Article Text

PDF

Brain areas involved in acupuncture needling sensation of de qi: a single-photon emission computed tomography (SPECT) study
  1. Jia-Rong Chen1,
  2. Gan-Long Li2,
  3. Gui-Feng Zhang2,3,
  4. Yong Huang2,
  5. Shu-Xia Wang4,
  6. Na Lu5
  1. 1First School of Clinical Medicine, Southern Medical University, Guangzhou, Guangdong Province, China
  2. 2School of Traditional Chinese Medicine, Southern Medical University, Guangzhou, Guangdong Province, China
  3. 3Xinxing Traditional Chinese Medicine School of Guangdong Province, Xinxing, Guangdong Province, China
  4. 4Department of Nuclear Medicine, Guangdong Provincial People's Hospital, Guangzhou, Guangdong Province, China
  5. 5Teaching and Research Section of Physics, Guangzhou Medical College, Guangzhou, Guangdong Province, China
  1. Correspondence to Dr Yong Huang, School of Traditional Chinese Medicine, Southern Medical University, Guangzhou 510515, Guangdong Province, China; nanfanglihuang{at}163.com

Abstract

Background De qi is a sensory response elicited by acupuncture stimulation. According to traditional Chinese medicine (TCM), de qi is essential for clinical efficacy. However, the understanding of the neurobiological basis of de qi is still limited.

Objective To investigate the relationship between brain activation and de qi by taking a single-photon emission computed tomography (SPECT) scan while applying acupuncture at TE5.

Methods A total of 24 volunteers were randomly divided into 4 groups, and received verum or sham acupuncture at true acupuncture point TE5 or a nearby sham point according to grouping. All subjects then received a 99mTc-ethylcysteinate dimer (ECD) SPECT scan.

Results All six subjects in the verum acupuncture at true acupuncture point group experienced de qi sensation; in contrast, all six subjects in the sham acupuncture at the sham point group responded with nothing other than non-sensation. Compared to the scan results from subjects who experienced non-sensation, SPECT scans from subjects with de qi sensation demonstrated significant activated points mainly located in brodmann areas 6, 8, 19, 21, 28, 33, 35, 37, 47, the parahippocampal gyrus, lentiform nucleus, claustrum and red nucleus; deactivated points were seen in brodmann areas 9 and 25.

Conclusions Verum acupuncture at true acupuncture points is more likely to elicit de qi sensation. De qi sensations mainly resulted in brain area activations, but not deactivations. These brain areas are related to the curative effect of Te5. The acupuncture needle sensations of de qi and sharp pain are associated with different patterns of activations and deactivations in the brain.

Statistics from Altmetric.com

Introduction

De qi is a sensory response elicited by acupuncture stimulation, characterised by a minimal muscle contraction around the needle accompanied by feelings of numbness, aching, spreading, heaviness or other sensations around the needle, even spreading along the meridian.1 It is thought to literally mean the arrival of vital energy from the perspective of traditional Chinese medicine (TCM),2 and it is considered as an important prerequisite for generating a therapeutic effect. De qi is a response indicating a positive effect on meridians under the stimulation of acupuncture. Takeda and Wessel found that the experience of de qi could be treated as a predictor of significant improvement.3

According to TCM, de qi is essential for clinical efficacy. It is generally believed that de qi is essential for producing acupuncture analgesia and anesthesia.4

De qi has recently drawn the attention of many scientific researchers for its important therapeutic effect.3 ,5–8 Previous studies have mainly focused on the sensation description and quantification of de qi. Questionnaires were distributed to subjects at the end of each tactile stimulation or acupuncture procedure in order to quantify the total intensity of de qi experienced, and the intensity was rated on a scale of 1–10, implying a rating of no unpleasantness to unbearable unpleasantness.9 The Massachusetts General Hospital Acupuncture Sensation Scale has been used to balance breadth and depth of sensations as well as the number of different sensations chosen by subjects.10 Other similar studies include the Subjective Acupuncture Sensation Scale (SASS),11 visual analogue scales (VAS)12 and the Southampton Needle Sensation Questionnaire (SNSQ).13

Mechanisms of de qi have been more and more frequently examined in the central nervous system. A close relationship has been found between acupuncture therapeutic effects and brain areas with corresponding functions.

Many brain regions have been found to be affected by acupuncture stimulation with elicitation of de qi. Hui et al found that attenuated activity was noted during acupuncture stimulation with elicitation of de qi in the limbic and paralimbic structures of the cortical and subcortical regions of the diencephalon, the brainstem and cerebellum.14 Hsieh et al found significant brain activation in the hypothalamus and insula with an extension to the midbrain during the elicitation of de qi at L14.15 Qiu et al found that de qi was correlated in intensity with deactivation of the angular gyrus in women, but not in men, and no statistical significance for brain activities was observed at either the angular gyrus or visual cortex (Brodmann areas (BAs)18/19) and other types of psychophysical responses such as dull pain or aching in de qi sensations.16 Dougherty et al described the change of fMRI imaging signal in the orbitofrontal cortex, insula and positron emission tomography (PET) signal changes in the orbitofrontal cortex, medial prefrontal cortex, insula, thalamus and anterior cingulated cortex, which includes brain regions within the lateral and medial pain networks.17

Though most the experiments on acupuncture were developed in a state of de qi, the understanding of the neurobiological basis of de qi is still limited, and the essence of de qi still needs to be explained. Meanwhile, most of the previous studies were tested at the LI4 and ST36 points, so research at other acupuncture points is limited.

We hypothesise that brain regions could be affected by acupuncture stimulation with elicitation of de qi, which should be different from elicitation of non-sensation or sharp pain. We further postulate that the brain regions affected by de qi sensation would vary depending on the stimulated acupuncture points, which may be connected to the therapeutic effect of the acupuncture points.

In this study, we attempted to find out the relationship between brain activation and de qi with single-photon emission computed tomography (SPECT), and we attempted to discover the discrepancy between de qi and sharp pain. Acupuncture was performed at Waiguan (TE5), a main acupuncture point that is commonly used for treating a variety of medical conditions according to TCM treatment.

Methods

Subjects

A total of 24 healthy volunteers, composed of 12 men and 12 women, were recruited from different universities in Guangzhou City. The inclusion criteria were: (1) undergraduate or postgraduate aged 21–28 years old studying a non-medicine speciality; (2) taking regular meals without excessive consumption of cigarettes, tea or coffee; normal sleep patterns and body structure; (3) right-handedness; and (4) must qualify with all the above after taking the screening test, which was conducted 3 months before the experiment. (All volunteers received the screening test by true and sham needling at the true acupuncture point as follows. Firstly, we blocked the volunteers’ vision so that the volunteers could not see what kind of needling they would receive. Secondly, a skilled acupuncturist (the same one in each case) stimulated the volunteers in the order of sham needling and true needling at the true acupuncture point. Thirdly, the volunteers were asked to report their sensation. Subjects with obvious de qi during sham needling and/or non-de qi during true needling at the true acupuncture point were rejected.5 All subjects agreed to sign informed consent before the experiment. The exclusion criteria were: (1) having pain (including dysmenorrhea) within 3 months prior to study; (2) having congenital diseases, psychiatric diseases, central/peripheral nervous system diseases, endocrine diseases, immunological diseases, or severe problems in the heart, liver and kidneys; (3) having thrombocytopenia, haemophilia or other coagulation disorders; and (4) having received acupuncture treatment within 3 months of the study.

The participant group was made up of 12 men and 12 women, with an age range of 21–28 years old (24.30±3.26), body weight of 55.60±10.23 kg and height of 168.60±11.12 cm. The study programme was approved by the local Ethics and Radiation Safety committee. All volunteers agreed to sign the informed consent forms before the experiment, and the protocols were approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou University and conducted in strict accordance with guidelines of the Administrative Regulations on Medical Institution and State Council of the People's Republic of China.18

The 24 volunteers were randomly divided into 4 groups by a random digit table. The groups were: verum acupuncture at true acupuncture point group, sham acupuncture at true acupuncture point group, verum acupuncture at sham point group, and sham acupuncture at sham point group. Each group had six volunteers.

Study design

Acupuncture point locations

TE5 is located on the forearm, 2 cun (cun is a unit of length used in TCM that refers to the width of the interphalangeal joint of the patient's thumb) above the transverse crease of the dorsum of the wrist, between the radius and ulna.19

The sham point in the experiment was located at the same level as TE5 and right on the midline between the TE and SI meridians. Both the true and sham points used in the experiment were on the right hand.

Acupuncture administration

The volunteers in the verum acupuncture at true acupuncture point group received verum acupuncture at the true acupuncture point TE5; the volunteers in the sham acupuncture at true acupuncture point group received sham acupuncture at the true acupuncture point TE5; the volunteers in the verum acupuncture at sham point group received verum acupuncture at a sham point; and those in the sham acupuncture at sham point group received sham acupuncture at a sham point. None of the volunteers knew what kind of stimulation they would receive.

Manipulation of verum acupuncture

The doctor placed the guide tube over the true/sham point, inserted the needle and then tapped the end of the needle to insert the tip of the needle. After removing the tube the needle was advanced into the skin to a depth of 15±2 mm. After the needling sensation was perceived, the doctor manually stimulated the point by rotating the needle±180° at a rate of 60 times/min for 3 min in every 5-min period. This procedure was repeated three times. The stimulation lasted for a total of 15 min.

Manipulation of sham acupuncture

The insertion procedure was the same as above. First, the blunt needle was inserted and touched the skin for 3 min, and then the needle was held for 2 min in this way; the stimulation was repeated three times and lasted for a total of 15 min.

The needles with their matched tubes used for true needling and blunt needles for sham needling were 0.30×40 mm, manufactured by Dongbang AcuPrime Co., UK.

Verum and sham acupuncture were conducted by the same doctor who had conducted the test stimulation 3 months earlier.

Sensation measures

Inquiries were made as to the needling sensation of all subjects at the end of the experiment. Subjects had to report if they noted any skin sensation, including aching, fullness, pressure, numbness, dull pain and sharp pain, during or immediately after the treatment and (if applicable) to quantify the occurrence of de qi-like sensations at the points and their intensity on a 10-fold visual analogue scale (VAS) (with 0 being no sensation and 10 being the strongest sensible sensation).

SPECT scan

All subjects received a 99mTc-ethylcysteinate dimer (ECD) SPECT scan. 99mTc-ECD25mCi was intravenously injected 10 min after the needle insertion while the scans were performed at 0.5 h after the needle insertion.

The SPECT imaging was performed with a Precedence 6 SPECT/CT from Philips (Amsterdam, The Netherlands) with the probe configured with the low energy high resolution collimator and the radius of gyration set at 13.5 cm. Data acquisition conditions were as follows: zoom=1.73, 128×128 matrix, orbitomeatal line vertical ground, 360° circular trace, 64 frames of projection image, 5.6° each frame, collection time: 20 s for each frame, 20% energy window width (peak 140 keV), more than 80 kcounts each frame of count. Conditions of image reconstruction were as follows: first, Astonish filtering was applied to deal with the data. Second, the cut-off frequency parameter was fc=1.0. Third, the iterative method was used to reconstruct the image three times, and to reconstruct horizontal, coronal and sagittal images. Thickness was set at 5.4 mm. The scan was designed based on the methods of Káplár et al and Chung et al respectively.20 ,21

Statistical analysis

The SPECT scan data were processed with the software Statistical Parametric Mapping (SPM2, http://www.fil.ion.ac.uk) and the corresponding operating platform of Matlab 6.1.22 The datasets were divided into three groups: (1) de qi, (2) sharp pain and (3) no sensations (neither de qi nor sharp pain) according to the sensations recorded at the end of experimental procedure. The slight movements of the subjects’ heads were corrected by the realignment, and then the images were normalised to the Montreal Neurological Institute space and smoothed spatially by a Gaussian kernel of 5 mm × 5 mm× 5 mm. The smoothed data were analysed using the generalised linear mode voxel by voxel. The corresponding t value of each voxel was counted by two-sample t test, and the statistical parametric mapping was determined based on the t values (p<0.01, uncorrected, K>30). The changes in the different cerebral regions under different stimulation conditions were obtained and superimposed to the standard cerebral image mode that comprehended every subject's anatomic images. The results were reported using coordinates of Talairach space and the active areas were shown in red.

Result

Frequency of de qi sensation

According to the results, when giving a score of between 3 and 6 on the VAS most subjects felt sensations of aching, fullness, pressure, numbness and dull pain, which are widely considered as the representative feelings of de qi. Therefore, we defined sensation scoring of under 2 as non-sensation, sensation scoring of between 3 and 6 as de qi, and sensation scoring of above 7 as sharp pain9 ,12 ,23 (Table 1).

All six subjects in the verum acupuncture at true acupuncture point group experienced de qi sensation; in contrast, all six subjects in the sham acupuncture at sham point group responded with nothing other than non-sensation. Subjects in the sham acupuncture at true acupuncture point group mainly experienced sharp pain besides de qi and non-sensation. Four subjects in the verum acupuncture at sham point group experienced de qi sensation, and the other two subjects experienced sharp pain (table 1, online supplementary table S1).

Table 1

Sensation measures of different groups

After the point was treated, subjects had to report if they had noted any skin sensation, including aching, fullness, pressure, numbness, dull pain and sharp pain, during or immediately after the treatment and, if applicable, to quantify the occurrence of de qi-like sensations at the points and their intensity on a 10-fold VAS (10 being the strongest sensible sensation).

According to the result, while VAS scoring between 3 and 6, most subjects felt the sensation of aching, fullness, pressure, numbness and dull pain, which were widely considered as the representative felling of de qi. Therefore, we defined sensation scoring of under 2 as non-sensation, sensation scoring of between 3 and 6 as de qi, and sensation scoring of above 7 as sharp pain.

Comparison of SPECT scans between subjects with de qi sensation and non-sensation

Compared with the scan results of subjects who experienced non-sensation, SPECT scans on subjects with de qi sensation demonstrated that significant activated points were mainly located in Brodmann areas 6, 8, 19, 21, 28, 33, 35, 37, 47, the parahippocampal gyrus, lentiform nucleus, claustrum and red nucleus (table 2, online supplementary table S2, figure 1A), and deactivated points were seen in Brodmann areas 9 and 25 (table 2, online supplementary table S3, figure 1B).

Table 2

Single-photon emission computed tomography (SPECT) signal contrast in brain: de qi versus non-sensation and de qi versus sharp pain

Figure 1

(A) The activated brain areas of needling with de qi versus non-sensation and (B) the deactivated brain areas of needling with de qi versus non-sensation. Subjects with de qi sensation demonstrated that significant activated points are mainly located in Brodmann areas 6, 8, 19, 21, 28, 33, 35, 37, 47, parahippocampal gyrus, lentiform nucleus, claustrum and red nucleus, and deactivated points were shown in Brodmann areas 9 and 25.

Comparison of SPECT scans between subjects with de qi sensation and sharp pain

Compared with the scan results of subjects who experienced sharp pain, SPECT scans from subjects with de qi sensation showed that there were significant activated points in Brodmann areas 2, 6, 8, 9, 11, 25, 37, 47, the lentiform nucleus, caudate body, thalamus and ventral anterior nucleus (table 2, online supplementary table S4, figure 2A), and deactivated points were seen in Brodmann areas 5, 6, 7, 10, 11, 19, 31 and 39 (table 2, online supplementary table S5, figure 2B).

Figure 2

(A) The activated brain areas of needling with de qi versus sharp pain and (B) the deactivated brain areas of needling with de qi versus sharp pain. Subjects with de qi sensation exhibited that there were significant activated points in Brodmann areas 2, 6, 8, 9, 11, 25, 37, 47, lentiform nucleus, caudate body, thalamus and ventral anterior nucleus, and deactivated points were shown in Brodmann areas 5, 6, 7, 10, 11, 19, 31 and 39.

Discussion

De qi sensation is obviously more likely to be elicited with verum acupuncture. All the subjects in the verum acupuncture at true acupuncture point group responded with de qi sensation. Furthermore, four of the six subjects in the verum acupuncture at sham point group experienced de qi sensation, while only one subject who had sham acupuncture experienced de qi. Differences between verum acupuncture and sham acupuncture have been found in some other studies, 24–26 which also found subjects were more likely to experience de qi with verum acupuncture.

Most previous studies on de qi sensation compared the brain imaging between true and sham acupuncture point needling, or just compared verum acupuncture and sham acupuncture, a non-penetrating, cutaneous stimulation.26 ,27 These comparison may lack rationality, since the elicitation of de qi may also exist while stimulating at sham points or through sham acupuncture (table 1, online supplementary table S1). Thus, we redistributed the subjects when analysing the SPECT imaging data into one of three groups (de qi group, sharp pain group and non-sensation group) based on the sensation that subjects actually felt but not according to the mode of stimulation they received in the study.

According to our study, the activated areas between subjects with de qi sensation and non-sensation can be divided into the following groups according to their functions: (1) cerebral areas for motor function, BA 6, 8, 47, red nucleus, lentiform nucleus, which govern the planning and execution of volitional movement and the control of saccadic eye movements; (2) cerebral areas connected to the eyes, BA 8, 19, 21, 37, parahippocampal gyrus, claustrum, which process visual information and the control saccadic eye movements; (3) cerebral areas connected to olfactory sensation, BA 28, which are involved in olfaction and hippocampal processing; and (4) cerebral areas connected to pain and emotion, BA 33, 35, which are involved in emotional processing and the affective dimensions of pain.

The deactivated cerebral areas represent management of mental activities. BA 9 and 25 govern executive functions and control emotional processing, facial expressions and affective dimensions of pain.

In other studies, no significant activations were reported for the de qi group alone, and the acupuncture needle sensation of de qi is associated with different patterns of activations and deactivations in the brain. Significant increase of blood flow in the hypothalamus and insula resulting from the elicitation of de qi at LI4 was shown via positron emission tomography by Hsieh et al.15 However, more studies found that deactivations rather than activations in the brain regions were shown for de qi sensation. Asghar et al found significant deactivations rather than activations occurred for the de qi group with obvious blood oxygen level-dependent signals via brain fMRI when stimulating LI4. However, a mixture of activations and deactivations were shown for sharp pain group.6 Hui et al reported a predominance of signal decreases in various limbic/subcortical structures including the insula, hippocampus, amygdale, thalamus, posterior cingulate gyrus and cerebellum during acupuncture needling at ST36 with de qi.14 Later research described clusters of decreased activity in limbic and paralimbic regions including the medial prefrontal, medial parietal and medial temporal lobes, along with increased activity in the sensorimotor cortices and select paralimbic structures.28 However, these studies were focused on de qi at LI4 and ST36 with fMRI, and there are limited studies on other acupuncture points.

Differences may be explained by variations in the psychophysical methods used to classify scanning datasets into de qi group, or use of different acupuncture points, or they may have arisen from differences in the statistical methodology and level of thresholding used.

Recently, limited evidence has defined whether the therapeutic effect of acupuncture was elicited through activating or deactivating the cerebrum. We argue that different acupuncture points may have a differing impact on the cerebrum while applying the acupuncture, and activated and deactivated impacts to the brain regions are patterns of manifestation related to the therapeutic effect during the elicitation of de qi.

De qi sensation is generally defined as dull pain, which is described as being completely different from the sensation of sharp pain, which is considered to be unrelated and noxious to treatment. Hence we hypothesise different activation sites in brain regions could be detected and interpreted through the elicitation of pain in the application of acupuncture.

Attention is especially drawn to the parietal lobe (including angular gyrus, precuneus, inferior parietal lobule), the frontal lobe (including middle frontal gyrus, inferior frontal gyrus, paracentral lobule and precentral gyrus) and the cingulate gyrus/posterior cingulate, where brain activation was shown only or mainly through the comparison of the scan images under de qi versus sharp pain conditions (table 2).

With the use of magnetic resonance imaging and positron emission tomography, brain regions related to pain have been widely researched. Recently, a number of neuroimaging studies on pain processing have mainly demonstrated brain regions related to pain were primarily on the anterior cingulate, secondary somatosensory and primary somatosensory cortices.29 Interestingly, these brain areas did not play a critical role in the elicitation of de qi or sharp pain according to the results of our study.

Failure to observe activation of these brain areas, which may relate to pain as reported in some other research29 (mainly including the anterior cingulate, secondary somatosensory and primary somatosensory cortices), could be considered as an explanation; previous research has argued that even actual noxious stimuli yield inconsistent results in the neuroimaging literature, and interpretations may vary due to the different tasks used across experiments in terms of the attention required from the subjects.30

Different tasks used across these experiments vary in terms of the type and the focus of the attention required from the subjects, which may affect the imaging results. Bushnell et al argued that the somatosensory cortex (SI) would be involved when subjects are required to make a fine discriminating judgment about the location or the type of pain.31 In our study, sensation of needling was reviewed at the end of the experiment, and the attention of the participants was not diverted from the process. This may be one interpretation of this inconsistency.

Conclusions

Verum acupuncture at a true acupuncture point is much more likely to elicit de qi sensation. De qi sensations mainly resulted in brain activations, but not deactivations. These brain areas are related to the theraputic effect of TE5. The acupuncture needle sensations of de qi and sharp pain are associated with different patterns of activations and deactivations in the brain.

Summary points

  • Computerised tomography was performed on volunteers with and without de qi.

  • Brain regions that were activated were clearly identified.

References

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:

Footnotes

  • Contributors YH: conception and design; G-LL, G-FZ, TC: acupuncture and experiment operation; S-XW, NL, J-RC: acquisition of data or analysis and interpretation of data; J-RC: drafting the article; J-RC, YH: final approval of the version published.

  • Funding This work was supported by National 973 Programme of China, grant number (No. 2006CB504505, No. 2012GB518504).

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou University of Chinese Medicine and registered at http://www.chictr.org (registration no.: ChiCTR-ONRC-08000255).

  • 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.