Article Text


Immediate effect of three different electroacupuncture protocols on fasting blood glucose in obese patients: a pilot study
  1. Maria Belivani1,
  2. Thomas Lundeberg2,
  3. Mike Cummings3,
  4. Charikleia Dimitroula1,
  5. Nicole Belivani1,
  6. Dimitris Vasilakos4,
  7. Apostolos Hatzitolios1
  1. 1First Propaedeutic Department of Internal Medicine, AHEPA Hospital, Aristotle University of Thessaloniki, Thessaloniki, Greece
  2. 2Foundation for Acupuncture and Biological Treatment Methods, Sabbatsbergs Hospital, Stockholm, Sweden
  3. 3British Medical Acupuncture Society, Royal London Hospital for Integrated Medicine, University College London Hospitals NHS Foundation Trust, London, UK
  4. 4Anaesthesiology and ICU Department, AHEPA Hospital, Aristotle University of Thessaloniki, Thessaloniki, Greece
  1. Correspondence to Maria Belivani, First Propaedeutic Department of Internal Medicine, AHEPA Hospital, Aristotle University of Thessaloniki, Thessaloniki 54646, Greece; maria.belivani{at}


Background Obesity is an increasing global health problem, and current methods of management are limited. Preliminary research data suggest that acupuncture may have an influence on metabolic parameters related to obesity.

Objective To determine the electroacupuncture (EA) protocol to be used in a future clinical trial examining the effect of acupuncture on metabolic parameters related to obesity and to examine whether a single EA treatment can change fasting blood glucose in obese subjects.

Methods 16 obese women aged 30–52 years with body mass index >30 kg/m2 were assigned consecutively into three groups and their fasting blood glucose was measured before and after administering a single session, lasting 30 min, of one of three EA treatment protocols. The Dorsal group received EA to dorsal segmental acupuncture points BL18–23 bilaterally (corresponding to the segmental levels innervating the pancreas); the Ear group received EA to ear points in the cavum conchae; and the Limb group received EA to points in the arms and legs (LI10–LI11, ST36–Zongping).

Results After a single session of EA there was a statistically significant decrease in fasting blood glucose in the Dorsal and Limb groups, but there was no change and even a trend towards an increase in the glucose level in the Ear group.

Conclusions The findings of this small pilot study suggest that EA to either dorsal segmental points corresponding to the pancreas or to muscle points in all four limbs may exert a beneficial effect on glucose metabolism in obese women.

Statistics from


Overweight and obesity are the fifth leading risk for global deaths. At least 2.8 million adults die each year as a result of being overweight or obese. In addition, 44% of the diabetes burden, 23% of the ischaemic heart disease burden and 4–7% of certain cancer burdens are attributable to overweight and obesity.1 Current conventional therapeutic strategies are limited.

A review of the literature on the treatment of obesity suggests that acupuncture (in different forms) exerts beneficial effects. Apart from a reduction in body weight, acupuncture seems to affect many biochemical markers of obesity such as insulin resistance, glucose and lipid metabolism, obesity-related peptides and inflammatory markers. However, further prospective clinical trials are needed to establish the effectiveness of this complementary method for obesity treatment.2

Clinical trials studying the effect of acupuncture on glucose metabolism and insulin levels have had mixed results, showing both an increase3 and a decrease after treatment.4 In one study, electroacupuncture (EA) decreased serum glucose levels through the increase of serum insulin and C-peptide levels compared with sham EA or diet restriction.3 In contrast, a recent study showed that manual acupuncture decreased insulin levels and the authors suggest that acupuncture improves insulin sensitivity and normalises insulin levels.4

The purpose of this pilot study was to determine the type of EA protocol to be used in a future clinical trial on the use of EA in the management of obesity. We chose to measure the change in fasting blood glucose as a marker for physiological activity of the intervention. We analysed the pre/post intervention change in fasting blood glucose in obese non-diabetic subjects, where the intervention was one of three different EA protocols: dorsal segmental points in the paraspinal muscle, ear points in the cavum conchae and points in muscles of all four limbs.

Materials and methods


Eligible participants were drawn from the population of consecutive patients who were referred to the Cardiovascular Risk Outpatients’ Clinic of the 1st Propaedeutic Internal Medicine Department of AHEPA University Hospital between March and September 2011. Inclusion criteria were: (1) body mass index >30 kg/m2; (2) age 25–55 years; (3) female; and (4) consent to participate. All subjects fulfilling the inclusion criteria underwent a medical history assessment, physical examination and routine laboratory tests in order to exclude patients with diabetes, hypertension, heart disease, endocrinopathy, nephropathy or those being treated with any medication.

Acupuncture treatment

We assigned the volunteers to the three treatment groups in turn, in alphabetical order of the treatments (back, ear and limb). This method of assignment was used until there were five patients in each group, after which the 16th patient was assigned by random group selection (a lottery). There were three treatment groups. The Dorsal group (5 patients) received EA in the back from T9 to L2 acupuncture points BL18, BL20, BL21 and BL23 bilaterally, corresponding to the segmental levels innervating the pancreas. The Ear group (5 patients) received EA in both ears in the Lung and Heart points. The Limb group (6 patients) received EA to points LI10, LI11, ST36 and Zongping in the arms and legs bilaterally.

Disposable stainless steel needles with guide tubes were used (Ener-Qi, Novasan, Madrid, Spain). For the ear points, the needle size was 0.16×13 mm and the needles were inserted superficially while, for the limb and dorsal points, the needle size was 0.26×40 mm or 0.30×75 mm depending on the fat tissue of the subject, and the length was chosen to permit the needles to be inserted into the underlying muscles so that EA would produce strong muscular stimulation. The needles were attached to an electric stimulator (CEFARACUS 4, Cefar Medical AB, Lund, Sweden) with a low frequency of 2 Hz with 0.1 s 80 Hz burst pulses. Intensity was adjusted to produce local muscle contractions (with similar intensity but no muscle contraction in the ears) and varied between 1.2 and 2.5 mA. The EA treatment, which lasted for 30 min, was performed by a medical doctor with 14 years training and experience in Western medical acupuncture. All subjects were asked to fast for at least 8 h before blood collection, to avoid any heavy physical exercise the day before and not to take supplementary medication. Two blood samples were taken from each subject, at the beginning and end of EA in the morning.

Outcome measurement

Forearm venous blood samples (1.5 mL) were collected by a professional nurse between 08:00 and 09:00 before breakfast. The blood samples were injected into heparinised tubes as quickly as possible and centrifuged for 3 min at 12 500 g and 4°C to obtain plasma. Plasma was spotted onto a slide which contained a reagent layer (glucose oxidase and peroxidase; YSI 7100 Multiparameter Bioanalytical System) for glucose concentrations.

Statistical analysis

As this was a pilot study it was estimated that a sample size of about five women per group would be sufficient to provide results that would answer our question. Non-parametric statistics were used as the sample size is small. The results are expressed as estimated medians and CIs. The Wilcoxon signed rank test was used to compare fasting glucose levels before and after EA. Analyses were performed using the statistical software MINITAB.


A total of 16 obese women, mean±SD age 37.9±9.5 years, mean body mass index 41±5.1 kg/m2, mean systolic blood pressure 128.5±12.8 mm Hg and mean diastolic blood pressure 80±9.1 mm Hg were recruited to the study. No differences were found between the groups in terms of these parameters.

There was a statistically significant within-group decrease in the glucose levels (p<0.05, Wilcoxon test) only in the Dorsal and Limb points groups, with no statistically significant difference between these two groups. There was no decrease in glucose levels in the Ear group. The results are shown in figures 1, 2 and table 1.

Table 1

Median change in fasting blood glucose levels in three groups before and after acupuncture and significance of within-group change using the Wilcoxon signed rank test

Figure 1

Fasting blood glucose before and after a single session of electroacupuncture for each participant (the y-axis has units in mg/dL and mmol/L).

Figure 2

Difference in fasting blood glucose after a single session of electroacupuncture for each participant (the y-axis has units in mg/dL and mmol/L).


The purpose of this pilot study was to determine the type of EA protocol to be used in a future clinical trial aiming to influence metabolic parameters related to obesity. We analysed the pre/post intervention change in fasting blood glucose in obese subjects and found a significant decrease in the glucose levels in groups given EA in dorsal segmental or limb points, while there was no decrease in the glucose level after acupuncture in ear points.

Experimental evidence suggests that EA has favourable effects on glucose metabolism. Abdominal EA in diabetic and obese rats, although inducing a sustained hypoglycaemic effect, did not influence insulin levels compared with EA in non-specific points.5 Moreover, in experimental models of diabetic rats, EA was shown to regulate blood glucose levels by increasing insulin sensitivity,6 ,7 inducing secretion of β-endorphin8 ,9 or stimulating cholinergic nerves.10 ,11 Regarding insulin resistance, EA has been reported to improve insulin sensitivity via several mechanisms (eg, lowering plasma free fatty acid levels,11 ,12 increasing plasma insulin growth factor-1 levels or stimulation of glucose transport in skeletal muscle independently of insulin).13 ,14 As already mentioned, clinical trials studying the effect of acupuncture on glucose metabolism and insulin levels have had mixed results3 ,4 but, to our knowledge, this is the first clinical study evaluating the immediate hypoglycaemic effect of EA in fasting obese humans.

The primary mechanism for the clinical effects of acupuncture is activation of somatic afferent nerves innervating the skin and muscles, which may modulate somatic and autonomic nervous system activity and endocrine and metabolic functions. It is widely accepted that there is no evidence for acupuncture point specificity and that needles can be inserted anywhere in appropriate segments.15 ,16 The physiological responses to acupuncture depend on many factors, including the number and placement of needles, the type of stimulation (manual or electrical with different frequencies) and the number and frequency of treatments. We placed the needles in back and limb muscles and stimulated with a low frequency of 2 Hz burst (eight pulses of 80 Hz within the burst) to evoke muscle twitches. We used this stimulation modality rather than needle penetration without electrical stimulation because EA parameters can be precisely characterised and the results are more or less reproducible. Low-frequency EA with repetitive muscle contractions activates physiological processes similar to those resulting from physical exercise and could thereby hypothetically influence metabolic variables.17 Peripherally, stimulation of acupuncture points in muscle tissue causes the release of a number of neuropeptides from peripheral nerve terminals into the surrounding area.18 The release of neuropeptides results in increased microcirculation.19 In addition, electrical stimulation increases skeletal muscle glucose uptake in rats.20 Through sympathetic reflexes, acupuncture at the segmental (spinal) level may modulate organs (eg, the pancreas) located in the same innervation area as the dorsal acupuncture points used in this study.21 In our study, one session of EA at either dorsal or limb points produced a significant decrease in fasting glucose levels, probably because EA induces muscle contractions that may mimic aspects of strong exercise. Another possible mechanism for the decrease in glucose in the Dorsal group is the modulation of pancreatic endocrine function through sympathetic reflexes, as described above. Acute effects of low-frequency EA in rats improved glucose tolerance and insulin sensitivity and decreased free fatty acids.11 ,12 The abovementioned mechanisms may explain the absence of a change in glucose level in the Ear group (Heart and Lung points).

Our study has some limitations that must be acknowledged. It was an uncontrolled pilot study with the aim of determining the best protocol in terms of point location for a future clinical study of obese patients. The subjects were allocated consecutively rather than randomly, but the groups were statistically similar at baseline. There was no allocation concealment but, since this was a novel study, we were experimenting to determine which protocol, if any, would be successful. It is therefore unlikely that there was any psychological influence on the part of the practitioners and, moreover, the outcome was an objective measure. In addition, we measured only glucose levels before and after the EA session, but insulin and free fatty acids could have been measured in order to determine whether our obese patients were insulin resistant and thus prove the underlying mechanism. Finally, our study population was relatively small and gender-based, so it will be of interest to perform a larger study that includes both genders and subsequently generalise our findings.

In conclusion, low-frequency EA in back and limb points appears to reduce plasma glucose levels in obese fasting patients. Larger randomised studies are needed to confirm the above findings and to evaluate the underlying mechanisms.

Summary points

  • In preparation for a clinical trial, we explored the effect of electroacupuncture (EA) at different locations on fasting blood glucose.

  • EA to the dorsum and the limbs lowered blood glucose levels.

  • Auricular EA did not lower blood glucose levels.


View Abstract


  • Contributors TL and MC conceived the project and designed the study. AH and DV advised on the design of the study. NB and CD assigned the patients and drafted parts of the paper. MB was the acupunturist and drafted the first version of the paper. All authors contributed to the analysis and interpretation of the data, refinement of the study protocol and approved the final manuscript.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics approval was obtained from the research ethical committee of the University of Thessaloniki and was carried out in accordance with the standards set out in the WMA Declaration of Helsinki.

  • 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.