Ketamine therapy, explained

What ketamine is

Ketamine is an anaesthetic, first approved in the United States in 1970, where it is a Schedule III controlled substance licensed for injection into a vein or muscle. It blocks the NMDA receptor, a receptor for the brain chemical glutamate; how this affects depression is not fully understood.

Standard ketamine is a mix of two mirror-image molecules, arketamine and esketamine. Even at the low doses used for depression, most people feel dissociation: a sense of detachment from body, surroundings or time.

How it is given

Esketamine nasal spray, sold as Spravato, is approved for depression in the US and the EU. The US Food and Drug Administration (FDA) approved it in March 2019, with an oral antidepressant, for treatment-resistant depression, usually defined as depression that has not improved after at least two antidepressants. A 2020 approval added adults with major depression and acute suicidal thoughts or behaviour, again alongside an oral antidepressant, and since January 2025 it may be used alone for treatment-resistant depression. In the US it is given only in certified clinics, where staff observe the person for at least two hours.

The EU authorised it on 18 December 2019 for treatment-resistant depression in adults, with an SSRI or SNRI antidepressant. It is also authorised, with an oral antidepressant, for short-term use when depressive symptoms amount to a psychiatric emergency. In England, NICE decided in December 2022 not to recommend it for treatment-resistant depression, so the NHS generally does not offer it.

Infusions use generic ketamine off-label; the FDA has not approved ketamine for any psychiatric disorder. The best-studied dose is 0.5 mg for every kilogram of body weight, infused over 40 minutes. Muscle injections, lozenges (troches) and oral liquids are much less studied. In October 2023 the FDA warned that compounded oral and under-the-tongue ketamine, supplied through telemedicine for home use, varies in dose and leaves nobody present to watch for sedation, dissociation or blood pressure changes.

Ketamine-assisted psychotherapy (KAP) adds talking therapy. In one model, a course of infusions is followed by structured therapy such as cognitive behavioural therapy. In another, preparation sessions come first, therapy happens during dosing, and later integration sessions help the person make sense of the experience. Here the altered state is part of the therapy, not a side effect.

What the research shows

A 2021 Cochrane review of 64 randomised trials of ketamine and similar drugs, with 5,299 participants, found that ketamine may improve response and remission over placebo at 24 hours, but rated this evidence very low certainty. The evidence that esketamine increases remission at 24 hours was rated moderate certainty.

In the open-label ELEKT-D trial, 403 people with treatment-resistant depression without psychosis were randomly assigned to three weeks of ketamine twice weekly or electroconvulsive therapy (ECT) three times weekly. Response rates were 55.4% and 41.2%, so ketamine was judged no worse than ECT. ECT was linked with a temporary dip in memory recall, ketamine with dissociation.

Only one of the three main four-week esketamine trials met its main statistical test. In ESCAPE-TRD, an open-label trial of 676 adults with blinded raters, 27.1% were in remission at week 8 on esketamine against 17.6% on add-on quetiapine. In people in stable remission, relapse by 24 weeks was 32% on continued esketamine and 46% after a switch to placebo.

Blinding is a weak point: people can usually tell they have had ketamine. When 40 adults with depression received ketamine or saline under surgical anaesthesia, so that neither they nor the staff knew which, both groups improved similarly over three days.

A 2026 review of 11 KAP studies in treatment-resistant depression found falling scores, but the three with a comparison group showed no significant difference. In a placebo-controlled trial of 96 adults with severe alcohol use disorder, three weekly ketamine infusions led to about 10 percentage points more days without alcohol at six months, most clearly alongside mindfulness-based relapse prevention therapy. Relapse rates did not differ.

Compounded ketamine is marketed in the US for anxiety, PTSD and OCD, yet in 2017 an American Psychiatric Association (APA) task force noted how little evidence existed outside depression. Long-term data are limited, although an open-label study followed 1,148 adults on esketamine for an average of 42.9 months and found no new safety concerns.

What a course and a session look like

The APA task force advised offering ketamine only after adequate trials of standard treatment, with consent covering the limited evidence, likely repeat doses, and concerns about thinking and memory, the bladder and misuse. Oxford’s NHS-run self-pay service, for example, requires at least two antidepressants tried for six weeks each, one psychological therapy and a GP or psychiatrist’s referral.

At an infusion, a pump delivers the dose into a hand or forearm vein over about 40 minutes while the person lies or reclines, with a nurse present. Effects can last 30 to 120 minutes after the pump stops, and a responsible adult should take the person home. Oxford asks people not to cycle, drink alcohol, sign legal documents or care for dependants until the next morning.

Trials of repeated dosing usually gave four to six infusions, twice a week; the APA task force suggested stopping if there is no meaningful benefit after two weeks. In Oxford, about half of patients continue after a first course of up to six to eight infusions, usually with infusions every four to eight weeks.

Before esketamine, the person avoids food for two hours and drinks for 30 minutes, then uses two or three sprayers five minutes apart. Blood pressure is checked before the dose and about 40 minutes after. Sessions run twice a week for four weeks, weekly in weeks five to eight, then weekly or fortnightly.

Oxford’s 2026 self-pay prices, for example, are £265 for the first assessment and for each infusion, and £591 or £754 per esketamine session, depending on dose.

Side effects and risks

In esketamine trials the most common side effects were dizziness (31%), nausea (27%), dissociation (27%), headache (23%), sleepiness (18%), altered taste (18%), vertigo (16%), numbness (11%), vomiting (11%) and raised blood pressure (10%).

In clinical trials, 3% to 19% of people on esketamine, against 1% to 4% on placebo, had a rise of at least 40 mmHg systolic or 25 mmHg diastolic in the first four weeks, and 0.3% to 0.4% lost consciousness. Slowed breathing has been reported since approval, mostly with other sedating drugs or in people with heart, lung or weight-related conditions.

The Oxford service warns of panic during a dose, of depression and suicidal thoughts occasionally worsening for up to two weeks afterwards, and of mania. Esketamine has not been shown to prevent suicide and does not replace hospital care when it is needed.

Painful bladder inflammation (ulcerative or interstitial cystitis) has been reported after long-term misuse or off-label use of ketamine; esketamine trials found more urinary symptoms than placebo but no such cases. Memory problems have been reported with repeated misuse, although thinking tests stayed stable in esketamine studies of up to three years. Liver damage has been reported with chronic use.

Who should not have it

• Formal contraindications. Esketamine must not be used by people who are allergic to ketamine or esketamine, have aneurysmal disease of the aorta or of arteries in the brain or limbs, or have had a bleed into the brain. The US also lists arteriovenous malformations, and the EU a heart attack or other cardiovascular event in the past six weeks.
• High blood pressure. A reading above about 140/90 mmHg before a dose (150/90 from age 65, in the EU) means the prescriber should weigh up a delay. Unstable heart or lung disease needs resuscitation equipment and trained staff at hand.
• Psychosis and bipolar disorder. Current or past psychosis, mania or bipolar disorder calls for careful assessment, and treatment only if benefits outweigh risks. Oxford normally declines people with current or recent psychosis or past mania triggered by drugs or medicines.
• Substance use. A history of drug misuse raises the risk of misusing ketamine. Oxford normally excludes anyone who has used illegal drugs in the past two years.
• Pregnancy and breastfeeding. Esketamine is not recommended in pregnancy, as animal studies of ketamine showed harm to the developing brain. Contraception is advised, and the US label advises against breastfeeding.
• Other medicines. Benzodiazepines, opioids and alcohol add to sedation. Stimulants and MAOI antidepressants can push blood pressure higher.

Choosing a clinic or practitioner

In the EU a psychiatrist should decide on esketamine. For infusions, the APA task force advised a prescriber licensed for controlled drugs and ready to manage heart and behavioural emergencies, ideally with advanced cardiac life support training. The clinic needs heart and breathing monitoring and a plan for transfer to hospital.

A careful service has written rules on blood pressure before dosing and on when to stop an infusion. It reviews response early, keeps maintenance at the lowest frequency that works, checks thinking, bladder function and substance use during long-term treatment, and keeps the person’s own GP or psychiatrist informed. The APA task force strongly advised against taking ketamine home to self-administer. For KAP, useful questions are who is present during dosing, what training they have and who handles medical monitoring. Guaranteed results, or treatment without medical screening, are reasons to look elsewhere.