Functional Magnetic Resonance Imaging of the Brain in the Investigation of Acupuncture

Z.-H. Cho. C.-S. Na. E. K. Wang. S.-H. Lee I-K. Hong

5.1

Introduction

 Medical imaging techniques that allow noninvasive observation of the structure and function of the human brain have improved dramatically during the past few decades. The relatively poor resolution of the biohazardous two-dimensional x-ray technique has been effectively replaced by safer and far more sensitive optical scanning techniques, including x-ray computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI). These modern advances in brain imaging routinely provide critical diagnostic information in such conditions as stroke, multiple sclerosis, and Parkinson’s disease as well as basic insights into how we experience, respond to, and even think about the world [2, 5-8, 12, 15, 18, 22—24, 27].

The application of this state of the art medical technology to the low tech, millennia-old practice of acupuncture seems at first to represent scientific overkill. But several preliminary explorations in this "West meets east" field of inquiry have yielded extremely promising results. For example, just as brain imaging is used on acute ischemic stroke patients to assess the need for administering so-called clot busting drugs [21], CT scans also have been effectively used to predict which stroke patients are likely to benefit from acupuncture treatment [20]. In addition, noninvasive imaging techniques have obvious potential for corroborating the brain pathways mediating acupuncture analgesia that have been identified in animals by microelectrode recordings, focal application of pharmacological agents, and ablation studies(see chapters 1.3).

The present chapter will focus in detail on the recent use of functional magnetic resonance imaging (fMRI) to test oriental medical theory [9, 10]. These studies have detected activity in the visual and auditory lobes of the brain during needling of distal acupoints on the leg and foot that Oriental medicine selects for treating eye and ear dysfunctions.

5.2

Brain Imaging for Identifying Neural Correlates of Acupuncture Analgesia 

Several imaging techniques, including PET, its related procedure single photon emission computed tomography (SPECT), and fMRI are used to measure neural activity as reflected in the uptake of cerebral blood glucose or increased cerebral blood flow. (Unlike most other cells, neurons have little stored glucose and rely on uptake of blood glucose as well as blood oxygen to sustain their high levels of activity.) The first two of these techniques are based on the use of radioisotopes that undergo predictable and detectable decay that can be resolved to specific brain regions. In a pilot study of five chronic pain patients, SPECT was employed to examine patterns of cerebral blood flow in various brain regions before and after acupuncture treatment [1]. The most striking finding was that four of the five patients had a marked left/right asymmetry in pretreatment blood flow in the thalamus, a major site in the neural integration of pain sensation. Following acupuncture treatment that resulted in all five patients reporting pain relief, the thalamic asymmetry was greatly reduced. Control subjects showed no blood flow asymmetries either before or after acupuncture.

The main advantage of MRI (for anatomical imaging) and fMRI (for physiological imaging) is their production of images without the use of radioisotopes. Since neural activity is critically dependent on oxygen derived from cerebral blood supplies, fMRI is an extremely sensitive means of measuring deoxyhemoglobin, the paramagnetic, oxygen-depleted carrier molecule. (Oxyhemoglobin is, fortuitously, nonmagnetic.) Functional MRI has been widely used for mapping changes in brain activity in response to specific sensory stimuli, motor tasks, or cognitive challenges as well as for detecting brain hemorrhages associated with pathological conditions [7, 8, 18, 23, 27].

Laboratories in Boston [16, 17] and Taiwan [33] have begun using fMRI to examine brain activity in response to needling at acupoints LI.4 and St.36 in normal subjects. Significantly greater levels of brain activity were detected during periods of acupoint stimulation than during periods of rest, shallow needling, or superficial pricking on the leg. A wide variety of brainstem, midbrain, and cerebral cortical structures showed reproducible patterns of increased or decreased activity; changes were especially related to structures associated with ascending nociceptive and descending antinociceptive pathways. Once stronger magnets become available, the activity patterns and the pathways they reflect will be identifiable with greater spatial and temporal resolution.

5.3

Functional Magnetic Resonance Imaging for Examining Correlations Between Brain Cortical Activity and Acupoint Function 

Several of the studies seeking anatomical bases for acupuncture points and meridians support the possibility that meridians, the classically defined “energy transporting channels’ are largely related to peripheral nerves [3, 4]. In fact, comparisons of acupoints with the peripheral nervous system given in many anatomical books [30] show that many acupoints correspond with the sites where small nerve bundles penetrate the fascia [14]. According to recently published reports, as many as 300 acupoints are situated on or very close to nerves, while an almost equal number are on or very close to major blood vessels that are surrounded by small nerve bundles [3]. This study, which also confirms that the acupoints lie along the peripheral nerves, leads us to

hypothesize that acupuncture signals are projected to the brain via the spinal cord and brainstem. Such signals could terminate in subcortical areas, while many are likely to reach the higher cortical areas, including the sensory cortex.

Acupoint/brain cortex relationships have been described in the Oriental acupuncture literature [11, 19, 31] and observed by experienced acupuncturists. More specifically, it has been hypothesized that the disease treatment claimed for acupuncture points may have somatotopic or sensory-related cortical correspondence (Fig. 1). We became interested in experimental testing of this hypothesis using fMRI to monitor cerebral cortical activity in areas functionally related to sensory conditions classically treated by selected acupoints. For example, visual dysfunctions that are localized and treated by Western medicine at sites along the retina/optic-nerve/occipital-lobe axis can, according to Oriental medicine, often be diagnosed by alterations in radial pulses corresponding to the urinary bladder (UB) and gallbladder (GB) channels.Such conditions can be treated, in turn, by needling along these meridians at distal acupoints localized on specific aspects of a toe, foot, and lower leg [11, 19]. More specifically, the UB meridian starts at the inner canthus of the eye, has 67 acupoints along its route, and ends on the lateral side of the little toe, while the GB channel starts at the outer canthus of the eye, has 44 acupoints, and ends at the lateral side of the fourth toe. The fMRI technique can therefore be used to explore quantitative correlations between acupoint stimulation and activation of functional areas of the brain. If aspects of Oriental medicine theory not predicted by the biomedical view of the body can be validated, e.g., correlations between sites of acupuncture stimulation and cerebral cortical activity not linked by known neural pathways, then an expanded theory of physiology may be required to combine aspects of both the Oriental and allopathic medical models.

To test our hypothesis that sensory-related acupoints have brain cortical correspondence, fMRI signals were sought in the visual cortex following needling of acupoints GB.37 (used to treat eye-related diseases such as itchiness or pain in the eyes, cataracts, night blindness, and optic atrophy) and in the auditory cortex following needling of GB.43 (known to be effective for treating ear-related diseases such as

deafness and tinnitus). We examined brain activity associated with stimulation of both acupoints and compared the results to our initial findings with another eye-

Table 1. Acupoints, target organs, and functionally activated areas

Fig. 2. Relation of acupoint stimulation and fMRI activity in visual cortex. a Two eye-related acupoints and a nonacupoint overlaid on the nervous system. b Activation maps of the brain due to (i) direct retinal stimulation by flashing light, (ii) acupuncture stimulation of UB.67, (iii) acupuncture stimulation of GB.37, and (iv) acupuncture-like stimulation of a nonacupoint

related acupoint, UB.67 [9]. For each of these three acupoints, the disease-related information and the cortical area expected to be activated are listed in Table 1. Distal acupoints on the lower leg and foot were chosen for ease of access, since subjects undergoing fMRI have their head, torso, and upper legs inside the magnet.

In our initial studies, we tested whether needling of UB.67, an acupoint traditionally used for treating eye disorders, would produce brain activity in the visual cortex that is detectable by fMRI (Fig. 2). Surprisingly, needling of this point led to reproducible increases in blood flow, i.e., increased fMRI signals, in Brodmann's areas 17, 18, and 19 of the visual cortex (Fig. 2b, ii). The effects were comparable to the changes in blood flow in the visual cortex produced by stimulation of the retina with flashing light (Fig. 2b, i). Needling of proximal acupoints UB.66 and UB.65 on the same channel also produced visual cortex activation [9]. In contrast, no visual cortex activity was detectable following needling either nonacupoints on the foot 2—5 cm from the vision-related acupoints (Fig. 2b, iv) or acupoint Sp.l on the large toe (Fig. 2a), which is irrelevant to the treatment of eye disorders. Of considerable interest, needling of GB.37, on a separate meridian but another of the most effective acupoints for the treatment of eye disorders (Fig. 3), again produced strong fMRI activity in the

Fig. 3. Acupuncture points overlaid on the peripheral nervous system as seen from anterior and posterior views

Fig. 4. Activation results of acupuncture stimulation of vision-related acupoint GB.37 observed by fMRI optical slice imaging. Cortical activation is shown due to (a) direct retinal stimulation by flashing light and (b) needling of the vision-related acupoint GB.37

Fig. 5. Activation results of acupuncture stimulation of a hearing-related acupoint GB.43. Cortical activation due to (a) flashing light, (b) music, and (c) acupuncture at GB.43. In (c), note the primary activation of the auditory cortex similar to that detected during listening to music (b) but also the small activation in the visual cortex similar to that detected during flashing light (a). This, we believe, is a secondary effect, since GB.43 is also used for the treatment of eye-related disease (Table 1)

visual cortex (Fig. 2b, iii). Activation of the visual cortex following stimulation of GB.37 is also shown in a series of “optical slices” (Fig. 4).

  As a further exploration of this phenomenon in other cortical areas, activity in the auditory cortex detectable by fMRI was examined following stimulation of GB.43, one of the best-known acupoints for the treatment of ear-related disease (Fig. 5). Interestingly, needling this acupoint resulted in strong activation of the auditory cortex in a manner similar to direct auditory stimulation with music but also led to weak activation of the visual cortex. It has been empirically observed and noted in acupuncture texts that some acupoints are relatively specific while others have more diverse functions. According to experienced acupuncturists, such primary and secondary responses may also be dependent on the health status of the patient.

5.4

Implications and Hypotheses

 Our results provide the first scientific evidence that acupuncture “signals” are projected to neocortical areas of the brain for central processing. These observations of the cortical projections of “signals” following stimulation of several acupoints strongly support the notion that many effects of acupuncture are mediated through the central nervous system. This concept of CNS involvement is supported by a considerable body of experimental evidence in the area of acupuncture analgesia, where correlations have been observed between acupoint stimulation, analgesia onset, and release of a variety of neurotransmitters, endogenous opioids, and hormones in the brain, spinal cord, and peripheral circulation [11, 13, 25, 26, 32] (see chapters 1.3).

Projections to the brain’s sensory cortices and the eventual effect on diseased organs by the brain’s higher centers may occur in concert with other functional centers in the body. The well-known homunculus of the human cortex illustrates just such a somatotopic possibility [28, 29]. To gain additional support for the theory of a CNS-mediated mechanism relating acupoints with disease sites as proposed from our preliminary findings, it seems necessary to study effects of many more prominent acupoints related to somatosensory and visceral organ systems. However, current observations may lead to some useful new hypotheses.

For example, we may hypothesize that stimulation of a specific acupoint delivers information to the corresponding cortical area(s), enabling the higher centers of the brain to make necessary decisions to regulate activities controlled by the endocrine and autonomic nervous systems. For this to occur, we believe that the hypothalamus and the amygdala play a key role, both in mediating the sensory input to the prefrontal cortex by integrating limbic information and in retrieving it from the prefrontal cortex. Information thus received from the amygdala would then be acted upon by the hypothalamus. The hypothalamus has unusually rich connections (both afferent and efferent) to many higher cortical areas such as the prefrontal cortex (both directly and via the amygdala), the limbic areas, and the brainstem and spinal cord.              One of the hypotheses postulated to explain acupuncture phenomena in terms of these neurobiological mechanisms is known as the beta-endorphin theory, which is presented schematically in Fig. 6 [13, 25, 32]. It is interesting to note that, in this model, the hypothalamus interacts with the higher cortical areas, especially the prefrontal cortex. A further extension of this acupuncture therapy model depicts the sen-

Fig. 6. A pain relief hypothesis involving higher cortical areas. In this example, the hypothalamus and other higher cortical areas are involved in secretion of beta-endorphin as well as other opioids and neurotransmitters. This hypothesis suggests involvement of higher cortical areas such as the sensory cortex as well as the frontal cortex, especially the prefrontal cortex and the limbic areas [11]

sory hierarchy coupled to the motor hierarchy via the limbic-hypothalamus or great limbic system (Fig. 7). In this schema, the final executive center is the hypothalamus where three survival-related systems involving endocrine, autonomic (ANS), and neuromodulatory functions are controlled.

The thrust of the present findings is that acupuncture stimulation is projected to higher brain areas such as the visual and auditory cortices. It is postulated that information is relayed from these sites to other key processing areas including the prefrontal cortex and limbic system. It is likely that acupuncture signals projected to these higher cortical areas will induce pain modulation as previously postulated but may also affect other survival-related functions. These latter mechanisms may shed light

Fig. 7. Functional brain hierarchies of motor and sensory functions. To illustrate the possible involvement of the acupuncture disease treatment effect, the limbic system is inserted as a mediator between the prefrontal cortex and the multimodal sensory association cortex. Here we hypothesize that acupuncture signals are projected to the sensory cortex and that the hypothalamus controls three major survival-related systems: endocrine, ANS, and diffuse modulatory neurochemical functions

on what has been a mystery for many acupuncture investigators, namely how acupuncture treats various diseases beyond the level of pain relief. In Fig. 8, an overall acupuncture disease treatment model is shown that contains but goes beyond current endorphin-mediated theories of pain control.

We may conclude that clues to the basic mechanisms underlying the several thousand-year-old practice of acupuncture will be revealed through modern scientific imaging techniques such as PET and fMRI. The careful and systematic examination of the hundreds of currently known acupoints and the mapping of corresponding cortical activation may well reveal evidence of homeostatic regulatory mechanisms not yet understood by Western physiology and medicine. Furthermore, future research with stronger magnetic fields will allow greater temporal and spatial resolution so that more subtle acupuncture signals can be detected. Such research will also contribute significantly to creating more accurate and reliable treatment for the millions of patients who may benefit from alternative medical therapies such as acupuncture.

 Acknowledgements. The present work is the product of many coworkers and their generous support. We are especially indebted to Drs. Heoung-Keun Kang and GwangWu Chung in the Department of Diagnostic Radiology of Chun Nam University School of Medicine, Kyangju, Korea for their support of our use of MRI scanners.

References

 1. Alavi A, LaRiccia P, Sadek H, Lattanand C, Lee L, Reich H, Mozley PD (1996) Objective assessment of the effects of pain and acupuncture on regional brain function with Tc99 mm HMPAO SPECT imaging. J Nuci Med 37 [Suppl 5]:278

2. Bandettini PA, Wong EC, Hinks RS, Tikofsky RS, Hyde JS (1992) Time course EPI of human brain function during task activation. Magn Reson Med 25:390—397

3. Chan SH (1984) What is being stimulated in acupuncture: Evaluation of the existence of a specific substrate. Neurosci Behav Rev 8:25—33

4. Chiang CY, Chang CT, Chu HL, Yang IL (1973) Peripheral afferent pathways for acupuncture analgesia. Sci Sin 16:210—217

5. Cho ZH, Chan JK, Ericksson L (1976) Circular ring transverse axial positron camera for 3D reconstruction of radionuclides distribution. IEEE, Trans Nuci Sci 23:613—622

6. Cho ZH, Ro YM, Lim TH (1992) NMR venography using the susceptibility effect produced by deoxyhemoglobin. Magn Reson Med 28:25—38

7. Cho ZH, Ro YM, Chung SC (1995) Susceptibility effect - enhanced functional MR imaging using tailored RF gradient echo (TRFGE) sequence. Int I Imag Syst Technol 6:164—170

8. Cho ZH, Ro YM, Park ST, Chung SC (1996) NMR functional imaging using a tailored RF gradient echo sequence: A true susceptibility measurement technique. Magn Reson Med 35:1—5

9. Cho ZH, Chung SC, Jones IP, Park JB, Park HI, Lee HJ, Wong EK, Mm BI (1998) New findings of the correlation between acupoints and corresponding brain cortices using functional MRI. Proc Natl Acad Sci, USA 95:2670—2673

10. Cho ZH, Lee SH, Hong IK, Wong EK, Na CS (1999) Further evidence for the correlation between acupuncture stimulation and cortical activation. Proceedings, New Directions in the Scientific Exploration of Acupuncture. University of California, Irvine

11. Filshie J, White A (1998) Medical Acupuncture: A Western scientific approach. Churchill Livingstone, Edinburgh, pp 225-294

12. Gilman 5 (1998) Imaging the brain. New Engl I Med 338:812—820, 889—896

13. Han JS (1993) Acupuncture and stimulation-produced analgesia. Handbook Exp Pharmacol 104/ 11:105—125

14. Heine H (1988) Anatomical structure of acupoints. I Trad Chin Med 8:207—2 12

15. Hennig J, Janz C, Speck 0, Ernst T (1995) Functional spectroscopy of brain activation following a single light pulse: Examinations of the mechanism of the fast initial response. Int I Imag Syst Technol 6:203—213

16. Hui KKS, Liu I, Wu M-T, Wang KKK (1996) Functional mapping of the human brain during acupuncture with magnetic resonance imaging. Proc Fourth World Conf Acupunct 4:71

17. Hui KK, Liu J, Makris N, Gollub RL, Chen AJ, Moore CI, Kennedy DN, Rosen BR, Kwong KK (2000) Acupuncture modulates the limbic system and subcortical gray structures of the human brain: evidence from fMRI studies in normal subjects. Human Brain Mapping 9:13—25

18. Kwong KK, Belliveau JW, Chester DA, Goldberg IE, WeisskoffRM et al (1992) Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proc Natl Acad Sci USA 89:5675—5679

19. Liu G (1996) Acupoints and meridians. Huaxia Publishing House, Huaxia, China

20. Naeser MA, Alexander MP, Stiassny-Eder D, Galler V, Hobbs J, Bachman D (1992) Real vs. sham acupuncture in the treatment of paralysis in acute stroke patients: A CT scan lesion site study. I Neurol Rehab 6:163—173

21. National Institute of Neurological Disorders and Stroke rI-PA Study Group (1995) Tissue plasminogen activator for acute ischemic stroke. New Engl I Med 333:1581—1587

22. Ogawa 5, Lee TM, Kay AR, Tank DW (1990) Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc NaIl Acad Sci USA 87:9868—9872

23. Ogawa 5, Tank DW, Menon R, Ellermann JM, Kim SG et al (1992) Intrinsic signal changes accompanying sensory stimulation: Functional brain mapping with magnetic resonance imaging. ProcaIl Acad Sci USA 89:5951—5955

24. Phelps ME, Hoffman EJ, Mullani NA, Ter-Pogossian MM (1975) Application of annihilation coincidence detection to transaxial reconstruction tomography. I Nucl Med 16:210—224

25. Pomeranz B (1987) Scientific basis of acupuncture. In: Stux G, Pomeranz B (eds) Acupuncture. Textbook and atlas. Springer, Berlin, pp 1—34

26. Pomeranz B (1996) Scientific research into acupuncture for the relief of pain. I Alt Compl Med 2:53—60

27. Raichle ME (1998) Behind the scenes of functional brain imaging: A historical and physiological perspective. Proc Natl Acad Sd USA 95:765—772

28. Ramachandran VS (1994) Phantom limbs, neglected syndromes, repressed memories, and Freudian psychology. Int Rev Neurobiol 37:291—333

29. Ramachandran VS (1998) Phantoms in the brain. William Morrow, New York

30. Rohen JW, Yokochi C, Lutjen-Drecoll E (t987) Color atlas of anatomy. Fourth edn. Williams and Wilkins, Baltimore

31. Stux G, Pomeranz B (1987) Acupuncture. Textbook and atlas. Springer, Berlin

32. Takeshige C, Sato T, Mera T, Hisamitsu T, Fang J (1992) Descending pain inhibitory system involved in acupuncture analgesia. Brain Res Bull 29:617—634

33. Wu MT, Hsieh JC, Xiong J, Yang CF, Pan HB et al (1999) Central nervous pathway for acupuncture stimulation: Localization of processing with functional MR imaging of the brain — preliminary experience. Radiol 212:133—141