Few treatments have been met with more criticism and hostility than electroconvulsive therapy (ECT). For decades, ECT was the subject for intense controversy and opposition, chiefly by civil rights groups, but also some psychologists and social workers were against the treatment due to the idea that mental symptoms should not be treated with biological methods.

Despite this public and professional opposition, ECT is widely accepted by psychiatrists in Scandinavia and in many other countries as indispensable in the treatment for severe depression and other severe psychiatric disorders such as delirious conditions, catatonia, and malignant neuroleptic syndrome on the grounds that it is effective, safe, and gives better and quicker results than antidepressant drugs, [1] and, finally, side-effects are usually minor.

However, the inevitable, transient, cognitive side-effects occurring after a series of ECT have played a major role for the image of ECT, and for decades, studies have been made to diminish ECT-induced memory impairment—to an extent so that treatment modifications have been recommended which sacrifice therapeutic effect for milder side-effects.

Since the 1960s, placing electrodes over the non-dominant hemisphere (right unilateral ECT, RUL-ECT) has in many studies shown that this modification induces less memory impairment than the conventional bitemporal electrode (BT-ECT) placement, but a larger number of treatment sessions are necessary to obtain clinical remission. If the current in RUL-ECT is delivered with higher energy levels, 8–12x the seizure threshold (ST), which is the minimum current that elicits a seizure—the therapeutic effect increases, but at the same time, the cognitive advantage disappears.

The efforts to optimize ECT using different electrode placements have for decades alone focused on the memory aspect, as have many attempts to minimize the cognitive side-effects using alternating current stimulation with a pulse width of 0.1–0.3 ms. This is much lower than the conventional 1.0–5.0 ms pulse width used in many Scandinavian and American studies.

A serious disadvantage of stimulation with ultra brief pulse width is that it often generates submaximal seizures with impaired antidepressive effect, and thus, the reduction in memory impairment will be at the cost of therapeutic effect.

Data from the studies by the Consortium for Research in ECT (CORE)—the most comprehensive ECT studies to date—in the present issue of this journal [2] throw light on some of the controversial issues regarding ECT—in particular the memory problem.

The CORE review underlines ECT's speed of action whereby suicidality is diminished—a highly important element when treating severe depression. Patients with symptoms of melancholic depression (having vegetative symptoms) responded dramatically to BT-ECT with a remission rate of up to 84%, and for those harboring delusions or being hallucinated, the remission rate was over 95%—a rate hardly surpassed by any treatment in the whole of medicine.

Although our knowledge of the mechanism of ECT is far from complete, it is evident that elicitation of generalized seizures is a prerequisite for therapeutic effect [3, 4] both in melancholic/psychotic depression and in delirious conditions and catatonia. Elicitation of generalized seizures is best obtained using BT-ECT as has been shown in several neurophysiological studies.

The effect of RUL-ECT in patients with the same illness characteristics as those from the BT-ECT of the CORE group was studied by researchers from Columbia University (CUC) [5].

The remission rate in the CUC study was 55%, in the CORE study 86%. These different rates may well be ascribed to the different electrode placement, pointing to BT-ECT superiority.

The idea that it is the intensity of the electrical stimulation that drives the severity of the cognitive side-effects of ECT has inspired to identifying at the first ECT session the seizure threshold (ST)—and giving the patients low doses of electricity just above this threshold to minimize cognitive impairment. In the CORE study, the vast majority of patients showed no rise of ST over time, and the widely applied determination of ST thus does not seem clinically justified [2].

After successful ECT, the risk of relapse within a year is over 80% making continued stabilizing treatment necessary.

The CORE studies further demonstrated that the use of continuation ECT at increasing intervals was equal to a combination of lithium and nortriptyline with respect to relapse prevention during a 6-month period, a finding of great value for many patients who do not tolerate an adequate drug regime.

Again, the most debated issue regarding ECT is the memory impairment. The transient memory disturbance is a side-effect of ECT, which does not contribute to the antidepressive effect.

The subjective experience of memory impairment following a course of ECT is sometimes described by patients as being long-lasting, in exceptional cases permanent, but this experience stands in contrast to measurements of objective memory loss. Thus, a recent meta-analysis of 2981 patients treated with ECT found that the cognitive side-effects mainly comprised the first 3 days after the treatment and that all cognitive functions improved compared with pretreatment conditions irrespective of stimulus parameters [6].

The solid findings from the CORE studies substantiate the superiority of ECT in the therapy of severe (melancholic/psychotic) depression, where it should be considered first-line treatment, and not a last resort intervention as suggested in many treatment algorithms.

Other clear indications for ECT where there are no comparable alternatives comprise delirious mania, catatonic stupor, postpartum psychosis cycloid psychosis lethal catatonia, and neuroleptic malignant syndrome [7]. Using BT-ECT favors therapeutic effect over a transient cognitive advantage, and in choosing treatment parameters, this should be borne in mind when treating critically ill patients.


  1. Top of page
  2. References