Several studies conclude that right unilateral (RUL) electroconvulsive therapy (ECT) given with ultrabrief pulse-width stimulus packages causes less cognitive impairment than other techniques.^1-4 I review this evidence and discuss how these findings can be interpreted to direct optimal clinical practice.

While ECT remains a remarkably safe and effective treatment for severe depression, its broad application has been hampered by concerns—both perceived and real—about its cognitive effects.⁵ Worries about memory loss make some patients reluctant to undergo this therapy and some practitioners reluctant to refer patients for it. Within the field of ECT itself, there has been tension for some years between the wish to maximize (the already excellent) antidepressant and antipsychotic efficacy of ECT and the competing wish to minimize any effects on memory.

This tension has resulted in several important enhancements in ECT technique which, taken together, have made contemporary ECT far better than the ECT of decades past. Major enhancements to ECT technique include RUL electrode placement, brief pulse stimulus currents, improved an­esthesia techniques, the “dose titra­tion” technique to individualize treatment stimulus dosing, and the discovery of the dose-response relationship between electrical stimulus dosing and antidepressant outcome (largely with RUL electrode placement.)^6-11

Is ultrabrief pulse stimulus ECT to be the next advancement added to this list?

It has long been established that the induction of a cerebral seizure is central to the neurobiological changes that result in the antidepressant effects of ECT.⁵ For decades, an electrical stimulus has been used to initiate this seizure event, replacing earlier methods that used chemicals. Interestingly, the characteristics of the electrical stimulus affect the effectiveness and cognitive outcomes of the seizure.⁷ Sine wave stimuli have been replaced with rectangular pulses on modern ECT devices, leading to a substantial decrease in cognitive effects. The width of these pulses, measured in milliseconds, may be important too. Contemporary ECT devices can be set to deliver pulse-widths between 0.25 and 2.0 milliseconds. Stimuli between 0.5 and 2.0 milliseconds are called “brief,” and those less than 0.5 millisecond are called “ultra­brief.” Sackeim and colleagues² state that the “optimal width for neuronal depolarization is estimated to be at most 0.1 to 0.2 millisecond” and cite earlier work in this area.^12,13 They also believe that wider pulse-widths are less efficient at inducing seizures, thus raising seizure threshold and perhaps contributing to cognitive effects.

Sienaert and associates^3,4 randomized 81 patients to either bifrontal (BF) (n = 40) or RUL (n = 41) ECT; all given with ultrabrief (0.3 millisecond) stimuli. The BF was given at 1.5 times seizure threshold and the RUL at 6 times seizure threshold. At the end of the ECT course, the response rate (measured by decrease in the Hamilton Rating Scale for Depression [HRSD] score of 50% or greater) in the BF and RUL groups was equal (78%). Using remitter criteria (HRSD score of 10 or below), the rates were 59% in the BF group and 72% in the RUL group. Using even stricter remitter criteria (HRSD score of 7 or lower), the rates fell to 34% in the BF group and to 44% in the RUL group. None of the group differences was significant.

A cognitive battery of tests was administered at baseline and at 1 and 6 weeks after ECT. Global cognitive function, verbal memory, attention, executive function, and autobiographical memory improved from base-line to after ECT, with no significant differences between the electrode placements. While patients in the RUL group needed fewer treatments (mean = 7.76) than did those in the BF group (mean = 10.08) to reach response criteria, there was no difference in number of treatments needed to reach the 2 levels of remission (HRSD, 10 and 7).

Loo and colleagues1 carried out a naturalistic study in which patients were given ECT with either standard brief pulse RUL ECT (pulse-width = 1.0 millisecond; n = 22) or ultrabrief RUL ECT (pulse-width = 0.3 milli­second; n = 74). In this context, “naturalistic” meant that the treating psychiatrist decided the type of ECT (either of the 2 groups), the number of treatments, and when to switch to bilateral placement because of lack of antidepressant response. Remission rates were very low (RUL = 36%, RUL-ultrabrief [RUL-UB] = 27%), both absolutely and when compared with those reported by Sackeim and coworkers2 and Sienaert and colleagues.^3,4 There were no significant differences in the Montgomery Asberg Depression Rating Scale (MADRS) scores between the 2 groups across the treatment course. However, MADRS scores declined more slowly in the RUL-UB group, and more treatments were required (10.3 vs 7.6). RUL-UB was superior to RUL on 1 measure of retrograde amnesia, the Autobiographical Memory Inventory–Short Form (30-item). Full results of the other tests in the cognitive battery were not reported.