Negative potential shifts and the prediction of the outcome of neurofeedback therapy in epilepsy
Introduction
Birbaumer et al. (1991) hypothesized that patients with intractable epilepsy are characterized by an impaired ability to regulate their level of cortical activation using cortico-thalamic feedback loops. The cortical activation level is assumed to manifest itself in oscillations of slow cortical potentials (SCP), with negative SCP shifts reflecting an increase, and positive SCP shifts, a decrease in the excitability of the underlying cortical networks (Birbaumer et al., 1990, Birbaumer et al., 1992). Studies with healthy subjects demonstrated the ability of humans to learn the cortical self-regulation (Elbert et al., 1980, Lutzenberger et al., 1980). This led to the idea that epilepsy patients can acquire the lacking cortical self-regulation during learning. According to this hypothesis, a neurofeedback method was developed in which actual SCP changes are presented to epilepsy patients in the form of a moving object on a screen. In fact, using this method, most patients with drug-resistant epilepsy could learn to control their SCP, resulting in a significant and lasting decrement of the seizure rate (Birbaumer et al., 1991, Daum et al., 1993, Rockstroh et al., 1993, Kotchoubey et al., 1996, Kotchoubey et al., 1997).
Some important questions, however, remained unanswered. Patients in this paradigm learned bi-directional SCP responses, producing either cortical negativity or cortical positivity in different trials. Further, they learned to transfer these skills to a condition in which they received no continuous feedback of their SCP. The results of the studies cited above were not consistent with respect to the factors accounting most for seizure reduction. This could be the ability to generate cortical positivity, or to differentiate between the positivity and negativity waves, or to apply any of these skills to a condition without feedback. About a third of the patients treated with this technique did not profit from SCP-neurofeedback training; factors which distinguish them from patients with clinical improvement are as yet unknown. One may speculate that different clinical outcomes were related to pre-existing individual differences in SCP responsivity, however, such consistent relationships were not found in the previous studies, probably because of the small number of patients or training sessions, or both.
Section snippets
Methods
In the present study, 28 patients with focal epilepsy (16 females, mean age 32.4 years (range 21–45), mean seizure history 22.7 years (range 6–40)) participated. As one patient failed to complete his seizure diary, the data of 27 patients were analyzed. Nine of them had a right temporal focus, 7 patients had a left temporal focus, and the remaining suffered from multi-focal epilepsy. All patients were regarded as drug-resistant with at least 2 years of well-balanced pharmacological treatment
Results
As shown in Fig. 1, differences between the SCP obtained with negativity task versus positivity task increased across sessions, particularly in trials with feedback. In FB trials, the amplitude difference between the two tasks (SCP differentiation) increased linearly as a function of session number (F(2,33)=20.3, P<0.001). In TF trials, this linear trend did not reach significance (F(2,33)=1.9, P=0.18), while the cubic trend was significant (F(3,31)=3.0, P=0.045), indicating that periods of
Discussion
In the present study, epilepsy patients learned to control their SCP (main effect of Task), with the amount of control increasing across sessions (significant linear and cubic trends). They were more successful with than without feedback (interaction Task×Trial-type). The average seizure rate decreased, on average, by 25% (about 60% in the improvement group), but no tendency for clinical improvement was observed in about a third of the sample. These results closely replicate those obtained in
Acknowledgements
The study was supported by the German Research Society (DFG). The authors are very much indebted to Michaela König for her valuable technical help.
References (14)
- et al.
Biofeedback of slow cortical potentials
Electroenceph clin Neurophysiol
(1980) - et al.
Ictal EEG wave forms from epidural electrodes predictive of seizure control after temporal lobectomy
Electroenceph clin Neurophysiol
(1992) - et al.
The clinical and prognostic relevance of the postictal slow focus in the electroencephalogram
Electroenceph clin Neurophysiol
(1995) - et al.
Self-regulation of slow cortical potentials in epilepsy: a retrial with analysis of influencing factors
Epilepsy Res
(1996) - et al.
Benzodiazepine sensitivity testing in the management of intractable seizure disorders in childhood
Electroenceph clin Neurophysiol
(1987) - et al.
Biofeedback of slow cortical potentials
II. Analysis of single event-related slow potentials by time series analysis, Electroenceph clin Neurophysiol
(1980) - et al.
Cortical self-regulation in patients with epilepsies
Epilepsy Res
(1993)