Relationships between seizure duration and seizure threshold and stimulus dosage at electroconvulsive therapy: implications for electroconvulsive therapy practice.

Recent studies cast doubt on the usefulness of seizure duration as an index for stimulus dosage adjustment at electroconvulsive therapy (ECT). A retrospective analysis on the relationships between seizure duration and seizure threshold and stimulus dosage was performed for 54 patients who received a standardized ECT dose titration and treatment protocol. Subjects receiving stimulus dosage at seizure threshold had a highly variable motor seizure duration, ranging from 15 to 89 s, and 7.4% of subjects had seizures longer than 60 s. Regression analysis showed that initial seizure threshold and height were independent predictors of motor seizure duration during dose titration and accounted for 34% of its variance. Stimulus dosage was increased by 40-67% between the first and second ECT session; however, 60% of subjects had no lengthening of seizures and motor seizure duration was significantly shorter at the second treatment (mean +/- SD: 35.7 +/- 12.3 vs 31.0 +/- 10.6 s; paired t = 2.9, d.f. = 34, P = 0.006). At the first and second ECT treatment, motor seizure duration was significantly related to stimulus dosage (r = -0.45 and r = -0.41, respectively). Given the negative correlation between seizure duration and stimulus intensity, a policy that recommends an increase of stimulus dosage and its reduction for patients with short and long seizures, respectively, does not necessarily result in 'optimal' seizure duration. A better dosing policy that places less emphasis on seizure duration is proposed.