Neuroscience Letters 464 (2009) 15

Contents lists available at ScienceDirect

Neuroscience Letters
journal homepage: www.elsevier.com/locate/neulet

An fMRI study of neuronal specicity of an acupoint: Electroacupuncture stimulation of Yanglingquan (GB34) and its sham point
Byung-jo Na a, , Geon-Ho Jahng b , Seong-uk Park c , Woo-sang Jung d , Sang-kwan Moon d , Jung-mi Park c , Hyung-sup Bae c
a Department of Internal Medicine, Kang-Nam Oriental Medicine Hospital, College of Oriental Medicine, Kyung-Hee University, 994-5 Dachil-dong, Gangnam-gu, Seoul, Republic of Korea b Department of Radiology, East-West Neo Medical Center, Kyung-Hee University, Seoul, Republic of Korea c Stoke and Neurologic Disorder Center, East-West Neo Medical Center, Kyung-Hee University, Seoul, Republic of Korea d Department of Cardiovascular and Neurologic Diseases (Stroke Center), College of Oriental Medicine, Kyung-Hee University, Seoul, Republic of Korea

a r t i c l e

i n f o

a b s t r a c t
The neuronal specicity of acupoints has not been entirely supported by the results of fMRI studies. The objective of this study was to investigate the neuronal specicity of an acupoint with electroacupuncture stimulation (EAS) using functional magnetic resonance imaging (fMRI). Functional MR imaging of the entire brain was performed in 12 normal healthy subjects during EAS of GB34 (Yanglingquan) and its sham point over the left leg in counter-balanced order. Anatomically, both GB34 and its sham point belong to the L5 spinal segment. EAS at the left GB34 specically activated the right putamen, caudate body, claustrum, thalamus, cerebellum, as well as the left caudate body, ventral lateral thalamus, and cerebellum, all related to motor function. EAS at the sham point of the left GB34 specically activated the right BA6, BA8, BA40, BA44, thalamus, as well as the left thalamus and cerebellum. Taken together, these ndings suggest that EAS at an acupoint and its sham point, in the same spinal segment, induced specic cerebral response patterns. These ndings support neuronal specicity of the acupoint studied. EAS at GB34 appears to be more related to motor function than EAS at its sham point, suggesting specicity of the GB34 acupoint. The results of this study provide neurobiological evidence for the existence of acupoint specicity, although further studies are necessary to better understand this phenomenon. 2009 Elsevier Ireland Ltd. All rights reserved.

Article history: Received 25 April 2009 Received in revised form 31 July 2009 Accepted 4 August 2009 Keywords: fMRI Electroacupuncture Neuronal specicity Acupoint GB34 (Yanglingquan)

Acupuncture has been gaining popularity among practitioners of modern medicine as an alternative and complementary treatment [5]. According to the theory of Oriental Medicine, each acupoint has functional specicity; the specic acupoints are carefully selected when acupuncture is used to treat disorders [12,16,20]. However, the underlying mechanisms of acupoint specicity are still not well understood. Since the 1990s, advances in noninvasive brain imaging techniques such as positron emission tomography and functional magnetic resonance imaging (fMRI) have enabled direct study of the human brain [2,4,6,7,9,10,11,1315, 18,2124]. Studies conducted using fMRI have provided evidence of acupoint specicity by demonstrating a correlation between acupoints and brain activation [4,14,15,18]. These ndings have indicated that acupuncture, at disease-implicated acupoints, can modulate the activity of the disease-related neuromatrix. For example, it has

Corresponding author. Tel.: +82 2 3457 9024; fax: +82 2 3457 9100. E-mail address: hani114@paran.com (B.-j. Na). 0304-3940/$ see front matter 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.08.009

been reported that specic cerebral sites are activated in response to stimulation of vision-associated acupoints (BL67, BL66, BL65, and BL60) [14,18], hearing-associated acupoints (GB43) [4] and language-associated acupoints (SJ8 and Du15) [15]. However, the specicity of acupuncture stimulation has not been entirely supported by the results of fMRI studies [3,7]. For example, Gareus et al. reported that stimulation of the vision-related acupoint did not induce a signicant blood oxygenation level dependent (BOLD)effect in the visual cortex [7]. Therefore, in this study, we chose the GB34 (Yanglingquan) acupoint and its sham point, not located in the meridian but close to each other in the same segment, to evaluate acupoint specicity. Anatomically, both GB34 and the sham point of GB34 belong to the L5 spinal segment. We rstly hypothesized that electroacupuncture stimulation (EAS) at the acupoint and its sham point, in the same spinal segment, would induce specic central responses, based on acupoint specicity. In fact, the acupoint GB34 is used to treat hemiplegia, a variety of muscle disorders, and knee pain [12,16,20]. We secondly hypothesized that EAS at GB34 was more likely related to motor function than EAS at its sham point. To test these hypotheses, we used blood

2 Table 1 Denition of the two types of electroacupuncture (EA). Real EA (Yanglinquan) Location Needle depth EA, current (mA) EA (Hz) Duration GB34 lateral area of left leg About 2 cm Subjectively optimal 2 30 s 3 blocks

B.-j. Na et al. / Neuroscience Letters 464 (2009) 15

Sham EA (for Yanglinquan) 2.0 cm lateral in the nonmeridian, to GB34 About 2 cm The same current as real EA 2 30 s 3 blocks

oxygenation level dependent (BOLD) fMRI during EAS of GB34 and its sham point. Twelve healthy right-handed male volunteers (mean and standard deviation = 33.6 6.2, range = 2643 years) participated in this study. The study protocol was approved by the local institutional review board. Informed consents were obtained from all subjects. The subjects had some knowledge of acupuncture due to cultural exposure, but had never received such treatment. None of the subjects had a history of psychiatric problems, neurological disorders or head trauma. Electroacupuncture stimulation (EAS) has been widely used as a substitute for classical acupuncture [8,13,21]. We applied actual electroacupuncture (EA) at acupoint GB34 and the sham EA of GB34 to the left leg of the subjects in counter-balanced order. The denition of the two types of electroacupuncture is provided in Table 1. The acupoint GB34 is located at the bular aspect of the leg, in the depression anterior and distal to the head of the bula [12,16,20]. The sham point was 2.0 cm lateral in the nonmeridian, to GB34. The other needle at the negative electrode was placed 1 cm proximally down along the gallbladder meridian for real EA, or along an imaginary line, for the sham EA. The other needle for the real EA and the sham EA was located in the L5 spinal segment (Fig. 1). Prior to the start of the fMRI experiments, all subjects were informed that different models of EAS, at various points, might be used for comparison. However, the precise locations of needle placement, the expected acupuncture effects, and the stimulation paradigm were not divulged. In addition, prior to the actual fMRI experiments, an EAS practice session was performed to determine the intensity of the EAS at which the subject could feel Deqi, which is the acupuncture effect of needle-manipulation characterized by sensations of numbness, heaviness, distention, and soreness. One experienced acupuncturist performed all electroacupuncture procedures using an EA device (ES-160, Ito Co. Ltd., Tokyo,

Japan). For all stimulations, a stainless steel acupuncture needle with a diameter of 0.25 mm and a length of 30 mm was inserted into the skin to a depth of approximately 2 cm at the acupoints or sham points. The needles were then adjusted to obtain Deqi. Of additional importance here is that sharp pain remained a specic concern and was avoided. Next, the needle, electrode, and electroacupuncture device were connected to a wire lead modied for safe use in the high magnetic eld environment of MRI. For each subject, the specic stimulation intensity was determined by gradually increasing and adjusting the voltage applied at each point, until the subject reported a mild to moderate sensation of Deqi. This was done at the beginning of each session, prior to fMRI scanning. The intensity was adjusted to a reasonable, but comfortable level that ranged from 2.4 to 3.2 mA (average, 2.82 mA) at which the subject could feel Deqi. The electroacupuncture stimulation was delivered using a continuous rectangular waveform with a pulse width of 0.4 ms at a frequency of 2 Hz. After each scan, subjects were asked to quantify their sensations about both acute pain and anxiety. The subjects rated the intensity of each sensation and their anxiety using a 5-point Likert scale: none, mild, moderate, severe and unbearable. No subjects reported experiences of sharp pain during the experiment. All the participants rated the intensity of each sensation and their anxiety as mild or moderate. There were no signicant differences in the intensity of each sensation and their anxiety evoked by stimulating the acupoint or its sham point under Fishers exact test (p > 0.05). All fMRI experiments were conducted using a 3.0 T whole body scanner (Philips Achieva, Best, the Netherlands) with an 8channel phase array head coil. During the experiment, subjects were instructed to remain relaxed with their eyes closed. First, T1-weighted spin-echo images were obtained to provide an anatomic reference (repetition time [TR]/echo time [TE] = 9.335/5 ms, eld of view [FOV] = 240 240 mm, ip angle = 90 , imaging matrix = 128 128). The images covered the entire brain and were parallel to the anterior commissureposterior commissure (ACPC) line. Next, functional images were obtained using a gradient-echo echo planar imaging (EPI) sequence. The parameters used were: TR = 3000 ms, TE = 35 ms, ip angle = 90 , FOV = 230 230 mm, imaging matrix = 64 64, slice thickness = 4.5 mm and gap between slices = 0 mm, and voxel size = 3.43 mm 3.43 mm 4.5 mm. A three-dimensional anatomic image was also acquired to reconstruct functional areas into brain

Fig. 1. Anatomical locations of the electroacupuncture stimulation points of GB34 (left) and its sham acupoint (right). The upper electrode is the positive pole and the lower electrode is the negative pole.

B.-j. Na et al. / Neuroscience Letters 464 (2009) 15 Table 2 Foci of brain activation in the multisubject analysis of the real electroacupuncture (EA) at the left GB34 and sham EA at the left GB34. Brain areas Side BA Real EA > rest Coordinates x, y, z Lentiform nucleus, putamen R R Caudate, caudate body L Claustrum Precentral gyrus R R 44 6 8 Inferior parietal lobule Thalamus Thalamus, Pulvinar Thalamus, ventral lateral nucleus R R L R L R Cerebellum, inferior semi-lunar lobule L 20, 10, 14 20, 70, 38 18, 62, 40 12, 66, 38 30, 64, 42 22, 68, 40 22, 50, 34 3.84 3.82 3.71 3.60 3.57 3.31 24, 64, 40 R Cerebellum, cerebellar tonsil BA: Brodmann area. L 3.41 24, 48, 38 32, 58, 38 4.03 3.71 3.73 40 20, 16, 14 3.63 6, 24, 4 4, 30, 6 3.48 3.56 24, 4, 20 28, 0, 12 16, 10, 20 20, 14, 22 12, 10, 22 14, 2, 22 28, 18, 22 Z-max 4.14 3.13 3.71 3.95 3.80 3.37 3.94 44, 4, 8 52, 14, 6 10, 14, 58 10, 22, 56 2, 20, 48 48, 32, 32 4.80 3.80 3.80 3.75 3.25 4.58 Sham EA > rest Coordinates xyz Z-max

Superior frontal gyrus

regions (TR = 9.9 ms, TE = 4.6 ms, imaging matrix = 240 240, FOV = 240 240 mm, ip angle = 7 , slice thickness = 1 mm, voxel size = 1 mm 1 mm 1 mm). Each subject was then subjected to two 3-min fMRI scans, with a 10 min interval between scans. During each scanning session, the EAS was delivered on one of the pairs of acupoints in the left leg. The block design for the two stimulations was R30 A30 R30 A30 R30 A30 , in which three stimulation periods (A, electrical manipulation stimulation for 30 s) were interposed with three rest periods (R, rest period during which time there was no stimulation for 30 s). Post-processing of the fMRI data from all subjects was performed using Statistical Parametric Mapping software (SPM2, www.l.ion.ud.ac.uk/spm/). To avoid the non-equilibrium effects of magnetization, we removed the rst 2 scans from the data of each subject, which left 240 scans for each subject to be analyzed. The images corresponding to each subject were realigned to correct for head motion and registered to the rst scan. After the realigned functional MR images were coregistered to the corresponding anatomical images, the three-dimensional anatomical images and the coregistered functional images were normalized to the Montreal Neurological Institute space and then spatially normalized functional images were smoothed using a 9-mm full-width half-maximum Gaussian kernel. The statistical mapping included two levels. First, the smoothed images were used for the xed-effect analysis based on the general linear model using a reference waveform adopted boxcar convolved with the canonical hemodynamic response function. The cerebral areas activated during stimulation (relative to baselines) were then identied (p < 0.001 uncorrected, spatial extent threshold, 100 voxels). Next, the group-level activation during stimulation

(relative to baselines) was determined using a random-effect analysis based on a one-sample t-test model using the results of the rst-level analysis for within-group analysis (p < 0.001 uncorrected, spatial extent threshold, 100 voxels). The resulting coordinates were then transformed into Talairach space. The results of the multisubject analysis are summarized in Table 2 and Fig. 2. The comparison of EAS, at the left GB34 versus the resting state, showed that EAS specically activated the right putamen, caudate body, claustrum, thalamus, cerebellum, as well as the left caudate body, ventral lateral nucleus, and cerebellum (Table 2; Fig. 2). The comparison of the EAS at the sham point of the left GB34 versus resting state showed that EAS specically activated the right BA6, BA8, BA40, BA44, thalamus, as well as the left thalamus and cerebellum (Table 2; Fig. 2). The purpose of the present study was to determine whether a different but close acupoint GB34 and its sham point, in the same spinal segment, would cause different fMRI responses. If no acupoint specicity exists, they should produce virtually similar responses. Conversely, if signicantly different responses resulted from these stimulations, this would provide evidence to support the specicity of an acupoint [24]. The brain regions activated by EAS at GB34 were different from those activated by EAS at its sham point (Table 2; Fig. 2). In addition, the results demonstrated that real EAS at GB34 had a greater effect and broad neuromatrix responses that involved limbic-related brain structures including the putamen, caudate body, and claustrum compared to the EAS at its sham point (Table 2; Fig. 2). The regions of the brain activated by EAS at GB34 were the basal ganglia, thalamus, and cerebellum, regarded as motor structures. Damage to these brain regions produces well-described alterations in motor function such as tremor, rigidity, akinesia, or

B.-j. Na et al. / Neuroscience Letters 464 (2009) 15

Fig. 2. Areas of activation in the multisubject analysis: (a) activation by electroacupuncture stimulation (EAS) at GB34 and (b) activation by EAS at the sham point of GB34.

dysmetria [1,17]. These activated brain regions suggest the indications for acupoint GB34 stimulation, such as hemiplegia and various muscle disorders [12,16,20]. Therefore, these results support neuronal specicity associated with EAS of GB34 as indicated by the activation of basal ganglia and cerebellar loops with motor areas of the cerebral cortex, all related to movement. EAS at GB34 might be helpful for stroke patients with motor disturbances caused by damage to the basal ganglia and cerebellar components of circuits associated with the motor areas of the cortex. However, additional studies are needed to conrm these ndings. These ndings support the results of previous fMRI studies on acupuncture or electroacupuncture at GB34. Jeun et al. reported that acupuncture stimulation at GB34 modulates the cortical activities of the somatomotor area in humans [11]. Zhang et al. reported that EAS at GB34 and BL57 induced deactivation in MI/PMC, as well as activation at the dorsal thalamus and putamen, known to be involved in motor functions [24]. A common region of the brain activated by EAS at its sham point as well as the GB34 acupoint was the cerebellum. This nding suggests that acupuncture performed at sites not located on meridians can have varying degrees of physiological and clinical effects, and that the cerebellum coordinates many functions of the brain such as autonomic control, cognition and affect, as well as in sensorimotor control with its complicated afferent and efferent connections with the cerebrum, brain stem and other regions [10,19,23]. Recent fMRI investigations of electroacupuncture and manual acupuncture have demonstrated increased cerebellar activity in response to treatment [10,22,23]. Yoo et al. found that cerebellar activity occurred in response to acupuncture stimulation of acupoint PC6 [23]. In addition, Yan et al. found that one of the common activation areas, in response to Liv3 or LI4 acupuncture, was the cerebellum, which was related to motor function [22]. Recently Hui et al. also suggested that modulation of the cerebro-cerebellar and limbic system activity may constitute an important pathway of acupuncture action [10]. One limitation of this study was the design of the controlled experiments. It is known that acupuncture performed at sites that are not located on meridians can have varying degrees of physiological and clinical effects [6]. In our study, the sham acupoint was

innervated by the same spinal nerve as the real acupoint, but was located 2 cm away from the meridian. In addition, we applied the sham EA, at nonmeridian points, using the same needle depth, stimulation intensity, frequency and pulse wave as those used for the real EA, because we were interested in the effects of EA at different locations. Furthermore, the sham EA used in our study elicited similar Deqi as the real EA; this was different from previous studies that used a placebo or minimal acupuncture devised to elicit no Deqi in the controls [2,9]. Therefore, there is the possibility that different results of this study compared to other studies occurred due to the different types of sham EA used. We thought it was more reasonable to base the detection of functional activity on the comparison of an activation task such as EAS with the baseline (without stimulation) as in our study. Therefore, we used the comparison of an electrical stimulation with the baseline (without stimulation) to investigate the neuronal specicity of an acupoint. Another limitation of this study was the sample size and statistical power. Conrmation of our ndings is needed with a larger sample size with greater statistical power. In addition, only the acupoint GB34 and its sham point were investigated; further study at other acupoints in other meridians will be needed to conrm our ndings. In conclusion, we have shown that EAS at an acupoint and its sham point, in the same spinal segment, induced specic cerebral response patterns, which provides evidence for neuronal specicity of an acupoint. We also have shown that EAS at GB34 may be more related to motor function than EAS at its sham point, and this is correlated with the clinical indications for acupoint stimulation of GB34. The results of this study may offer some primary neurobiological evidence for the existence of acupoint specicity, although further studies are necessary to better understand this phenomenon.

References
[1] K.P. Bhatia, C.D. Marsden, The behavioural and motor consequences of focal lesions of the basal ganglia in man, Brain 117 (1994) 859876. [2] G. Biella, M.L. Sotgiu, G. Pellegata, E. Paulesu, I. Castiglioni, F. Fazio, Acupuncture produces central activations in pain regions, Neuroimage 14 (2001) 6066.

B.-j. Na et al. / Neuroscience Letters 464 (2009) 15 [3] Campbell, Point specicity of acupuncture in the light of recent clinical and imaging studies, Acupunct. Med. 24 (3) (2006) 118122. [4] Z.H. Cho, I.K. Hong, C.K. Kang, J.S. Kim, C.S. Na, K.J. Park, K.W. Jeong, E.K. Wong, Acupuncture-stimulated auditory-cortical activation observed by fMRI: a case of acupoint SJ5 stimulation, Proc. Int. Soc. Mag. Reson. Med. 8 (2000) 327. [5] D. Diehl, G. Kaplan, I. Coulter, D. Glik, E.L. Hurwitz, Use of acupuncture by American physicians, J. Altern. Complement. Med. (1997) 119126. [6] J.L. Fang, T. Krings, J. Weidemann, I.G. Meister, A. Thron, Functional MRI in healthy subjects during acupuncture: different effects of needle rotation in real and sham acupoints, Neuroradiology 46 (2004) 359362. [7] I.K. Gareus, M. Lacour, A.C. Schulte, J. Hennig, Is there a BOLD response of the visual cortex on stimulation of the vision-related acupoint GB 37? J. Magn. Reson. Imag. 15 (2002) 227232. [8] J.S. Han, Acupuncture: neuropeptide release produced by electrical stimulation of different frequencies, Trends. Neurosci. 26 (2003) 1722. [9] J.C. Hsieh, C.H. Tu, F.P. Chen, M.C. Chen, T.C. Yeh, H.C. Cheng, Y.T. Wu, R.S. Liu, L.T. Ho, Activation of the hypothalamus characterizes the acupuncture stimulation at the analgesic point in human: a positron emission tomography study, Neurosci. Lett. 307 (2001) 105108. [10] K.K. Hui, J. Liu, O. Marina, V. Napadow, C. Haselgrove, K.K. Kwong, D.N. Kennedy, N. Makris, The integrated response of the human cerebrocerebellar and limbic systems to acupuncture stimulation at ST 36 as evidenced by fMRI, Neuroimage 27 (3) (2005) 479496. [11] S.S. Jeun, J.S. Kim, B.S. Kim, S.D. Park, E.C. Lim, G.S. Choi, B.Y. Choe, Acupuncture stimulation for motor cortex activities: a 3T fMRI study, Am. J. Chin. Med. 33 (2005) 573578. [12] R. Johns, The Art of Acupuncture Techniques, North Atlantic Books, California, 1996. [13] J. Kong, L. Ma, R.L. Gollub, J. Wei, A pilot study of functional magnetic resonance imaging of the brain during manual and electroacupuncture stimulation of acupuncture point (LI-4 Hegu) in normal subjects reveals differential brain activation between methods, J. Altern. Complement. Med. 8 (2002) 411419.

[14] G. Li, R.T. Cheung, Q.Y. Ma, E.S. Yang, Visual cortical activations on fMRI upon stimulation of the vision-implicated acupoints, Neuroreport 14 (2003) 669673. [15] G. Li, Liu, R.T. Cheung, An fMRI study comparing brain activation between word generation and electrical stimulation of language-implicated acupoints, Hum. Brain Mapp. 18 (2003) 233238. [16] D. Liangyue, G. Yijun, H. Shuhui, J. Xiaoping, L. Yang, W. Rufen, W. Wenjing, W. Xuetai, X. Hengze, X. Xiuling, Y. Jiuling, Chinese Acupuncture and Moxibustion, Foreign Language Press, Beijing, 2004. [17] F.A. Middleton, P.L. Strick, Basal ganglia and cerebellar loops: motor and cognitive circuits, Brain Res. Rev. 31 (2000) 236250. [18] C.M. Siedentopf, S.M. Golaszewski, F.M. Mottaghy, C.C. Ruff, S. Felber, A. Schlager, Functional magnetic resonance imaging detects activation of the visual association cortex during laser acupuncture of the foot in humans, Neurosci. Lett. 327 (2002) 5356. [19] R.S. Snell, Clinical Neuroanatomy, Lippincott Williams & Wilkins, 2006, pp. 300305. [20] WHO, WHO Standard Acupuncture Point Locations in the Western Pacic Region, Geneva, 2008. [21] M.T. Wu, J.M. Sheen, K.H. Chuang, P. Yang, S.L. Chin, C.Y. Tsai, C.J. Chen, J.R. Liao, P.H. Lai, K.A. Chu, H.B. Pan, C.F. Yang, Neuronal specicity of acupuncture response: a fMRI study with electroacupuncture, Neuroimage 16 (2002) 10281037. [22] B. Yan, K. Li, J.Y. Xu, W. Wang, Acupoint-specic fMRI patterns in human brain, Neurosci. Lett. 383 (2005) 236240. [23] S.S. Yoo, E.K. Teh, R.A. Blinder, F.A. Jolesz, Modulation of cerebellar activities by acupuncture stimulation: evidence from fMRI study, Neuroimage 22 (2) (2004) 932940. [24] W.T. Zhang, Z. Jin, F. Luo, L. Zhang, Y.W. Zeng, J.S. Han, Evidence from brain imaging with fMRI supporting functional specicity of acupoints in humans, Neurosci. Lett. 354 (2004) 5053.