Survey design

The questionnaire was created collaboratively by the clinical acupuncturists and research staff at the Athinoula Martinos Center of Biomedical Imaging (Charlestown, MA, United States). Instead of the traditional sensations such as soreness, numbness and heaviness, we focused on sensations that are often overlooked (ie dull pain, sharp pain and throbbing).

Paper questionnaires were answered by licensed Chinese medicine practitioners on a self-reported and voluntary basis. Participants were asked whether they considered a specific type of pain-related sensation elicited from manual needling was deqi and whether it was beneficial or harmful to the patient. The questions were designed in a multiple-choice format and tailored to specifically address the nature of acupuncture sensations and its clinical significance based on the experience of the patient and the acupuncturist. Besides definitive answers, the participant was allowed to select "unsure" and allocated space for comments.

Survey administration

Fifty questionnaires were distributed to licensed acupuncturists attending workshop seminars directed by one of the co-authors (LL): 30 in Chicago (USA) and 20 in Shanghai (China). Questionnaires were also distributed to acupuncturists in Chinese medicine teaching institutions: 20 to the School of Traditional Chinese Medicine in San Francisco (United States) and 6 to Guang'anmen Hospital of the China Academy of Traditional Chinese Medicine in Beijing (China). Another ten questionnaires were distributed via email to the Beijing University of Chinese Medicine (China). Questionnaires were bilingual in English and Chinese. Survey participation was on a voluntary basis and did not involve patients; no consent form was needed. The acupuncturists' names were kept confidential. The distributors did not participate in the design of the questionnaires but in data analysis.

Statistical methods

A test of two proportions using two-tailed Fisher's Exact test was used to compare the sample proportions from the two countries using the conventional criterion for statistical significance, alpha = 0.05. Analyses were performed using Minitab v15 (Minitab^® Statistical Software, USA).

Participation rate

Of the 86 questionnaires distributed, 47 were returned, representing a 55% response rate. Of the 47 respondents, 26 were from the US and 21 from China. Four of the returned questionnaires from the US were excluded from analysis because the respondents indicated that the questions were unclear to them. The total of 43 returned questionnaires pooled from both countries for the final analysis represented 50% of the 86 questionnaires distributed and 92% of the 47 respondents. All the respondents in this study identified themselves in the survey as those practicing acupuncture of traditional Chinese style.

Participants

The training experience for participants pooled from the two countries averaged 6.4 years (range 2-18). Experience in practice averaged 12 years, (range 1-32). A larger range was observed for acupuncturists in China in years of practice (1-32 years) than those in the US (1.5-17 years).

Relative importance in clinical efficacy: subject's vs acupuncturist's sensations

Relationship between acupuncturist's needle grasping sensation and patient's deqi response

Participants were asked if there was a relationship between the needle grasping sensation felt by the acupuncturist and the patient's deqi experience (Table 1). Most respondents indicated that the acupuncturist's needle grasping and the patient's deqi experience occurred concurrently (41%) or that sometimes they occurred concurrently (31%). Almost a quarter of respondents (24%) was unsure or not providing an answer.

Patient's attitude towards deqi sensations

Participants were questioned about their patient's attitudes towards deqi sensations (Table 1). A total of 40% of participants reported that patients liked the deqi sensations whereas 19% reported "dislike" and rest neutral. There was an interesting difference between the two countries regarding the deqi sensation. Of the 17 respondents who indicated that the patients liked the sensations, only 2 were from the US and all the rest from China (P < 0.0001). In contrast, of all returned questionnaires that indicated "dislike" (19%), all were from the US and none from China (P = 0.0020).

Correlation between deqi response and efficacy

Participants were asked whether there was a correlation between deqi and clinical efficacy in their acupuncture practice (Table 1). A total of 30 respondents (73%) indicated a positive correlation; no cultural differences among the respondents.

Types of pain and deqi

Participants were asked to indicate "deqi" or "not deqi" for sensations often associated with pain, such as throbbing, dull pain and sharp pain and whether these sensations were beneficial or harmful. Of all participants, 72% agreed that dull pain was characteristic of deqi and beneficial to clinical efficacy (53%). Only 2-3% indicated dull pain was "not deqi" and harmful. About 20% of returned questionnaires did not have the answer for the question.

Very few participants considered sharp pain to be deqi or beneficial (7% and 14% respectively) and 50% classified it as "not deqi" and 42% believed it was harmful.

Respondents were less certain about the nature of throbbing and its relation to efficacy than they were about dull pain or sharp pain. Responses were almost equally divided between "deqi" (28%) and "not deqi" (21%). While over two-thirds answered "unsure" or did not provide an answer, 14% considered throbbing as "beneficial" and 21% "harmful".

Limitations

The sample size of this study was small and the scope of this survey is narrow. The exclusion of the timing of deqi assessment and sensations is also a limitation of this study. We consider a participant's reliance on memory in respect of sensation timing may result inaccuracy in a survey; therefore, we advocate a detailed exploration of dosage and patient attitude in an apposite clinical lab setting.

Relation between deqi and clinical efficacy

Most participants (77%) in both countries believed there was a positive correlation between deqi and clinical efficacy. Our fMRI studies found a positive correlation between a subject's psychophysical and hemodynamic response: ie strong deqi sensations induced strong deactivation of the limbic system [27, 28]. Such correlation provides strong evidence, albeit indirect, that neurophysiological effects of acupuncture deqi are mediated via specific brain networks.

Most participants indicated that both the patient's deqi and the acupuncturists' grasping sensation were important. However, this pilot survey found that a small percentage of participants disagreed with the majority, with 12% favoring the patient's experience and 5% favoring the acupuncturist's sensory experience.

Dull pain, sharp pain and throbbing

Of the three types of pain related sensations, dull pain was generally considered to be deqi and beneficial while sharp pain was considered to be unrelated and noxious instead. In regard to throbbing, about half of the respondents were uncertain; the rest were almost equally divided between the two categories. Dull pain is seldom described in acupuncture literature. Park and Lee mentioned a 'dull' sensation but did not specify whether it was pain-related [23]. We first called attention to dull pain as a characteristic component of deqi and noted that it often occurred before or independent of sharp pain instead of occurring after sharp pain as described in the literature [26, 27]. This pilot survey found that most respondents indicated dull pain as deqi and beneficial and sharp pain as noxious and harmful. These sensations and their relationship to deqi are often overlooked/neglected sensations. We regard their importance as we encounter them in our studies. With fMRI monitoring, we demonstrated that dull pain deactivated the limbic system and that sharp pain activated it [27, 28]. Our findings are in agreement with MacPherson's report [22].

The survey results on throbbing were less definitive than dull pain or sharp pain; most participants were unsure of its nature and the rest were equally divided between deqi and not deqi as well as between beneficial and harmful. There is little information on throbbing in the literature. Occasional reports suggest that it occurs more often in electro- than in manual acupuncture; it's association with neuronal activity is unclear [5, 23]. Our fMRI studies found that this sensation could be associated with either predominant deactivation or activation of the limbic system and pain matrix; throbbing may represent an intermediate stage in the excitation of the fine pain conducting fibers; a progressive increase in the intensity of mechanical stimulation may cause transition from the more innocuous dull pain to throbbing and finally to overt noxious sharp pain.

The significant difference between the two countries in respect of the patients' attitudes towards deqi sensation may have clinical implications. Most respondents in China reported that patients liked the deqi sensations while those from the US disliked them. Interestingly, this was the only response that showed significant difference between the two countries. A difference in patients' attitudes may affect how acupuncturists adjust their techniques and intensity of stimulationto the sentiments of his patients. This raises an important question as to how TCM practices are translated and practiced across cultures and uncovers confounds in meta-analysis of clinical efficacy studies. It is possible that acupuncturists in the US adopt less intensive and shorter durations of stimulations out of consideration for the patients' attitudes towards the sensory experience, resulting in lower dosage levels and lower therapeutic effects. Further investigation with large clinical sample sizes is warranted.