Treatment-resistant depression & electroconvulsive therapy — from failed medication trials to durable recovery
Treatment-resistant depression (TRD) is not a diagnosis of hopelessness — it is a clinical checkpoint that signals it is time to broaden the treatment plan. The standard research and clinical definition requires failure to achieve adequate symptom relief despite two or more antidepressant trials, each given at an adequate dose for an adequate duration. Reaching this threshold is precisely the point at which electroconvulsive therapy (ECT) becomes a reasonable, evidence-based option to discuss.
Two ingredients must both be present before a medication trial can be judged to have truly failed:
• Adequate dose: the medication was titrated to a dose within the recognized therapeutic range for that agent, not stopped at a subtherapeutic starting dose • Adequate duration: the patient remained on that dose for long enough to see an effect — typically 6–8 weeks, since antidepressant response is gradual rather than immediate
A trial that was stopped after two weeks because of impatience, or one that never left a low starting dose, does not count as a failed trial in the technical sense. Distinguishing genuine non-response from an inadequate trial matters enormously, because it changes what happens next: optimize the current medication, or move on to a different strategy such as ECT.
The two-trial threshold is a pragmatic compromise. Requiring only one failed trial would label many people "treatment-resistant" who simply had not yet found their right medication — antidepressants from different classes and even within the same class can have meaningfully different individual response profiles. Requiring many more trials, on the other hand, prolongs suffering and delays effective options.
Large naturalistic studies (such as the STAR*D trial) showed that with each successive medication trial after the first, the probability of remission on the next medication step drops substantially. By the second and third medication failure, many patients have a lower likelihood of remitting on yet another standard antidepressant than they would from switching strategy altogether — which is exactly why guidelines position ECT as a rational option once TRD criteria are met, rather than something to be delayed indefinitely.
ECT is not reserved only for people who have exhausted every other option over years. Current practice increasingly recognizes it as a legitimate, high-efficacy option to raise as soon as a patient meets treatment-resistance criteria — particularly when illness severity, safety concerns, or psychotic features make speed of response clinically important.
Use the "Prior failed antidepressant trials" slider in the panel to see how a patient's trial history moves them from "not yet resistant" toward a clear indication to discuss ECT. Each red block in the timeline animation represents one completed medication trial at an adequate dose and duration that did not achieve remission. Crossing the two-trial line is the moment TRD criteria are met — the panel's resistance-status metric updates accordingly.
Among all treatments available for severe depression, electroconvulsive therapy has one of the highest documented response rates of any modality — commonly cited in the 60–80%+ range for appropriately selected patients with treatment-resistant or severe depression. That figure compares favorably with the more modest incremental benefit typically seen from switching to yet another antidepressant medication trial after prior trials have failed.
The core clinical logic is straightforward: as documented in large effectiveness studies, response probability to a given antidepressant medication trial tends to fall with each prior failed trial. A patient who has already failed two or three adequate medication trials is, on average, less likely to remit on a fourth or fifth medication than a treatment-naive patient would be on their first.
ECT works through a different mechanism than oral antidepressants — inducing a generalized, controlled seizure under general anesthesia — and this different mechanism is precisely why it can succeed where multiple medication classes have not. Because its response rates do not decline the way sequential medication trials tend to, ECT often represents the single highest-probability path to relief once TRD criteria are met.
ECT remains, by many measures, the most effective acute treatment available for severe major depressive episodes — a fact sometimes underappreciated because of stigma linked to outdated historical portrayals. Modern ECT is delivered under general anesthesia with a muscle relaxant, brief pulse-width stimulation, and continuous physiological monitoring.
"Appropriately selected" matters — response rates are highest in patients with:
• Severe major depressive episodes, particularly with psychotic features or catatonia • Depression with significant safety concerns (active suicidality, inability to sustain nutrition/hydration) where rapid response is clinically urgent • Confirmed treatment resistance to adequate medication trials • No major medical contraindication to brief general anesthesia
ECT is typically delivered two to three times per week, with a full acute course usually spanning roughly six to twelve sessions, adjusted to the individual trajectory of improvement.
The bar visualization in the canvas contrasts illustrative response-rate ranges: ECT response in appropriately selected TRD patients versus the expected response to yet another standard medication trial versus placebo. These are population-level, illustrative figures meant to convey relative magnitude — individual outcomes vary and should always be discussed with a treating clinician.
Not every predictor of depression severity predicts ECT outcome the same way — in fact, some clinical features that make depression especially disabling are also the features most strongly associated with excellent ECT response. Psychotic features, older age at presentation, a shorter duration of the current episode, and catatonic features are each individually linked to a higher likelihood of robust improvement with ECT.
Major depressive episodes with psychotic features (delusions or hallucinations congruent with depressive themes) respond to ECT at particularly high rates, and ECT is often considered a preferred — not merely a fallback — treatment in this population because psychotic depression tends to respond poorly to antidepressant monotherapy alone.
Catatonia, whether occurring alongside depression or another underlying condition, is one of the most ECT-responsive syndromes in all of psychiatry. Catatonic symptoms — immobility, mutism, posturing, waxy flexibility — can improve rapidly and dramatically over the course of just a few ECT sessions, sometimes providing near-immediate relief of a severely impairing and medically risky state.
Because psychotic and catatonic depression are both severe, high-risk presentations that also happen to be strongly ECT-responsive, some guidelines support earlier use of ECT in these specific subgroups rather than waiting through multiple medication trial failures first.
Older age at the time of the depressive episode is generally associated with a more favorable ECT response, an association observed consistently across outcome studies, though ECT remains effective and is used safely across the adult lifespan including in medically complex older adults.
A shorter duration of the current depressive episode before starting ECT is likewise associated with better outcomes than a long-standing, chronic episode. This mirrors a broader pattern in mood disorder treatment: earlier, more decisive intervention tends to be associated with better response than treatment delayed after years of unremitting illness.
The canvas visualizes four predictor factors — psychotic features, older age, shorter episode duration, and catatonic features — each contributing to an overall illustrative response-likelihood gauge. In the simulation these factors animate in sequence to show how multiple favorable predictors combine; the gauge is a simplified teaching illustration, not a validated clinical prediction tool.
ECT does not typically deliver its full benefit in a single session. For most patients, symptom improvement begins to emerge within the first several treatments, and continues to accumulate across a full acute course — commonly around six to twelve sessions delivered two to three times per week — with the rate of week-to-week improvement often most pronounced in the early-to-middle portion of the course.
Clinically, depression-severity ratings tracked across an ECT course typically show a declining curve: severity scores fall session by session, often with the steepest early gains appearing within the first handful of treatments, followed by continued — sometimes more gradual — improvement through the remainder of the course.
This trajectory has practical implications: clinicians reassess severity ratings throughout the course rather than waiting until a fixed session count is reached. A patient improving quickly may need fewer sessions than initially anticipated; a patient improving more slowly but steadily may benefit from extending the course beyond the typical range, always weighed against side-effect burden.
Early improvement is an encouraging sign, but stopping too soon risks an incomplete response that is more vulnerable to early relapse. Completing a full, individualized course — rather than halting treatment the moment symptoms first begin to lift — is associated with deeper and more durable remission. This is analogous to completing a full course of antibiotics rather than stopping once symptoms first improve: the trajectory needs time to reach its plateau.
The session-by-session curve in the canvas is illustrative rather than a guarantee for any individual — some patients improve faster, some more slowly, and a minority do not respond even with a full course. The treating team monitors validated depression-severity ratings throughout to guide real decisions about course length.
Move the "ECT session number in course" slider to animate the declining depression-severity curve session by session. The "cumulative improvement" metric in the panel rises correspondingly, and the "response category" metric reflects illustrative clinical thresholds — no response yet, partial response, or full response — based on that cumulative improvement at the current session number.
Achieving response or remission at the end of an acute ECT course is a major clinical milestone — but it is the beginning of a new phase, not the end of the story. Without continuation treatment, relapse risk in the months following acute-course completion is significant. Continuation antidepressant medication or maintenance ECT sessions meaningfully reduce that risk, and continuation planning should be discussed before the acute course even ends.
The biological and clinical processes that responded to acute ECT do not become permanently self-sustaining the moment the last acute session ends. Depression is often a recurring illness, and the vulnerability that produced the treatment-resistant episode in the first place does not disappear simply because that particular episode has resolved.
Studies following patients after a successful acute ECT course consistently find that relapse rates are substantial when no continuation treatment is provided, with the risk concentrated especially in the weeks and months immediately following the final acute session — precisely the window when continuation planning matters most.
Two evidence-supported continuation strategies are commonly used, sometimes in combination:
• Continuation/maintenance pharmacotherapy: an antidepressant regimen (often including an agent not previously failed, or a combination strategy) started or restarted around the time of the acute ECT course and continued afterward • Maintenance ECT: spaced, less-frequent ECT sessions (for example, tapering from weekly to monthly) continued after the acute course, used particularly for patients who relapsed previously despite medication continuation, or who cannot tolerate/tolerate poorly continuation medications
The choice between these — or combining them — is individualized, weighing prior treatment history, side-effect tolerance, relapse history, and patient preference. What is not in question is that some form of continuation planning substantially outperforms stopping all treatment at the end of the acute course.
Discharge from an ECT course without an explicit continuation plan is one of the most preventable contributors to early relapse. Continuation planning — deciding on medication, maintenance ECT, or both — should be finalized before, not after, the final acute session.
The canvas illustrates the post-course period as a branch point: one path shows relapse risk accumulating without continuation treatment, while parallel paths show the protective effect of continuation medication and maintenance ECT. The panel's relapse-risk metric is an illustrative, informational figure meant to reinforce the clinical importance of continuation planning — not a personalized risk prediction.