Electroconvulsive Therapy: Stimulus Electrode Placement
For many years, stimulus electrodes were applied bitemporally, with midpoints of the electrodes approximately 1 inch above the midpoint of a line connecting the external canthus with the upper tip of the tragus (the small cartilaginous structure in the anterior region of the external ear). This type of electrode placement is called bilateral (BL) ECT. In 1958, Lancaster and colleagues developed an alternative placement in which the stimulus electrodes were both placed over the cerebral hemisphere opposite the one believed to be dominant for speech, thereby relatively sparing verbal memory function (Lancaster et al. 1958). This technique is called right unilateral (RUL) ECT.

Since that study, there have been dozens of additional investigations focusing on the efficacy and safety of various RUL types versus the more traditional BL ECT (reviewed in Abrams 1997a). These studies have, for the most part, tended to demonstrate three general types of findings: 1) memory impairment, particularly left hemisphere function, is less likely, less severe, and less persistent with RUL ECT; 2) the antidepressant effect of BL ECT may be more rapid or pronounced than with RUL ECT, at least under certain circumstances; and 3) optimization of RUL technique can maximize its therapeutic potency. It is now known, for example, that RUL ECT administered using a frontotemporal to high centroparietal (just lateral to the vertex of the scalp) configuration (d’Elia 1970) in combination with a stimulus intensity that is at least moderately suprathreshold has significant efficacy, though not necessarily equivalent to BL ECT in some patients (Sackeim et al. 1991, 1993). More recent preliminary data indicate that when RUL ECT is administered at a stimulus intensity that is markedly above the seizure threshold in intensity (500% above threshold), the efficacy may become equivalent to that of BL ECT (Sackeim et al. 2000). Interestingly, although such increases in RUL ECT stimulus intensity are associated with a rise in level of amnesia, this effect is less than that observed with BL ECT. Further studies are needed to indicate how to optimally administer RUL ECT and to determine when BL ECT may be needed. In the meantime, each practitioner must choose the electrode placement and stimulus intensity on the basis of a case-by-case analysis of anticipated benefits versus likely risks (American Psychiatric Association Committee on ECT 2000).

Recently, a bifrontal electrode position has been suggested as a means to optimize risk-benefit considerations (Letemendia et al. 1993), although insufficient data are yet available to ascertain the merits of this technique. A similar situation exists with regard to a proposed left frontal to right frontotemporal positioning (Swartz 1994).

Regardless of the type of electrode placement, it is important that the impedance of the contact zone between the stimulus electrodes and the scalp be kept as low as possible to ensure adequate current flow during the stimulus (American Psychiatric Association Committee on ECT 2000). This end can be accomplished by cleansing the underlying scalp with alcohol, rubbing it with some rough gauze or abrasive material, and applying a conductive gel.

Electrical Stimulus

Electrical Principles Pertinent to ECT
The electrical stimulus is applied after maximal muscular relaxation has been attained and good electrical continuity (i.e., a relatively low interelectrode impedance) is present. With most contemporary ECT devices made in the United States, the electrical continuity can be tested before stimulation via a self-test feature. This feature involves passing an extremely low-intensity stimulus (below the sensory threshold) across the entire path followed by the stimulus, including the patient. The interelectrode impedance measured during the self-test procedure is termed the static impedance (typically around 400-3,000 ohms), as opposed to the dynamic impedance (ranging from 150 to 350 ohms), which is the interelectrode impedance during actual stimulus delivery.

At present, ECT devices are produced by three companies in the United States: Medcraft, Inc. (Darien, CT); MECTA, Inc. (Lake Oswego, OR); and Somatics, Inc. (Lake Bluff, IL). The signal used for the actual ECT stimulus, at least in the United States, is the bidirectional brief pulse. This signal, which is relatively efficient in its ability to induce a generalized seizure, is characterized by pulse width (typically 0.5-2.0 msec), pulse frequency (typically 40-90 Hz, or pulse pairs per second), duration of the entire train of pulses (e.g., 0.5-8.0 seconds), and the peak current of each pulse (e.g., 0.5-1.0 amp). A single composite stimulus intensity parameter is charge (in millicoulombs), which is equal to two times the product of pulse width, pulse frequency, duration, and peak current.

Stimulus Dosing
The minimum stimulus intensity required to induce a generalized seizure is the seizure threshold. Seizure threshold varies as much as 50-fold across patients and is influenced by factors such as gender (is higher for men), age (increases with age), treatment number (increases with a higher number of treatments), and stimulus electrode placement (is lower with RUL, at least with the d’Elia placement) (American Psychiatric Association Committee on ECT 2000; Sackeim et al. 1991). Increasing evidence indicates the importance of the seizure threshold in stimulus dosing, particularly for RUL ECT. As mentioned above, recent studies indicate that, for RUL ECT, efficacy increases as a function of how much the stimulus intensity exceeds the seizure threshold, with RUL ECT possibly becoming equal in efficacy to BL ECT at 500% above threshold (Sackeim et al. 1993, 2000). As noted above, increasing stimulus intensity is associated with greater cognitive impairment for both RUL ECT and BL ECT (Sackeim et al. 1993, 2000).

Based on these relationships, the most precise method for stimulus dosing is to determine the seizure threshold with a titration procedure at the first treatment and to then administer a stimulus at a chosen degree above the threshold at subsequent treatments. A dosing schedule using known relationships between factors such as gender, electrode placement, and/or age can be used to minimize the number of restimulations at the first treatment with this technique (Coffey et al. 1995). As an alternative, fixed formula-based dosing techniques have also been proposed, utilizing factors influencing seizure threshold as noted above (American Psychiatric Association Committee on ECT 2000); however, because of the high interindividual variability in seizure threshold, such techniques are not able to accurately match the stimulus to the patient’s seizure threshold. Further studies are needed to identify a strategy that will allow optimal stimulus dosing for each patient.