The question “Which is better?” can conceal the details that make a treatment comparison useful. Ketamine and electroconvulsive therapy have been compared in randomized studies, but those studies did not all recruit the same patients or ask the same primary question. Reading them together is more informative than choosing a winner from a headline.
ELEKT-D: keep the population in the conclusion
ELEKT-D randomized 403 adults with treatment-resistant major depression without psychosis. In the group who began assigned treatment, response occurred in 55.4% receiving intravenous ketamine and 41.2% receiving ECT. The study met its prespecified noninferiority criterion. Its primary response outcome concerned a substantial reduction in self-reported depressive symptoms; it was not a finding that every participant recovered or that ketamine was superior for all severe depressive presentations. [1]
“Without psychosis” is a substantive boundary. The trial cannot resolve the relative value of the treatments in psychotic depression, catatonia or every psychiatric emergency. Its treatment course, assessment schedule and attrition also form part of the result. A number belongs to the protocol and population that produced it before it belongs in a conversation about another patient. [1]
Why KetECT does not tell exactly the same story
The Swedish KetECT trial studied hospitalized patients with unipolar depression and used remission as its main outcome. Remission occurred in 63% receiving ECT and 46% receiving ketamine; the trial did not establish ketamine’s noninferiority on that outcome. Differences in population, illness characteristics, protocol and outcome definition make a simple side-by-side comparison with ELEKT-D misleading. [2]
The lesson is not that one trial cancels the other. It is that a broad treatment category contains clinically different situations. Inpatient status can reflect severity and service context; response and remission answer different questions; and treatment delivery can vary. A useful journal club begins by reconstructing those differences rather than averaging the percentages as if the studies were interchangeable. [2]
Response, remission and recovery are different destinations
A response threshold captures improvement from baseline. A remission threshold asks whether symptoms have fallen below a defined level. Recovery additionally concerns sustained wellbeing and function. Someone can respond meaningfully while remaining significantly unwell. Comparing a response rate in one trial with a remission rate in another therefore changes both the clinical destination and the statistical question.
Noninferiority also has a precise meaning: the comparison is assessed against a margin selected before the results are known. It is not a general declaration of equivalence, identical risks or equal suitability. For decisions, the choice of margin, who was analyzed and the confidence interval deserve as much attention as the headline percentage. These are tools for interpretation, not obstacles to using evidence.
Cognition deserves a conversation of its own
A secondary ELEKT-D analysis explored whether baseline features were associated with different improvement patterns. Such analyses can suggest clinically interesting hypotheses, but they do not automatically provide a validated tool for assigning an individual to one treatment. A predictor becomes more useful when its performance is tested prospectively outside the data in which it was found. [3]
Indian ECT guidance emphasizes baseline assessment, informed consent and monitoring, including cognitive effects. Discussing memory as a single yes-or-no adverse event is inadequate: the patient’s experience, objective assessment, functional consequences and evolution over time all matter. Changes in ECT technique may alter the balance, and a treatment course should be considered with its monitoring and continuation plan rather than as an isolated procedure. [4]
Do not substitute esketamine evidence for ketamine evidence
Intranasal esketamine is a specific formulation with its own clinical program and product requirements. The current US Spravato label includes treatment-resistant depression in adults and a separate indication concerning depressive symptoms in major depression with acute suicidal ideation or behavior. The indication, concomitant treatment requirements and supervision arrangements need to be read as written; they cannot be assumed from a trial of intravenous racemic ketamine. [5]
The label also states that prevention of suicide or reduction of suicidal ideation or behavior has not been demonstrated, and improvement does not remove a clinically indicated need for hospitalization. This distinction matters whenever a rapid reduction in depression scores is described as proof that someone is safe. Symptom change is one part of continuing clinical assessment, not a substitute for it. [5]
The setting is part of the intervention
The APA consensus statement on ketamine highlights the need for careful assessment, appropriate clinical expertise, monitoring and consideration of longer-term uncertainty. A monitored research intervention should not be equated with an unsupervised or differently formulated product. Route, exposure, concomitant substances and the ability to respond to an adverse event all affect what is being evaluated. [6]
For ECT, the corresponding discussion includes anesthetic assessment, consent, the treatment technique and a plan to assess benefit and adverse effects over the course. The Indian guideline provides a locally relevant framework for these elements. Access and logistics are also clinical realities: travel, an accompanying person, time away from work and follow-up can influence whether an otherwise reasonable option is workable. [4]
Sources & study types
- Anand et al. Ketamine versus ECT for nonpsychotic treatment-resistant major depression. NEJM (2023).ELEKT-D randomized noninferiority trial · Accessed 27 October 2026
- Ekstrand et al. Racemic ketamine as an alternative to ECT: KetECT. International Journal of Neuropsychopharmacology (2022).Randomized inpatient noninferiority trial · Accessed 27 October 2026
- Jha et al. Ketamine vs ECT: secondary analysis of ELEKT-D. JAMA Network Open (2024).Secondary trial analysis · Accessed 27 October 2026
- Thirthalli, Sinha and Sreeraj. Clinical practice guidelines for ECT. Indian Journal of Psychiatry (2023).Indian Psychiatric Society guidance · Accessed 27 October 2026
- Spravato. US prescribing information, September 2026 revision.Official product information · Accessed 27 October 2026
- Sanacora et al. A consensus statement on ketamine in mood disorders. JAMA Psychiatry (2017).APA research task-force consensus · Accessed 27 October 2026
About this review
- Research checked
- 27 October 2026
- Clinical review
- Not independently reviewed
- Method
- Focused narrative review of selected primary studies and official guidance; not a systematic review.
International evidence is discussed in its source context. Indian authorization, availability and individual clinical decisions require separate assessment.