Objective To evaluate the efficacy of the five ear acupuncture points (Shen-men, Spleen, Stomach, Hunger, Endocrine), generally used in Korean clinics for treating obesity, and compare them with the Hunger acupuncture point.
Methods A randomised controlled clinical trial was conducted in 91 Koreans (16 male and 75 female, body mass index (BMI)≥23), who had not received any other weight control treatment within the past 6 months. Subjects were divided randomly into treatment I, treatment II or sham control groups and received unilateral auricular acupuncture with indwelling needles replaced weekly for 8 weeks. Treatment I group received acupuncture at the five ear acupuncture points, treatment II group at the Hunger acupuncture point only and the sham control group received acupuncture at the five ear acupuncture points used in treatment I, but the needles were removed immediately after insertion. BMI, waist circumference, weight, body fat mass (BFM), percentage body fat and blood pressure were measured at baseline and at 4 and 8 weeks after treatment.
Results For the 58 participants who provided data at 8 weeks, significant differences in BMI, weight and BFM were found between the treatment and control groups. Treatment groups I and II showed 6.1% and 5.7% reduction in BMI, respectively (p<0.004). There were no significant differences between the two treatment groups.
Conclusions This finding suggests that the five ear acupuncture points, generally used in Korean clinics, and the Hunger point alone treatment are both effective for treating overweight people.
Statistics from Altmetric.com
Excess body weight is the sixth most important factor contributing to the overall burden of disease worldwide.1 While the average life expectancy has decreased, owing in part to obesity, the major adverse effects include cardiovascular disease, type 2 diabetes and several cancers.2 ,3 WHO describes obesity as one of the most blatantly visible, yet most neglected, public health problems.4
Auricular acupuncture therapy is based on the understanding that the external ear represents all parts of the human body, including the internal organs, and provides acupuncture points corresponding to these parts.5 Auricular acupuncture therapy was discovered by Dr Paul Nogier based on observations that backaches were cured after the patient received a burn on their ear.6 Dr Nogier pictured an auricle as a fetus that was curled up and upside down; he used his anatomical knowledge to treat diseases by applying pressure to the reaction points that represent the internal organs. This treatment method was first reported at the International Acupuncture Medical Academy in 1956 (Marseilles, France).6 It is used for obesity,7 ,8 smoking,9 ,10 drug addiction11 and other conditions.12–14 Auricular acupuncture therapy for obesity has been reported to be relatively safe, economical, effective and to reduce body weight by decreasing the desire to eat.15 ,16
In previous studies, the treatment groups received acupuncture at points such as Shen-men (M55) and Stomach (M87),17 Stimulating Vagus Nerve,18 Hunger (M18) and Stomach,19 Lung (M101), Stomach and Hunger,20 immediately in front of external auditory meatus,21 and Stomach and Mouth (M84).22 Treatment of the Hunger point, in particular, has been reported to reduce both eating desire and body weight.19 However, the points Shen-men, Spleen, Stomach, Hunger and Endocrine are generally used in Korean clinics to treat obesity.23 ,24 Thus, we aimed to test the efficacy of these five ear acupuncture points in comparison with the Hunger acupuncture point alone, which has been reported previously.19
Materials and methods
This was a randomised controlled clinical trial (figure 1). Each patient was allocated to receive unilateral auricular acupuncture at five ear acupuncture points once a week for 8 weeks (treatment I), at the Hunger acupuncture point alone (treatment II), or assigned to the control group. Three sets of group numbers were generated by computer and placed in opaque sealed envelopes. Subjects drew put one envelope which was then opened by an independent investigator who did not participate in the study process. Subjects were instructed to eat the same number of meals regularly and not to eat any snacks, meat or wheat flour meal. Meals comprised one bowl of rice (210 g) for subjects >70 kg and two-thirds of a bowl of rice (140 g) for those <70 kg, with instructions to eat side dishes balanced with the rice. Also, subjects were instructed not to take exercise other than that required for their daily work.
Subjects and setting
The trial was conducted from April to September 2006. Ninety-one Koreans (16 male and 75 female) with a body mass index (BMI) ≥23, who answered advertisements posted in daily newspapers and in hospital pamphlets, were included in the study. The Asia-Pacific Region of WHO defines BMI≥30 as severe obesity, 30>BMI≥25 as obesity and 25>BMI≥23 as overweight for the Asian population. In this study, we defined BMI≥23 as obesity in a broad sense.25
The sample size calculation was based on a previous study which used a similar intervention in similar patient groups and the same primary outcome instrument.20 A total of 123 individuals volunteered to participate and their suitability for the study was verified using the inclusion and exclusion criteria through telephone interview and a preliminary examination before the study. The inclusion and exclusion criteria are listed in table 1.
The 91 participants were allocated randomly into the treatment I (n=31), treatment II (n=30) or control (n=30) group. The study was approved by the institutional review board of the department of applied Korean medicine of Kyung Hee University and all subjects provided informed consent (AKMC IRB 0601-01-A3).
Treatment I subjects received unilateral acupuncture with indwelling needles at five ear acupuncture points (Shen-men, Stomach, Spleen, Hunger, Endocrine), which are generally used in Korean clinical treatments (see online supplementary figure S1).19 ,20 Treatment II subjects received an indwelling needle unilaterally at the Hunger acupuncture point only, whose specificity has been reported.19 Treatment groups received acupuncture in one ear on the first visit. One week later, the applied needles were removed and the same treatment was performed on their other ear. The needles used for the treatment groups were tack-like, 0.2×2 (head diameter×point length) mm-sized Dong Bang acupuncture needles. The heads of tack-like needles are fixed on surgical tape so that when the steel points are inserted, the needles remain attached to the skin (Suwon, Korea). The depth of insertion was 2 mm. The control group received sham acupuncture at the five ear acupuncture points used in treatment I group at their first visit. The needles used for the control group were fixed on surgical tape, but were removed immediately after insertion (2 mm), while the surgical tape remained on the acupuncture points (figure 2). After 1 week, the surgical tape was removed and the same procedure was performed on their other ear. A single doctor of traditional Korean medicine performed all treatments.26 Treatment was carried out once a week for 8 weeks; subjects were blinded to their group assignment.
All three groups were subject to a reduced-intake diet regimen with exercise forbidden (see “Study design” section), to exclude any confounding factor that might affect the treatment.
The outcomes were BMI, waist circumference (WC), weight, body fat mass (BFM), percentage body fat (PBF) and blood pressure (BP). Weight was measured once a week from before the start of treatment through 8 weeks and after removing the needles at 8 weeks (nine measurements in total) using the Fatness Measuring System DS-102 (Dong Sahn Jenix, Korea). Height was measured twice at the beginning of the study using the Fatness Measuring System DS-102 and the mean was recorded. WC, BFM, PBF and BP were measured at 4-week intervals, before treatments at 1 and 5 weeks and after removing the needles at 8 weeks (three measurements in total). WC was measured half way between the lowest subcostal line and the top of the iliac crest. BFM and PBF were measured by the bioelectrical impedance method using the InBody330 (Biospace, Korea). BP was measured after resting for 5 min, using a sphygmomanometer (HICO, Japan). One trained nurse performed all measurements and was blinded to the participants’ group. Compliance with diet and exercise regimen was recorded by the patients and checked weekly at the clinic. We also provided instructions to any subjects with problems in complying with the regimen.
All measurements are presented as frequencies; descriptive statistics and continuous variables are presented as mean±SD. Analysis of variance was conducted to compare changes in physical measurements. Any measurement that showed a significant change in F value was processed using multiple comparisons by least significant difference. To identify the factors associated with changes in the measurements, a multiple regression analysis was conducted. In all analyses, p<0.05 was considered to indicate a significant difference, using the SPSS V.12.0 software. Only the participants who provided information at 8 weeks were included in the statistical analysis.
As shown in figure 1, 91 participants were randomised to three groups. Nine participants dropped out of each of the two treatment groups and 15 from the control group.
Groups were similar at baseline (table 2) except for a significant difference in mean age (p=0.029) of the three groups.
Body mass index
There were significant differences in BMI among the three groups from 4 weeks. The treatment and control groups differed significantly, from 3 weeks (figure 3). Treatment groups I and II showed 6.1% and 5.7% reduction in BMI (p<0.004) at 8 weeks. No significant difference in BMI was found between the treatment I and treatment II groups (table 3).
Age and baseline BMI were significantly associated with the reduction in BMI. With increased age and higher baseline BMI, the percentage reduction of BMI was greater. Although age differed significantly among the three groups, the effects according to treatment group remained significant after controlling for age and baseline BMI (see online supplementary table S1).
WC decreased in all three groups from 4 to 8 weeks and the reduction was greatest in the treatment I group. There were no significant differences between the three groups after 4 weeks (F=2.20; p=0.145); however, the three groups differed significantly after 8 weeks (F=4.58; p=0.018). Further, after 8 weeks, there was a significant difference between the treatment I and control groups, but not between the treatment II and control groups (table 3).
Only gender was significantly associated with baseline WC; the mean baseline WC of women was 10.6 cm less than that of men. Only age affected WC at 4–8 weeks: with increased age, the percentage reduction of WC was greater. However, after controlling for age, an increase in the percentage reduction in the treatment I group was not significant (see online supplementary table S2).
Weight differed significantly between the treatment groups and the control group after 4 weeks, whereas there was no significant difference between treatment groups I and II (table 3).
BFM, body fat
There were no significant differences in BFM between the three groups after 4 weeks (F=1.30; p=0.279), whereas at 8 weeks, the three groups differed significantly (F=4.12; p=0.002). Also, there was a significant difference in BFM between the treatment I and control groups, but not between the treatment II and control groups (table 3). There were no significant differences between the three groups at 4 or 8 weeks (table 3).
There were no significant differences between the three groups at 4 or 8 weeks (table 3).
After 8 weeks of treatment, BMI, WC and weight differed significantly between the control group and the treatment groups. This difference was supported by differences in WC and BFM, but not BP. The difference between treatment groups in WC did not persist when corrected for age.
In this study, withdrawals occurred mostly from 3 to 5 weeks owing to dissatisfaction with weight loss, difficulty in following the dietary regimen, schedule conflict and unknown reasons. The most common reasons, dissatisfaction with, and difficulty following, the dietary regimen, were particularly notable in the control group. This supports the suggestion that the treatment groups found it easier than the control group to regulate the desire to eat. Also, compliance was high after 1–2 weeks, but not thereafter. Therefore, more positive management may be necessary to enhance the compliance from 3 weeks onward (see online supplementary table S3).
We aimed to blind participants using the same acupuncture points for treatment I group and the control group. However, even if the acupuncture needle is removed immediately, the temporary stimulation might produce an effect in the control group. Despite this possibility, significant differences between the treatment I group and the control group remained. In future studies, the effect according to duration of stimulation should be investigated. Also, only single blinding of the subjects was conducted because it was not possible to blind the operator by use of a sham needle.27
Significant differences among the treatment I, treatment II and control groups were evident from 4 weeks. Thus, in future studies, treatment should be continued for at least 4 weeks. The percentage reduction of BMI was greater with increased age and baseline BMI. However, the effects according to the treatment groups remained significant after controlling for age and baseline BMI.
WC measurements were reduced in treatment I compared with the treatment II and control groups, though this trend disappeared when corrected for age. Also, we recognised that measurement of WC is subject to greater measurement error than body weight or BMI. It has been reported that the effectiveness of acupuncture for patients with obesity is related to metabolic function.28 ,29 Increased metabolic function promotes metabolisation of fat and results in weight loss.30 Through this mechanism, our results suggest that treatment I is more useful than treatment II in metabolising abdominal fat. Thus, continuous stimulation of the five acupuncture points generally used in Korean clinics resulted in a significant reduction in central obesity in comparison with continuous stimulation of the Hunger acupuncture point only, or temporary stimulation of the five acupuncture points. If the trend we found is supported by other studies, stimulation of the Hunger acupuncture point is more convenient and a good choice; whereas if the patients have central obesity, then the five acupuncture points generally used in Korean clinics should be used. Age and baseline BMI may influence the effectiveness of auricular acupuncture treatment of obesity. Further studies are necessary to clarify the correlation between the effectiveness of auricular acupuncture treatment and age and baseline BMI.
One limitation of our study is the statistical difference in age between treatment groups I and II, which might have affected the results. Also, the study sample might have been too small to identify a small difference between the acupuncture treatments. The dropout rate was high at 36%. As obesity is a long-term problem, future studies need to test the effectiveness of acupuncture for longer follow-up periods. Finally, we did not test the blinding of the control group; participants in this group might have examined the adhesive dressings in their ears and found no needles.
Both five-needle acupuncture treatment generally used in Korean clinics and one-needle treatment at the Hunger point appear to be effective in reducing body weight in the short term. The five-needle treatment may be better for reducing WC.
Auricular acupuncture is often used to induce weight loss.
We randomised overweight people to a five-point treatment, a single-point treatment and a sham control over 8 weeks.
Both treatments were more effective than the sham control.
Review history and Supplementary material
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
SY and KSK contributed equally.
Contributors SY and KSK designed the study protocol and drafted the manuscript. SL participated in the study design and conducted the literature review. All the authors read and approved the final manuscript.
Funding This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No 2007-0054931).
Competing interests None.
Ethics approval Institutional review board of the department of applied Korean medicine of Kyung Hee University.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.