THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINEVolume 12, Number 7, 2006, pp. 633–637© Mary Ann Liebert, Inc.

Acupuncture Needle Sensations Associated with De Qi: A Classification Based on Experts’ Ratings

HUGH MACPHERSON, Ph.D.,1,2 and AZIZ ASGHAR, Ph.D.3,4

ABSTRACT

Background: Many English language words have been used to describe the acupuncture needle sensationknown as de qi, words such as dull, aching, and spreading. However, there is little agreement on which actualwords are acceptable as descriptors. In experimental trials of acupuncture in which the needle sensation is animportant variable, a quantitative measure is needed to monitor and control for variability in de qi.

Methods: An established scale in the literature provides a list of 25 sensations associated with acupunctureneedling that patients might experience. An international group of acupuncture experts rated these 25 sensa-tions in two categories: those predominantly associated with de qi and those with acute pain at the site ofneedling. For each category, sensations were classified into hierarchic clusters, one for de qi and one for acutepain, and the results were presented in dendrograms.

Results: Twenty-nine international experts were invited to participate; 22 (76%) responded and 20 completedthe questionnaire. On average, they had 21 years’ experience in acupuncture practice (range 10–30 years). Sevensensations were found to be in the cluster associated with de qi: aching, dull, heavy, numb, radiating, spread-ing, and tingling. Nine sensations were found to be in a cluster associated with acute pain at the site of needling:burning, hot, hurting, pinching, pricking, sharp, shocking, stinging, and tender. The experts also raised a num-ber of issues regarding the limitations of the questionnaire used, providing useful data for future research.

Conclusions: Data from experienced acupuncturists have been analyzed to provide two separate clusters ofsensations associated with acupuncture needling: a de qi cluster and an acute pain cluster. In the design of ex-perimental trials involving acupuncture needling, researchers will find these two clusters of sensations usefulfor monitoring and controlling for variation in needle sensation.

633

BACKGROUND

One of the fundamental characteristics of acupuncture,according to acupuncturists whose practices are based

on traditional principles, is the “obtaining of qi,” more com-monly known by the pinyin de qi (pronounced duh chee).1

Patients and more experienced practitioners feel a sensationwhen the de qi arrives at an acupoint. For example, the prac-titioner may feel “a moderate sinking or tight sensation un-der the needle tip,” and the patient may feel “a sensation of

soreness, numbness, heaviness, or distension around thepoint, or even a sensation travelling to a certain place.”2

From earliest times this sensation has been seen as thera-peutically important, and without this sensation, many prac-titioners will assume that there can be no clinical benefit.1

However, there has been little agreement about the preciseEnglish language words that are best used to describe thede qi sensation.

In experimental research in which the therapeutic effectsof acupuncture are being evaluated in controlled trials, this

1Foundation for Traditional Chinese Medicine, York, UK.2Department of Health Sciences, University of York, UK.3Hull York Medical School, York, UK.4Department of Biological Sciences, University of Hull, UK.

de qi sensation is a variable that is a potential confounder,on the basis that different experiences of needle sensationmay be associated with different outcomes. For this reason,attempts have been made to develop a scale to quantify theneedle sensation. The first of such studies was conducted byVincent et al. (1989).3 They adapted the McGill Pain Ques-tionnaire4 with the help of 10 practicing acupuncturists andreduced the 78 adjectives describing various forms of painto create a new scale of 20 adjectives describing possible sen-sations that patients might feel at the site of needling. Scor-ing was on a four-point scale with “not at all,” “mildly,”“moderately,” and “severely” (scored 0, 1, 2, and 3, respec-tively). More recently, Park et al. (2002)5 modified Vincentet al.’s3 scale by adding a further five sensations, namely,boring, pinching, pressing, flickering and tender, with scor-ing on a four-point scale with “none,” “slight,” “moderate,”and “strong” (scored from 0 to 3). In terms of quantifyingde qi, both of these scales have limitations, because they mea-sure a range of sensations that include pain as well as de qi.

For experimental studies of acupuncture, there is a needto establish these two dimensions of needle sensation moreclearly, separating de qi from acute pain at the site ofneedling. Evidence of the importance of this comes fromtwo brain imaging studies by Hui and her colleagues whofound different hemodynamic responses to needling whenparticipants experienced either de qi or acute pain.6,7

In the current study, we sought to build on the researchof Vincent et al.3 and Park et al.5 With the help of interna-tional acupuncturist experts, and using quantitative methods,we aimed, for the first time, to separate the sensations as-sociated with de qi from those associated with acute pain atthe site of needling. Given that much of current acupunc-ture practice associates the de qi sensation explicitly with atherapeutic effect, this separation is essential, in that it willfacilitate the de qi sensation being controlled for in experi-mental studies.

METHODS

A sample of 29 international acupuncture experts wereidentified; those who participated are listed in our ac-knowledgements below. They had a high profile in terms oftheir authorship of textbooks, clinically oriented papers, oracupuncture research, as well as all being practicing and ex-perienced acupuncturists. We sent them a covering letter ex-plaining the purpose of the research and a questionnairewithin which we asked questions about de qi and acute painat the site of needling using the 25 sensations of Park et al’sscale.5 We also collected demographic data. We ascertainedthe importance of de qi to this group of acupuncturists byasking: “In the majority of your patients, do you normallyseek to obtain de qi when using LI-4 (Hegu)?” with optionsfor rating on a five-point scale of “Never” to “All the time,”which we scored 0–4.* We then asked two questions about

sensations that patients might experience without linkingthem to any specific acupuncture point. First, we asked:“Which of the following (25) sensations would you associ-ate with de qi?” The rating options were from “Never” (0)to “Always” (8). Second, we asked: “Which of the follow-ing (25) sensations would you associate with acute pain andnot de qi?” with the same rating options as above. And fi-nally, we provide an empty box at the end of the question-naire for comments.

Data from the questionnaires was entered into SPSS(SPSS Inc. Chicago, IL). In our analysis—separately for deqi and acute pain—we classified the sensations using a hi-erarchical clustering algorithm based on within-groups link-age and Euclidian distance. In this type of analysis, everysensation is initially considered a cluster and then the sen-sations that are quantitatively closest in terms of the expert’sratings are combined sequentially into successively largerclusters. We present the results in two dendrograms withineach of which we sought a predominant cluster of sensa-tions, with one dendrogram showing a de qi cluster and theother an acute-pain cluster.

RESULTS

Experts’ background

Of the 29 international experts who were invited to par-ticipate, 22 (76%) responded. Two (2) provided a writtenresponse but did not complete the questionnaire and 20 com-pleted and returned the questionnaire. The average age was51 (range 36–65), 18 were male, and the mean number ofyears in clinical practice was 21 (range 10–30). On the scalethat experts (n � 20) rated whether they normally sought deqi at LI-4 (Hegu), the average score was 3.6 (median 4, range2–4).

Sensation cluster data

In the cluster analysis, seven sensations were found to bein the predominant cluster associated with de qi: aching;dull; heavy; numb; radiating; spreading; and tingling (Fig.1). The median for these sensations was always greater thanfive (range 5.5–7). Nine sensations were found to be in apredominant cluster associated with acute pain at the site ofneedling: burning; hot; hurting; pinching; pricking; sharp;shocking; stinging; and tender (Fig. 2). These sensations hada median of 4 or � (range 4–8). When comparing the twodendrograms, no sensation was found to be associated withboth predominant clusters. Nine sensations were not asso-

MACPHERSON AND ASGHAR634

*One expert identified a potential ambiguity in this question,stating that” “In the majority I always seek de qi. In the minorityI never seek de qi. The strictly correct answer (“Always”) does notrepresent the truth!”

ciated with either de qi or acute pain at the site of needling:boring; electric; flickering; intense; penetrating; pressing;pulling; pulsing; and throbbing.

Experts’ comments

There was a range of comments from among the 22 re-sponding experts. Some concerns were raised includingthose associated with: (1) the potential variability of the deqi sensation between patients, for different conditions andbetween points; (2) the inextricability of different sensa-tions; (3) the subtlety of the sensations; (4) the limitationsof the terms and scoring used; and (5) the reductionist as-sumptions behind the questionnaire itself.

1. The potential variability of the de qi sensation betweenpatients for different conditions and between points—Oneexpert discussed variability among patients, stating that,in his experience, when needling older patients, the nee-dle is less likely to “grab.” He also stated that: “Acupointsvary in their ability to produce de qi.” Another respon-dent reported: “For me, de qi can be very mild (like a flyon the skin) which we use for nerve problems like sciat-ica up to very strong (like a pair of pliers) which we usemore for muscle tendon problems.” He went on to write:“For internal problems we use a moderate de qi. If it istoo strong, it turns to pain.” Others identified acupuncturepoints at which de qi was more or less likely to be felt.

2. The inextricability of different sensations—One expert“wrestled for some time with the acute pain sensationquestion, as it asked for scoring of two different attributes(‘acute pain’ and ‘not de qi’) in one compound charac-teristic. He wrote: “Some patients, for example, find mostpoints to be painful but I have no reason to think that deqi is absent.” For another expert, there were some sensa-tions that he could “equally associate with either de qi oracute pain.” Another expert reinforced this point by see-ing the threshold between strong de qi and pain as indi-vidual for each patient. Another expert wrote that know-ing only one sensation was not sufficient to know whethera patient was experiencing de qi, and that he would havewanted to ask the patient some more questions.

3. The subtlety of the sensations—One expert wrote: “Mak-ing these associations was hard. I usually think of de qi asa subtle sensation that I recognize in an intuitive way.” An-other expert wrote that the descriptors did not tally withthose given by her patients who often described more sub-tle sensations such as a “cool flowing stream,” with “sen-sations going from one point to another.” She explainedthat for her patients, de qi sensations might be felt at theother end of the same channel (meridian), on different chan-nels, or on the same point on the opposite side of the body.

4. Limitations of the terms and scoring used. An expertwrote that patients often experience “warmth,” although“the term ‘hot’ in the questionnaire seemed a little harsh.”For another, the terms “shocking” and “penetrating” were

ACUPUNCTURE NEEDLE SENSATIONS IN DE QI 635

Rescaled Distance Cluster Combine

0 5 10 15 20 25

PinchingSharpPrickingStingingBurningPulsingTenderHotHurtingFlickeringPenetratingBoringPullingShockingIntenseThrobbingElectricPressing

AchingRadiatingHeavySpreadingDullTinglingNumb

0.50.500132.521123414345

777775.57

Sensation Median

FIG. 1. Hierarchic clusters of sensations associated with de qi,based on ratings by 20 experts. Dendrogram using average link-age (between groups).

Rescaled Distance Cluster Combine

0 5 10 15 20 25

FlickeringTinglingDullNumbPullingPressingElectricHeavyIntensePenetratingBoringPulsingRadiatingSpreadingAchingThrobbing

HurtingSharpBurningStingingPinchingTenderHotShockingPricking

1201333144431224

887664574

Sensation Median

FIG. 2. Hierarchic clusters of sensations associated with acutepain, not de qi, based on ratings by 20 experts. Dendrogram usingaverage linkage (between groups).

not clear. For another respondent, the word “flickering”was assumed to mean “fasciculation.” The questionnairedid not allow the additional option of responding with“never experienced.” The sensation of de qi following achannel in the direction of the needle was identified asan important indicator, yet the questionnaire did not coverthis beyond the terms “spreading” and “radiating.”

5. The reductionist assumptions behind the questionnaireitself. Two acupuncturists responded to the questionnairebut without rating the sensations, because they felt thatde qi did not lend itself to quantification. To quote one:“I find the parameters of this questionnaire to be sim-plistic and reductionist, very inadequately representingmy approach to acupuncture needling . . . It would belike asking a pianist how he gets a crescendo, asking ifthis is by hitting the piano harder.” This concern led therespondent to write: “Research based on this approachmight not be adequately representative of the actual prac-tice of acupuncture.”

DISCUSSION

This study is the first to use quantitative methods for an-alyzing experts’ ratings to determine the sensation they as-sociate with de qi separately from the sensation they asso-ciate with acute pain at the site of needling. A cluster ofseven sensations was found to be associated with de qi,namely, aching, dull, heavy, numb, radiating, spreading, andtingling, and this list is largely consistent with those reportedin the literature.3,5 The face validity of this set of seven isreinforced by the fact that none of these are included in theacute pain (not de qi) cluster. That de qi sensations are dis-tinct from pain sensations is a finding supported by studiesof both UK3 and Korean8 patients. Our results are based onthe expert opinion of an international group of expertacupuncturists with an average of 21 years’ experience. In-terestingly, the five sensations added by Park et al.5 to theVincent scale,3 namely, boring, pinching, pressing, flicker-ing, and tender, do not appear to be associated with de qi.

The absence of a one-to-one correspondence between thede qi sensation and a single descriptive word in English canbe understood in a postulated context that “the sensation dif-fers from individual to individual, and is affected by the con-dition of the patient or the anatomical location of the point.”1

If this variability is to be expected as normal, then the com-plexity of controlling for the needle sensation in experi-mental studies of acupuncture, and ensuring that similar lev-els of de qi are experienced and that there are no acute painsensations, are major challenges.

Subsequent to conducting this study, we became aware ofresearch at Massachusetts General Hospital5,9 in which a nee-dle sensation scale was developed for use in conjunction withimaging studies. They state that their “verbal descriptors wereselected on the basis of Traditional Chinese Medicine de-

scriptions of the de qi sensation in previous literature and ourown experience.”9 Their scale had nine sensations: stabbing,throbbing, tingling, burning, heaviness, fullness, numbness,soreness, and aching. A blank space was provided for anyadditional sensation not listed. Comparison of these sensa-tions with our expert opinion and quantitative cluster analy-sis shows that only four of our de qi sensations overlap,namely, aching, heaviness, numbness, and tingling.†

An assumption behind our study is that de qi is an im-portant and probably necessary component of acupuncture,although acupuncturists do not universally hold this as-sumption. There is no widespread agreement on how strongthe de qi sensation should be for a therapeutic effect.10 Inexperimental research, the needle sensation is often verystrong in order to demonstrate an impact neurologically.10

In everyday practice, however, sensations elicited are usu-ally much milder. In our study, we did not ask practitionersabout their experience as acupuncturists if the de qi sensa-tion arrived at an acupuncture point. Acupuncture texts de-scribe this in many ways, including “the sensation of get-ting a bite on a fishing line.”1 For some experiencedpractitioners, the objective sensations felt by the acupunc-turist may be a more reliable guide to the arrival of the deqi sensation than the patient’s subjective experience.

In terms of limitations, it should be noted that we haveonly sought acupuncturists’ opinions, and have not involvedpatients as experts on their subjective experiences of needlesensation. It is likely, however, that all of the 22 acupunc-turists who participated have experienced acupuncture them-selves, and if so, they are likely to have responded in a dualrole. Not all 20 of the experts who completed the question-naire found it easy. The Results section above details theirconcerns about the methods and questionnaire as a basis forclassifying these complex needle sensations. De qi, when itis propagated along a channel, a technique described in manytexts, fell outside the scope of our inquiry.1,2 Finally, be-cause hierarchic cluster analysis is an exploratory statisticalmethod, the clusters we have identified should be treated astentative until they are confirmed by further research. Suchresearch could usefully incorporate a stronger patient per-spective, perhaps initially using qualitative methods to fur-ther develop the list of descriptors of the de qi sensation.

CONCLUSIONS

Data from international experts have been used withquantitative methods to separate out two clusters of acupunc-ture needle sensations: a de qi sensation cluster and an acutepain cluster. In the design of experimental trials involvingacupuncture needling, researchers will find these two clus-

MACPHERSON AND ASGHAR636

†This team of researchers has recently extended their scale to 12sensations as well as an “other” option (R. Gollub, personal com-munication, February 12, 2005).

ters of sensations useful when controlling for variation inneedle sensation.

ACKNOWLEDGMENTS

We acknowledge the contributions of our expert acupunc-turists: Terje Alrek, Panos Barlas, Richard Blackwell, MarkBovey, Charlie Buck, Wainwright Churchill, Peter Deadman,Nadia Ellis, Andrew Flower, Roisin Golding, Nick Haines,Val Hopwood, Ted Kaptchuk, Nicholas Lampert, Lixing Lao,Giovanni Maciocia, Albrecht Molsberger, Arya Nelsen, Jong-bae Park, Adrian White, Peter White, and Chris Zaslawski.We also acknowledge the encouragement of George Lewithto publish this study and the help of statistician Jeremy Mileswith the hierarchic cluster analysis. Hugh MacPherson is sup-ported by a Department of Health postdoctoral award in com-plementary and alternative medicine.

REFERENCES

1. O’Connor J, Bensky D. Acupuncture: A Comprehensive Text.Seattle: Eastland Press, 1992.

2. Qiu M-L. Chinese Acupuncture and Moxibustion. Edinburgh:Churchill Livingstone, 1993.

3. Vincent C, Richardson PH, Black JJ, Pither CE. The signifi-cance of needle placement site in acupuncture. J PsychosomRes 1989;33:489–496.

4. Melzack R. McGill pain questionnaire: Major properties andscoring methods. Pain 1975;1:275–299.

5. Park J, White A, Stevinson C, et al. Validating a new non-penetrating sham acupuncture device: Two randomised con-trolled trials. Acu Med 2002;20:168–174.

6. Hui K, Liu J, Makris N, et al. Acupuncture modulates the lim-bic system and subcortical gray structures of the human brain:Evidence from fMRI studies in normal subjects. Hum BrainMapping 2000;9:13–25.

7. Hui KK, Liu J, Marina O, et al. The integrated response of thehuman cerebro-cerebellar and limbic systems to acupuncturestimulation at ST 36 as evidenced by fMRI. Neuroimage2005;27:479–496.

8. Park H, Park J, Lee H, Lee H. Does de qi (needle sensation)exist? Am J Chin Med 2002;30:45–50.

9. Kong J, Fufa DT, Gerber AJ, et al. Psychophysical outcomesfrom a randomized pilot study of manual, electro-, and shamacupuncture treatment on experimentally induced thermalpain. J Pain 2005;6:55–64.

10. Birch SJ, Felt RL. Understanding Acupuncture. Edinburgh:Churchill Livingstone, 1999.

Address reprint requests to:Hugh MacPherson, Ph.D.

Department of Health SciencesUniversity of York

Seebolm Rowntree Building, First FloorHeslington, York, YO10 5DD

United Kingdom

E-mail: hm18@york.ac.uk

ACUPUNCTURE NEEDLE SENSATIONS IN DE QI 637

This article has been cited by:

1. Bin Yan, Yu Lei, Li Tong, KeWei ChenFunctional Neuroimaging of Acupuncture 142-155. [CrossRef]

2. Alex Benham , Mark I. Johnson . 2012. Characterization of Painful and Non-painful Acupuncture Needle Sensations DuringBidirectional Rotation of a Single Needle Inserted 15–25 mm at Large Intestine 10. Medical Acupuncture 24:3, 172-180.[Abstract] [Full Text HTML] [Full Text PDF] [Full Text PDF with Links]

3. Jiliang Fang, Xiaoling Wang, Hesheng Liu, Yin Wang, Kehua Zhou, Yang Hong, Jun Liu, Lei Wang, Chao Xue, Ming Song,Baoyan Liu, Bing Zhu. 2012. The Limbic-Prefrontal Network Modulated by Electroacupuncture at CV4 and CV12. Evidence-Based Complementary and Alternative Medicine 2012, 1-11. [CrossRef]

4. Kehua Zhou , Jiliang Fang , Xiaoling Wang , Yin Wang , Yang Hong , Jun Liu , Lei Wang , Chao Xue , Ping Wang , BaoyanLiu , Bing Zhu . 2011. Characterization of De Qi with Electroacupuncture at Acupoints with Different Properties. The Journalof Alternative and Complementary Medicine 17:11, 1007-1013. [Abstract] [Full Text HTML] [Full Text PDF] [Full TextPDF with Links]

5. Pei-Chi Chou , Heng-Yi Chu , Jaung-Geng Lin . 2011. Safe Needling Depth of Acupuncture Points. The Journal of Alternativeand Complementary Medicine 17:3, 199-206. [Abstract] [Full Text HTML] [Full Text PDF] [Full Text PDF with Links]

6. Irene Moy, Magdy P. Milad, Randall Barnes, Edmond Confino, Ralph R. Kazer, Xingqi Zhang. 2011. Randomized controlledtrial: effects of acupuncture on pregnancy rates in women undergoing in vitro fertilization. Fertility and Sterility 95:2, 583-587.[CrossRef]

7. Chang Shik Yin , Ju-Hong Kim , Hi-Joon Park . 2011. High-Velocity Insertion of Acupuncture Needle Is Related to LowerLevel of Pain. The Journal of Alternative and Complementary Medicine 17:1, 27-32. [Abstract] [Full Text HTML] [Full TextPDF] [Full Text PDF with Links]

8. Ji-Sheng Han, Yuh-Shan Ho. 2011. Global trends and performances of acupuncture research. Neuroscience & BiobehavioralReviews 35:3, 680-687. [CrossRef]

9. Hyangsook Lee, Heejung Bang, Youngjin Kim, Jongbae Park, Sangjae Lee, Hyejung Lee, Hi-Joon Park. 2011. Non-penetrating sham needle, is it an adequate sham control in acupuncture research?#. Complementary Therapies in Medicine19, S41-S48. [CrossRef]

10. Daniel Pach, Cynthia Hohmann, Rainer Lüdtke, Frank Zimmermann-Viehoff, Claudia M. Witt, Claudia Thiele. 2011. GermanTranslation of the Southampton Needle Sensation Questionnaire: Use in an Experimental Acupuncture Study. ForschendeKomplementärmedizin / Research in Complementary Medicine 18:6, 321-326. [CrossRef]

11. Seungmin Lee, Younbyoung Chae, Seung-Nam Kim, Song-Yi Kim, Ji-Yeun Park, Young-Rong Kwark, O-Seok Kang,Hyejung Lee, Chang Shik Yin, Hi-Joon Park. 2010. Short term effects by acupuncture to SP3 on the autonomic blood flowcontrol. Neurological Research 32:1, 37-42. [CrossRef]

12. Chang Shik Yin, Hi-Joon Park, Song-Yi Kim, Jae Min Lee, Mee Sook Hong, Joo-Ho Chung, Hyejung Lee. 2010.Electroencephalogram changes according to the subjective acupuncture sensation. Neurological Research 32:1, 31-36.[CrossRef]

13. Aziz UR Asghar, Gary Green, Mark F. Lythgoe, George Lewith, Hugh MacPherson. 2010. Acupuncture needling sensation:The neural correlates of deqi using fMRI. Brain Research 1315, 111-118. [CrossRef]

14. Dietrich H.W. Groenemeyer , Lei Zhang , Sven Schirp , Jürgen Baier . 2009. Localization of Acupuncture Points BL25 andBL26 Using Computed Tomography. The Journal of Alternative and Complementary Medicine 15:12, 1285-1291. [Abstract][Full Text HTML] [Full Text PDF] [Full Text PDF with Links]

15. CHANG SHIK YIN, JONGBAE PARK, JI-YOUNG LEE, YOUNBYOUNG CHAE, WOO-CHANG JANG, SEUNG-TAEKIM, HYEJUNG LEE, HI-JOON PARK. 2009. ACUPUNCTURE PERCEPTION (DEQI) VARIES OVER DIFFERENTPOINTS AND BY GENDER WITH TWO DISTINCT DISTRIBUTION PATTERNS OF DULLNESS AND PAIN. Journalof Sensory Studies 24:5, 635-647. [CrossRef]

16. Sala Horowitz. 2009. Acupuncture for Treating Mental Health Disorders. Alternative and Complementary Therapies 15:3,135-141. [Citation] [Full Text PDF] [Full Text PDF with Links]

17. Lesley Brown, Mike Holmes, Alice Jones. 2009. The application of transcutaneous electrical nerve stimulation to acupuncturepoints (Acu-TENS) for pain relief: a discussion of efficacy and potential mechanisms. Physical Therapy Reviews 14:2, 93-103.[CrossRef]

18. Luke D. Rickards. 2009. Therapeutic needling in osteopathic practice: An evidence-informed perspective. InternationalJournal of Osteopathic Medicine 12:1, 2-13. [CrossRef]

19. Zhi-Qi Zhao. 2008. Neural mechanism underlying acupuncture analgesia. Progress in Neurobiology 85:4, 355-375.[CrossRef]

20. Peter White , Felicity Bishop , Henry Hardy , Sam Abdollahian , Adrian White , Jongbae Park , Ted J. Kaptchuk , GeorgeT. Lewith . 2008. Southampton Needle Sensation Questionnaire: Development and Validation of a Measure to GaugeAcupuncture Needle Sensation. The Journal of Alternative and Complementary Medicine 14:4, 373-379. [Abstract] [FullText PDF] [Full Text PDF with Links]

21. Hugh MacPherson, Gary Green, Angel Nevado, Mark F. Lythgoe, George Lewith, Ross Devlin, Robyn Haselfoot, AzizU.R. Asghar. 2008. Brain imaging of acupuncture: Comparing superficial with deep needling. Neuroscience Letters 434:1,144-149. [CrossRef]

22. Jian Kong , Randy Gollub , Tao Huang , Ginger Polich , Vitaly Napadow , Kathleen Hui , Mark Vangel , Bruce Rosen , TedJ. Kaptchuk . 2007. Acupuncture De Qi, from Qualitative History to Quantitative Measurement. The Journal of Alternativeand Complementary Medicine 13:10, 1059-1070. [Abstract] [Full Text PDF] [Full Text PDF with Links]