Page 204 - ILAE_Lectures_2015

Chapter 20

Neurophysiological investigation of epilepsy

SHELAGH J.M. SMITH1 and ROBIN P. KENNETT2

1National Hospital for Neurology and Neurosurgery, Queen Square, London, and Epilepsy
Society, Chalfont St Peter, Buckinghamshire, and 2John Radcliffe Hospital, Oxford

Routine EEG recording

Since its discovery in the late 1920s, the electroencephalogram (EEG) has developed into an
array of digitally-based techniques, integrated with video and other investigative modalities
that retain a central role in diagnosis and management of patients with seizure disorders. In
contrast to the technological advances, relatively little progress has been made in
understanding the cerebral generators of EEG signals, in part because of their anatomical
complexity. Much of what appears on scalp EEG recordings represents the summation of
synchronised excitatory or inhibitory post-synaptic potentials in apical dendrites of neurones
in superficial neocortex, while deep generators may produce little or no change at the surface.

EEG is not always used appropriately in epilepsy care; evidence-based guidance on its role
is limited by a paucity of high quality data and methodological deficiencies in published
studies1. Furthermore, the limitations of scalp EEG may not be fully appreciated by non-
specialists, and EEGs can be misinterpreted if there is insufficient knowledge of the range of
normal and non-specific phenomena – this being a common reason for over-diagnosis of
epilepsy2.

In general, sensitivity of routine inter-ictal EEG for detecting epilepsy and its specificity for
distinguishing epilepsy from other paroxysmal disorders are both limited. Published figures
for diagnostic sensitivity of EEG range between 25 and 55%; up to about half of patients with
epileptic disorders may have one normal inter-ictal EEG, and around 10% of patients with
epilepsy never show epileptiform discharges3. Hence, a normal or negative EEG cannot be
used to rule out the clinical diagnosis of an epileptic seizure.

Importantly, demonstration of epileptiform abnormalities in the EEG does not in itself equate
to epilepsy or indicate that the patient has a seizure disorder. Non-epileptic individuals show
epileptiform abnormalities in the EEG in a number of circumstances. A large study of
standard EEGs in healthy mostly male adults with no declared history of seizures showed
epileptiform discharge in 0.5% of the subjects4. A slightly higher incidence of 24% is found
in healthy children and in non-epileptic patients seen in hospital EEG clinics. The incidence
rises substantially to 1030% in patients with cerebral pathologies such as tumours, previous
head injury, cranial surgery or congenital brain injury5, and patients with pure psychogenic
type non-epileptic seizures have a higher incidence of epileptiform EEGs than normal
subjects6. Thus, great caution is necessary when assessing the significance of epileptiform
activity in such circumstances, particularly if the history offers little or no indication that the
patient has epilepsy on clinical grounds.