Topography of diprenorphine binding in human cingulate gyrus and adjacent cortex derived from coregistered PET and MR images

Abstract

AbstractPositron emission tomography (PET) studies of ligand binding lack sufficient anatomical detail to evaluate topographical variations in binding within each of the lobes of the human cerebral cortex. This study employed PET to localize [11C]diprenorphine binding to opioid receptors and magnetic resonance (MR) imaging for defining medial surface structures. Continuous arterial sampling for metabolite corrected [11C]diprenorphine levels and CNS blood flow were used to model the volume of distribution (VDtot) of binding for three subjects. The PET images of VDtot were coregistered to the MR images for each case and 37 regions of interest were used to calculate VDtot. The VDtot was averaged for the three cases and coregistered with an MR reconstruction of the medial surface and plotted onto a flat map of this region.The average VDtot showed that binding was highest in anterior cingulate, rostral cingulofrontal transition, and prefrontal cortices, while binding in caudal parts of anterior cingulate and superior frontal cortices, and posterior cingulate cortex varied from high to low. Three statistical levels of binding were defined in relation to the high binding in perigenual area 24: high and equal to area 24, moderate and significantly lower than area 24 (p <0.01), or low (p <0.001). These levels of binding were plotted onto an unfolded map of the medial cortex. The VDtot was high in rostral cortex, and a strip of high binding continued caudally on the dorsal lip of the cingulate gyrus. There were patches of high binding in cinguloparietal transition, posterior parietal, and supplementary motor cortices. Four regions had low binding: (1) areas 29 and 30 in the callosal sulcus, (2) fundus of the cingulate sulcus likely involving the cingulate motor areas, (3) fundus of the superior cingulate sulcus involving two divisions of supplementary motor cortex, and (4) sensorimotor cortex on the paracentral lobule. Variations in binding may reflect functional specializations such as low binding in the cingulate motor and visuospatial areas and high levels in areas involved in processing information with affective content. The higher sensitivity of three‐dimensional scanning and coregistration of PET and MR images makes it feasible to analyze single individuals and, by performing pixel‐by‐pixel spectral analysis and generation of parametric maps, statistical analyses are possible. © 1995 Wiley‐Liss, Inc.