Hui KKS, Marina O, Liu J, Papadimitriou G, Howard J, Napadow V, Nixon E, Kwong KK, Makris N.

Correlates of hemodynamic and psychophysical responses in acupuncture fMRI.

Athinoula A Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, 149, 13th St, Rm 2301, Charlestown, MA 02129. hui@nmr.mgh.harvard.edu

INTRODUCTION: We seek to understand the relationships between the hemodynamic response in fMRI and "deqi" (a composite of acupuncture sensations related to clinical efficacy according to Chinese medicine), the correlation between the response of different brain regions, and the difference in central effects between acupoints. We hypothesize that modulation of the limbic system constitutes the neurocorrelate of deqi and contributes to its therapeutic actions. We performed an automated analysis of the average signal changes in the regions of interest in order to minimize potential investigator bias.

METHOD: Acupuncture was performed at LI4, ST36 and LV3 in 43 healthy adults. Two periods of needle twisting, each 2 min, were interleaved with periods of needle at rest in a 10 min scan. Tactile stimulation over the acupoint served as control. The subject was interviewed for sensations experienced at the end of each scan. Whole brain fMRI was performed on a 1.5 T Siemens Sonata using T2*-weighted gradient echo sequence. Anatomical masks defining regions of interest were constructed based on the cohort's averaged dataset of high resolution structural images. The BOLD response for each region was determined as the sum of the percent signal change from baseline (increase or decrease) for all voxels thresholded at p<0.003. AFNI was employed for data analysis,

RESULTS: Both psychophysical and hemodynamic response to acupuncture showed distinct differences from sensory control. Aching, soreness and dull pain were common in acupuncture but rare or absent in sensory control. Correlations of intensity were seen between aching and soreness, and between heaviness, fullness and numbness. In sensory control, the tapping sensation was occasionally associated with mild tingling, numbness or light pressure. When acupuncture induced the typical deqi without pain, signal attenuation occurred bilaterally in the amygdala, hippocampus, parahippocamupus, hypothalamus, basal forebrain, anterior and posterior cingulate, temporal pole, frontal pole, orbitofrontal and ventromedial prefrontal cortices. The pattern of signal change showed correlation across the hemispheres and across the limbic/paralimbic structures. The secondary somatosensory cortices demonstrated robust signal increases, indicating that the extensive limbic signal attenuation was not the result of global deactivation. Considerable overlap in response patterns was seen between different acupoints. The signal change differed in extent, with LI4, ST36 and LV3 in decreasing order. Tactile stimulation activated the thalamocortical structures, with limited effect on the limbic system.

CONCLUSIONS: An automated analysis of acupuncture fMRI data demonstrated general agreement with our previous reported results by conventional data analysis. The human brain responds in concert to acupuncture at multiple levels, with the limbic system playing a central role. Acupuncture with deqi produced prominent signal decreases. Modulation of the limbic network may constitute an important mechanism of acupuncture action and the neurocorrelate of the deqi response.

 

Back