1) The study investigated the differences in cerebral activity and clinical efficacy between acupuncture and sham acupuncture treatment for functional dyspepsia (FD).
2) 72 FD patients were randomly assigned to receive either real acupuncture or sham acupuncture for 4 weeks. 10 patients from each group also underwent PET scans before and after treatment.
3) Results showed acupuncture was more effective at reducing dyspeptic symptoms compared to sham acupuncture based on clinical scores. Acupuncture also produced greater deactivation of brain regions like the brainstem, anterior cingulate cortex, insula, thalamus, and hypothalamus compared to sham acupuncture. Deactivation of these regions correlated with improvement in clinical scores for
2. 2 Zeng et al.
health-care and social issue because of its high prevalence (3), unclear Chinese Medicine from December 2008 to May 2010. Patients
pathology, unsatisfactory treatment options (4), large medical were enrolled if they met the following criteria: (i) were right-
burden (5), and serious reduction in quality of life (QOL) (6). handed and aged 20 to 30 years; (ii) matched the Rome III diagno-
FUNCTIONAL GI DISORDERS
Hence, both patients and practitioners desire effective alternative sis criteria for FD; (iii) matched the Rome III diagnosis criteria for
therapies. postprandial distress syndrome; (iv) were acupuncture-naïve; and
In China and some other Asian countries, acupuncture, a major (v) signed an informed-consent form. Participants were screened
medical resource, has been used to treat gastrointestinal symptoms out if they (i) were pregnant, might become pregnant, or were lac-
for several millennia. In Western countries, acupuncture is being tating; (ii) suffered from or had a history of serious neurological,
increasingly accepted as an alternative treatment for functional cardiovascular, respiratory, or renal illnesses; (iii) had a history of
gastrointestinal disorders (7,8). During the past decade, a consid- head trauma with loss of consciousness; (iv) suffered from men-
erable number of clinical and experimental studies have indicated tal disorders including major depressive disorder such as anxi-
that acupuncture is able to relieve gastric symptoms such as belch, ety disorder, bipolar disorder, schizophrenia, or claustrophobic
abdominal distension, and stomachache and to alter gastrointesti- syndrome; (v) were using aspirin, nonsteroidal anti-inflamma-
nal motility functions (9–12), as well as that acupuncture points on tory drugs, steroids, phenothiazines, selective serotonin reuptake
the stomach meridian were the most effective sites for gastric dis- inhibitors, medications affecting gastrointestinal motility, or cer-
orders (11,12). However, it is not fully understood how needling at tain other drugs; (vi) were currently participating in other clini-
acupuncture points works and how real acupuncture differs from cal trials; or (vii) had any contraindications to acupuncture (e.g.,
sham acupuncture. anticoagulation therapy).
Many studies have demonstrated that most of the clinical ben- After a 2-week baseline period, the enrolled patients were ran-
efits of acupuncture are mediated by the central nervous system domly assigned to either the acupuncture group or the sham
(13,14), but the specific effects of acupuncture on the human acupuncture group using a computer-generated randomization
brain remain uncertain. With the development of neuroimaging sequence. The sequence was concealed from the care providers
techniques, the use of positron emission tomography (PET) and through the use of sealed, opaque, sequentially numbered enve-
functional magnetic resonance imaging to explore the central lopes. Patients were blinded to group assignment.
mechanism of acupuncture has been an active area of research. This study was performed according to the principles of the
However, most current studies are centered on the healthy state or Declaration of Helsinki (Edinburgh version, 2000). The study
organic diseases, and little work has been undertaken with regard protocol was approved by the Ethics Committee of the 1st Teach-
to functional disorders, which is where the advantages of acupunc- ing Hospital of the Chengdu University of Traditional Chinese
ture treatment lie. Medicine.
In a previous study, we found that FD patients showed an
extensive increasing cerebral glycometabolism, especially in Acupuncture interventions
the homeostatic afferent processing network, compared with Each group’s treatment consisted of 20 sessions of acupuncture
healthy subjects. The anterior cingulate cortex (ACC), insula, and treatment, with a duration of 30 min, each administered over
thalamus/hypothalamus might be the regions most closely related a period of 4 weeks (five sessions per week). The acupuncture
to the severity of FD (15). We hypothesize that successful acupunc- treatment was performed on four classic acupuncture points
ture therapy will reduce the increase in cerebral glycometabolism for gastric disorders: ST34 (Liangqiu), ST36 (Zusanli), ST40
and regulate the activity of these key regions. Our previous work (Fenglong), and ST42 (Chongyang). The sham acupuncture treat-
also suggested that affecting cerebral activity might be the mecha- ment was performed on four non–acupuncture points (points
nism through which short-term manual acupuncture treatment 1–4), which were selected in accordance with the findings of
relieves FD (16). However, the differences in efficacy and cerebral previous studies (17–19) (Figure 1).
responses between acupuncture and sham acupuncture were not All acupuncture points and non–acupuncture points were punc-
addressed in our previous study. Hence, the present study aimed to tured unilaterally and alternated between left side and right side.
(i) assess the efficacy of acupuncture treatment for FD by compar- After the skin was cleaned with tincture iodine and alcohol, sterile
ing the differences in therapeutic effects between acupuncture and acupuncture needles (0.25 mm in diameter, 25 or 40 mm long,
sham acupuncture and (ii) investigate the influence of acupunc- Hwatuo, Suzhou, China) were inserted for 15–25 mm and gently
ture and sham acupuncture on cerebral glycometabolism and ana- twisted, lifted, and thrusted in even amplitude, force and speed
lyze the possible correlations between clinical variables and brain four to six times until a deqi response (soreness, numbness, disten-
responses in order to explore the potential central mechanism of sion, and heaviness) was obtained. Then, auxiliary needles were
acupuncture treatment. inserted in the proximal limbs or trunk laterally to the acupunc-
ture points and non–acupuncture points to a depth of 2 mm with-
out stimulation. Each acupuncture needle and its auxiliary needle
METHODS were connected to the electrical leads of the HAN Acupoint Nerve
Participants Stimulator (HANS, Model LH 200A TENS, Nanjing, China) for
FD patients were recruited from the outpatient department in the 30 min, with a stimulation frequency of 2/100 Hz and a stimulation
1st Teaching Hospital of the Chengdu University of Traditional intensity varying from 0.1 to 1.0 mA.
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com
4. 4 Zeng et al.
to provide 30 patients in each group in order to meet the require- In this way, we could gain regions of interest for correlation anal-
ment of minimum sample size. ysis. After voxels not belonging to the same anatomical region
within the cluster were discarded, the activities of the survived
FUNCTIONAL GI DISORDERS
Clinical variables. The clinical variables were analyzed with voxels were extracted and averaged. Pearson coefficients were cal-
SPSS 16.0 (SPSS, Chicago, IL) by two blinded evaluators. The culated between the mean activity of the cluster and the increase
data analysis was based on an intention-to-treat population. We in NDI score, the mean activity of the cluster, and the decrease in
omitted the cases with missing data, which reflected only the SID score.
baseline measurement. Analysis of variance and the Kruskal– Based on individual cerebral activity changes before and after
Wallis test were used for numerical variables. A χ2 test was used real/sham acupuncture treatment, we performed a two-sample
for categorical variables. A two-sided test was applied for all t-test to further detect between-group differences using the
available data. A P value < 0.05 was considered statistically sig- following contrast: real acupuncture group (each FD patient: after
nificant. treatment minus at baseline) minus sham acupuncture group (each
FD patient: after treatment minus at baseline).
PET-CT data. The PET-CT data were processed with the statisti- For visualization, all results were transformed into the Talairach
cal parametric mapping technique (SPM5; http://www.fil.ion.ucl. stereotactic space and overlaid on MRIcro (http://www.sph.s.c.
ac.uk/spm) and MATLAB (The Math Works, Sherborn, MA). The edu/comd/rorden/mricro.html) for presentation.
data preprocessing entailed the following steps. (i) Individual PET
images were subjected to coregistration onto their corresponding
CT images to improve the accuracy of the spatial normalization. RESULTS
(ii) The PET images from each subject were spatially normalized From December 2008 to May 2010, 72 FD patients were randomly
to the standard SPM-PET template and resliced to 2-mm isotropic assigned. In all, 64 patients (34 in the acupuncture group and
resolution. The first step of the normalization was to determine 30 in the sham acupuncture group) comprised the pre-proto-
the optimal 12-parameter affine transformation. The affine reg- col population. Twenty patients (10 in each group) finished the
istration was followed by estimation of nonlinear deformations, PET-CT scans.
whereby the deformations are defined by a linear combination
of three-dimensional discrete cosine transform basis functions. The baseline characteristics
(iii) The resultant transformation was applied to the coregistered Demographics, including age, sex, weight and height, and dis-
PET images so that all subjects matched the same spatial template. ease status as indicated by, for example, duration of symptoms,
(iv) The normalized data set was spatially smoothed with a 6-mm NDI score, and SID score, did not differ between the two groups
full width at half maximum Gaussian kernel. (P > 0.05) (Table 1).
To detect the cerebral activity changes in FD patients after
treatment, we compared the differences in the cerebral glycome- The therapeutic effects
tabolism pattern between FD patients at baseline and after treat- In the acupuncture group, the SID score for postprandial disten-
ment. Statistical parametric maps were constructed by computing sion significantly decreased, from 1.912±0.571 to 0.880±0.640
a paired t-test, which was defined as FD patients after treatment (P < 0.05); the SID score for early satiety significantly decreased,
minus FD patients at baseline. To explore the possible correlation from 1.352±0.810 to 0.530±0.661 (P < 0.05); and the NDI score
between cerebral responses and clinical efficacy, we employed significantly increased, from 77.812±10.121 to 90.028±8.910
correlation analysis to investigate the correlations of cerebral (P < 0.05) after treatment (Table 2).
glycometabolism changes with the decrease in SID score (end In the sham acupuncture group, the SID score for postpran-
of treatment minus baseline) and increase in NDI score (end of dial distension significantly decreased, from 1.930±0.641 to
treatment minus baseline). The selection of regions of interest was 1.400±0.721 (P < 0.05), and the NDI score significantly increased,
based mainly on our previous study (15) and involved the within- from 78.212±9.223 to 86.040±9.210 (P < 0.05) after treatment. The
group subtraction analysis. In our previous study (15), we found decrease in SID score for early satiety was not significant (P > 0.05)
that the glycometabolism increase in the ACC, middle cingulate (Table 2).
cortex, insula, thalamus, cerebellum, hypothalamus, prefrontal The decrease in SID scores (end of treatment minus baseline) in
cortex, brainstem, hippocampus, and parahippocampal gyrus was the acupuncture group was significantly greater than that in the
significantly related to or tended to be associated with the symp- sham acupuncture group (P<0.05). The increase in NDI score (end
tom severity of FD patients. In the present study, the within-group of treatment minus baseline) between the two groups was not sig-
analysis showed that the activity of these regions changed to a dif- nificant (P > 0.05). Only the increase in NDI score in the acupunc-
ferent degree after treatment. Hence, we chose these 10 regions for ture group was clinically significant because it was greater than the
correlation analysis. For each patient, we chose the MNI coor- minimal clinically important difference (Table 3).
dinates of the maximally abnormal voxel within an anatomical
area as the center to draw a sphere 8 mm in diameter (25). Within Cerebral glycometabolism changes
this sphere, the voxels located in the white matter and ventricles In the acupuncture group, a decrease in cerebral glycometa-
were removed to ensure the integrity of its structure and function. bolism was observed after treatment in the bilateral brainstem,
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com
6. 6 Zeng et al.
Table 3. Comparison of the therapeutic effects between acupuncture group and sham acupuncture group
Items Acupuncture group Sham acupuncture group Statistical value P value
FUNCTIONAL GI DISORDERS
No. of patients 34 30
NDI score
End of treatment, mean (95% CI) 90.028 (86.920; 93.136) 86.040 (82.600; 89.480) 368.000 0.056
End of treatment—baseline, mean (95% CI) 12.216 (8.533; 15.898) 7.828(4.912; 10.743) 376.500 0.072
SID score: postprandial distension
End of treatment, mean (95% CI) 0.880 (0.660; 1.110) 1.400 (1.130; 1.670) 321.000 0.003
End of treatment—baseline, mean (95% CI) − 1.029 ( − 1.333; − 0.726) − 0.533 ( − 0.768; − 0.299) 343.000 0.014
SID score: early satiety, mean (95% CI)
End of treatment, mean (95% CI) 0.530 (0.300; 0.760) 0.930 (0.620; 1.240) 365.000 0.030
End of treatment—baseline, mean (95% CI) − 0.824 ( − 1.102; − 0.546) − 0.167 ( − 0.428; 0.094) 277.500 0.001
CI, confidence interval; NDI, Nepean Dyspepsia Index; SID, Symptom Index of Dyspepsia.
SID or with NDI were not even might be attributable to the (Table 2). Second, the alleviations in postprandial distension and
different numerical ranges in the SID and NDI scores. early satiety in acupuncture group were greater than those in the
In the sham acupuncture group, the increase in NDI score sham acupuncture group (P<0.05) (Table 3). Furthermore, only
tended to be associated with the glycometabolism decrease in the improvement in QOL caused by acupuncture treatment was
the brainstem (r = − 0.720) and thalamus (r = − 0.724) (P<0.1, clinically valuable (Table 2). Therefore, the differences in clini-
corrected) (Figure 3a). cal efficacy between the two groups indicated that the efficacy of
acupuncture treatment was not nonspecific and that acupuncture
treatment was more effective than sham acupuncture treatment for
DISCUSSION FD patients.
This is the first neuroimaging study that focuses on the poten- Our results were partly in line with those of a study by Park
tial central mechanism of real and sham acupuncture treatments et al. (28), who found that both acupuncture at classic points (real
for FD. It demonstrates the similarities and differences in clinical acupuncture) and acupuncture at nondefined points (sham acu-
efficacy and brain responses between real and sham acupuncture puncture) decreased dyspepsia symptoms and improved the QOL
treatments. of FD patients and that there were no significant differences between
the real acupuncture group and sham acupuncture group. Some
Similarities and differences in clinical efficacy between real methodology issues, including the method of acupuncture stimu-
and sham acupuncture treatments lation (manual acupuncture vs. electro-acupuncture), the length of
In this study, clinical improvements were found in both groups time the needle was kept inserted, duration of treatment, and loca-
after treatment. Both treatments remarkably alleviated post- tion of non–acupuncture point, probably contribute to the differ-
prandial distension and improved the QOL (P<0.05) (Table 2); ences in results. Furthermore, the differences in treatment frequency
in addition, the improvement in QOL between the two groups might be a factor causing the efficacy difference (29). In our study,
did not differ (P > 0.05) (Table 3). These similarities might result the electro-acupuncture treatment was performed five times per
from a placebo effect. In fact, placebo response rates are found to week on each patient for 4 weeks, consistent with the standard prac-
be high in functional gastrointestinal disorder (26). A recent sys- tice in China, whereas in the study by Parket al., the manual acu-
tematic review (27) demonstrated that in all trials of acupuncture puncture treatment was performed on patients three times per week
treatments for gastrointestinal diseases, the significant improve- for 2 weeks. Hence, the stimulating quantity in our study was greater
ments of QOL were independent of real or sham acupuncture. than that in the study by Park et al. However, the influence of treat-
The efficacy of acupuncture in improving QOL in irritable bowel ment frequency on clinical efficacy needs further investigation.
syndrome and inflammatory bowel disease may be explained by
nonspecific effects, whereas specific acupuncture effects may be Similarities and differences in cerebral responses to real and
shown by clinical scores. sham acupuncture treatments
However, based on the results, we found that acupuncture treat- The potential common mechanism of real acupuncture and
ment was significantly superior to sham acupuncture treatment, sham acupuncture treatments for FD. In the present study, both
especially in alleviating symptoms. First, sham treatment was acupuncture treatment and sham acupuncture treatment elicited
effective only for postprandial distension, whereas acupuncture cerebral glycometabolism changes to different degrees. The com-
treatment alleviated both postprandial distension and early satiety mon areas responding to acupuncture and sham acupuncture
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com
8. 8 Zeng et al.
Table 4. Continued
Region Side Acupuncture group Sham acupuncture group
FUNCTIONAL GI DISORDERS
Talairach t value BA Sign Talairach t value BA Sign
X Y Z X Y Z
Temporal pole L − 44 15 − 30 − 7.59 BA38 ↓
R 47 21 − 23 10.53 BA38 ↓
Middle temporal L − 56 −2 − 25 − 7.59 BA21 ↓
gryrus
R 58 −4 − 24 − 8.04 BA21 ↓ 59 1 − 20 − 13.22 BA21 ↓
Inferior temporal L − 40 − 15 − 20 − 9.63 BA20 ↓
gryrus
R 51 − 10 − 35 − 8.27 BA20 ↓
Middle occipital L − 35 − 65 26 − 24.58 BA39 ↓
gryrus
R 32 − 94 12 − 9.89 BA18 ↓
Postcentral gyrus R 48 − 21 39 16.65 BA3 ↑
Precuneus L − 11 − 64 34 16.88 BA7 ↑
R 9 − 69 45 15.97 BA7 ↑ 5 − 45 18 − 8.04 BA 30 ↓
Parietal inferior L − 29 − 54 45 18.01 BA7 ↑
lobe
ACC, anterior cingulate cortex; BA, Brodmann area; FD, functional dyspepsia; L, left; MCC, middle cingulate cortex; PCC, posterior cingulate cortex; R, right.
Up or down arrow (↑/↓) indicates whether the structure showed a signal increase or decrease, respectively.
P < 0.05, family-wise error corrected with a minimal cluster size of 50 voxels.
2 7 –7 –32 –6
Acupuncture group
12 6 –4 –14 –6
Sham acpuncture group
–35 –6 +6 +35
Figure 2. Cerebral glycometabolism changes in functional dyspepsia (FD) patients after treatment. Acupuncture and sham acupuncture elicited cerebral
glycometabolism changes to different degrees, but acupuncture evoked more pronounced changes in cerebral activity, especially in the homeostatic
afferent processing network. P<0.05, family-wise error corrected with a minimal cluster size of 50 voxels.
included the brainstem, thalamus, some regions in the prefrontal acupuncture group (P<0.1, corrected). The results suggested
cortex, somatosensory cortex, and visual-related cortex. Among that the improvement of QOL might be associated with the
these areas, the glycometabolism decrease in the brainstem deactivations of the brainstem and thalamus in both groups.
and thalamus were significantly related to the increase in NDI The brainstem is the pathway for all fiber tracts passing up and
score in the acupuncture group (P<0.05, corrected) and tended down from the peripheral nerves and spinal cord to the highest
to be associated with an increase in NDI score in the sham parts of the brain, and it serves as a lower center in functions such
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com
10. 10 Zeng et al.
a b × 1000 c
× 10000
× 10000
0 0 0
Insula (post–pre)
Insula (post–pre)
Insula (post–pre)
–0.3 –0.3
r = –0.85
FUNCTIONAL GI DISORDERS
–0.6 –0.6
Acu –5 Acu
–0.9 –0.9
r = 0.77
–1.2 –1.2
–1.5 –10 –1.5
–5 0 5 10 15 20 25 30 35 Acu Sham –5 –4 –3 –2 –1 0
NDI (post–pre) SID (post–pre)
× 1000
Thalamus (post–pre)
Thalamus (post–pre)
Thalamus (post–pre)
× 10000
× 10000
0 0 0
–0.3
r = –0.72* –0.3
Acu
–0.6 –0.6
–5 r = 0.83 Acu
–0.9 Sham –0.9
r = –0.87
–1.2 –1.2
–1.5 –10 ** –1.5
–5 0 5 10 15 20 25 30 35 Acu Sham –5 –4 –3 –2 –1 0
NDI (post–pre) SID (post–pre)
× 1000
× 10000
× 10000
0 0 0
BS (post–pre)
BS (post–pre)
BS (post–pre)
–0.3 –0.3
r = –0.72* Acu
–0.6 –0.6
r = –0.85 –5 Acu
–0.9 –0.9
r = 0.72*
Sham
–1.2 –1.2
–1.5 –10 –1.5
–5 0 5 10 15 20 25 30 35 Acu Sham –5 –4 –3 –2 –1 0
NDI (post–pre) SID (post–pre)
× 1000
0 × 10000 0
× 10000
0
ACC (post–pre)
ACC (post–pre)
ACC (post–pre)
–0.3 –0.3
r = –0.84
–0.6 Acu –0.6
–5 r = 0.78 Acu
–0.9 –0.9
–1.2 ** –1.2
–1.5 –10 –1.5
–5 0 5 10 15 20 25 30 35 Acu Sham –5 –4 –3 –2 –1 0
NDI (post–pre) SID (post–pre)
× 1000
0 0
× 10000
0
× 10000
HYPO (post–pre)
HYPO (post–pre)
HYPO (post–pre)
–0.3 0 –0.3
r = –0.76
–0.6 Acu –0.6
–5 r = 0.79 Acu
–0.9 –0.9
–5
–1.2 –1.2
–1.5 –10 –1.5
–5 0 5 10 15 20 25 30 35 Acu Sham –5 –4 –3 –2 –1 0
–10
NDI (post–pre) SID (post–pre)
Figure 3. The correlation coefficients of brain responses and clinical variables. (a) In the acupuncture group, the increase in Nepean Dyspepsia
Index (NDI) score was significantly related to the glycometabolism decrease in the insula, thalamus, brainstem, anterior cingulate cortex (ACC), and
hypothalamus (P<0.05, corrected); in the sham acupuncture group, the increase in NDI score tended to be associated with the glycometabolism decrease
in the brainstem and thalamus (P<0.1, corrected). (b) Compared with the sham acupuncture group, the glycometabolism in the thalamus and ACC in
the acupuncture group significantly decreased (P<0.05). (c) In the acupuncture group, the decrease in Symptom Index of Dyspepsia (SID) score was
significantly related to the glycometabolism decrease in the insula, thalamus, ACC, and hypothalamus (P<0.05, corrected); the decrease in SID score
tended to be associated with the glycometabolism decrease in the brainstem (P<0.1, corrected). BS, brainstem; Hypo, hypothalamus; r, correlation
coefficient; *P<0.1, corrected; **P<0.05.
The majority of these deactivated regions in the acupuncture fiber activation. Nonpainful and painful visceral and somatic
group belong to the homeostatic afferent processing network as stimuli, as well as emotional stimuli, can activate this network
well as the corticolimbic network. (33). Our previous study (15) indicated that, as compared with
The homeostatic afferent processing network is a brain network healthy subjects, FD patients showed higher glycometabolism
that is consistently activated in response to homeostatic afferent in the key regions of the homeostatic afferent processing network,
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com
12. 12 Zeng et al.
3. Ghoshal UC, Singh R, Chang FY et al. Epidemiology of uninvestigated and 20. Ang D, Talley NJ, Simren M et al. Review article: endpoints used in func-
functional dyspepsia in Asia: facts and fiction. J Neurogastroenterol Motil tional dyspepsia drug therapy trials. Aliment Pharmacol Ther 2011;33:
2011;17:235–44. 634–49.
4. Kandulski A, Venerito M, Malfertheiner P. Therapeutic strategies for the 21. Tian X, Li Y, Ma T et al. Review of efficacy index of acupuncture for
treatment of dyspepsia. Expert Opin Pharmacother 2010;11:2517–25. functional dyspepsia. Liaoning J Tradit Chin Med 2008;25:1584–6.
FUNCTIONAL GI DISORDERS
5. Brook RA, Kleinman NL, Choung RS et al. Functional dyspepsia impacts 22. Talley NJ, Meineche-Schmidt V, Par P et al. Efficacy of omeprazole in
absenteeism and direct and indirect costs. Clin Gastroenterol Hepatol functional dyspepsia: double-blind, randomized, placebo-controlled trials
2010;8:498–503. (the Bond and Opera studies). Aliment Pharmacol Ther 1998;12:1055–65.
6. Chang JY, Locke##3rd GR, McNally MA et al. Impact of functional gas- 23. Talley NJ, Verlinden M, Jones M. Validity of a new quality of life scale for
trointestinal disorders on survival in the community. Am J Gastroenterol functional dyspepsia: a United States multicenter trial of the Nepean
2010;105:822–32. Dyspepsia Index. Am J Gastroenterol 1999;94:2390–7.
7. Ouyang H, Chen JD. Review article: therapeutic roles of acupunc- 24. Tian XP, Li Y, Liang FR et al. Translation and validation of the Nepean
ture in functional gastrointestinal disorders. Aliment Pharmacol Ther Dyspepsia Index for functional dyspepsia in China. World J Gastroenterol
2004;20:831–41. 2009;15:3173–7.
8. Takahashi T. Acupuncture for functional gastrointestinal disorders. J Gas- 25. Takahashi H, Kato M, Matsuura M et al. When your gain is my pain and
troenterol 2006;41:408–17. your pain is my gain: neural correlates of envy and schadenfreude. Science
9. Yin J, Chen JD. Gastrointestinal motility disorders and acupuncture. Auton 2009;323:937–9.
Neurosci 2010;157:31–7. 26. Enck P, Vinson B, Malfertheiner P et al. The placebo response in functional
10. Xu S, Hou X, Zha H et al. Electroacupuncture accelerates solid gastric dyspepsia, reanalysis of trial data. Neurogastroenterol Motil 2009;21:370–7.
emptying and improves dyspeptic symptoms in patients with functional 27. Schneider A, Streitberger K, Joos S. Acupuncture treatment in gastrointesti-
dyspepsia. Dig Dis Sci 2006;51:2154–9. nal diseases: a systematic review. World J Gastroenterol 2007;13:3417–24.
11. Lin YP, Yi SX, Yan J et al. Effect of acupuncture at Foot-Yangming Merid- 28. Park YC, Kang W, Choi SM et al. Evaluation of manual acupuncture at
ian on gastric mucosal blood flow, gastric motility and brain-gut peptide. classical and nondefined points for treatment of functional dyspepsia: a
World J Gastroenterol 2007;13:2229–33. randomized-controlled trial. J Altern Complement Med 2009;15:879–84.
12. Yi SX, Yang RD, Yan J et al. Effect of electro-acupuncture at Foot-Yangming 29. Harris RE, Tian X, Williams DA et al. Treatment of fibromyalgia with
Meridian on somatostatin and expression of somatostatin receptor genes in formula acupuncture: investigation of needle placement, needle stimula-
rabbits with gastric ulcer. World J Gastroenterol 2006;12:1761–5. tion, and treatment frequency. J Altern Complement Med 2005;11:663–71.
13. Takahashi T. Mechanism of acupuncture on neuromodulation in the gut, a 30. Cho ZH, Oleson TD, Alimi D et al. Acupuncture: the search for biologic
review. Neuromodulation 2011;14:8–12. evidence with functional magnetic resonanceimaging and positron emis-
14. Han JS, Xie GX, Zhou ZF et al. Enkephalin and beta-endorphin as media- sion tomography techniques. J Altern Complement Med 2002;8:399–401.
tors of electro-acupunctureanalgesia in rabbits: an antiserum microinjec- 31. Qiu YH, Wu XY, Xu H et al. Neuroimaging study of placebo analgesia in
tion study. Adv Biochem Psychopharmacol 1982;33:369–77. humans. Neurosci Bull 2009;25:277–82.
15. Zeng F, Qin W, Liang FR et al. Abnormal resting brain activity in patients 32. Eippert F, Bingel U, Schoell ED et al. Activation of the opioidergic descend-
with functional dyspepsia is related to symptom severity. Gastroentology ing pain control system underlies placebo analgesia. Neuron 2009;63:
2011;141:499–506. 533–43.
16. Zeng F, Song WZ, Liu XG et al. Brain areas involved in acupuncture treat- 33. Mayer EA, Naliboff BD, Craig AD. Neuroimaging of the brain-gut axis:
ment on functional dyspepsia patients: a PET-CT study. Neurosci Lett from basic understanding to treatment of functional GI disorders.
2009;456:6–10. Gastroenterology 2006;131:1925–42.
17. Assefi NP, Sherman KJ, Jacobsen C et al. A randomized clinical trial of 34. Mayer EA. Gut feelings: the emerging biology of gut–brain communication.
acupuncture compared with sham acupuncture in fibromyalgia. Ann Intern Nat Rev 2011;12:453–66.
Med 2005;143:10–9. 35. Mayer EA, Tillisch K. The brain-gut axis in abdominal pain syndromes.
18. Chang CH, Huang JL, Ting CT et al. Atropineinduced HRV altera- Annu Rev Med 2011;62:381–96.
tion is not amended by electroacupuncture on Zusanli. Am J Chin Med 36. Mayer EA, Aziz Q, Coen S et al. Brain imaging approaches to the study of
2005;33:307–14. functional GI disorders: a Rome working team report. Neurogastroenterol
19. Brinkhaus B, Hummelsberger J, Kohnen R et al. Acupuncture and Chinese Motil 2009;21:579–96.
herbal medicine in the treatment of patients with seasonal allergic rhinitis: 37. Craig AD. How do you feel? Interoception: the sense of the physiological
a randomized-controlled clinical trial. Allergy 2004;59:953–60. condition of the body. Nat Rev Neurosci 2002;3:655–66.
The American Journal of GASTROENTEROLOGY VOLUME 104 | XXX 2012 www.amjgastro.com