Electroconvulsive Therapy

Abstract

Electroconvulsive therapy is a form of convulsive therapy: the deliberate induction of a generalized seizure, under general anesthesia and muscle relaxation, to treat severe psychiatric illness. It remains the most effective acute treatment for major depression, and it is a first-line option for depression with psychosis, catatonia, and life-threatening refusal of food or fluids. Its two operating parameters, electrode placement and stimulus dose relative to the seizure threshold, trade antidepressant efficacy against memory side effects, and most of the modern technique is an effort to widen that margin. Its cognitive costs are largely transient, though retrograde autobiographical memory loss can persist. Its therapeutic mechanism is still debated, with neuroplasticity and neuroendocrine theories now dominant.

Keywords: electroconvulsive therapy, seizure, treatment-resistant depression, electrode placement, retrograde amnesia

Electroconvulsive therapy (ECT) passes a brief electrical current through the brain to trigger a controlled generalized seizure, the seizure rather than the current being the therapeutic event (Lisanby, 2007). Introduced in 1938, it is the oldest somatic treatment still in routine psychiatric use, and across eight decades of refinement it has never been displaced as the most effective option for the most severe mood disorders (Espinoza & Kellner, 2022).

What Electroconvulsive Therapy Is

ECT is the therapeutic induction of a generalized tonic-clonic seizure by passing an electrical stimulus between electrodes on the scalp. The procedure is performed under brief general anesthesia with a muscle relaxant, so the visible convulsion is suppressed while the electrographic seizure proceeds; a course typically runs two to three treatments per week for six to twelve sessions (Lisanby, 2007). Its parent category in the Medical Subject Headings, convulsive therapy, reflects the founding insight that the seizure itself is curative, an idea that predates the electrical method: chemically induced seizures were used first, and the electroshock stimulus simply proved a more controllable trigger.

Ugo Cerletti and Lucio Bini administered the first electrically induced therapeutic seizure in Rome in 1938, building on Ladislas Meduna's chemical convulsive therapy of the preceding years (Fink, 2001). The primary indications today are major depressive disorder that has resisted pharmacotherapy, depression with psychotic features, bipolar depression and mania, and catatonia, where response can be both rapid and dramatic (Espinoza & Kellner, 2022). Because response is often faster than with medication, ECT is favored when illness is immediately dangerous, as in acute suicidality or refusal of food and fluids.

The electrode-placement trade-off

Choose a placement. Each buys antidepressant efficacy against cognitive burden, and the electrodes light up on the schematic head. The bars are ordinal illustrations of the direction of the trade-off, not clinical figures.

Both temples. Fastest and most effective; current crosses both temporal lobes, so the memory cost is greatest.

Efficacy and Indications

The systematic evidence places ECT above every alternative for acute severe depression. A meta-analysis by the UK ECT Review Group pooled the randomized trials and found ECT significantly more effective than pharmacotherapy, with real ECT superior to simulated ECT, bilateral placement superior to unilateral, and higher electrical dose superior to lower (UK ECT Review Group, 2003). A later network meta-analysis of non-surgical brain stimulation for acute depression ranked ECT among the most effective interventions of any kind (Mutz et al., 2019).

Response is not uniform. The strongest clinical predictor of a good outcome is the presence of psychotic features, and older age also forecasts higher remission; a meta-analysis of predictors found both to be reliable, while medication resistance modestly lowers the odds of response (van Diermen et al., 2018). Efficacy also varies with setting: remission rates in tightly controlled research protocols exceed those observed when ECT is delivered in ordinary community practice, where technique and dosing are more variable (Prudic et al., 2004).

A single course does not end treatment. Relapse after successful ECT is high without continuation therapy, and a multisite trial from the Consortium for Research in Electroconvulsive Therapy (CORE) found that continuation ECT and continuation pharmacotherapy were comparably effective at preventing relapse over six months (Kellner et al., 2006). Table 1 summarizes how the three placements compare on the efficacy-versus-cognition trade-off that governs the technique.

Why unilateral ECT must be dosed high

Slide the stimulus dose in multiples of the seizure threshold. Bilateral ECT is already effective at 1.5x, but right unilateral is nearly inert until roughly 6x, where it catches up. This interaction is the core of modern dosing.

At 1.5x threshold: bilateral efficacy 65, right unilateral efficacy 1 (illustrative scale).

The central technical finding of modern ECT is that dose and placement interact. A landmark randomized trial showed that right unilateral ECT is ineffective at low dose but becomes as effective as bilateral ECT when given at roughly six times the seizure threshold, while producing markedly less cognitive impairment (Sackeim et al., 2000). This is why unilateral ECT must be dosed high relative to threshold to work, and why threshold titration at the first session is standard practice.

Cognitive Effects

Memory side effects are the reason ECT is feared and the target of most of its refinement. They fall into two kinds: anterograde amnesia, difficulty forming new memories during a course, and retrograde amnesia, loss of memories formed before treatment, which is most pronounced for autobiographical events near the time of the course. A meta-analysis of objective cognitive performance across 84 studies found that measurable deficits are concentrated in the first three days after treatment, that nearly all cognitive domains return to or exceed baseline within about fifteen days, and that some measures improve beyond baseline thereafter as the depression itself lifts (Semkovska & McLoughlin, 2010).

The persistent exception is autobiographical memory. In community practice, a prospective study found that a subset of patients showed retrograde amnesia detectable months after treatment, and that persistent deficits were associated with older technique, specifically bilateral placement and sine-wave (rather than brief-pulse) stimulation (Sackeim et al., 2007). This finding drove two changes now standard: the abandonment of sine-wave devices and the shift toward unilateral placement and shorter pulse widths. The PRIDE study demonstrated that right unilateral ultrabrief-pulse ECT, which narrows the pulse to about 0.3 milliseconds, produces robust remission in geriatric depression with a favorable cognitive profile (Kellner et al., 2016).

Cognitive recovery after a course

Slide the number of days after the last treatment. A composite of objective cognitive measures dips in the first days, returns to baseline by about two weeks, and can rise above it as the depression itself resolves. The curve illustrates the shape of the meta-analytic finding; the vertical scale is a standardized difference.

Day 2: composite cognition is impaired relative to baseline (standardized difference -0.43).

Mechanisms

How ECT works is genuinely unsettled, and the leading theories are not mutually exclusive (Bolwig, 2011). Neuroplasticity accounts point to the reliable finding that ECT increases the volume of the hippocampus and amygdala. A structural MRI study documented a hippocampal volume increase after a course of ECT (Nordanskog et al., 2010), and a systematic review and meta-analysis confirmed that both the hippocampus and amygdala enlarge across studies (Takamiya et al., 2018). Whether that volume change is neurogenesis, synaptogenesis, angiogenesis, or edema, and whether it even correlates with clinical response, remains contested.

More recent neuroimaging complicates the hippocampus-centered story. A large multisite analysis found that ECT-induced structural changes are broadly distributed across the cortex rather than confined to the medial temporal lobe, suggesting that the volume increase is a marker of a whole-brain neuroplastic response rather than the mechanism itself (Ousdal et al., 2020). Competing accounts emphasize the seizure's neuroendocrine effects, the release of neurotrophic factors, anticonvulsant adaptations that raise the seizure threshold over a course, and normalization of functional connectivity. No single account yet explains why a generalized seizure is antidepressant.

Worked Example

Consider a clinician titrating right unilateral ECT. At the first session the patient's seizure threshold is measured at 60 millicoulombs of charge. Because unilateral ECT is only effective at a high multiple of threshold, the treatment dose is set at six times that value: 60 × 6 = 360 millicoulombs. Had the same patient been assigned bilateral ECT, the standard dose would be only 1.5 times threshold, or 60 × 1.5 = 90 millicoulombs. The unilateral patient thus receives four times the charge (360 versus 90) precisely to compensate for the placement that spares memory: the higher dose recovers the efficacy that unilateral placement would otherwise sacrifice, while keeping current away from both temporal lobes. This is the arithmetic behind the Sackeim finding that dose and placement must be considered together, never in isolation (Sackeim et al., 2000).

Key Researchers

Lucio Bini (1908-1964). University of Rome La Sapienza; with Cerletti he co-invented electroconvulsive therapy in 1938 and built the first ECT apparatus. Wikipedia

Tom G. Bolwig (b. 1937). Professor Emeritus of Psychiatry at the University of Copenhagen; author of widely cited reviews on the mechanism of ECT and its effects on hippocampal neuroplasticity. Wikidata

Ugo Cerletti (1877-1963). University of Rome La Sapienza; the neurologist who introduced electroconvulsive therapy in 1938 with Lucio Bini. Wikipedia

Max Fink (1923-2025). Professor of Psychiatry and Neurology Emeritus at Stony Brook University; a leading authority on ECT and catatonia and co-leader of the CORE trials. Wikipedia

Charles H. Kellner (contemporary). Professor Emeritus at the Medical University of South Carolina; lead investigator of the CORE and PRIDE electroconvulsive therapy trials. ORCID

Sarah H. Lisanby (contemporary). National Institute of Mental Health and Duke University; a researcher on ECT, magnetic seizure therapy, and transcranial magnetic stimulation. ORCID

W. Vaughn McCall (contemporary). Medical College of Georgia, Augusta University; a researcher on ECT dosing, depression, insomnia, and suicide risk. ORCID

Declan M. McLoughlin (contemporary). Trinity College Dublin and St Patrick's University Hospital; leads the EFFECT-Dep and KARMA-Dep ECT trials and co-authored the definitive cognitive-effects meta-analysis. ORCID

Harold A. Sackeim (contemporary). Professor Emeritus of Psychiatry and Radiology at Columbia University; an authority on ECT electrode placement, stimulus dosing, and cognitive side effects. ORCID

Maria Semkovska (contemporary). University of Southern Denmark; lead author of the meta-analyses of objective cognitive performance and autobiographical memory after ECT. ORCID

Discussion

ECT occupies a paradoxical place in psychiatry: it is the field's most effective treatment for its most severe conditions and also its most stigmatized. The stigma is partly historical, rooted in the era of unmodified treatment before anesthesia and muscle relaxants, and partly a reasonable response to the reality of memory side effects. The modern technical program, threshold titration, unilateral placement, ultrabrief pulses, exists to shrink those side effects without surrendering efficacy, and it has largely succeeded for anterograde memory while leaving retrograde autobiographical loss as the residual concern (Semkovska & McLoughlin, 2010). For the cognitive psychologist, ECT is also a natural experiment in memory consolidation: its selective disruption of recently formed autobiographical memories, sparing remote ones, is among the clearest human evidence for a time-graded consolidation process.

Current Directions

The active research front is dominated by three questions. The first is mechanistic: whether the distributed structural changes now visible on neuroimaging are causes of recovery or merely correlates, and whether the hippocampal volume increase can be dissociated from clinical response (Ousdal et al., 2020). The second is comparative: how ECT stacks up against the newer non-surgical brain stimulation methods, and against ketamine, for treatment-resistant depression, with network meta-analyses beginning to place these interventions on a common efficacy scale (Mutz et al., 2019). The third is optimization: ongoing trials such as EFFECT-Dep and KARMA-Dep, led by McLoughlin's group, test whether pulse width, placement, and adjunctive agents can be tuned to preserve efficacy while further reducing cognitive cost. Personalized dosing guided by individual seizure physiology, rather than fixed multiples of threshold, is the likely next step.

Glossary

Anterograde amnesia.

Difficulty forming and retaining new memories during a course of treatment; in ECT it resolves within days to weeks after the course ends.

Autobiographical memory.

Memory for personally experienced events; its retrograde loss near the treatment period is the principal persistent cognitive risk of ECT.

Bifrontal placement.

An electrode configuration with both electrodes on the forehead, intended to spare temporal memory structures while retaining efficacy.

Bilateral ECT.

The bitemporal placement, with one electrode over each temple; the most effective and fastest-acting configuration, and the one with the greatest cognitive burden.

Brief-pulse stimulation.

Delivery of the stimulus as short rectangular pulses (roughly 0.5 to 2 milliseconds), which induces a seizure more efficiently and with less cognitive cost than the obsolete sine-wave current.

Catatonia.

A syndrome of motor and behavioral dysregulation, often with immobility or mutism, that frequently responds rapidly and dramatically to ECT.

Continuation ECT.

Tapered maintenance treatments given after a successful acute course to prevent relapse, an alternative to continuation pharmacotherapy.

Convulsive therapy.

The broader treatment class, the deliberate induction of seizures for psychiatric benefit, of which ECT is the electrical form.

Electroconvulsive therapy.

The induction of a generalized seizure by an electrical stimulus, under anesthesia, to treat severe psychiatric illness.

Generalized seizure.

A seizure engaging both cerebral hemispheres; its adequate induction, not the electrical current itself, is the therapeutic event in ECT.

Retrograde amnesia.

Loss of memories formed before treatment; in ECT it is time-graded, affecting recent memories more than remote ones.

Right unilateral ECT.

A placement with both electrodes over the non-dominant right hemisphere; effective only at high dose relative to threshold, but with the lowest cognitive burden.

Seizure threshold.

The minimum electrical charge needed to induce an adequate seizure in a given patient, measured by titration and used to set the treatment dose.

Stimulus dosing.

Setting the electrical charge as a multiple of the individual seizure threshold; the parameter that, with placement, determines both efficacy and side effects.

Treatment-resistant depression.

Major depression that has failed to respond to adequate trials of antidepressant medication; the most common indication for ECT.

Ultrabrief pulse.

A pulse width near 0.3 milliseconds, shorter than conventional brief pulse, which further reduces cognitive side effects while retaining efficacy in unilateral placement.

Frequently Asked Questions

Is the patient awake during ECT?
No. Modern ECT is given under brief general anesthesia with a muscle relaxant, so the patient is unconscious and the body's convulsion is suppressed while the therapeutic seizure proceeds (Lisanby, 2007).

Does ECT cause permanent brain damage?
Structural imaging shows the opposite of damage: ECT increases hippocampal and amygdala volume rather than reducing it, and objective cognitive measures largely recover within weeks (Takamiya et al., 2018).

How effective is ECT compared with antidepressant drugs?
It is more effective. Pooled randomized trials show ECT superior to pharmacotherapy for acute severe depression, and network meta-analysis ranks it among the most effective of all brain-stimulation treatments (UK ECT Review Group, 2003).

Which memory problems are lasting?
Anterograde memory recovers within days to weeks; the potentially persistent effect is retrograde loss of autobiographical memories from around the treatment period, more common with older bilateral and sine-wave technique (Sackeim et al., 2007).

Why must unilateral ECT be given at a high dose?
Right unilateral ECT is ineffective near the seizure threshold and becomes as effective as bilateral only at about six times threshold, so high dosing is what recovers its efficacy while sparing memory (Sackeim et al., 2000).

What happens after a successful course to prevent relapse?
Relapse is high without maintenance, so patients receive continuation ECT or continuation medication; a multisite trial found the two comparably effective over six months (Kellner et al., 2006).

Who responds best to ECT?
Depression with psychotic features and older age predict the highest remission rates, while established medication resistance modestly lowers the odds of response (van Diermen et al., 2018).

Do we know how ECT actually works?
Not fully. The leading theories center on seizure-induced neuroplasticity and neuroendocrine effects, but no single account yet explains why a generalized seizure is antidepressant (Bolwig, 2011).

References