PSYREFLECT
RESEARCHMarch 2, 20263 min read

ECT vs. Ketamine for Treatment-Resistant Depression: The ELEKT-D Trial Changes How We Talk About Both

Key Findings
  • Randomized non-inferiority trial (N=403, 5 US academic centers) comparing IV ketamine (3x/week × 3 weeks) vs ECT (3x/week × 3 weeks) in nonpsychotic treatment-resistant major depression
  • Ketamine met non-inferiority for depression symptom response (55.4% vs 41.2%, 95% CI crossing threshold) — ketamine is a non-inferior alternative to ECT for response, not remission
  • ECT was superior for remission (46.7% vs 29.3%, p<.001) — clinically important distinction; ketamine's response advantage is partial improvement, ECT's advantage is full remission
  • Cognitive side effects significantly lower with ketamine; ECT showed measurable memory and processing speed impairment at end of treatment; both groups improved at 6-month follow-up

Electroconvulsive therapy is the most effective acute treatment for severe, treatment-resistant depression. It is also the most stigmatized and the most cognitively costly. Ketamine — which produces rapid antidepressant effects via NMDA receptor antagonism within hours of infusion — has been used increasingly as a less invasive alternative, but without a head-to-head RCT against ECT, clinicians and patients were making the comparison in an evidential vacuum.

The ELEKT-D trial is the first adequately powered randomized comparison of the two. Its findings are clinically important and more nuanced than either camp will prefer.

What the Trial Found

The primary finding — ketamine non-inferior to ECT for response — will likely drive adoption. But the secondary finding that ECT was superior for remission (47% vs 29%) deserves equal weight in clinical decision-making.

The distinction between response and remission is not semantic. Response (≥50% reduction in MADRS score) means meaningful improvement. Remission (MADRS ≤10 or equivalent) means effectively no clinically significant depression. For patients with moderate treatment-resistant depression who are functional and managing safety, a response may be adequate. For patients with severe psychomotor retardation, inability to work or care for dependents, or active suicidality requiring rapid sustained improvement, the remission gap matters substantially.

The cognitive side effect profile diverges sharply. ECT produced measurable impairment in autobiographical memory and processing speed that was detectable at end of treatment. Ketamine produced negligible cognitive impairment. Both groups had normalized to similar cognitive profiles by 6-month follow-up, which supports the long-standing clinical observation that ECT's cognitive effects are largely transient — but the transience doesn't eliminate the intratreatment burden for patients who are working, driving, or managing caregiving responsibilities during treatment.

What This Changes in Practice

For the ECT vs. ketamine conversation: Clinicians now have RCT data to structure the discussion rather than extrapolating from mechanism and case series. The question is no longer "does ketamine work as well as ECT?" but "what outcome does this specific patient need, and how does the risk-benefit profile differ for them?"

For patients with psychomotor and functional severity: ECT's remission advantage is strongest in severe presentations. The 18-point remission gap (47% vs 29%) is clinically significant; for patients where full remission is the functional threshold that makes resuming work or safe parenting possible, ECT's additional efficacy warrants prioritizing it despite the cognitive and logistical costs.

For patients for whom cognitive safety is a primary concern: Ketamine's equivalent response with lower cognitive impact supports it as a first-choice neuromodulation strategy when the patient or family identifies cognitive continuity as a priority — especially for patients managing executive function-dependent roles, or older adults with pre-existing cognitive vulnerability.

For the stigma conversation: The ELEKT-D result provides a platform for reframing ECT's role. It is not that ECT is replaced by ketamine — it is that the two are tiered by outcome goal. ECT for remission. Ketamine for rapid response with lower cognitive cost. Neither is a treatment of last resort; both are evidence-based acute interventions that should be discussed earlier in the treatment sequence.

The 6-Month Picture

Both groups maintained improvement at 6-month follow-up at comparable rates, and cognitive differences resolved. This provides some reassurance that ECT's intratreatment cognitive effects do not translate to long-term cognitive disadvantage — an important point for informed consent discussions.

The ELEKT-D RCT (N=403) showed ketamine is non-inferior to ECT for treatment response (55% vs 41%) but inferior for remission (29% vs 47%). The two are not equivalent alternatives — they are different tools for different outcome goals. ECT for full remission in severe presentations. Ketamine for rapid response with a more favorable cognitive profile.

Limitations

Non-inferiority design is appropriate for establishing equivalence but not superiority; the trial was powered for non-inferiority on response, not equivalence on remission. Study excluded psychotic features, which are among the strongest indications for ECT. All participants were at academic centers — generalizability to community settings is unknown. IV ketamine protocol (not intranasal esketamine/Spravato) — the two are not bioequivalent and findings do not directly translate to esketamine. Open-label design (neither blinding of patients nor raters was possible) introduces performance and measurement bias. Maintenance treatment after the acute phase was not standardized.

Source
New England Journal of Medicine
Ketamine versus ECT for Nonpsychotic Treatment-Resistant Major Depression
2023-08-31·View original
Tags
treatment-resistant-depressionectketamineneuromodulationclinical-trialpharmacotherapycognitive-effects
Related
Industry
Medicare Is Expanding TMS Coverage — But the Eligibility Criteria Still Block the Patients Who Need It Most
Centers for Medicare & Medicaid ServicesRead →
Research
A Single Dose of Methylone Reduces PTSD Symptoms Within 10 Days — No Psychotherapy Required
JAMA PsychiatryRead →
Research
Two Conditions, One Stimulation Target: rTMS for Chronic Low Back Pain With Insomnia
The Journal of PainRead →
PsyReflect · Free · Mon & Thu
Get analyses like this every Monday and Thursday.
Only what matters for practice. Curated by a clinical psychologist. 5 minutes instead of 4 hours of monitoring.
← Previous
In Group Therapy, Alliance and Cohesion Predict Outcome Equally — and That Changes What You Should Measure
Next →
How Genocide Trauma Passes Between Generations: A Systematic Review of 36 Studies Shows Shared Mechanisms Across Cultures