Clinical Psychopharmacology · Paper A
ECT mechanism of action
ECTElectroconvulsive therapy, a treatment that uses a brief controlled seizure to relieve severe depression.Read more → looks, on the surface, like the electricity is the treatment — but decades of controlled research point somewhere else entirely. What actually produces the antidepressant/antipsychotic benefit is the generalised seizure itself, not the electrical stimulus, the anaesthetic, or the ritual of being treated — a fact proven by studies that gave patients everything except a real seizure and watched them not improve — with increased monoamine signalling and increased BDNF/hippocampal neurogenesis as downstream, complementary consequences of that seizure, not competing explanations for it. The same seizure-or-no-seizure question is also exactly what separates ECT from every other physical brain treatment in psychiatry, and a genuine predictor of ECT response (psychotic depression, catatoniaA syndrome of abnormal movement and behaviour, ranging from being frozen and unresponsive to being excessively agitated.Read more →) is a separate, equally testable fact from a plausible-sounding non-predictor (electrode placement, prior episode count, family history).
Doctors have used since the 1930s, long before anyone understood how it worked, because it visibly helped patients nothing else could reach — the severely depressed, the catatonic, the psychotically depressed. That gap between "it works" and "why" persisted for most of the twentieth century, and it still shapes how the topic is examined: the exam cares far less about whether you know ECT helps than about whether you can name what, mechanistically, is actually doing the helping, and tell that apart from treatments that look superficially similar.
means deliberately inducing a generalised seizure (a burst of synchronised, uncontrolled electrical activity across the whole brain, causing a brief convulsion) under general anaesthesia and a muscle relaxant, using an electrical stimulus delivered through scalp electrodes. The "mechanism of action" question asks: once that seizure happens, what actually changes in the brain that lifts a severe depression or resolves ? Three linked answers exist — the seizure itself, biochemical change, and structural change — and none of them stands alone.
Here is the trap: assuming the electrical current itself, or perhaps the anaesthetic and the ritual of a "treatment", is what makes work. It is not. The generalised seizure is what matters, and this was proven, not just assumed. In the studies that settled the question, one group of patients received a full anaesthetic and muscle relaxant but no electrical stimulus at all — sham ECT — while another received a genuine stimulus, strong enough to trigger a full seizure. Only the real-seizure group improved. A separate line of evidence, sub-convulsive stimulation (delivering current below the level needed to trigger a seizure), showed the same thing: current without a seizure produces no benefit. The seizure is the active ingredient; everything else is scaffolding around it.
Once you accept that the seizure itself is necessary, the next question is what a seizure actually changes. Three theories answer this at three different levels, and the exam expects you to place a finding on the correct one rather than treating them as competing explanations.
The monoamine theory says a seizure increases the turnover of serotonin, noradrenaline and dopamine — broadly the same biochemical direction antidepressant drugs aim for, but reached faster and, in severe depression, often more completely.
The neurotrophic theory goes structural instead: a seizure raises brain-derived neurotrophic factor (BDNF, a protein that supports neurons surviving and growing new connections), and a course of measurably increases hippocampal volume on MRI, alongside animal evidence of new neurons and forming there — neurogenesis (the birth of new neurons). Chronic depression is associated with hippocampal shrinkage, so this theory frames ECT as partly reversing physical damage, not just adjusting chemistry.
A third, less central theory notes that the seizure threshold (the electrical dose needed to trigger a seizure) rises across a course of treatment — the anticonvulsant effect — and proposes that this rising threshold itself tracks with improvement. Treat this as a minor, secondary theory, not the leading explanation. None of the three needs to "win" against the others: they describe the same underlying event, an adequate generalised seizure, at the level of an electrical phenomenon, a biochemical change, and a structural change respectively.
Imagine a patient part-way through a course of bilateral for severe, psychotic depression. Session by session, her clinician records three things. First, seizure duration and quality on EEG — this session's seizure lasted 28 seconds and looked fully generalised, so it counts as adequate; a 4-second, one-sided response would not, and the stimulus would be repeated at a higher dose that same session. Second, a brief post-ictal blood test shows a rise in (a hormone released after a seizure) and cortisol — this is not itself the mechanism, it is a marker confirming an adequate seizure occurred, the way a receipt confirms a purchase happened without being the item you bought. Third, after the full course, a repeat MRI shows a small increase in hippocampal volume compared with the scan taken before treatment began.
Now suppose this same patient had instead received sub-convulsive stimulation throughout — current delivered every session, but never enough to cross her seizure threshold. On the evidence above, you would correctly predict that none of the three markers would show the pattern linked to improvement: no adequate seizure on EEG, no meaningful post-ictal hormone surge, and no expected change on repeat imaging. The stimulus was real; the seizure was not; and the seizure is exactly what the whole system depends on.
is one of several physical treatments used when medication alone has not worked, and the exam's favourite trick is describing one of the others using ECT-adjacent language.
| Looks similar | How to tell it apart |
|---|---|
| Transcranial magnetic stimulation (TMS) | Magnetic pulses over the scalp, patient awake, no anaesthetic, no seizure induced — a seizure here is a rare side effect, never the mechanism |
| Vagus nerve stimulation (VNS) | Implanted device, continuous nerve signalling not a seizure, works on the same monoamine pathways indirectly, benefit builds over months |
| Deep brain stimulation (DBS) | Implanted electrodes in a specific circuit, continuous targeted stimulation, needs neurosurgery, still investigational, no seizure induced |
| Sham | Historical research control — same anaesthetic and routine, but no stimulus and no seizure; how the seizure's necessity was proven, not a treatment offered today |
The one dividing line that sorts all of these at a glance: does the treatment deliberately produce a generalised seizure? Only does. Every other option in this family is specifically engineered to avoid one.
There is no diagnostic test here in the usual sense, but the exam treats "which patients respond best" as its own examinable fact, separate from mechanism. Mood-congruent psychotic depression (severe depression with that fit the depressed mood, such as guilt or worthlessness) and (a syndrome of , abnormal posturing, and , whatever its underlying cause) are the two presentations most consistently linked to a strong, often fast, response.
Pulling the other way, a longer current episode and a comorbid personality disorder both predict a poorer response. Two things that sound like they should matter, but the evidence does not support as reliable predictors, are the exact number of previous depressive episodes and a family history of mood disorder. One more thing genuinely matters clinically but is a treatment choice, not a patient trait: electrode placement. Bilateral electrodes are generally more reliably effective but carry a greater cognitive side-effect burden; unilateral electrodes spare cognition more but need careful dosing to match that efficacy.
A course of in the UK is planned and monitored under guidance from NICE and the Royal College of Psychiatrists (RCPsych), covering consent requirements, the use of ECT as an option after other treatments have failed or when a rapid response is genuinely needed (severe depression with high suicide risk, psychotic features, , or a dangerous physical state from refusing food and fluids), and the monitoring standards for anaesthesia and seizure adequacy at each session.
In practice this means every session is a small, checked experiment: was the seizure adequate, does the response justify continuing, and is the cognitive side-effect burden acceptable — decisions genuinely driven by the mechanism this page has walked through, not by habit or routine.
If a vignette describes any brain-stimulation treatment, ask one question before anything else: is a generalised seizure being deliberately produced? If yes, you are reading about and its mechanism — seizure, monoamine change, BDNF and neurogenesis. If no, you are reading about TMS, VNS or DBS, and the seizure-necessity evidence you just learned does not apply to whichever one it is. Before you move on, try explaining to a friend why a surge after a seizure is a receipt, not the mechanism itself.
ECT's benefit depends on the generalised seizure itself — proven by sham- and sub-convulsive-stimulation trials — with monoamine and BDNF/neurogenesis changes as complementary downstream consequences, and that same seizure is exactly the feature TMS, VNS and DBS are each designed to avoid.