Ketamine for depression: what the evidence really shows

Ketamine is an anaesthetic that, at low ("subanaesthetic") doses, produces rapid antidepressant effects. Standard antidepressants — SSRIs and SNRIs — act on serotonin and take weeks to declare themselves. Ketamine works through a different system entirely.^[6]

Mechanistically, ketamine primarily blocks the NMDA glutamate receptor. This reduces the activity of inhibitory GABA interneurons, which produces a burst of glutamate release. That glutamate activates AMPA receptors, triggering downstream cascades — including BDNF–TrkB and mTOR signalling — that promote synaptogenesis: the growth and strengthening of synaptic connections thought to be lost in depression. Neuroimaging also shows ketamine shifts activity across large-scale brain networks (default mode, salience, central executive) that are disrupted in depression. The full mechanism is still being worked out and likely involves more than NMDA blockade alone.^[6]

This "regrow the connections" model is why a brief exposure can produce effects lasting days to weeks: the drug is a trigger for change, not something that has to stay in the system to keep working.

The signature of this drug class is speed. In the pivotal 2025 monotherapy trial — esketamine given alone versus placebo in 378 adults with treatment-resistant depression — esketamine separated from placebo on the MADRS depression scale by about 24 hours after the first dose, and the effect grew over four weeks.^[2]

At 24 hours the MADRS advantage over placebo was 3.8 points for the 56 mg dose and 3.4 points for the 84 mg dose. By day 28 it was 5.1 points (56 mg) and 6.8 points (84 mg), all statistically significant, with observed effect sizes of 0.48 and 0.63 for the two doses.^[2]

Contrast that plainly with conventional treatment: antidepressants typically show no clear benefit over placebo for the first one to two weeks. Same-day movement is the clinical novelty here.

Intravenous (racemic) ketamine is given as an infusion, usually 0.5 mg/kg over about 40 minutes. It is used off-label for depression and has been given in specialist clinics for years.^[4]

Esketamine (Spravato) is the S-enantiomer of ketamine, delivered as a nasal spray. It was FDA-approved for treatment-resistant depression in 2019 (taken with an oral antidepressant) and, in 2020, for depressive symptoms in major depression with acute suicidal ideation. In 2025 it became the first drug FDA-approved as monotherapy for treatment-resistant depression, on the strength of the trial described above.^[1][2]

Both are given under medical supervision, with a monitoring period after each dose. Direct head-to-head randomised comparisons of the two forms are still limited; each has a strong evidence base against placebo or an active comparator, and the choice in practice usually comes down to access, route and monitoring requirements rather than a proven efficacy gap between them.

ELEKT-D, published in the New England Journal of Medicine in 2023, was an open-label randomised non-inferiority trial in 403 patients with non-psychotic treatment-resistant depression comparing IV ketamine with electroconvulsive therapy (ECT). Response — a 50% or greater reduction in symptoms — was 55.4% with ketamine versus 41.2% with ECT; ketamine was non-inferior to ECT. ECT was associated with a decline in memory recall during treatment that gradually recovered, while ketamine caused transient dissociation instead.^[4]

ESCAPE-TRD, also 2023, was a phase 3b randomised trial in 676 patients comparing esketamine nasal spray with extended-release quetiapine — an active comparator — both added to an SSRI or SNRI. Remission at week 8 (MADRS ≤10) was 27.1% with esketamine versus 17.6% with quetiapine (P=0.003); patients on esketamine were about 1.5 times more likely to reach remission.^[3]

For wider context, a 2024 network meta-analysis of 72 randomised trials and 12,105 participants ranked ECT, ketamine, esketamine and psilocybin among the most effective, best-balanced options for treatment-resistant depression, while quetiapine and brexpiprazole did not significantly beat placebo on response — though esketamine and psilocybin had lower tolerability.^[7]

Durability is the harder question for any rapid treatment. In ESCAPE-TRD, of the patients who reached remission at week 8 on esketamine, most stayed well: 21.7% of the whole esketamine group were in remission at week 8 and relapse-free through week 32, versus 14.1% on quetiapine.^[3]

This is why these drugs are given as a course rather than a one-off: an induction phase — typically twice weekly — followed by tapering maintenance dosing to hold the response. Stopping abruptly after a single good response commonly leads to relapse.

One honest limitation: robust evidence largely covers the first several months to a year, and very long-term efficacy and safety data are still accumulating.

A distinct use is the rapid reduction of symptoms in people in acute suicidal crisis. In the ASPIRE II trial — 230 patients with major depressive disorder and active suicidal ideation with intent — esketamine plus standard care reduced MADRS depression scores at 24 hours significantly more than placebo plus standard care, a 3.9-point greater reduction (P=0.006).^[5]

The nuance matters, and it deserves to be stated plainly: both groups — including those receiving intensive standard hospital care — showed similar rapid improvement in the specific measure of suicidality itself, so the between-group difference on suicidality was not statistically significant. Ketamine's clearest and fastest effect is on depressive symptoms; it is an adjunct to comprehensive crisis care, not a stand-alone fix for suicidality.^[5]

If you or someone you know is in crisis, contact your local emergency services or a suicide prevention line such as 988 (US & Canada).

In the monotherapy trial, the common acute effects during and after dosing were nausea (about 25%), dissociation (about 24%), dizziness (about 22%) and headache (about 19%). These generally appear shortly after dosing and resolve within roughly one and a half to two hours, which is why patients are monitored on site.^[1][2]

There is also a transient rise in blood pressure and heart rate during dosing, which is monitored for. Dissociation — the feeling of being detached from your body or surroundings — is expected and temporary in supervised use.

Ketamine is a controlled substance with real misuse potential. Supervised, in-clinic dosing and the Spravato REMS programme, which does not permit take-home dosing, exist to manage that risk. Heavy recreational or repeated non-medical use is separately linked to bladder damage (ketamine cystitis) and cognitive effects — a further reason medical use is dose-limited and supervised.

The bottom line: side effects are mostly short-lived and manageable under supervision, but this is emphatically not a self-administered treatment.

Ketamine and esketamine have earned a real place in the care of treatment-resistant and acute severe depression: fast, effective for many people who have failed other drugs, and now — in esketamine's case — approvable as a stand-alone therapy.^[2]

They are not first-line for typical depression, they are not a cure, and they are not a supplement. They require medical assessment, supervised dosing, a maintenance plan and ongoing monitoring. If you want to understand how our physicians approach complex cases, our clinical team and philosophy is a good starting point — and because sleep so strongly shapes mood, Peak Human's guide to sleep optimization is worth reading alongside any depression treatment plan.

The most reasonable framing is this: a powerful, rapid tool for a specific, hard-to-treat group, best used inside a comprehensive treatment plan. The evidence is strong and growing — but the longer-term chapters are still being written.