Article Text


Differential effects of variable frequencies of manual acupuncture at ST36 in rats with atropine-induced inhibition of gastric motility
  1. Li-Li Gao1,
  2. Yi Guo1,
  3. Tao Sha1,
  4. Yang-Yang Liu1,
  5. Jia-Bei Tang2,
  6. Fang Yuan3,
  7. Tao Zhou4,
  8. Shou-Hai Hong1,
  9. Dong Ming2
  1. 1College of Acupuncture and Massage, Tianjin University of Traditional Chinese Medicine, Tianjin, China
  2. 2School of Precision Instruments & Optoelectronics Engineering, Tianjin University, Tianjin, China
  3. 3Acupuncture Department, Shenzhen Traditional Chinese Medicine Hospital, Shenzhen, China
  4. 4Department of Physiology, College of Integrated Traditional and Western Medicine, Tianjin University of Traditional Chinese Medicine, Tianjin, China
  1. Correspondence to Yang-Yang Liu, College of Acupuncture and Moxibustion, Tianjin University of Traditional Chinese Medicine, No. 312, Anshan West Road, Nankai District, Tianjin 300193, China; lyy02211{at}


Objective The ‘intensity-response’ relationship between acupuncture stimulation and therapeutic effect is currently the focus of much research interest. The same needling manipulation with different frequencies can generate differential levels of stimulus. This study aimed to examine the effects on gastric motility induced by four twirling frequencies based on relatively standardised manual acupuncture (MA) manipulations.

Methods Twirling manipulations at 1, 2, 3, and 4 Hz were practised before the experiments by a single operator using an MA parameter measurement device and stability was evaluated through time-frequency analysis. Forty-eight Sprague-Dawley rats were randomly divided into six groups (n=8 each): Control, Model, Model+MA (1, 2, 3, and 4 Hz). Rats in the five Model groups received injections of atropine into the tail vein to inhibit gastric motility, which was continuously recorded by a balloon in the gastric antrum. Rats in the four Model+MA groups received MA at 1, 2, 3 and 4 Hz, respectively, for 70 s and needles were retained for a further 5 min.

Results The amplitude of waveforms produced by the four twirling frequencies was relatively consistent and reproducible. The gastric motility amplitude in all groups decreased after modelling (injections of atropine) (p<0.01). Twirling manipulation at 1, 2, and 3 Hz (but not 4 Hz) increased gastric motility amplitude (p<0.05). The increase in gastric motility amplitude induced by MA at 2 Hz was greater than for all other frequencies (p<0.05).

Conclusions Acupuncture at ST36 helped recover gastric motility amplitude in rats with atropine-induced gastric inhibition and the effects induced by 1–3 Hz frequency were greater than those induced by 4 Hz.


Statistics from


Acupuncture is a traditional Chinese treatment that is used in 183 of 202 countries recently surveyed.1 The therapeutic effectiveness of acupuncture, which is the main focus of many scientific studies, can be influenced by multiple factors in Traditional Chinese Medicine (TCM), such as the selection of acupuncture points,2 the de qi sensation,3 and different acupuncture modalities (eg, electroacupuncture, acupressure, and transcutaneous electrical acupoint stimulation).4 Among all the acupuncture modalities, manual acupuncture (MA) is most frequently used in clinical practice.5 Traditional MA includes several different forms of manipulation, such as lifting-thrusting, twirling, pressing, scraping, plucking, flying, shaking, and trembling.6 Different manipulation techniques may induce different therapeutic effects.7 Furthermore, the same type of needling manipulation but at different frequency, intensity, and/or duration can also generate variable levels of stimulus,8 which may lead to differential therapeutic effects. However, this assertion is largely based on opinion and lacks verification through sufficiently standardised scientific experiments. Previous studies have been performed to determine the relationship between different MA manipulations and their therapeutic effects.7 ,9 However, MA manipulation is less reproducible than electroacupuncture, which uses standardised stimulation parameters. Thus, whether the discrepancy in the therapeutic effects of MA manipulation is due to its instability or different manipulations remains unclear. However, it is possible to examine MA stimulation parameters quantitatively using the ATP-II acupuncture parameter analyser, a device that may also help improve the reproducibility of acupuncturists’ needling manipulations.10

Acupuncture is indicated for several gastrointestinal diseases.11 Numerous clinical trials and animal studies support the efficacy of acupuncture in treating gastrointestinal motility disorders and ST36 is one of the most commonly used acupuncture points.12 Previous studies have shown that MA at ST36 can effectively regulate gastric motility in rats13 and stimulate gastric motility in a rat model of inhibited motility.14 ,15

The aim of the present study was to investigate the effects on gastric motility induced by MA at ST36 at different frequencies in rats with inhibited gastric motility.


This study was conducted in two stages. Firstly, the acupuncturist practised different frequencies of twirling manipulation at 1, 2, 3, and 4 Hz, guided by the ATP-II acupuncture manipulation parameter analyser. Manipulation waveforms were recorded and subjected to time-frequency analysis to evaluate manipulation stability. Secondly, a rat model of inhibited gastric motility was created by injection of atropine into the tail vein. Gastric motility was continuously recorded via a balloon in the gastric antrum and the amplitude and frequency of gastric motility waves were recorded.

MA manipulation recording

An ATP-II device (Shanghai University of TCM) was used to measure various MA manipulation parameters through electrical resistance-sensing technology. When a needle is inserted into the pinhole and moved, the silicone rubber inside produces friction that drives the slide-arm of the sensor. The resultant change in electrical resistance is transformed into voltage signals, which are amplified by single-chip and operational amplifiers and converted into digital signals that are processed by a computer. MA manipulation waveforms appear in real-time on the computer screen.

A single acupuncturist practised twirling manipulations at 1, 2, 3, and 4 Hz using the ATP-II acupuncture manipulation parameter analyser. During the practising phase he regulated his manipulations according to the waveforms shown on the computer screen until the waveforms became reproducible. Thereafter, the waveforms of each specific frequency (guided by the use of a metronome) were recorded 10 times (for a total of 70 s each time) by the same acupuncturist. During the testing phase, the acupuncturist was prevented from looking at the MA manipulation waveforms on the screen, which were recorded and analysed to assess reproducibility objectively.

Data processing and analysis

The data collected by the ATP-II device were processed using MATLAB software (V.2014a) based on the short-time Fourier transform method. The peak points of amplitude on the time-frequency spectrum for each moment were obtained. The amplitude and frequency of each peak point was recorded as the raw data. Finally, frequency scatter and amplitude scatter diagrams were created, with the frequency and amplitude value of the first moment (t[n]) as the abscissa, and the next moment (t[n+1]) as the ordinate. Given that the use of the metronome served to stabilise the rhythm, assessment was focused on the amplitude.

Experimental animals

Forty-eight healthy male Sprague-Dawley rats weighing 210–250 g were randomly divided into six groups: Control, Model, Model+MA (1 Hz), Model+MA (2 Hz), Model+MA (3 Hz), and Model+MA (4 Hz) (n=8 rats each). All rats were provided by the Experimental Animal Center of Academy of Chinese Military Medical Sciences. Animals were housed in polypropylene cages and kept in a controlled environment (room temperature 23±1°C, relative humidity 55±5%). Rats were acclimatised for 7 days under a 12 h light–dark cycle with full and free access to food and water before the experiment. This study was carried out in accordance with the Guidance Suggestions for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of China. All experimental procedures were approved by the Animal Ethics Committee of Tianjin University of Traditional Chinese Medicine and were conducted in a manner that minimised the number of animals used and the number of procedures per animal. A sample size calculation was performed based on the results of a pilot study using gastric motility amplitude as the primary outcome measure. Based on mean and SDs of 0.31 to 0.38 and 0.07 to 0.16, respectively, seven rats per group were needed to detect differences between group at a significance level of p<0.05 and power of 90%. Group sizes were further inflated to eight rats to account for a 10% anticipated mortality rate.

Gastric motility recording

Animals were fasted overnight with free access to water and anaesthetised using an intraperitoneal injection of 20% urethane (1.5 g/kg). Gastric motility was recorded by inserting a small balloon via a duodenal incision into the pyloric area as previously described.2 ,16 The abdomen was shaved before preparing the skin and making a 3 cm longitudinal incision along the medioventral line. Then, a small longitudinal incision (3 mm) was made in the duodenum, about 5 mm from the pylorus. Through this incision, a small balloon (5 mm in diameter) made of an elastic rubber condom was inserted into the gastric antrum. The balloon was set on one end of a thin polyethylene tube (inner diameter 2 mm, outer diameter 2.5 mm) and kept in position by tying the tube to the duodenum. Afterwards, the incision was covered with warm gauze. The balloon was filled with 0.30–0.35 mL water at 37°C to keep the pressure inside the stomach around 80–120 mm Hg. The other end of the polyethylene tube was connected via a T coupling to a transducer and physiological signal acquisition system (Powerlab 8/35, AD Instruments Co, Australia) and Chart V.7.3 software. The sampling frequency was set at 1000 Hz. The rectal temperature was kept constant at around 37°C using a feedback-controlled heating blanket.

Recording was continued for 30 min to allow gastric motility waves to stabilise. After normal recordings were established (baseline), the rats in the Model and four Model+MA groups underwent modelling of inhibited gastric motility by slow and steady injection of 0.01% atropine (0.25 mL/100 g) into the tail vein, as previously described.15 In the Control group, 0.9% normal saline (0.25 mL/100 g) was injected at the same rate. In the four Model+MA groups, needles were inserted 10 min after atropine was injected, and needling manipulations were performed for 70 s. Needles remained in situ for 5 min before being withdrawn. Gastric motility recording was continued for a further 30 min, as shown in figure 1. The same observations were made in the Control and Model groups (but without MA treatment). Relatively stable gastric motility waves over a 5 min period were extracted during the normal stage (baseline), after modelling, during needle retention, and 10, 20, and 30 min after withdrawal of the needles. Two 2 min recording periods were selected from the total 5 min section and the average amplitude and frequency values of the two last 2 min waves were taken as the raw data. The average gastric motility amplitude and frequency values during the 70 s needle manipulation period were calculated.

Figure 1

Timeline of gastric motility recording ± manual acupuncture treatment at ST36.

MA manipulation method

The left ST36 acupuncture point was selected for this study. ST36 in the rat is located at the lateral side of the posterior knee, 5 mm below the fibular head. The acupuncture point was shaved before insertion of a sterile single-use stainless steel needle (15 mm in length and 0.3 mm in diameter; HuaTuo Acupuncture Needle Factory, Suzhou, China). A twirling manipulation 180° clockwise and anticlockwise was steadily performed at 1, 2, 3, or 4 Hz at a constant depth of 5 mm for 70 s. All manipulations were performed by the same acupuncturist, using a metronome to stabilise the rhythm.

Statistical analysis

Data were statistically analysed using SPSS V.16.0 software and expressed as mean±SD. The investigators performing the data analysis were kept blind to treatment allocation. Within each group, the seven different time points were compared by repeated measures analysis of variance (ANOVA). Between group comparisons were made using one-way ANOVA followed by post-hoc test of least significant difference. Differences were considered to be significant at the value of p<0.05.


Validation of method

The scatter diagrams shown in figure 2 include 10 sets of data per one specific manipulation. The abscissa is the frequency and amplitude values of the previous moment (t[n]), and the ordinate is the next moment (t[n+1]).

Figure 2

Scatter diagrams of amplitude (A) and frequency (B) of the waveforms produced by four different twirling manipulations (1–4 Hz). Each includes 10 sets of data per one specific manipulation. The abscissa is the frequency and amplitude values of the previous moment (t[n]), and the ordinate is the next moment (t[n+1]).

As shown in figure 2A, amplitude values of the MA manipulation waveforms in four frequencies were closely distributed along the 45° diagonal for every moment, indicating that the amplitude values of the previous moment and the next moment were almost identical, without any major change. The twirling manipulation amplitude values of 1, 2, 3, and 4 Hz were mostly concentrated at 0.05–0.10 V for every moment, demonstrating that the twirling manipulation amplitudes of the different frequencies were reproducible and did not fluctuate heavily. Similarly, as shown in figure 2B, the frequency values of the four twirling manipulations were concentrated at 1, 2, 3, and 4 Hz for each moment, indicating a reasonable degree of reproducibility.

Gastric motility

As shown in figure 3, the frequency of gastric motility within each group did not significantly change over time (p>0.05, repeated measures ANOVA) and there were no significant differences between the groups at any stage (p>0.05, one way ANOVA). Notably, there was no statistically significant difference in gastric motility frequency after atropine injection in the Model group when compared with the Control group.

Figure 3

Recordings of gastric motility frequency (measured via a balloon in the gastric antrum) at seven time points in healthy rats (Control group, n=8) and rats injected with atropine to inhibit gastric motility (n=40) and left untreated (Model group, n=8) or treated with ST36 manual acupuncture at 1, 2, 3 or 4 Hz (four Model+MA groups, n=8 each).

Figure 4 shows the influence of the twirling MA manipulation at four different frequencies (1, 2, 3, and 4 Hz) on the amplitude of gastric motility. The amplitude in the Control group did not significantly change over time; however, in the Model and four Model+MA groups, amplitude of gastric motility decreased after injection of atropine (all p<0.01). No baseline differences in amplitude were found between the six groups, and there were no significant differences in amplitude after atropine injection (but prior to MA treatment) between the Model and Model+MA groups.

Figure 4

Recordings of gastric motility amplitude (measured via a balloon in the gastric antrum) at seven time points in healthy rats (Control group, n=8) and rats injected with atropine to inhibit gastric motility (n=40) and left untreated (Model group, n=8) or treated with ST36 manual acupuncture at 1, 2, 3 or 4 Hz (four Model+MA groups, n=8 each). p<0.01 compared with amplitude at baseline within same group; *p<0.05 compared with amplitude after modelling within the same group; p<0.05 compared with amplitude when twirling the needles within Model+MA (2 Hz) group.

In the Model+MA groups treated at 1, 2, and 3 Hz (not 4 Hz), gastric motility amplitude following the twirling MA manipulation significantly increased compared with that observed immediately after atropine injection within each group (p<0.05). Compared with 0.26±0.07 mm Hg after modelling, amplitudes in the Model+MA (1 Hz) group tended to be higher when twirling the needles (0.32±0.13 mm Hg, p=0.07) then increased to 0.32±0.09 mm Hg during needle retention (p=0.003), to 0.31±0.08 mm Hg at 10 min (p=0.001), 0.33±0.07 mm Hg at 20 min (p=0.011), and 0.35±0.07 mm Hg at 30 min (p=0.012) after needle withdrawal, respectively. Amplitudes in the Model+MA (2 Hz) group increased from 0.32±0.08 mm Hg after modelling to 0.52±0.23 mm Hg when twirling the needles (p=0.014) to 0.44±0.14 mm Hg during needle retention (p=0.009), to 0.40±0.07 mm Hg at 10 min (p=0.008), 0.42±0.12 mm Hg at 20 min (p=0.041), and 0.44±0.12 mm Hg at 30 min (p=0.015) after needle withdrawal, respectively. Compared to 0.32±0.13 mm Hg after modelling, amplitudes in the Model+MA (3 Hz) did not differ significantly during needle twirling (0.35±0.13 mm Hg, p=0.327), but thereafter increased to 0.38±0.13 mm Hg during retention of the needles (p=0.029), to 0.39±0.14 mm Hg at 10 min (p=0.048), 0.40±0.13 mm Hg at 20 min (p=0.009), and 0.44±0.14 mm Hg at 30 min (p=0.003) after withdrawal of the needles, respectively. By contrast, MA at 4 Hz did not significantly alter the amplitude when compared to values immediately after modelling.

Furthermore, the gastric motility amplitude was significantly higher in the Model+MA (2 Hz) group than in the Model and Model+MA (1, 3, and 4 Hz) groups when twirling the needles (p=0.003, 0.010, 0.026, and 0.014, respectively).


The relationship between stimulus quantity and acupuncture effect is currently the focus of scientific research both in traditional Chinese and Western medical acupuncture. The aim of the present experiment was to study objectively the relationship between MA frequency and effects on gastric motility using a relatively standardised MA manipulation. We employed the ATP-II acupuncture manipulation parameter analyser to evaluate the reproducibility of twirling manipulations at different frequencies. The amplitude of the waveforms produced by the four twirling frequencies was reproducible, confirming that the variation between the four MA manipulations was purely one of frequency (and not amplitude). The results also indicate that training of acupuncturists to be more consistent in the amplitude of their needling stimulation is feasible using the ATP-II analyser.

We also showed that gastric motility amplitude can be enhanced by MA at ST36 following inhibition by injection of atropine in rats. Amplitude decreased significantly after atropine injection, although the frequency of gastric motility did not change significantly after modelling, which is in agreement with a previous study.17 A decrease in either the frequency or amplitude (or both) can contribute to the inhibition of gastric motility.15

It has been shown that acupuncture at ST36 exerts a regulatory influence on the stomach, and the nervous system plays an essential role.18 A series of studies on somato-autonomic reflexes have been performed focusing on gastric function and motility, which suggest that the excitatory effects are mediated via parasympathetic efferents.19 MA at ST36 may stimulate gastric motility via somato-parasympathetic reflexes, wherein the afferent nerve pathway is composed of the cutaneous and muscle afferent nerves at ST36, the efferent nerve pathway is the gastric vagal efferent nerve, and the reflex centre is located in the brain.16 ,20 The stimulatory effect of MA at ST36 may also be mediated via opioid pathways, and the modulation of such pathways may contribute to the longer-lasting effects of acupuncture on gastric motility.13

A previous study has suggested that there may be an ‘intensity-response’ relationship between electroacupuncture stimulation and effects on gastric motility,21 and that the differential activation of somatic afferent fibres may be the underlying peripheral neural mechanism. In the present study, an ‘intensity-response’ relationship between different twirling manipulations and the gastric motility effect was also shown. Previous studies have shown that different stimulation intensities of MA manipulation can activate different types of peripheral afferent fibres22 and that the latency period varies between nerves. This may lead to a discrepancy in nerve conduction velocity23 resulting in different therapeutic effects on gastric motility via somato-parasympathetic reflexes. In addition, the release of opioid peptides evoked by electroacupuncture is frequency-dependent, as demonstrated by Han.24 As the stimulatory effects of MA can also be mediated via opioid pathways, the discrepancy in the therapeutic effects on gastric motility with different frequencies of twirling manipulation at ST36 may be caused by activation of different somatic afferent fibres and/or release of different types of opioid peptide.

This study found that twirling manipulations with frequencies of 1, 2, and 3 Hz had better therapeutic effects than a frequency of 4 Hz on the recovery of the gastric motility amplitude. Among the four frequencies, 2 Hz was the most effective, and 4 Hz was the least. Therefore, as the frequency of twirling manipulation increases, the therapeutic effects appear to increase at first and then decrease. This may reflect the fact that the nervous system displays characteristics of adaptability and saturability to MA stimulation.25 Given that neuroelectrical signal transmission is complex, nerve discharge frequency is unlikely to be directly proportionate to MA manipulation frequency. Thus, when acupuncture stimulation reaches a certain frequency limit, nerve discharge frequency may no longer increase.

Our study has several limitations. As animals were the experimental subjects, it is unknown whether these results can be applied to humans. Therefore, clinical trials need to be performed to prove extrapolability. Furthermore, the neural mechanism underlying the differential effects of different frequencies of twirling MA manipulation is still not fully explained and needs further exploration.

In conclusion, acupuncture at ST36 may be helpful in recovering gastric motility amplitude following inhibition in atropine-injected rats, and the therapeutic effects induced by 1–3 Hz frequency of twirling manipulation appear to be better than 4 Hz.


The authors would like to thank BAI Yang for her valuable help in the implementation of the study.


View Abstract


  • L-LG, YG and TS contributed equally.

  • Contributors Y-YL and YG made substantial contributions to conception and design; L-LG, TS, Y-YL, J-BT made contributions to acquisition, analysis and interpretation of data; L-LG and TS drafted the article; FY, TZ, S-HH and DM gave important intellectual advice on this article.

  • Funding This work was supported by the National Natural Science Foundation of China (nos. 81001551 and 81330088).

  • Competing interests None declared.

  • Ethics approval This study was prospectively approved by the Animal Ethics Committee of Tianjin University of Traditional Chinese Medicine (reference number TCM-LAE2011004) and was conducted in accordance with local and national guidelines for animal welfare equivalent to the National Research Council ‘Guide for the Care and Use of Laboratory Animals’.

  • 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.