Introduction: Animal models indicate that many acupuncture effects are mediated through deep structures in the brain. Direct evidence in humans is, however, limited. Demonstration of regionally specific, quantifiable acupuncture effects on relevant structures of the human brain would help elucidate the scientific basis of acupuncture and promote integration of this traditional therapeutic modality into the practice of modern medicine. The limbic system plays an important role in the regulation of multiple physiological functions. We postulate that modulation of this neuronal network by may constitute the initiating steps by which acupuncture achieves its diverse therapeutic effects. We used non-invasive functional magnetic resonance imaging (am) in normal human subjects to test this hypothesis.

Methods: Subjects were scanned during acupuncture and tactile stimulation in a l.5 Tesla GE Signa system equipped for echoplanar imaging. Acupuncture was performed at Hegu (LI l4) on either of the left (n=6) or the right hand (n=4) or sequentially on both hands (n=3) in 13 right-handed normal volunteers, age range 27-52. The needle was rotated manually at the rate of l20 times per minute with a balanced tonifying and reducing technique. The needle was left in place for minutes and manipu1ation was repeated for another 2 minutes. Superficial tactile stimulation was delivered over the Hegu area with a matched paradigm to compare its effects with acupuncture. High-resolution structural images were acquired by T1-weighted echo-planar recovery sequence. The structural and motion-corrected functional images were transformed into Talairach space for individual and group analyses. Scans of subjects who experienced pain instead of deqi were not included in-group averages. The time-course of signal change was examined for the putative activations identified on Statistical maps. The signal intensities during needle at rest served as baseline for measuring the signal changes during need1e manipulations.

Results: Eleven of the 13 subjects experienced varying degrees of deqi (acupuncture sensation) and 2-experienced pain. Both acupuncture and tactile stimulation elicited signal increases in the primary and secondary somatosensory cortices (SI, SII) bi1aterally, but the changes were less pronounced with acupuncture than with tactile stimulation. Importantly, acupuncture needle manipulation, but not tactile stimulation, produced signal decreases in multiple limbic and paralimbic structures in subjects Who experienced deqi. Signal decreases with P<10-5 was observed in the nuc1eus accumbens amygdala, hippocampus, parahippocampus, hypothalamus, ventral tegmental area, anterior cingulate, caudate nuc1eus, putamen, anterior insula and temporal pole. Signal decreases with P

Discussion / Conclusion : The consistent and synchronous effects of acupuncture on the limbic and paralimbic structures suggest that acupuncture exerts a coordinated modulatory action on this highly connected neural network. While there are multiple interpretations for decreases in BOLD fMRI signal, the decrease signal during acupuncture needle manipulation most likely reflects decreased regional cerebra1 blood flow resulting from suppression of neuronal metabolic activity. The limbic system plays a central role in the processing of emotion and in the regulation of sensory endocrine and autonomic nervous system functions. Modu1ation of the limbic system may constitute initiating steps by which acupuncture regulates multip1e physiological functions and achieves diverse therapeutic effects.

　　By Hui KKS ,Liu Jing,Gollub RL ,Chen AJM,Markris NT, Moore CI ,Kennedy DK ,Rosen
　　1MGH-NMR Center, Dept. Of Radiology; 2Center for Morphometric Analysis, Dept of Neurology; 3Dept of Psychiatry Massachusetts General Hospital & Harvard Medical School, Boston, MA; 4East-West Immune Institute, Waltham, MA, U.S.A.