The relative efficiency of altering pulse frequency or train duration when determining seizure threshold.

In 12 depressed inpatients referred for bilateral electroconvulsive therapy (ECT), each patient was titrated at the first treatment session by using an ascending method-of-limits procedure with a step-wise increase in pulse frequency (frequency titration) or train duration (duration titration). At the second treatment session, seizure threshold was redetermined by using the method (frequency or duration titration) not used at the first treatment. Frequency or duration was maintained at the lowest level when the other parameter was titrated. Seizure threshold was significantly lower with duration titration (mean, 90 mC; SD, 27.3) than frequency titration (mean, 114 mC; SD, 35.6; p = 0.03). On average, patients in the duration-titration group required 1.2 (SD, 0.6) subconvulsive stimulations before a seizure was elicited, and patients in the frequency-titration group required 1.7 (SD, 0.9) subconvulsive stimulations before a seizure was elicited, a nonsignificant difference. These findings suggest that to elicit a seizure during ECT, increasing train duration may be slightly more efficient than increasing frequency. Basic and other clinical research findings indicate that increasing pulse width may be an inefficient way to elicit a seizure. Therefore the following sequence in the determination of seizure threshold is worth considering when using dose-titration or related techniques: the train duration should be increased first before increasing pulse frequency, and the decision to increase pulse width should be reserved for patients who do not seize at the maximal duration and frequency settings. Further empiric research is needed to establish the utility of this approach.