Electroconvulsive Therapy: Seizure Monitoring

Monitoring of Motor Response
Seizure monitoring has long been an integral component of ECT technique (Weiner et al. 1991). Traditionally, this monitoring has consisted of the detection of the presence and duration of the motor convulsive response. Because of the effects of muscle relaxation, it is usually necessary to prevent the flow of the muscle relaxant to a hand or foot, by means of a blood pressure cuff inflated above systolic pressure, just before infusion of the drug. Generally, it is best to place the cuff on the right ankle when RUL ECT is given, because monitoring on the right will help ensure that the seizure has generalized to both sides of the brain.

Monitoring of Electroencephalographic Response
The seizure produced with ECT involves the intense and highly synchronous discharge of neurons throughout the cerebral hemispheres. It is now recognized that the best way to monitor this process is the use of scalp-recorded electroencephalography (EEG) (American Psychiatric Association Committee on ECT 2000; Weiner et al. 1991). The EEG seizure is typically longer in duration than the ictal motor response. Although the mean difference is only 10-15 seconds, the difference in duration in certain individuals may be much greater, and prolonged seizures can easily be missed when EEG monitoring is not available. For these reasons, single- or double-channel EEG monitoring capability is now incorporated into most ECT devices made in the United States. For a single EEG channel, one recording electrode should be placed with its center located approximately an inch above the midpoint of the patient’s left eyebrow, and the second electrode should be placed on the left upper mastoid region. This placement allows the practitioner to ensure that a RUL ECT seizure has generalized to both cerebral hemispheres. With two EEG channels, the additional two electrodes can be placed over the right hemisphere in an analogous fashion. Use of two channels allows seizure symmetry to be observed and provides backup in cases when there are substantial EEG artifacts.

The EEG seizure typically begins by the end of the stimulus, although it can be delayed for up to 10-15 seconds, particularly if the stimulus is barely suprathreshold. On the EEG, the tonic portion of the ictal motor response is usually characterized by a polyspike format, whereas the clonic portion shows a polyspike-and-slow-wave pattern. The end of the EEG seizure may be either abrupt or gradual. The immediate postictal EEG is usually grossly suppressed and may appear “flat”; the degree of suppression is less with RUL ECT and with barely suprathreshold seizures (Krystal et al. 1993; Nobler et al. 1993).

Determination of Seizure Adequacy
Until recently, seizure duration was assumed to be the primary measure of seizure adequacy, with a typical cutoff criterion of 25 seconds being used clinically on a widespread basis (Weiner et al. 1991). More recent work by Sackeim et al. (1993), showing that barely suprathreshold RUL ECT is not therapeutically effective, has established that the assurance of an “adequate” seizure duration in itself does not ensure that the seizures are therapeutically adequate. Alternative EEG-based measures sensitive to the extent to which the stimulus is suprathreshold would guide the practitioner in delivering the minimum effective stimulus intensity at each treatment, thereby maximizing risk-benefit considerations (Krystal 1998). A number of recent studies suggest that attributes of EEG data recorded during the induced seizures, such as the amplitude and morphological regularity of slow waves and the degree of postictal suppression, have promise in this regard (Krystal 1998). It should be noted, however, that no EEG-based models have yet been validated for use in clinical dosing, though validation studies are under way.

Management of Missed or Inadequate Seizures
Missed seizures are lacking in therapeutic efficacy and should be followed by an increase in stimulus intensity (e.g., 50%) after an interval of approximately 20 seconds (to allow for the possibility of a delayed ictal response) (American Psychiatric Association Committee on ECT 2000). If the patient’s seizure threshold has exceeded the maximum stimulus intensity available with the ECT device used, the threshold can be decreased by lowering the dose of anesthetic agent (if possible), by switching to ketamine as the anesthetic agent (1-2.2 mg/kg iv), or by stopping or decreasing the dose of any concurrent medications with anticonvulsant properties. Brief (abortive) seizures, or those that are deemed to be otherwise inadequate, should be followed by restimulation at a higher stimulus intensity after an interval of 45 seconds (to allow for the effects of a relative refractory period to pass).

Management of Prolonged Seizures
Prolonged seizures, defined as lasting longer than 3 minutes, should be aborted pharmacologically (American Psychiatric Association Committee on ECT 2000). If two trials of ultrabrief anesthetic agent or benzodiazepine (e.g., midazolam 1-2 mg iv) are ineffective, a longer-acting benzodiazepine should be used, together with a loading dose of phenytoin, intubation, and neurological consultation (Weiner et al. 1991). The practitioner should be sure that the seizure is in fact continuing, because a poorly suppressed postictal EEG or an inappropriately high EEG gain setting can be misinterpreted as a prolonged seizure. One rule of thumb is that the seizure is almost certainly over if the patient has resumed spontaneous respiration.