DC-EEG recording: a paradigm shift in seizure localization?

Interictal or ictal discharges are the basis of EEG interpretation when seizure localization is the goal, but a new report suggests an alternative. It has been known for over 40 years that, when DC recordings are made, seizures may be associated with localized, very slow, sustained voltage changes, which do not oscillate (ictal DC shift).1 These voltage changes occur more or less simultaneously with the rhythmic oscillating (AC) voltages that are the conventional EEG correlate of seizures. In this issue of Neurology, Vanhatalo et al.2 report prolonged DC-EEG recordings from five patients with temporal lobe epilepsy who were monitored for seizure localization by video/ EEG. They conclude that DC-EEG recordings at the bedside are practical and that ictal DC shifts are consistently observed in scalp recordings and reliably lateralize temporal lobe seizures. The major impediment to this type of recording is that it requires a genuine DC-EEG amplifier and nonpolarizable (silver–silver chloride) electrodes capable of recording for several days without excessive baseline drift. Advances in amplifier technology have overcome these hurdles: electrode holders containing a large volume of electrode gel make airtight contact between the holder and skin, and slow potentials generated from the skin are short-circuited. DC-EEG amplifiers are no more expensive to build than EEG amplifiers now in use, and electrode application is nearly as quick as with conventional electrodes. When proper equipment is available, trained technicians apply the electrodes, and electroencephalographers familiar with various sources of artifact interpret the recordings, good DC recordings are feasible. We anticipate that this technology will be used most in comprehensive epilepsy centers, to localize seizure discharges for epilepsy surgery. Antiepileptic drug therapy fails to control seizures in 30 to 40% of patients with epilepsy.3 Patients with refractory partial seizures may be candidates for surgical treatment, and a recent randomized study in patients with such seizures found surgical management (58% seizure-free) superior to medical management (8% seizure-free).4 Is there a need for a new seizure localization technique? Noninvasive presurgical evaluation of patients with intractable partial epilepsy now requires use of neuroimaging techniques including structural MRI to identify pathologic conditions underlying the epileptogenic zone, such as tumors, heterotopias, or mesial temporal sclerosis, that correlate with long video-EEG recordings from the scalp.5 In some instances, peri-ictal SPECT, PET, or MRS may assist in localizing the site of seizure onset. Seizure localization still may be uncertain because studies are negative or equivocal, abnormalities are multifocal, or localization suggested by various studies is discordant. In such cases, video-EEG monitoring with implanted intracranial electrodes may be performed, but the procedure is not without risk. A new noninvasive technique that provides independent information about seizure localization might reduce the need for invasive monitoring. Is long-term DC-EEG recording such a technique? Vanhatalo et al.2 make a case for its feasibility, costeffectiveness, and reliability. They suggest that it may lateralize temporal lobe seizures even when the conventionally monitored ictal discharge appears to be bilateral. We agree. We also agree with the authors’ opinion that additional studies would help to bolster the case for adopting DC-EEG, particularly with respect to costeffectiveness. Prospective studies are needed to assess its reliability in more patients and to demonstrate how frequently it provides additional independent information in temporal lobe seizure patients whose scalprecorded ictal EEG gives unclear lateralization. No technique is 100% reliable, and there is the possibility that neurologists using DC-EEG may find that it provides information that is sometimes discordant with other techniques, including convention-