sample size estimate (‘‘half-sampling’’). Based onhis new sample size further patients were finally recruitedor the study.

In the planning phase of the study the relative reductionf body weight between baseline and end of treatment waselected as the primary outcome variable. Relative reduc-ion of body weight between baseline or end of treatmentnd follow-up as well as relative reduction in BMI and bodyat were defined as secondary outcome variables.

Due to asymmetric distribution of some of the variables,ontinuous baseline variables are presented as group spe-ific medians and quartiles and compared between groupssing Wilcoxon’s rank-sum test. Relative weight changesre calculated such that negative signs represent reduc-ion (e.g., of weight). Missing values are imputed usingultiple imputation (‘‘proc MI’’ of SAS with ‘‘monotone

eg’’ statement and 200 imputations). Boxplots of relativeeight changes use the full data set, i.e. available andveraged imputed values. Models with imputed values are

nalyzed using Rubin’s rules as implemented in proc mian-lyze of SAS. ANOVA models are used to compare relativehanges between groups, age and baseline BMI are useds co-variables for group adjustment. Results of univariate

Effects of auricular electroacupuncture on obesity

Re

lative

we

igh

t ch

an

ge

(%

)

-11

-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

start to FUend to FUstart to end

Figure 1 Boxplots of relative change in body weight (negative

S

RpTgc−(ep

ugr4

RTtg−T(

vad2

vam4

RTmw(f

vaTm(

vam7

S

NEight patients reported side effects related to the P-

values represent weight reduction), light boxes: verum, darkboxes: placebo. FU = follow-up.

models (i.e. without group adjustment) are very close toadjusted results and are not shown.

The reported p-values are the results of two-sided tests.P-values <0.05 were considered to be statistically signifi-cant. All computations were carried out using SAS softwareVersion 9.3 (SAS Institute Inc., Cary, NC, USA, 2010).

Results

Patients and demographic data

Fifty six healthy obese (BMI > 25) female patients wereenrolled in the study. Baseline characteristics are summa-rized in Table 1. Median age in the verum group was slightlyhigher than in the placebo group (54.1 vs. 49.7. years,p = 0.283). Groups did not differ significantly with regard toinitial weight, BMI, height and body fat. During the treat-ment period 6 patients in the verum group and 5 patients inthe placebo group refused further participation. Addition-ally, 3 patients (1 in the verum group, 2 in the placebo group)were lost during follow up (i.e. after end of treatment).

Primary outcome parameter

Relative change of body weight during treatmentBoxplots of relative change in body weight are shown inFig. 1. The mean change in body weight in the verum groupwas −3.73% (CI = −4.70 to −2.76) whereas in the placebo

group a reduction of only −0.70% (CI = −1.57 to +0.16) couldbe observed. The mean difference between both groups was3.03%-points (CI 1.72—4.33; p < 0.001).

Sed

23

econdary outcome parameters

elative change of body weight during the follow-uperiodhe difference in weight change between the treatmentroups was retained after the treatment phase. The meanhange between end of study and follow-up visit was1.37% (CI = −1.93 to −0.82) in the verum group and +0.54%

CI = −0.01 to +1.09) in the placebo group. The mean differ-nce between the groups was 1.91% points (CI = 1.13—2.70,

< 0.001).The mean change between begin of study and follow-

p visit was −5.08% (CI = −6.02 to −4.13) in the verumroup and −0.16% (CI = −1.06 to +0.74) in the placebo groupespectively. The mean difference between the groups was.92% points (CI = 3.61—6.22, p < 0.001).

elative change of body mass indexhe mean change between begin of study and end ofreatment was −3.62% (CI = −4.39 to −2.84) in the verumroup whereas in the placebo group a reduction of only0.82% (CI = −1.55 to −0.10) points could be observed.he mean difference between both groups was 2.79%-pointsp < 0.001).

The mean change between end of study and follow-upisit was −1.44% (CI = −1.97 to −0.92) for the verum groupnd 0.56% (CI = 0.01—1.11) for the placebo group. The meanifference between follow-up and end of treatment was.00% points (CI = 1.26—2.75) for both groups (p < 0.001).

The mean change between begin of study and follow-upisit was −4.99% (CI = −6.10 to −3.88) for the verum groupnd −0.26% (CI = −1.32 to 0.79) for the placebo group. Theean difference between begin of study and follow up was

.73% points (CI = 3.21—6.24) for both groups (p < 0.001)

elative change of body fathe mean change between begin of study and end of treat-ent was −5.93% (CI = −7.90 to −3.96) in the verum grouphereas in the placebo group only a reduction of −0.50%

CI = −2.42 to 1.42) points could be observed. The mean dif-erence between both groups was 5.43%-points (p < 0.001).

The mean change between end of study and follow-upisit was −2.02% (CI = −5.82 to -1.79) for the verum groupnd −0.44% (CI = −2.29 to 3.18) for the placebo group.he mean difference between follow-up and end of treat-ent was 2.46% points (CI = −2.22 to 7.13) for both groups

p = 0.302).The mean change between begin of study and follow-up

isit was −7.65% (CI = −11.86 to −3.45) for the verum groupnd −0.17% (CI = −3.45 to 3.10) for the placebo group. Theean difference between begin of study and follow up was

.48% points (CI = 2.17—12.78) for both groups (p = 0.005)

ide effects of acupuncture

o patient reported side effects related to acupuncture.

tim®/placebo device, namely skin irritations behind thear caused by the adhesive patch of the P-Stim®/placeboevice.

24 R.P. Schukro et al.

Table 1 Baseline characteristics.

Variable Verum (n = 28) Placebo (n = 28)

Lower quartile Median Upper quartile Lower quartile Median Upper quartile

Age (years) 44.5 54.1 60.7 36.7 49.7 55.7Weight (kg) 81.1 93.0 102.2 78.6 92.6 108.6BMI (kg/m2) 30.0 33.7 36.8 28.0 33.9 37.1Body fat (%) 35.8 43.1 51.6 32.8 45.4 52.4

D

Taa

scttPeofich

vattta

(fahtmaYiabiiptoops

eoie

wseasc

aeni

L

Atidlpa

aFaf

C

T

R

iscussion

his study demonstrates a significant effect of electricaluricular acupuncture on weight loss in obese women. Inddition, also the BMI and body fat decreased.

Auricular acupuncture was associated with some mildide effects that resolved immediately after end of appli-ation. The mechanism how acupuncture might be useful inhe treatment of obesity is still under investigation. Oneheory refers to a vagal effect on appetite and satiety.7

eripheral vagal nerve stimulation (VNS) via subcutaneouslectrodes positioned between the clavicle and the mastoidr near the esophagogastric junction has been performedor weight loss.7 A positive association between some block-ng algorithms and weight loss has been reported,7 however,omparative studies are lacking and the clinical value of VNSas still to be confirmed.8

Peuker and Filler provided an overview of the inner-ations pattern of the lateral surface of the auricle. Theuricular branch of the vagus nerve (ABNV) supplies mainlyhe cymba conchae (100%), the cavity of concha (45%), theragus (45%) and the Antihelix (73%).9 The vagal innerva-ion of the auricle might contribute to a possible effect ofuricular acupuncture on body weight.

In addition the arcuate nucleus of the hypothalamusARH), which is a crucial center for the modulation ofood intake, could play a role in the mode of action ofcupuncture. It contains two populations of neurons thatave opposite influence on appetite. One neuron expresseshe anorexigenic peptide alpha-melanocyte stimulating hor-one (�-MSH) (derived from proopiomelanocortin, POMC)

nd the other the orexigenic peptide NPY (neuropeptide). Tian et al. showed that electrical acupuncture (EA)

ncreased the expression of mRNA encoding for POMC as wells increased levels of �-MSH.10 Fei et al. found a reducedody weight gain associated with the inhibition of foodntake in obese rats.11 Electrical acupuncture stimulationncreased peptide levels of �-MSH and mRNA levels of itsrecursor POMC in the ARH neurons. Moreover, the CSF con-ent of �-MSH was increased by EA application.11 Lesionsf ARH by glutamate abolished the inhibition effect of EAn food intake and body weight, which suggests that ARHlays an important role in mediating the satiety effect of EAtimulation.11

Two recent systematic reviews investigated the clinical

ffects of acupuncture and Chinese herbal medicine (CHM)n obesity.4,5 Both found a significantly higher weight lossn the acupuncture/CHM groups, higher BMI reduction orven higher CRP decrease compared to placebo or common

estern pharmacological therapies such as metformin oributramine. Almost all reviewed studies reported sideffects of acupuncture, but all of them were classifieds mild. Limitations of most clinical trials were the smallample size or poor methodology (e.g. no placebo/shamontrol).

Our study supports these findings, which may show a newpproach to the clinical management of obesity. P-Stim® isasy to use; it does not require sophisticated training in Chi-ese Medicine12 and can be applied by any physician trainedn acupuncture.

imitations

s this study was planned as pilot project, sample size andhe duration of observation are limited. Our further stud-es on the effects of P-Stim® on obesity will involve a moreetailed assessment of food intake and quality of life, aarger sample size and a longer time of observation to dis-lay possible effects of auricular electrical acupuncture onppetite and satiety.

In conclusion electrical auricular acupuncture could be safe, additive, non-pharmacological treatment of obesity.urther studies including a larger number of patients and

more detailed assessment of food intake need to be per-ormed.

onflict of interest

here is no conflict of interest to declare.

eferences

1. Bray GA. Obesity: basic considerations and clinical approaches.Dis Mon 1989;35:449—537.

2. Cabyoglu MT, Ergene N, Tan U. The treatment of obesity byacupuncture. Int J Neurosci 2006;116:165—75.

3. Shiraishi T, Onoe M, Kojima T, Sameshima Y, Kageyama T.Effects of auricular stimulation on feeding-related hypothala-mic neuronal activity in normal and obese rats. Brain Res Bull1995;36:141—8.

4. Cho SH, Lee JS, Thabane L, Lee J. Acupuncture for obe-sity: a systematic review and meta-analysis. Int J Obes (Lond)

2009;33:183—96.

5. Sui Y, Zhao HL, Wong VC, Brown N, Li XL, Kwan AK, et al. Asystematic review on use of Chinese medicine and acupuncturefor treatment of obesity. Obes Rev 2012;13:409—30.

1

1

12. Sator-Katzenschlager SM, Wolfler MM, Kozek-Langenecker SA,

Effects of auricular electroacupuncture on obesity

6. Schindler K, Ludvik B. Assessment of body composition—–methods and practical aspects. Wien Med Wochenschr2004;154:305—12.

7. Pardo JV, Sheikh SA, Kuskowski MA, Surerus-Johnson C, HagenMC, Lee JT, et al. Weight loss during chronic, cervical vagusnerve stimulation in depressed patients with obesity: an obser-vation. Int J Obes (Lond) 2007;31:1756—9.

8. Verdam FJ, Schouten R, Greve JW, Koek GH, Bouvy ND. An

update on less invasive and endoscopic techniques mimickingthe effect of bariatric surgery. J Obes 2012;2012:597871.

9. Peuker ET, Filler TJ. The nerve supply of the human auricle. ClinAnat 2002;15:35—7.

25

0. Tian D, Li X, Niu D, Shi Y, Chang JK, Han J. Electroacupunc-ture up-regulated arcuate nucleus alpha-MSH expression in therat of diet-induced obesity. Beijing Da Xue Xue Bao 2003;35:458—61.

1. Fei W, Tian de R, Tso P, Han JS. Arcuate nucleus of hypothalamusis involved in mediating the satiety effect of electroacupunc-ture in obese rats. Peptides 2011;32:2394—9.

Sator K, Sator PG, Li B, et al. Auricular electro-acupunctureas an additional perioperative analgesic method during oocyteaspiration in IVF treatment. Hum Reprod 2006;21:2114—20.