EQUIVALENT ELECTRICAL SOURCE ANALYSIS OF PAIN-RELATED SOMATOSENSORY-EVOKED POTENTIALS ELICITED BY A CO2-LASER

The purpose of this study was to localize possible neural sources of pain-related cortical evoked potentials. A brain electrical source analysis was performed on late somatosensory evoked potential data (500-ms window was analyzed) elicited by short heat pulses produced by a CO2 laser. These stimuli activate pain and temperature pathways. The first, fairly small, negative response can be recorded on the scalp about 160 ms from the stimulus (N1 component). The major negativity on the scalp has its peak about 240 ms from the stimulus and is followed by a positivity (N2 and P2 components). A four-dipole model was developed using data following left-hand stimulation. Spatiotemporal source analysis suggested that the N1 component could be generated by contralateral activity in both primary and secondary somatosensory cortices and that N2 could be generated by bilateral activity mainly in secondary somatosensory cortices. A deep dipole in the midline was active during the time range of the negative potentials and seemed to be mostly responsible for the P2 component. This dipole was located too frontally to be thalamic, but it corresponded well to the location of the anterior cingulate gyrus. The model also yielded good fits for right-hand and left-foot stimulation data and, in addition, another set of left-hand data obtained with different electrode spacing in a different group of subjects (residual variances from 2.8% to 3.3%). The model explaining data sets from different body part stimulations varied very little, except with respect to the location of the dipole representing the activity of the primary somatosensory area. The models for the two left-hand stimulation data sets were similar, with dipole locations within a few millimeters of one another.