Pain processing in four regions of human cingulate cortex localized with co-registered PET and MR imaging.

Neurosurgical and positron emission tomography (PET) human studies and animal electrophysiological studies show that part of the anterior cingulate cortex (ACC) is nociceptive. Since the contribution of the ACC to pain processing is poorly understood, this study employed PET and magnetic resonance (MR) image co-registration in grouped and individual cases to locate regions of altered relative regional cerebral blood flow (rCBF). Seven right-handed, neurologically intact males were subjects; each received neuropsychological and pain threshold testing. Subjects were scanned during infusion of H2[15O]: four randomized scans during innocuous heat stimulation to the back of the left hand and four scans during noxious but bearable heat to the same place. The averaged rCBF values during innocuous stimuli were subtracted from those during noxious stimuli and statistical parametric maps (SPMs) for the group were computed to identify regions of altered relative rCBF. Finally, single-subject PET images of elevated and reduced rCBF were co-registered with MR images and projected onto reconstructions of the medial surface of the hemisphere. The SPM analysis of the group showed one site with elevated rCBF in the midcingulate cortex and one in the perigenual cortex predominantly contralateral to the side of stimulation. There were bilateral sites of reduced rCBF in the cingulofrontal transitional cortex and in the posterior cingulate cortex (PCC). Co-registered PET and MR images for individuals showed that only one case had a single, large region of elevated rCBF, while the others had a number of smaller regions. Six cases had at least one significant elevation of rCBF in the right hemisphere that primarily involved area 24b'; five of these cases also had an elevation in area 32', while the seventh case had elevated rCBF in these areas in the left hemisphere. The rostral site of elevated rCBF in the group was at the border of areas 24/24' and areas 32/32' although most cases had a site of elevation more rostral in the perigenual cingulate cortex. The ACC site of reduced rCBF was in areas 8 and 32 and that in the PCC included much of areas 29/30 in the callosal sulcus, areas 23b and 31 on the cingulate gyral surface and parietal area 7m. The localization of relative rCBF changes suggests different roles for the cingulate cortex in pain processing: (i) elevated rCBF in area 24' may be involved in response selection like nocifensive reflex inhibition; (ii) activation of the perigenual cortex may participate in affective responses to noxious stimuli like suffering associated with pain; and (iii) reduced rCBF in areas 8 and 32 may enhance pain perception in the perigenual cortex, while that in the PCC may disengage visually guided processes.