Article Text

other Versions


Benefits of acupuncture for diabetic gastroparesis: a comparative preliminary study
  1. Netanella Danielli Miller1,
  2. Elad Schiff2,
  3. Eran Ben-Arye3,4,
  4. Joelle Singer5,
  5. Tsachi Tsadok Perets6,7,
  6. Shlomit Flaut6,7,
  7. Nadav Sahar6,7,
  8. Yaron Niv6,7,
  9. Ram Dickman6,7
  1. 1The Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
  2. 2Department of Internal Medicine, Bnai Zion Medical Center, Haifa, Israel
  3. 3Integrative Oncology Program, Clalit Health Services, The Oncology Service and Lin Medical Center, Haifa, Israel
  4. 4Complementary and Traditional Medicine Unit, The Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
  5. 5Department of Endocrinology, Rabin Medical Center, Petach Tikva, Israel
  6. 6Department of Gastroenterology, Beilinson Hospital, Rabin Medical Center, Petach Tikva, Israel
  7. 7Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
  1. Correspondence to Dr Ram Dickman, Department of Gastroenterology, Rabin Medical Center, Beilinson Campus, Petach Tikva 49100, Israel; dickmanr1{at}


Objectives The aim of this preliminary study was to compare the effectiveness of domperidone and acupuncture for the management of diabetic gastroparesis.

Methods This was a preliminary, prospective non-randomised, unblinded case-crossover study conducted in patients with longstanding, uncontrolled diabetes mellitus and gastroparesis. All patients received domperidone (20 mg four times a day) for 12 weeks, followed by a 2–3 week washout period, and then biweekly acupuncture treatments for 8 weeks. Gastric emptying rate, glucose and glycated haemoglobin (HbA1C) levels were measured at start and end of each treatment period. At each of these timepoints patients completed the Gastroparesis Cardinal Symptom Index (GCSI), the Satisfaction with Life Scale (SWLS), and the Short-Form 36 Health Survey Update (SF-36).

Results The trial was curtailed after only eight participants could be recruited in 3 years. The mean age of patients was 57.1±9.9 years, the male:female ratio was 1:7 and mean body mass index (kg/m2) was 25.2±1.2. There was no change in any of the outcome parameters after treatment with domperidone. Acupuncture was associated with a decrease in scores for almost all cardinal symptoms of the GCSI, as well as in increased total score on the SWLS (p=0.002) and the social functioning domain of the SF-36 (p=0.054). Acupuncture did not lead to an improvement in gastric emptying, or glucose control from baseline.

Conclusions Acupuncture treatment may lead to symptomatic improvement in patients with diabetic gastroparesis. Within the limitations of this preliminary, non-randomised and unblinded study, it appears that this effect may be due to non-specific mechanisms.

  • Acupuncture
  • Diabetes
  • Internal Medicine

Statistics from


Gastroparesis is characterised by an abnormal rate of gastric emptying typically associated with nausea/vomiting, postprandial fullness, early satiety and bloating.1 ,2 Diabetes mellitus is one of the most common causes of gastroparesis (29%). The underlying mechanisms of diabetic gastroparesis are autonomic neuropathy, enteric neuropathy, abnormalities of the interstitial cells of Cajal, poor glucose control (manifested by acute fluctuations), use of incretin-based medications and psychosomatic factors.2 Most affected patients have longstanding diabetes, usually complicated by retinopathy, neuropathy and nephropathy.2 Gender may also be important given that females have a slower solid and liquid gastric emptying rate than males,3 ,4 possibly due to differences in oestrogen levels.3 Diabetes-induced gastroparesis may, in turn, further impair glucose control and lead to severe nutritional deficits, with high morbidity, poor quality of life and even death.2 ,5

The standard technique for diagnosing delayed gastric emptying is gastric scintigraphy.6 Our previous study of patients who were dyspeptic and healthy controls showed that the continuous breath test is a promising alternative tool, yielding comparable qualitative measurements to gastric scintigraphy.7 Early diagnosis of diabetic gastroparesis is important because in mild to moderate cases strict glucose control and dietary modifications frequently result in clinical improvement.2 ,6

When conservative measures are insufficient, a medical regimen of prokinetic and antiemetic drugs and pain relievers may be administered together with efforts to balance glucose levels. In Israel, the most common prokinetic agents used to treat gastroparesis are metoclopramide and domperidone. However, only a small number of studies have found these drugs to be more effective than placebo in improving the gastric emptying rate.8–10

Acupuncture has been found to affect acid secretion, gastrointestinal motility, neurohormonal levels and sensory pain thresholds.11–21 It has also been reported that acupuncture improves gastroparesis symptoms. In a study conducted in China, Xu et al21 described that electrical needle stimulation reduced dyspepsia in patients with gastroparesis. More recently, others reported that Chinese patients with type 2 diabetes mellitus treated with acupuncture showed a decrease in dyspeptic symptoms and accelerated gastric emptying of solids.22 Nevertheless, evidence from Western countries is still lacking. Data on the relationship between symptomatic improvement and anatomical or physiological changes in response to acupuncture treatment remain sparse, and there are no studies comparing acupuncture to pharmacological management. Therefore, the aim of the present preliminary trial was to compare the effects of domperidone and acupuncture on self-reported symptoms and quality of life, and on blood glucose control and gastric emptying rate in Israeli patients with longstanding, uncontrolled type 2 diabetes mellitus and gastroparesis.

Patients and methods

Study design

A preliminary prospective non-randomised, unblinded case-crossover study design was used. The study was performed in accordance with the principles of the Declaration of Helsinki and good clinical practice, and was approved by the Human Subjects Protection Program of the Rabin Medical Center. All participants provided written informed consent before enrolment and received a detailed explanation of the potential benefits and risks of acupuncture treatment.


The sample included eight selected adult patients (age >18 years) with diabetic gastroparesis. Inclusion criteria were type 2 diabetes mellitus of more than 3 years’ duration with poor glucose control in the past 6 months, as shown by a glycated haemoglobin (HbA1C) level >8% on two measurements during this period, or hypoglycaemic events three or more times per week or a major hypoglycaemic event, in addition to documented delayed gastric emptying. All eligible patients were referred by a gastroenterologist from the Gastroenterology Outpatient Clinic of the Rabin Medical Center. Exclusion criteria were presence of scleroderma, myopathy, multiple sclerosis, Parkinson’s disease, myelopathy, or severe liver, lung, renal, haematological or other comorbidity, patient fear of acupuncture, leg or limb amputation, patient sensitivity to domperidone or to acupuncture, needle phobia, treatment with narcotics, upper airway symptoms (such as hoarseness, wheezing or laryngospasm), symptoms of cold intolerance or cold extremities (to avoid the need for moxa, a traditional Chinese procedure for the treatment of cold hands and feet consisting of burning Artemisia vulgaris at acupuncture points), history of peptic ulcer disease, or gastrointestinal surgery, evidence of gastric outlet obstruction, patient unwillingness or inability to provide informed consent and patient inability to fully complete all phases of the study. Moreover, we excluded pregnant or lactating females and those who were not using a medically acceptable form of birth control.


The study consisted of a 12 week first-treatment phase during which patients were administered domperidone, followed by a 2–3 week washout period and then an 8 week second-treatment phase during which patients underwent acupuncture. Preliminary examinations before initiation of domperidone treatment included electrocardiography to ensure the presence of a regular, normal heart rate and QT intervals, and upper endoscopy to ensure the absence of gastric outlet obstruction.

Domperidone treatment

Domperidone (20 mg four times a day) was administered to all patients during the 12 week first-treatment phase. To assure compliance, patients received clear instructions on how to take the drug (2 pills of 10 mg, 15 min before a meal). They were provided with a 4-week supply, and were asked to return at the end of each month for the next 4-week supply.

Acupuncture treatment

The traditional Chinese medicine intervention and assessment used in the study were developed on the basis of our review of traditional Chinese medicine texts and a MEDLINE search.21 ,22 The treatment protocol was established according to the literature and by consensus of five Chinese medicine experts using an iterative Delphi process to assure a sufficiently consistent response.23–26 Prior to starting treatment, patients completed an Acupuncture Diagnostics Questionnaire (see online supplementary appendix I) so the protocol could be individually adapted to their specific clinical presentation. In adherence to the Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA),27 the basic protocol consistedof four acupuncture points selected primarily to ‘calm’ and ‘regulate’ the stomach from the perspective of the traditional Chinese medicine pattern diagnosis. The acupuncture points used were as follows: (1) PC6 (Neiguan), (2) ST36 (Shousanli), (3) LR14 (Qimen) and (4) CV12 (Zhangwan). Additional points (up to 3) were selected as necessary according to the symptom-questionnaire-based pattern diagnosis (see online supplementary appendix II).

A total of 16 acupuncture sessions were scheduled over 8 weeks. Treatments were administered twice weekly, usually 2 days apart by a graduate in Chinese medicine. Each graduate went through specific training for the clinical trial, and was supervised by a traditional Chinese medicine (TCM) expert. Each TCM expert was a Professor of TCM and had at least 15 years of clinical experience in Chinese medicine. Patients were informed that there would be minimal practitioner–patient interaction in order to isolate the effect of the acupuncture itself. The point locations were swabbed with 70% isopropyl alcohol according to clean-needle technique standards, and each needle (0.22 gauge) was inserted and stimulated until the patient reported de qi. Needles were inserted bilaterally, except at CV12. Length was determined by body size and varied from 30 to 40 mm; depth was guided by de qi. No reducing or tonifying methods were used (‘even’ needle stimulation). ST36 and PC6 were connected bilaterally to an electrical stimulator at a frequency of 2 Hz incontinuous waves of medium intensity (patient just aware of sensation). After the needles were inserted, the patient was left alone except for additional needle stimulation after 10 min that included an ‘even’ needle stimulation until de qi sensation was achieved. Total needle retention time was 20 min.

Tolerability and safety

Adverse events were recorded on the basis of practitioner observations, or patient responses during both phases of treatment.

Measures of treatment effectiveness

Physiological measures

To assess glucose control, fasting blood HbA1C levels were measured (normal range 3.5% to 6.0%) at baseline and at the end of each treatment phase. Gastric motility was evaluated at the same timepoints with a continuous breath test (the BreathID system; Exalenz Bioscience, Israel) that continuously samples patients’ exhaled air and calculates gastric emptying rate by integrating the sample data.7 Gastric emptying half-time (T0.5), the most practical measure of gastric emptying rate, was determined according to the percentage of labelled food in the stomach after 1, 2 and 4 h from initiation of the test. Gastric T0.5 >140 min was considered pathological.

Symptomatic measures

At each measurement timepoint, patients completed a battery of self-report questionnaires on the effects of treatment.

  1. The Gastroparesis Cardinal Symptom Index (GCSI), validated Hebrew version,28 consists of three subscales from the Patient Assessment of Upper Gastrointestinal Symptoms (PAGI-SYM) that measure cardinal symptoms of gastroparesis: nausea/vomiting (three items: nausea, retching, vomiting), postprandial fullness/early satiety (four items: stomach fullness, inability to finish a normal-sized meal, feeling excessively full after meals, loss of appetite) and bloating (two items: bloating, visibly larger belly after meals).1 Each item is rated on a scale from 0 (absent) to 5 (severe). A drop of more than 1 point for each cardinal symptom from the previous measurement is considered a valid indicator of symptom improvement. GCSI scores were assessed at baseline and at completion of each treatment phase and in addition, 6 months from completion of the acupuncture sessions.

  2. The Short-Form 36 Health Survey Update (SF-36)29 is a measure of health status that has been widely applied in patients with a range of diseases, including gastroparesis. The SF-36 is a single multi-item scale that assesses eight health domains: physical functioning, role–physical, bodily pain, general health, mental health, vitality, social functioning and role–emotional, in addition to two summary measures—physical health score and mental health score. Each domain is transformed into a scale from 0 to 100; a 5-point change in the SF-36 score represents a 5% change in health status. As a change of 5 points or more is clinically significant, the SF-36 may serve as a practical guide for the clinical approach and management of quality of life domains.

  3. The Satisfaction with Life Scale (SWLS)30 is a five-item instrument designed to measure global cognitive judgments of the subject's life. Items are rated on a scale from 1 (strongly disagree) to 7 (strongly agree). The responses to each item are averaged to yield a final score. Higher scores reflect greater satisfaction with life.

Personal interviews

Patients were interviewed by an investigator (RD) at baseline and at the end of each treatment phase. The investigator recorded demographic parameters and HbA1C levels, collected the questionnaires, assessed compliance by asking about adherence to treatment instructions, checked for adverse effects of treatment and verified complete numerical responses to the questionnaire items.

Study endpoints

The primary endpoint of the study was degree of improvement in mean scores on the GCSI (total and subscales). Additional endpoints were change in mean scores on the SF-36 (total and domains), the SWLS, and mean change in HbA1C level and gastric T0.5 from baseline.

Statistical analysis

Data analysis was performed with the SPSS V.20 (Chicago, Illinois, USA). Despite the small number of patients, it was possible to reach statistical significance (see online supplementary appendix III). A paired Student t test was used to analyse differences in the objective and subjective measures from baseline to completion of domperidone administration, and to completion of acupuncture treatment. A decline of more than 1 point in GCSI scores from completion to baseline, in each treatment phase, was also measured. For the subjective measures, an additional comparison was performed 6 months after completion of acupuncture treatment. All comparisons were two-tailed, with p<0.05 signifying statistical significance. All data are presented as mean±SD.


Demographic characteristics

The trial was curtailed after only eight participants meeting the criteria could be recruited in the 3-year study period. The study group comprised seven women and one man; the mean age was 57.1±9.9 years and mean body mass index (kg/m2) 25.2±1.2. Mean duration of diabetes was 6.6±6.5 years. None of the patients had heart, stomach or duodenal abnormalities on preliminary testing.

Cardinal symptoms

The baseline scores for the cardinal symptoms and the changes after treatment with domperidone and acupuncture are shown in table 1. Scores for all three subscales of the GCSI decreased significantly from baseline to completion of acupuncture treatment, indicating clinical improvement. Comparison of the baseline scores with scores after completion of domperidone treatment yielded no significant differences.

Table 1

Effects of treatment with domperidone and acupuncture on cardinal symptom scores

An additional comparison 6 months after completion of acupuncture treatment showed that the effects of acupuncture treatment were maintained. There were no significant changes in GCSI scores from completion of acupuncture treatment to 6 months later.

Quality of life

The baseline scores for the SF-36 and the changes after treatment with domperidone and acupuncture are shown in table 2. Between baseline and the completion of domperidone treatment, mean scores for the SF-36 domains, separately and total, either decreased or remained stable; the only exception was social functioning, which improved slightly. Similar findings were noted after completion of acupuncture treatment, which had a marginally significant effect only on social functioning (p=0.054).

Table 2

Effects of treatment with domperidone and acupuncture on patient quality of life scores

There was no significant change in SWLS scores between baseline and the completion of domperidone treatment (mean –0.55, p=0.16). Comparison of the change in scores from baseline to completion of acupuncture treatment yielded a significantly greater improvement (mean –1.13, p=0.002).

Gastric emptying

Mean gastric T0.5 was 148.5±56.7 min at baseline. An improvement was noted in approximately 50% of the patients at completion of acupuncture treatment (mean 163.14±45.7), but the group mean change from baseline did not reach statistical significance (p=0.315). There was also no significant change in gastric T0.5 between baseline and completion of domperidone treatment (139.29±28.35 min; p=0.44).

Glycaemic control

Mean HbA1c level was 9.2±2.4% at baseline. There was no significant change in HbA1C level at completion of either domperidone treatment (mean 9.4±3.2%), or acupuncture treatment (mean 9.34±2.9) (p=0.66 and p=0.69, respectively).

Tolerability and safety

The entire course of domperidone and acupuncture was completed by all patients. None reported any related adverse effects.


To the best of our knowledge, this is the first study to apply acupuncture for the treatment of diabetic gastroparesis in a Western country. The results of this preliminary study showed that domperidone was ineffective in improving either the objective or subjective parameters. Earlier studies have reported better results with domperidone than placebo, mainly in controlling nausea and vomiting.8 ,10 ,31–33 The poor outcome in our study may be at least partly explained by the small sample size and/or the female predominance. Acupuncture treatment was associated with a significant improvement in scores for almost all cardinal symptoms evaluated. This finding concurs with a previous study using an individualised approach, wherein acupuncture effectively reduced symptoms of diabetic gastroparesis.22

Several non-specific mechanisms may underlie the effect of acupuncture on diabetic gastroparesis symptoms: Response to observation and assessment (Hawthorne effect); response to the therapeutic ritual (placebo effect); and response to patient–practitioner interaction. In the present study, the Hawthorne effect was similar for acupuncture and domperidone, and thus ruled out. Regarding the placebo effect, we did not measure patient expectations for either treatment and, therefore, could not control for this variable. Nevertheless, we speculate that the ceremonial effect of acupuncture versus the routine procedure of simply taking a pill may have played a role.

We attempted to control for the effect of communication/psychotherapeutic interaction during acupuncture by limiting verbal contact between patient and practitioner. Nevertheless, the biweekly acupuncture encounters probably involved more patient attention than domperidone administration.

Acupuncture was also associated with a significant improvement in SWLS scores. However, marginally significant findings for the SF-36 were limited to the social functioning domain. The lack of improvement in the other SF-36 domains was unexpected given the known effect of poorly controlled diabetes on quality of life.5 This finding may also have been due to the small sample size and unequal sex distribution. Other studies that used a traditional individualised approach to the treatment of diabetic gastroparesis22 did not assess quality of life.

In our present study, neither domperidone nor acupuncture treatment led to an improvement in gastric emptying rate. Conversely, in a previous study of nine patients with diabetic gastroparesis treated by an individualised approach, acupuncture was found to accelerate gastric emptying.22 A recent case report in a diabetic haemodialysis patient reported similar results.34 The female predominance in our sample may explain this discrepancy, as delayed gastric emptying is known to be associated with female sex, independently of diabetes mellitus. Moreover, we evaluated gastric emptying time with a breath test, whereas Wang et al22 used gastric scintigraphy. Nonetheless, the mean T0.5 in this earlier study was not far below their 100-min cut-off differentiating normal from abnormal values. Therefore, the effect of acupuncture on this variable warrants further investigation in a larger trial. Overall, our findings suggest that acupuncture treatment may exert a positive effect on symptoms of gastroparesis, possibly via pain modulation and visceral hypersensitivity pathways, and less directly through effective augmentation of gastric motility. Symptom generation in patients with dyspepsia is also due to visceral hypersensitivity. The antinociceptive effects of acupuncture at PC6 and ST36 may be beneficial to patients with visceral hypersensitivity.35 In one study, the antinociceptive effect of acupuncture at ST36 was abolished by pretreatment with naloxone, but not by naloxone methiodide, suggesting that the effect on visceral pain is mediated via a central opioid pathway.36

Finally, neither domperidone nor acupuncture treatment improved blood HbA1C level. This finding may be attributable to the patients’ poor glycaemic control, in view of previous reports that poor glycaemic control in patients with diabetes slowed the rate of gastric emptying.2 ,9

The main limitations of the present study are the study design, being non-randomised and non-blinded, the small number of participants and uneven sex distribution. Despite an intense search of the Gastroenterology and Diabetes Clinics at our tertiary medical centre, we found it difficult to enrol additional patients due to the low cumulative incidence of gastroparesis (1%) in type 2 diabetes mellitus, and our rather stringent inclusion criteria. Thus, this study does not provide sufficient data to conclusively determine that acupuncture is a viable alternative to domperidone. However, it does demonstrate that acupuncture may be effective in improving symptoms and satisfaction with life in patients with poorly controlled diabetes mellitus and gastroparesis. The results suggest that a larger clinical study of the effects of acupuncture in this setting is feasible and worthwhile. Such a trial should also include sham acupuncture and natural history (no intervention) groups in order to control for non-specific effects and fluctuations in disease activity. It is suggested that the crossover design is then inappropriate because acupuncture can have long-lasting effects. Finally, at a late stage in revision of this work, clinical guidelines on gastroparesis were published that presented positive evidence on acupuncture, though not sufficient to recommend its use.37


Patients and researchers in the present study were compliant with the research protocol, and study outcomes were appropriate. Even with the small number of patients, acupuncture was shown to be associated with an alleviation of gastroparesis symptoms and improved general well-being. No significant difference from domperidone was found for physiological outcomes. These preliminary findings justify a larger prospective controlled trial. A cooperative study among centres may be advantageous because of the difficulty in patient recruitment.

Summary points

  • Gastroparesis is rare, and often treated with domperidone.

  • This preliminary crossover study in eight patients compared domperidone with acupuncture.

  • Symptoms improved following acupuncture, but not following domperidone administration; mean gastric emptying time did not improve with either group, though half the acupuncture patients responded.


View Abstract
  • Supplementary Data

    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:


  • NDM and ES contributed equally to this work.

  • Contributors NDM was responsible for study design, statistical analysis, helped to draft the paper and helped with final approval of the version to be submitted. ES was responsible for study conception and design, helped in recruitments of patients and drafted the paper. EB-A was responsible for study conception and design. JS participated in the design of the study and helped to draft the paper and in recruitment of patients. TTP carried out/was responsible for study design and helped in performing the breath tests and to draft the paper. YN carried out/was responsible for study design, wrote the statistical analysis plane and helped to draft the paper. SF carried out/was responsible for study design, statistical analysis, helped to draft the paper and helped with final approval of the version to be submitted. NS carried out/was responsible for study design, data analysis statistical analysis, helped to draft the paper and helped with final approval of the version to be submitted. RD carried out/was responsible for study conception and design and acquisition of data, recruitments of patients and drafted the paper and helped with final approval of the version to be submitted. He is guarantor. All authors read and approved the final manuscript.

  • Competing interests None.

  • Ethics approval Human Subjects Protection Program of the Rabin Medical Center.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

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.