Abstracts

Biofeedback is one behavioral intervention that allows representation of unconscious physiological processes in a consciously accessible form (e.g. by means of a visual display), allowing patients to be trained to modulate these processes. A number of biofeedback approaches have been suggested for the management of epilepsy. In a previous study, we have shown a robust effect of Galvanic Skin Response (GSR) biofeedback treatment in reducing frequency of seizures in patients with epilepsy. The rationale of this approach is based on a previous study in which we demonstrated an inverse relationship between GSR and the amplitude of the contingent negative variation (CNV) in healthy subjects, in that increases in peripheral sympathetic activity were associated with reduction in an EEG index of cortical neural excitation. Furthermore, an imaging study has demonstrated that generation of CNV involves a thalamo-cortical circuit and that peripheral GSR activity modulates this circuit. Together, these findings suggest that GSR biofeedback may influence, directly or indirectly, thalamo-cortical sensory regulation circuits, and modulate the seizure threshold by altering thalamic excitatory input to the cortex. Thus it is worthwhile to pursue such treatment related modulations of the CNV in patients with epilepsy. Eighteen patients with drug refractory epilepsy were randomly assigned either to an active GSR biofeedback group (n = 10) or to the sham control biofeedback group (n = 8). Patients participated in a total of 12 sessions over 4 weeks, during which they received either real GSR biofeedback training, where they were trained to decrease skin resistance using biofeedback, or sham training, where the feedback was unrelated to the subjects[rsquo] GSR. Contingent Negative Variation (CNV) was measured before and after the biofeedback and sham control treatments. Biofeedback training significantly reduced patients[rsquo] seizure frequency in the active biofeedback group (p = 0.004), but not the control group (p[gt]0.10). This was manifest as a significant between group differences in seizure reduction (p = 0.007). The CNV amplitude was significantly reduced after training in the GSR biofeedback (active) group, but not in the sham control group. Our findings support the hypothesis that GSR biofeedback training can lead to both a reduction in seizure frequency and CNV amplitude in patients with epilepsy. (Supported by Bial Foundation and Raymond Way Fund.)