Non-invasive evaluation of input-output characteristics of sensorimotor cerebral areas in healthy humans

Abnormalities of short-latency somatosensory evoked potentials in parkinsonian patients

Twenty-two patients (16 affected by parkinsonian syndromes, 6 by other neurological diseases) and 12 age-matched controls were examined. Short-latency somatosensory evoked potentials were recorded from 30 scalp electrodes in the 45-52 msec following separate left and right median nerve stimulation at the wrist. Bit-colour maps were generated on a 4096 pixel matrix via quadratic interpolation. Peak latencies and amplitudes of the parietal, central and frontal components were evaluated. Moreover, the amplitude ratios between parietal and frontal components on the same hemiscalp and between peaks on homologous right and left scalp districts were taken into account. The unique significant difference between parkinsonians and controls was represented by a depressed frontal N30 wave. This peak was absent in 3 and reduced in 7 out of 16 parkinsonians, with an overall abnormality rate of 47% of the examined arms. Average maps pooling data of parkinsonians and controls confirmed the presence of reduced evoked activity for the whole duration of wave N30 on those mid- and parasagittal frontal districts where this peak is maximally represented in normals. A similar abnormality was found in 1 of the 6 non-parkinsonian neurological patients suffering from a meningioma of the falx compressing the left supplementary motor area. Possible pathophysiology of such wave N30 abnormalities in parkinsonians is discussed.

Short latency somatosensory evoked responses to median nerve stimulation in healthy humans: electric and magnetic recordings

Somatosensory Evoked Potentials (SEPs) and Somatosensory Evoked magnetic Fields (SEFs) to median nerve stimulation at wrist were recorded in 5 healthy subjects and the components between 15 and 30 ms after the stimulus were evaluated on the hemiscalp contralateral to the stimulated wrist. SEPs were measured by means of a 32-channel recorder and compared with SEFs obtained via multiple measurements with a 4-channel sensor. Equivalent dipole localization was carried out for the magnetic components peaking at about 15, 20 and 24 ms. The scalp distribution of SEPs, illustrated by bit mapped color images, were qualitatively explained by three separate sources. The first is described as a tangentially oriented dipole placed behind the Central Sulcus and responsible for the parietal N20-"late P25" waves and for the frontal P20-N30 ones. The second is represented by a radieal dipole placed just in front of the Central Sulcus and pointing towards the motor strip, responsible for the rolandic P22 component. The third is just behind the Central Sulcus and is radieally oriented towards the surface of the postcentral sensory area for the "early P25" parietal wave. The SEFs distributions, illustrated by color isofield contour maps, were quantitatively explained by a unique tangential dipole localized, with good resolution, well behind the Sulcus for the 15 ms waves and slightly frontal to this site for the waves peaking at around 20 and 24 ms. The equivalent dipole has been localized at a depth of about 5 cm (15 ms component), 2 cm (20 ms components) and 4 cm (24 ms component), across the studied subjects. It is stressed that the dipole responsible for the magnetic pattern is likely to be the same tangential dipole responsible for a part of the electric pattern. Due to their radieal orientation, the other two dipoles, proposed for the SEPs maps, would be mostly undetectable by a magnetic investigation.

Scalp topography of somatosensory evoked potentials following electrical stimulation of femoral nerve

Up to 29 channels of somatosensory evoked potentials (SEPs) were recorded in 10 normal volunteers following unilateral femoral nerve (FN) and tibial nerve (TN) electrical stimulation. Typical short latency FN SEPs consisted of 6 components, P15, N19, P26, N34, P44 and N56. P15 and N19 were widely distributed on the scalp. The first localized scalp component, P26, was strictly postrolandic and distributed on the contralateral parietal scalp close to midline with a prerolandic phase reversal, N26. This scalp distribution is clearly different from the first localized potential of tibial nerve SEPs. N34 and P44 were maximal at the vertex with a distribution that spread to the ipsilateral central and parietal scalp. The amplitude of P26 increased and latency shortened with increasing stimulus intensity and both values plateaued after the stimulus intensity reached motor threshold. No correlation was found between the peak latency of P26 and body height.

Simultaneous motor output and sensory input: cortical interference site resolved in humans via neuromagnetic measurements

Neuromagnetic measurements were carried out during median nerve stimulation at the wrist in complete relaxation (a) and during active contraction of the hand muscles (b). Firstly, activity of the generator source responsible for the major component of the sensory evoked magnetic fields (SEFs) was mapped and localized during condition (a) and (b). Then the subtraction maps ((a)-(b)) were obtained and the virtual 'dipole' responsible for the 'interference' (less than or equal to SEFs amplitude) between the afferent input and the motor output was tridimensionally localized in a position compatible with the knee and the convexity of the postcentral gyrus.

Frontal distribution of early cortical somatosensory evoked potentials to median nerve stimulation

The topography of early frontal SEPs (P20 and N26) to left median nerve stimulation was studied in 30 normal subjects and 3 patients with the left frontal bone defect. The amplitudes of P20 and N26 were maximum at the frontal electrode (F4) contralateral to the stimulation and markedly decreased at frontal electrodes ipsilateral to the site of stimulation. There was, however, no latency difference of P20 and N26 between ipsilateral and contralateral frontal electrodes. These results suggest that the origin of the ipsilateral and contralateral P20 and N26 is the same. The wide distribution of P20 and N26 over both frontal areas could be explained by assuming a smearing effect from generators actually located in the rolandic fissure and motor cortex.

Methodology for non-invasive mapping of human motor cortex with electrical stimulation

Different techniques for mapping motor evoked potentials recorded from hand, upper arm, leg and mouth were analyzed. The best results were obtained when: (1) delivering constant voltage stimuli through a bipolar surface stimulator, (2) positioning the anode over the desired scalp location and the cathode 2.5 cm anterior to the anode, (3) maintaining low impedances, and (4) increasing the stimulus intensity over the theoretical motor representation area until a 500-1000 microV muscle response is achieved and then delivering the same stimulus over different scalp locations. This technique allows bilateral mapping of the different body part representations in 1-2 h. Areas so identified are small, clearly separate from each other and from the corresponding somatosensory areas.

Neuromagnetic functional localization: principles, state of the art, and perspectives

The neuromagnetic method has been providing impressive results in the understanding of the functions and of some pathologies of the human brain. The possibility of achieving three-dimensional source localization represents a fundamental step forward in the study of the organization of cortical areas, in that of focal disorders and, in general, in the investigation of brain information processing. The development of large multichannel systems to achieve real time functional localization is being carried on in several countries and the first prototypes are already operating for a full assessment of the benefits of the technique. The use of time-varying magnetic pulses provides the opportunity for non-invasive stimulation of the central nervous system toward the achievement of functional imaging of the motor cortices.

Monitoring of subcortical and cortical somatosensory evoked potentials during carotid endarterectomy: comparison with stump pressure levels

Monitoring of multichannel somatosensory evoked potentials (SEPs) has been performed in 40 cases of carotid endarterectomy (CEA). SEPs were obtained after median nerve stimulation at wrist, recording from 2nd cervical and from the scalp parietal (ipsi- and contralateral) and central (contralateral) positions. The reduction of CBF due to clamping of the carotid artery provoked SEP abnormalities in 10 of the 40 cases. None of the 30 patients with unmodified SEPs developed post-surgical neurological sequelae. SEP alterations were characterized exclusively by amplitude decrements and latency increases of the cortical components, the subcortical ones being unaffected. In 5 of these patients, SEPs returned to normal values before the end of the intervention and no neurological deficit was observed on awakening. In the remaining 5 cases SEPs retained their abnormalities and patients developed post-surgery neurological sequelae (4 immediately, 1 the day after). SEP alterations affected parietal and central components to a similar extent; however, in a few cases cerebral blood flow deficits provoked by carotid clamping modified differently the central P22 and the parietal N20-P25 waves. Comparisons with stump (back) pressure in the carotid artery revealed a higher sensitivity of the SEP technique in detecting vascularization problems due to carotid clamping. The time course of the appearance of SEP abnormalities seems to discriminate alterations secondary to collateral revascularization from those determined by embolization.