Enduring Interest: Generic Substitution | Genius | Groundbreaking Paradigm Shifts | Zmail


Electroencephalograms or EEGs are a basic screening measure of brain waves which is performed in neuropsychiatry both to detect abnormal firing in the brain and to find local (focal) abnormalities. Both sleep and wake EEGs with activating procedures such as hyperventilation and photic stimulation are useful as each can give valuable information and demonstrate abnormalities. Hyperventilation - overbreathing - which changes the proportion of carbon dioxide in the body and therefore the acidity / alkalinity level should be performed only in the absence of medical conditions contra-indicating it. EEGs appear reasonable in neuropsychiatric evaluations when there are seizure history possibilities or possible temporolimbic features (as reflected on the INSET), or with the episodic nature of symptoms or a history of atypical spells . Sleep records have been well demonstrated to more likely find focal pathology than waking EEGs which is why they are generally routinely performed now. However waking EEGs have also have a high pick up rate and sleep EEGs cannot be interpreted without the wake EEG justifying a wake record.

Prior to the development of the EEG, by the neuropsychiatrist, Hans Berger in the 1930's, all seizure disorders were classified with mental disorders. EEG technology still remains rather primitive and reflections of brain waves from the perspective of analysis of psychopathology somewhat limited. Nevertheless, the only definitive way of demonstrating that a symptom or physical sign such as, for example, an olfactory hallucination is definitely epileptic, is the demonstration while the person is having that experience of correlates of seizure phenomena on EEG, such as spike-wave paroxysms - episodes of half to several seconds of usually sharp abnormal brain waves, sometimes localized in the brain e.g. in the right temporal lobe. This demonstration is unusual unless the seizure phenomena are relatively uncontrolled, as an EEG is just a short cross-sectional measure for an hour or two of a patient's life-cycle. Even in the event of the patient having an experience which may be a seizure, the EEG correlate may not necessarily be of a spike wave kind but depending on location, it could be normal or show a marked slowing, with a non-specific theta rhythm generally of limited help unless focal or a delta rhythm, which is frankly abnormal unless the patient is asleep (theta is 4 to 7 cycles per second, delta is less than 4). It is occasionally extremely difficult to localize such features on scalp EEG even when firing is occurring because symptoms may occur from the mesial temporal or deep structures within the brain which do not easily manifest on surface EEGs.

Routine Electroencephalograms (EEGs) involve both waking records with special activating procedures such as hyperventilation and photic stimulation (in the absence of medical conditions contra-indicating these) as well as sleep records. EEGs should be ordered not only in possible seizure disorder , but appears reasonable given any possible temporolimbic features, episodic nature of symptoms or history of atypical spells. Sleep records may increase the potential delineation of focal abnormality such as a temporal lobe focus by approximately fourfold than waking EEGs. However waking EEGs have a high pick up rate and sleep EEGs cannot be interpreted without the wake EEG so both should be performed. A normal EEG does not imply absence of epilepsy.

EEGs are possibly under-used in psychiatry partly because electroencephalographers have a broader range of what constitutes normality searching mainly for focal and seizure phenomena. They are generally not psychiatrists and potentially valuable research and clinical information may be lost. For example, testable hypotheses are that relatively flat EEG tracings may be more common in certain personality disorders, with certain psychotropics, or in a subpopulation of schizophrenia. Seldom is this kind of background even reported on.

The administration of chloral hydrate (e.g. 1 gram as premedication) prior to the sleep record is useful as this induces sleep with little changes of significance in the electroencephalogram and does not prevent the demonstration of focal abnormalities. Certain medications should be particularly avoided in EEGs. The benzodiazepine group are the worst offenders as by virtue of their very strong anti-epileptic effects, they have profound effects in normalizing the EEG. Such effects at a receptor level may last weeks even with the apparent short acting benzodiazepines so that the yield of demonstrating epilepsy after the patient has had benzodiazepines administered apparently decreases substantially, although adequate data in this regard is not easily available.

Special electrode placements may increase yield by a few percent, but are seldom used today: With nasopharyngeal electrodes, the greater yield was insubstantial; and sphenoidal electrodes placement, unfortunately, requires time and expertise and cause discomfort limiting their use. A recent suggestion which I recommend has been the placement of electrodes on the buccal skin surface in the area of the submandibular notch - possibly as effective in picking up foci as sphenoidal placements. Specialized centers use cerebral cortical or submeningeal strip placements during neurosurgery procedures and these may show firing, for example, in patients with temporal lobe epilepsy and psychosis, in the region of the hippocampus. The direct placement of intracranial electrodes shows how commonly spike firing may be occurring in this area with no correlate of any kind on surface EEGs.

Developments in this regard have been rapid over the past few years. EEG Telemetry involves prolonged monitoring over periods of time varying from 12 hours to 2 weeks while the patient is generally confined to a particular room. Cable telemetry is most commonly used. This involves, for example, a 25 foot cable connected to the EEG montage on the patient's head. Very often no seizure manifestations are picked up for prolonged periods of time because seizures only occur paroxysmally. Moreover, those patients evaluated in a specialized center with EEG telemetry are invariably so atypical that the hypothesized seizure originates deep within the brain. The apparatus is very expensive and the costs involved in monitoring patients are thousands per day at times for two weeks. Instead, home ambulatory electroencephalograms are easily available and should in psychiatry become the state of the art.

Ambulatory Electroencephalogram

Home Ambulatory Electroencephalograms (EEG) with the patient not modifying medication is a valuable test as the patient's symptomatology can be monitored day and night in a natural environment of home using computerized filtering of artefact. The advantage of this technique is to establish if any scalp electrode can detect events such as atypical spells alerted to by pushbuttons which could be reflecting deep brain electrical activity . It has limited availability at this point, however, but our pick up rate for atypical spells (paroxysmal neurobehavioral disorder) and seizures is very high - a major advance over routine electroencephalography.

Recent advances in EEG technology may ultimately change the whole perspective in its use in psychiatry. Computerized EEG monitoring allows breakdown of wave forms and allows correlation with evoked potentials including cognitive evoked potentials. It also facilitates demonstrations of changes in particular areas of the brain which can be easily delineated at a visual level. This should prove to be a useful psychophysiological correlate of psychopathology. Indeed, this may be the beginning of an important new era. However, at this point in time it is still experimental.

Ambulatory Electroencephalogram (EEG) with the patient not modifying medication is a valuable test given episodic symptomatology which can be monitored day and night in a natural environment of home using computerized filtering of artefact. One advantage of this technique here is to establish if any scalp electrode can detect events such as atypical spells alerted to by pushbuttons reflecting deep brain electrical activity.




Copyright ©1997-2003 Pacific Neuropsychiatric Institute