Ketamine: What It Actually Treats, What Aftercare Is Real, and Why It Wears Off

Summary

Ketamine's best-evidenced use is the one nobody is selling — it has been on the World Health Organization's essential medicines list as an injectable anaesthetic since 1985, and in much of the world it is the only anaesthetic that works without piped oxygen or reliable electricity — while the psychiatric uses that generate all the money and all the content sit several rungs lower: a real, statistically significant but modest acute antidepressant effect in treatment-resistant depression from a registration programme in which only one of three short-term trials met its primary endpoint, a suicid

Why Moderate

Moderate overall, because the entry's centre of gravity is the psychiatric and recovery material, and that material sits squarely in the middle: one acute effect whose direction is well supported across three designs but whose registration programme went one-for-three on short-term primary endpoints, one strong finding about durability that is genuinely well supported from four directions, and then a rapid fall-off through Moderate and Emerging into fifteen-person proof-of-concept work. Packaging the whole entry higher would let the anaesthetic evidence launder the psychiatric claims; packaging it lower would misrepresent both the depression evidence and the durability finding, which are real.

NOT the tier above (Strong Evidence) because the uses people actually arrive here for are not strongly evidenced. The suicidality claim failed its own key secondary endpoint and the label refuses it. Two of the three short-term esketamine registration trials missed their primary endpoints. Chronic pain is moderate for one condition and weak or absent for everything else. PTSD, alcohol use disorder and cocaine use disorder are single unreplicated trials of thirty, ninety-six and fifty-five people. The aftercare and integration protocol that names this entry is almost entirely convention. And the mechanism, which is often narrated as settled, has one human test on record that contradicted its central animal finding.

NOT the tier below (Emerging) because two claims in this entry are genuinely solid and it would be dishonest to drag them down. Ketamine's status as essential anaesthetic infrastructure is beyond dispute — on the WHO essential medicines list since 1985, and in much of the world the only anaesthetic that works without piped oxygen or reliable electricity. And the durability finding — fast onset, no durable hold, every effective protocol a repeat-exposure protocol — is supported independently by a responder-durability median, a randomised-withdrawal trial with hazard ratios of 0.49 and 0.30, a multi-society pain guideline's explicit absence of long-term benefit, and the approved product's own maintenance dosing schedule.

The internal split — read this, not the headline:
• Anaesthesia, procedural sedation and acute analgesia: Foundational — decades of global use, essential medicines listing since 1985, and the one use with no commercial interest behind it.
• Esketamine produces a statistically significant acute improvement in treatment-resistant depression: Strong Evidence for the direction only, with the effect size (four MADRS points), the manufacturer sponsorship, and the one-of-three short-term primary-endpoint record all stated rather than rounded away. The magnitude is not settled and the registration package alone would not carry this tier.
• The effect is fast and does not hold without repeat dosing: Strong Evidence — supported from four independent directions, and the organising fact of the entry.
• Maintenance dosing prevents relapse in responders: Strong Evidence for the direction, from a randomised-withdrawal design that is simultaneously the cleanest available test, the design most flattering to any drug put through it, and the trial the original approval leaned on hardest.
• Supervised monitoring, at least two hours, no take-home dosing, no driving until the next day: Foundational as a regulatory requirement — this is the only part of the aftercare protocol that is mandated and specific.
• Esketamine reduces depressive symptoms rapidly in acutely suicidal patients: Moderate — primary endpoint met, on top of at least five days of hospitalisation that everyone in the trial received.
• Ketamine reduces suicidality itself: not supported at any tier — key secondary endpoint not met, and the label states effectiveness in preventing suicide or reducing suicidal ideation has not been demonstrated.
• Intravenous ketamine for complex regional pain syndrome: Moderate, bounded at up to twelve weeks.
• Intravenous ketamine for fibromyalgia, low back pain, migraine, phantom limb pain, postherpetic neuralgia, cancer pain, ischaemic pain: weak or no evidence by the multi-society guidelines' own grading, and no intermediate- or long-term benefit outside CRPS.
• Repeated infusions for chronic PTSD: Emerging — one thirty-person trial with a psychoactive control, replication status unverified.
• Ketamine plus relapse-prevention therapy increases days abstinent in severe alcohol use disorder: Emerging — one ninety-six-person phase 2 trial, one of two primary outcomes null, lower confidence bound near zero, unreplicated, authors' own words "proof of concept."
• A single infusion plus behavioural therapy increases abstinence in cocaine use disorder: Emerging — one fifty-five-person single-site trial.
• Ketamine for smoking cessation: no claim — no source located, and none appears in this entry.
• Ketamine for bipolar depression or OCD: Experimental — fifteen-person crossover trials with effectively impossible blinding.
• Severe uropathy in heavy chronic recreational use: Strong Evidence, with a characteristic symptom picture, a local mechanism, and a documented severity spectrum through fibrosis and contracted bladder to upper-tract involvement and surgery.
• Uropathy substantially improves on cessation in most but not all cases: Moderate — roughly three-quarters substantially improved within three months, a minority left with lasting damage.
• Therapeutic-dose ketamine does not appear to elevate urological risk: Moderate, and this is the independent review's own conclusion — bounded hard by four-week median follow-up, fifteen percent of studies at low risk of bias, and laboratory urinalysis in only two of twenty-seven.
• Ketamine cholangiopathy: Moderate, chronic heavy use only, on a low-barrier journal vehicle whose specific figures should be traced before being leaned on.
• Abuse liability, tolerance and physical dependence with prolonged use: Strong Evidence — on the label, in the boxed warning, and not softened by clinical framing.
• No transition to illicit use reported in supervised trials: Emerging — reassuring, in populations that mostly excluded active substance use disorder and were followed briefly, and against four clear tolerance-or-dependence signals across 2,174 depression patients.
• Higher benzodiazepine doses blunt the antidepressant response: Emerging — one retrospective analysis of forty-seven patients, with a contradicting time course reported elsewhere that we did not read.
• The mTOR-synaptogenesis mechanism: Emerging — established in rats, contradicted in the one human test.
• The "neuroplasticity window" as a clinical parameter: not tiered as evidence — clinical convention built on rodent timelines.
• Paid professional integration is necessary: not tiered as evidence — a commercial position with no trial isolating it.
• Racemic ketamine is more efficacious than esketamine: not settled — one indirect meta-analytic comparison favouring racemic, carrying an erratum and a published challenge, with the head-to-head randomised test not reporting until 2030.
• Real-world response rates around sixty to seventy-five percent for both agents: UNVERIFIABLE, cut — no primary source; traces to clinic and telehealth marketing.

Practical takeaway

The framing to hold. Ketamine is a controlled substance and every use described in this entry is a clinician-supervised one. This section exists to help someone who is considering, undergoing, or has just had supervised treatment understand what the aftercare is actually for and which parts of it rest on anything. It is not a route to obtaining ketamine, and it contains no dosing.

First: is your question one this drug has an answer to? Route by the map, not by the marketing.
• Anaesthesia, procedural sedation, acute analgesia — this is the settled ground and it is a decision made by an anaesthetist, not by you.
• Depression that has not responded to two or more adequate antidepressant trials — this is the psychiatric indication with real trial support, and the honest expectation is a fast, moderate improvement that will need maintaining. It is also worth knowing that only one of the three short-term registration trials met its primary endpoint, so the size of the benefit is less settled than the approval sounds. Ask about the maintenance plan before the first dose, not after the third.
• Acute suicidal ideation — ketamine reduces depressive symptoms rapidly in this population and has not been shown to reduce suicidality itself. The intensive standard of care that surrounded everyone in those trials — which included hospitalisation for at least five days — is not the optional part. If you are in crisis, that is an emergency route, not a treatment-shopping decision.
• Complex regional pain syndrome — the one chronic pain indication with moderate guideline support, bounded at around twelve weeks.
• Fibromyalgia, low back pain, migraine, mixed neuropathic pain, phantom limb pain, postherpetic neuralgia — the multi-society guidelines grade the evidence weak or absent for all of these. If this is your condition, fibromyalgia_lifestyle_and_management and low_back_pain_evidence_and_management are the better use of your time and money.
• PTSD — one thirty-person trial with a strong result and a twenty-seven-day median durability. Interesting, not established.
• Alcohol or cocaine use disorder — one phase 2 trial each, both unreplicated, one with a null co-primary outcome. Worth discussing with a clinician who knows the evidence. Not a substitute for the treatments that do have a mature evidence base, and the recovery-relevant signal from the alcohol trial is that the structured psychological therapy carried more of the effect than the drug did. The wider argument belongs to psychedelic_assisted_addiction_interruption.
• Bipolar depression, OCD, anything else — fifteen-person crossover trials or nothing. Not a basis for a treatment decision.

The aftercare protocol, with each element labelled by what it rests on.

Regulator-mandated, specific, non-negotiable for the approved product:
• Stay for the full monitoring period — at least two hours after each dose, followed by a clinical assessment of stability before you leave. Two hours is the floor, not the release time.
• Do not drive or operate machinery until the next day, after a restful sleep. This is the label's own instruction, and it is the single most concrete aftercare rule in existence. Arrange the lift before you go, not afterwards.
• No take-home dosing. The approved product is not dispensed for home use, at all. If a service is mailing you ketamine, you are outside the framework this rule belongs to, and the regulator has issued two alerts specifically about that.

Sensible, mechanistically coherent, and not tested:
• Hydrate on treatment days and void frequently. This is the clinical risk-reduction advice around bladder exposure, and it is mechanistically reasonable because the injury depends on cumulative contact between urine and the bladder lining. It is expert recommendation, not a trial result.
• Clear the rest of the day. Nothing about the "neuroplasticity window" is a measured human parameter, but sedation, dissociation, altered judgment and impaired reaction speed are all documented, so an empty afternoon is a straightforwardly good idea for reasons that have nothing to do with plasticity.
• Do not make consequential decisions the same day. Clinical convention, and consistent with the label's statement that attention, judgment and thinking are impaired. No trial has tested it. It costs nothing.
• Let the nervous system settle rather than driving it. Blood pressure rises peak around forty minutes and last around four hours; a quiet, low-stimulation, unhurried few hours matches that time course. The general down-regulation practices are at autonomic_nervous_system_balance and vagal_tone_practices, and this entry defers to them rather than reinventing them.

Commercial practice presented as necessity, and labelled as such:
• Structured paid integration sessions have never been shown to be necessary. Reflecting on the experience is worth doing, writing it down is worth doing, and doing that work with a therapist you already trust or with the treating clinician has never been shown to be inferior to buying a package — because the comparison has not been run. The one place structured psychological work demonstrably mattered alongside ketamine was manualised relapse-prevention therapy delivered inside a randomised trial. If you are being told that self-directed integration is inadequate, ask who is telling you and what they sell.
• The "window" as a priced product. Treat it as a reason to protect your calendar, not as a biological deadline you are failing to meet.

What to ask the person prescribing, before the first dose. These are the questions the evidence in this entry says are load-bearing, and none of them are confrontational.
1. What does maintenance look like, what will it cost, and what is the exit? The effect wanes; every effective protocol repeats. A plan that has no answer to "and then what" is a subscription, not a treatment plan.
2. Am I on a benzodiazepine, and at what diazepam-equivalent dose? There is a real signal that higher doses blunt the antidepressant response. Never change a benzodiazepine dose yourself — abrupt reduction carries seizure risk, which is far better established than the interaction.
3. How will my urinary symptoms be monitored, and how often? Clinicians who follow this literature screen urine periodically during maintenance. Ask whether yours does.
4. Do I have a history of depersonalisation or derealisation, and does that change the calculation? Dissociation is the mechanism, not an incidental effect.
5. Is what I am being given the approved product, or a compounded formulation, and what does that change about oversight?

The part that is not in the drug. The durable levers for depression, anhedonia and stress physiology are the ordinary ones, and they are the thing a window — if it exists — would be worth opening for. This entry does not re-argue them: see depression_lifestyle_interventions, anhedonia_reward_recovery, chronic_stress_management and hpa_axis_dysregulation.

Evidence detail

Why This Entry Exists

Ketamine is the hardest kind of topic to write honestly, because almost everyone describing it has a position that pays. The molecule is simultaneously a World Health Organization essential medicine, a Schedule III controlled substance in the United States and a Class B drug in the United Kingdom, a branded nasal spray sold under a restricted safety programme with a five-part boxed warning, a cheap generic infused off-label in hundreds of private clinics, a compounded lozenge mailed to people's homes by telehealth companies the regulator has twice warned about, and a recreational drug that in heavy chronic use destroys bladders. Every one of those is true at once. Any account that reads cleanly in one direction has dropped something.

This entry exists because someone whose life the drug might genuinely touch — through depression that has not responded to anything else, through chronic pain, or through trying to get out from under alcohol — deserves the version where the numbers are attached to the studies they came from, the sample sizes travel with the effect sizes, and the sentence "ketamine is FDA-approved for depression" is unpacked into what is actually approved (esketamine nasal spray, in a certified clinic, under direct observation) rather than left to do work for products it does not cover.

Two things in this entry are unusual enough to state up front. The first is the durability problem, and it is the organising fact. Ketamine produces one of the fastest antidepressant effects known — hours, not weeks — and it does not hold on its own. Among PTSD responders to a six-infusion course, the median time to losing that response was twenty-seven and a half days. Outside complex regional pain syndrome, the multi-society pain guidelines find no evidence of intermediate- or long-term benefit at all. The approved nasal spray's own dosing schedule is twice weekly, then weekly, then weekly or fortnightly — that is a maintenance schedule, not a course of treatment. So the honest framing is: fast, real, and not durable by itself. Everything downstream follows from that. It is why the treatment is structurally a subscription, why the clinic sector describes waning effect as "maintenance" rather than as a limitation, and why the long-term consequences of years of repeat exposure are the single most important unanswered question about the whole field.

The second is the aftercare and integration apparatus, which is where this entry's name comes from and where the evidence is thinnest. There is exactly one part of it that is genuinely mandated and specific: the approved product's safety programme requires administration in a certified healthcare setting under direct observation, monitoring for at least two hours after each dose, no dispensing for home use at all, and no driving or other hazardous activity until the next day following a restful sleep. Everything else — the rest of the day cleared, the hydration, the no-big-decisions rule, the journaling, the structured integration sessions, the "neuroplasticity window" you are told to exploit — ranges from sensible precaution to clinical convention to a commercial product. None of it is wrong. Almost none of it has been tested. The one place psychological work has been shown to matter alongside ketamine is a trial where the therapy was manualised relapse-prevention therapy delivered by a research team, which is not the same thing as a coaching package sold by the hour, and even there the confidence intervals were wide enough that the authors called their own result a proof of concept.

What bad advice this protects against, in all directions:
• "Without professional integration sessions the treatment is wasted — the drug only opens the door, the integration is what heals you." → This is the most confidently asserted unevidenced claim in the field, and it is asserted almost entirely by people who sell integration coaching. No trial in this entry isolates paid integration as a component and shows it adds benefit. The closest thing to supporting evidence runs the other way on specifics: in the alcohol trial, the arm that combined ketamine with a structured psychological therapy did produce the largest single cell effect, but the therapy was a manualised relapse-prevention protocol delivered inside a randomised trial, the abstract reports no formally significant interaction between the drug and the therapy, and the whole trial was ninety-six people across four arms. That supports "structured psychological work alongside the drug is worth taking seriously." It does not support "you must buy a package or you have wasted your money." Reflecting carefully on your own, with a therapist you already have, or with the treating clinician, has never been shown to be inferior — because it has never been tested against anything.
• "Three phase 3 trials proved it works — that is what approval means." → It is not what this approval means. Of the three short-term phase 3 trials in the esketamine registration programme, only one — TRANSFORM-2, the trial quoted throughout this entry — separated from its active control on its primary endpoint. The other two did not. A regulator normally wants two adequately powered positive short-term trials before approving an antidepressant, and here the approval leaned unusually heavily on a randomised-withdrawal maintenance trial instead — a design that asks whether taking a drug away from people already doing well on it causes relapse, which is a different question from whether starting it works. None of that makes the drug ineffective: the positive short-term trial is real, the withdrawal trial is real and strongly positive, and the trial in acutely suicidal patients met its depression endpoint. It does mean the phrase "FDA-approved" is doing less work than people think it is, and that anyone citing the approval as shorthand for a robust efficacy package is citing something that was argued about at the advisory committee.
• "There's a neuroplasticity window of about seventy-two hours (or two weeks) after dosing, and what you do in it determines the outcome." → The window is a clinical convention that has hardened into a stated fact. Its origin is rodent work: in rats, ketamine rapidly activated the mTOR pathway and increased synaptic signalling proteins within hours, and blocking that pathway in the animals' prefrontal cortex abolished both the synaptogenesis and the antidepressant-like behaviour. No human trial has randomised the timing or the content of post-dose activity and measured what it does to outcome. Worse for the tidy story, the one human test of the mechanism went the other way: giving depressed patients rapamycin, which blocks mTOR, did not block ketamine's antidepressant effect — it prolonged it. Treat the window as a reasonable organising heuristic for clearing your calendar. Do not treat it as a measured biological parameter, and be suspicious of anyone selling you a service priced against it.
• "Ketamine is FDA-approved for depression, so the lozenges my telehealth provider mails me are an approved treatment." → What is approved is esketamine nasal spray, in adults with treatment-resistant depression, administered under the direct observation of a healthcare provider in a certified setting, with at least two hours of monitoring, and with an explicit prohibition on dispensing it for home use. Compounded racemic ketamine — oral, sublingual, or nasal — is not approved for any psychiatric indication. The FDA has issued two separate alerts about compounded ketamine, in February 2022 and October 2023, the second stating that safe and effective dosing of ketamine for any psychiatric disorder has not been established and that at-home administration is risky specifically because no provider is present to monitor for sedation and dissociation.
• "One infusion and the depression lifted — it's a cure." → It is the opposite of a cure, structurally. The effect arrives fast and decays. The trial evidence for keeping it is a randomised-withdrawal design in which people who had responded to the nasal spray and then had it swapped for placebo relapsed substantially more often than those who continued. That is legitimate evidence that maintenance works, and it is also, by design, the study shape most likely to make a drug look indispensable — because it enriches for responders and then removes the drug from half of them. Both readings are correct simultaneously and you should hold both.
• "Ketamine stops suicidal thinking — it's the emergency treatment for suicidality." → This is the claim that fails hardest against its own source. The trial run specifically in people with major depression and active suicidal ideation with intent met its primary endpoint, which was depression severity at twenty-four hours. It did not meet its key secondary endpoint: the between-group difference on the clinician-rated suicidality scale was not statistically significant, and both arms improved rapidly — meaning the intensive standard of care, which for everyone in that trial included hospitalisation for at least five days, also worked. The approved label states in its own Limitations of Use that effectiveness in preventing suicide or in reducing suicidal ideation or behaviour has not been demonstrated, and that the product does not remove the need for hospitalisation where clinically indicated. The honest sentence is that ketamine rapidly reduces depressive symptoms in acutely suicidal people, which is worth a great deal and is not the same claim.
• "Bladder damage is an abuser problem — it doesn't happen at clinical doses" and "ketamine will destroy your bladder." → Both halves have to be said together or the entry misleads. The severe picture is unambiguously a heavy-chronic-recreational phenomenon: pooled across forty-five articles and nearly five thousand patients, urinary frequency in more than three-quarters, urgency in around seventy percent, suprapubic pain in around sixty percent, upper urinary tract involvement in three in ten, and a mean functional bladder capacity of about ninety-five millilitres, in a population using doses far above and far more often than any clinical protocol. The therapeutic literature does not show that. An independent 2025 systematic review of twenty-seven studies in psychiatric treatment found urological symptoms in zero to twenty-four and a half percent of patients, mostly mild to moderate, generally at or below control rates, and concluded that ketamine treatment does not appear to be associated with elevated risk of urological symptoms. But the same review says its median follow-up for urological adverse events was four weeks, only fifteen percent of the included studies were low risk of bias, eleven of them did nothing more systematic than record spontaneously reported side effects, and only two ran laboratory urinalysis. Short-course clinical ketamine has not been shown to wreck bladders and there is reasonable evidence it does not. Nobody has watched maintenance patients long enough, with good enough monitoring, to call years of repeat dosing proven safe. An earlier draft of this entry cited that review backwards — presenting the top of its zero-to-twenty-four-and-a-half-percent range as if it were a central tendency, and using a paper that concluded no elevated risk as evidence that the risk is elevated. That error is corrected here and named so it is not reintroduced.
• "Ketamine treats addiction." → It is the most personally important use case in this entry and the one where the gap between the marketing and the evidence is widest. The alcohol trial is genuinely good work with a real six-month endpoint, and its abstinence result is genuinely positive. It is also ninety-six people across four arms, one of its two primary outcomes was null, and the lower bound of the confidence interval on its main effect sits at about one percent — near zero. The cocaine trial's much-quoted contrast, roughly forty-eight percent abstinence versus eleven percent, comes from fifty-five people, one site, one infusion. Neither has been replicated. Report the sample in the same sentence as the number or do not report the number. The broader argument about whether psychedelic-class drugs can interrupt addictive patterns at all belongs to psychedelic_assisted_addiction_interruption and is not re-argued here.
• "Spravato is just overpriced ketamine — the generic molecule is identical and works the same" and its mirror, "the branded spray is the proven one, the generic is unregulated cowboy medicine." → Neither is established, and the field is finally testing it properly. The published head-to-head meta-analytic comparison does not find the two equivalent, and it favours racemic ketamine rather than the branded product — though that comparison is indirect, carries a published erratum, and drew a formal challenge in print. Meanwhile a Yale-led comparative-effectiveness trial funded by an independent public research institute, not by a manufacturer, is recruiting four hundred participants to randomise intravenous racemic ketamine against the branded nasal spray. It completes in December 2030. Anyone telling you this question is settled today is telling you something the people spending twelve and a half million dollars to answer it do not believe.
• "It's a Schedule III drug, so it's basically safe" and "it's a dangerous street drug with no legitimate medical use." → The first ignores a boxed warning naming sedation, dissociation, respiratory depression, abuse and misuse, and suicidal thoughts and behaviours, and ignores that physical dependence and tolerance have both been reported with prolonged use. The second ignores that this is essential surgical infrastructure for much of the world. When the UK Home Office asked whether ketamine should be moved up to Class A, the statutory advisory council reviewed it and said no — keep it Class B — partly because the evidence that reclassification reduces harm is limited and partly because harsher classification increases stigma and discourages people from asking for help. That recommendation was not unanimous, and the entry says so rather than borrowing more authority from it than it has.
• "Push through the dissociation, that's just the drug working." → It is the drug working, which is exactly the problem for some people. Dissociation is not an incidental side effect here; it is close to the mechanism of action, and it is named in the boxed warning alongside sedation. For anyone with a history of depersonalisation or derealisation, that makes ketamine a distinct and specific caution rather than a general one, and it is a conversation to have with the prescribing clinician before the first dose, not afterwards. See depersonalisation_derealisation_management.

This entry owns the honest use-case map with each indication's evidence tier attached, the durability and maintenance problem and what it implies about cost and exposure, the aftercare and recuperation protocol with each element labelled as regulator-mandated, sensible convention, or commercial product, the critique of paid integration as a necessity claim, the branded-versus-generic patent-economics controversy, and the urological and hepatobiliary risk boundary including where the therapeutic reassurance genuinely holds and where it runs out. It defers the general case for psychedelic-class drugs interrupting addictive patterns to psychedelic_assisted_addiction_interruption, the classic psychedelic-assisted therapy model and its trial architecture to psilocybin_assisted_therapy_clinical, the management of depersonalisation and derealisation to depersonalisation_derealisation_management, the non-drug levers for depression to depression_lifestyle_interventions and anhedonia_reward_recovery, the stress-physiology background to hpa_axis_dysregulation and chronic_stress_management, the post-dose autonomic settling practices to autonomic_nervous_system_balance and vagal_tone_practices, and the conservative management of the two pain conditions the guidelines explicitly found ketamine weak or useless for to fibromyalgia_lifestyle_and_management and low_back_pain_evidence_and_management. Where a question resolves to one of those, this entry hands off rather than re-arguing it.

Safety register, stated once and meant throughout. Nothing in this entry is a route to obtaining or self-administering ketamine, and no dose figure below is guidance. Where doses appear they are trial and label parameters reported so the evidence can be appraised. Every use case described here is a clinician-supervised one, and every clinical decision described here belongs with a clinician.

Evidence

Read the tiers, not the thesis. This entry genuinely spans from Foundational to Experimental, and the spread is the most important thing on the page: the strongest evidence attaches to the use with no commercial interest behind it, and the weakest evidence attaches to the uses generating all the revenue.

The use case map, best-established first.

1. Anaesthesia, procedural sedation and acute analgesia are ketamine's original and best-established uses (Foundational). Ketamine appears on the WHO Model List of Essential Medicines, 23rd list, 2023, under general anaesthetics, as an injectable medicine, and has been listed as an intravenous anaesthetic since 1985. The WHO describes its uses as induction and maintenance of anaesthesia, procedural sedation and analgesia, and notes its particular importance in resource-limited settings, where it is often the only feasible anaesthetic because it does not require piped oxygen, ventilators or reliable electricity. The clinical properties that earn it that place — relative haemodynamic stability compared with agents that drop blood pressure, and relative preservation of airway reflexes and spontaneous respiration — are standard anaesthetic pharmacology consistent with the WHO framing, though we did not verify a named textbook or guideline citation for those two properties specifically and are stating them generically. Note that the approved esketamine nasal spray's label explicitly states it is not approved as an anaesthetic; this is racemic ketamine's territory, not the branded product's. (WHO Model List of Essential Medicines, 23rd list, 2023, WHO/MHP/HPS/EML/2023.02. Foundational — decades of global surgical use, no serious dispute. Correction of record: an earlier draft of this entry dated the listing to 1984 in four places. The available sources give 1985, and the date is corrected throughout so a future reconciliation pass does not reinstate 1984. Cui bono note: there is no commercial interest here at all. Ketamine is generic and cheap, which is exactly why it is on the essential medicines list and exactly why nobody markets it as an anaesthetic. This is the honest anchor for the entire entry: the strongest evidence for ketamine attaches to the use nobody is selling. It cuts against prohibition-era framing that treats ketamine primarily as a drug of abuse. It also does not travel in the other direction — clinics sometimes borrow the safety record of supervised anaesthetic dosing to reassure patients about unsupervised dosing at home, which the WHO listing underwrites in no way whatsoever.)

2. Esketamine nasal spray produces a statistically significant but modest acute improvement in treatment-resistant depression — in the one short-term registration trial of three that met its primary endpoint (Strong Evidence for the direction; the magnitude is not settled and the registration package alone would not carry this tier). The trial the field quotes was a phase 3, double-blind, active-controlled study across thirty-nine outpatient referral centres. Patients whose current antidepressant was not working were switched to flexibly-dosed esketamine nasal spray plus a newly initiated oral antidepressant, or to a newly initiated oral antidepressant plus placebo nasal spray. Four hundred and thirty-five were screened, two hundred and twenty-seven were randomised, and one hundred and ninety-seven completed the twenty-eight-day double-blind phase. The primary outcome was change in the Montgomery-Åsberg Depression Rating Scale at day twenty-eight, and the result was a difference of least-squares means of minus four points (SE 1.69, 95% CI minus 7.31 to minus 0.64). Discontinuation for adverse events was seven percent versus 0.9 percent. The part that is usually left out: this was one of three short-term phase 3 trials in the registration programme, and it is the only one that separated from its control on its primary endpoint. The fixed-dose companion trial and the third short-term trial did not. Regulators normally expect two adequately powered positive short-term trials for a new antidepressant; here the approval leaned unusually heavily on the randomised-withdrawal maintenance trial described in the next claim. The current label carries the treatment-resistant depression indication, as monotherapy or alongside an oral antidepressant — the monotherapy indication was added by supplemental approval on 21 January 2025, six years after the original 5 March 2019 approval — together with a boxed warning naming sedation, dissociation, respiratory depression, abuse and misuse, and suicidal thoughts and behaviours, and access restricted through a formal risk-management programme. (Popova V, Daly EJ, Trivedi M, et al. "Efficacy and Safety of Flexibly Dosed Esketamine Nasal Spray Combined With a Newly Initiated Oral Antidepressant in Treatment-Resistant Depression." Am J Psychiatry 2019;176(6):428-438. PMID 31109201. On the 1-of-3 record: TRANSFORM-1, PMID 31290965, is the fixed-dose companion that did not reach statistical significance on its primary endpoint. Verification scope: the 1-of-3 fact and the reliance on the withdrawal trial were confirmed from the FDA advisory-committee briefing record and from independent published commentary, not by reading the TRANSFORM-1 and TRANSFORM-3 primary papers in this pass — so this entry states the fact and does not quote figures from those two trials. Strong Evidence is assigned to the direction — that the drug produces a rapid, modest acute improvement — because three separate designs produced positive primary endpoints (this trial, the withdrawal trial below, and the trial in acutely suicidal patients). It is not assigned to the magnitude, and it would not survive on the registration package alone. Sponsorship is Janssen Research and Development, the manufacturer, and must travel with the finding. Cui bono note: this is a manufacturer-run trial supporting its own product's approval, and a four-point MADRS difference is a real but modest result that is routinely reported as simply "significantly greater." Clinics and telehealth prescribers cite the approval to imply their own off-label intravenous or oral racemic products are approved; they are not. Critics who say esketamine was approved on thin evidence are pointing at a specific and documented regulatory argument, not a vibe — but the approval and the boxed warning both exist and neither should be waved away.) A commonly circulated pair of effect sizes for racemic infusion — around 1.46 at twenty-four hours and 0.68 at one week — could not be traced to a named primary meta-analysis and is deliberately absent from this entry. The qualitative shape those numbers describe (large and fast at twenty-four hours, substantially smaller by a week) is independently supported and is stated without them.

3. The effect is fast and does not hold, and every effective protocol is therefore a repeat-exposure protocol (Strong Evidence, and this is the finding that organises the entry). The shape is supported from four independent directions. Among PTSD responders to a six-infusion course, median time to loss of response was twenty-seven and a half days. In the randomised-withdrawal maintenance trial, two hundred and ninety-seven patients who had reached stable remission (176) or stable response without remission (121) on esketamine plus an oral antidepressant were randomised one-to-one either to continue the spray or to switch it for placebo while continuing the oral antidepressant. Relapse was substantially more common in the withdrawal arm in both strata: among stable remitters, 26.7 percent versus 45.3 percent (hazard ratio 0.49, 95% CI 0.29 to 0.84); among stable responders, 25.8 percent versus 57.6 percent (hazard ratio 0.30, 95% CI 0.16 to 0.55). The multi-society chronic-pain guidelines found no evidence for intermediate- or long-term pain improvement outside complex regional pain syndrome, and even the CRPS benefit is bounded at up to twelve weeks. And the approved product's own dosing structure — twice weekly for four weeks, then weekly, then weekly or every two weeks — is a maintenance schedule rather than a course. (Feder A, et al. Am J Psychiatry 2021;178(2):193-202, PMID 33397139, for the twenty-seven-and-a-half-day median; Daly EJ, et al. JAMA Psychiatry 2019;76(9):893-903, PMID 31166571, trial NCT02493868, for the randomised withdrawal; Cohen SP, et al., ASRA/AAPM/ASA Consensus Guidelines, Reg Anesth Pain Med 2018;43(5):521-546, for the pain bound; SPRAVATO prescribing information for the maintenance schedule. Strong Evidence for the shape. The specific figure "one to two weeks back to baseline" traces only to secondary summaries and is not used here. The withdrawal trial's relapse percentages and hazard ratios were previously withheld in this entry as unread; they have now been verified against the primary paper and are stated above. Cui bono note: this is where both biases converge and both are wrong. Clinics and telehealth providers have a direct financial interest in framing a waning effect as "maintenance" — an indefinite subscription rather than a limitation — and in not foregrounding that no trial has established the long-term safety of years of repeat exposure. The randomised-withdrawal design is simultaneously the cleanest way to demonstrate that maintenance works and the design most likely to make any drug look indispensable, because it enriches for responders and then removes the drug; that it also became the load-bearing trial for the original approval is a fact worth holding alongside its strong result. Prohibition-era framing errs the other way by treating any repeat exposure as inevitably harmful, when supervised maintenance has a genuine relapse-prevention evidence base.)

4. In acutely suicidal patients, esketamine improved depression faster — but the suicidality endpoint was not met, and the label says so (Moderate, and the popular framing has this backwards). Two hundred and thirty patients with major depressive disorder and active suicidal ideation with intent were randomised to esketamine eighty-four milligrams or placebo nasal spray, both twice weekly for four weeks, both on top of comprehensive standard of care including hospitalisation for at least five days and a newly initiated or optimised oral antidepressant; two hundred and twenty-seven were dosed and analysed. The primary endpoint was met: change in MADRS at twenty-four hours, least-squares mean difference minus 3.9 (SE 1.39), 95% CI minus 6.60 to minus 1.11, two-sided p equals 0.006 (esketamine minus 15.7 ± 11.56 versus placebo minus 12.4 ± 10.43). The key secondary endpoint was not met: patients in both groups experienced rapid reduction on the revised Clinical Global Impression of Severity of Suicidality, and the between-group difference was not statistically significant. The label's Limitations of Use states that effectiveness in preventing suicide or in reducing suicidal ideation or behaviour has not been demonstrated, and that the product does not obviate the need for hospitalisation where clinically indicated. (Ionescu DF, et al. "Esketamine Nasal Spray for Rapid Reduction of Depressive Symptoms in Patients With Major Depressive Disorder Who Have Active Suicide Ideation With Intent: Results of a Phase 3, Double-Blind, Randomized Study (ASPIRE II)." Int J Neuropsychopharmacol 2021;24(1):22-31. PMID 32861217. Plus SPRAVATO prescribing information, Limitations of Use. The companion trial, ASPIRE I, is named but no numeric effect is published here — we read a pooled and press-level summary of it, not the primary paper. Moderate — the primary endpoint is real and met; the endpoint everyone quotes it for is not. Cui bono note: manufacturer-sponsored trials with a manufacturer-favourable press framing that reads to a non-specialist as "reduces suicidality." The regulator's own label says otherwise. Clinics lean hard on the suicidality claim because it is the most emotionally urgent thing they can say. The opposite error is dismissing the whole indication as hype, when the primary endpoint genuinely was met — and the fact that both arms improved is itself a finding worth carrying: five days of hospitalisation and optimised antidepressant treatment also works, and the trial's own authors point at exactly that as a likely reason the suicidality endpoint did not separate.)

5. For chronic pain the multi-society guidelines are far narrower than "supports ketamine for chronic pain" (Moderate for CRPS specifically, weak or nothing for everything else). The 2018 consensus guidelines, approved jointly by the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine and the American Society of Anesthesiologists committees on pain medicine and on standards and practice parameters, graded the evidence as moderate for complex regional pain syndrome, with pain improvement for up to twelve weeks from infusion regimens in the range of 22 mg/h over four days or 0.35 mg/kg/h over four hours daily for ten days. Evidence was weak for spinal cord injury pain, short-term only. Evidence was weak or absent for mixed neuropathic pain, phantom limb pain, postherpetic neuralgia, fibromyalgia, cancer pain, ischaemic pain, migraine and low back pain. Critically: excluding CRPS, there was no evidence supporting ketamine infusions for intermediate- or long-term pain improvement. (Cohen SP, Bhatia A, Buvanendran A, et al. "Consensus Guidelines on the Use of Intravenous Ketamine Infusions for Chronic Pain Management." Reg Anesth Pain Med 2018;43(5):521-546, published 1 July 2018. Moderate — a multi-society consensus guideline is about as clean a source as this field offers, and it is notably unflattering to the treatment. Verification limit: the guidelines are dated 2018 with a stated review point of 2023, and we did not verify whether a superseding version exists. Cui bono note: infusion clinics cite the existence of consensus guidelines without ever reproducing the grading. If you have fibromyalgia or low back pain, this guideline is the reason this entry sends you to fibromyalgia_lifestyle_and_management and low_back_pain_evidence_and_management rather than toward an infusion.)

6. Repeated ketamine infusions improved chronic PTSD against an active psychoactive control (Emerging). Thirty individuals with chronic PTSD were randomised to six infusions of ketamine (0.5 mg/kg) or midazolam (0.045 mg/kg) as a psychoactive placebo control, over two consecutive weeks. The ketamine group improved significantly more on the CAPS-5 and the MADRS from baseline to week two: mean CAPS-5 total score 11.88 points lower (SE 3.96), effect size 1.13 (95% CI 0.36 to 1.91), response rate sixty-seven percent versus twenty percent, no serious adverse events. Among ketamine responders, median time to loss of response was twenty-seven and a half days after the two-week course. (Feder A, et al. "A Randomized Controlled Trial of Repeated Ketamine Administration for Chronic Posttraumatic Stress Disorder." Am J Psychiatry 2021;178(2):193-202, PMID 33397139. Emerging — thirty people at a single academic site. Replication status was not verified in this pass; a later multi-site trial exists in the literature which we did not read, so this entry asserts neither that the result has replicated nor that it has failed to. Cui bono note: the midazolam comparator is a genuine methodological strength, because it partially blinds for the psychoactive experience, which most psychedelic-adjacent trials fail to do. But a thirty-person trial with a large effect size is exactly the profile that shrinks on replication. Clinics market "ketamine for PTSD" as established; a promising single trial with a twenty-seven-day median durability is the honest headline.)

7. RECOVERY — in severe alcohol use disorder, ketamine increased days abstinent at six months, and one of its two primary outcomes was null (Emerging, and this is the trial that matters most here). A double-blind placebo-controlled phase 2 trial randomised ninety-six patients with severe alcohol use disorder in a two-by-two design: three weekly ketamine infusions (0.8 mg/kg intravenously over forty minutes) plus psychological therapy; three saline infusions plus therapy; three ketamine infusions plus alcohol education; or three saline infusions plus education. The two primary outcomes were self-reported percentage of days abstinent and confirmed alcohol relapse, both at six-month follow-up. Days abstinent were significantly higher in the ketamine group than placebo at six months, mean difference 10.1 percent (95% CI 1.1 to 19.0). The largest cell-level effect was ketamine plus therapy versus saline plus education, 15.9 percent (95% CI 3.8 to 28.1). There was no significant difference in relapse rate between the ketamine and placebo groups — the second primary outcome was null. The abstract does not report a formally significant interaction between the drug and the therapy, so the therapy claim rests on a cell comparison rather than a tested interaction, and the authors themselves describe the confidence intervals as wide and the study as proof of concept. It was well tolerated, with no serious adverse events attributed to study drug in a severe alcohol use disorder population. (Grabski M, et al. "Adjunctive Ketamine With Relapse Prevention-Based Psychological Therapy in the Treatment of Alcohol Use Disorder." Am J Psychiatry 2022;179(2):152-162, PMID 35012326. Emerging — ninety-six people across four arms is roughly twenty-four per cell, one primary outcome was null, the lower confidence bound on the main effect is 1.1 percent, and it is unreplicated. Cui bono note: this must be reported with both hands. It is a well-designed trial with a real six-month endpoint and a genuinely positive abstinence result, and ketamine clinics and "psychedelic-assisted recovery" providers cite it as evidence that ketamine treats addiction, which it does not establish. The opposite error is dismissing it because ketamine has misuse liability — the trial found no serious drug-related adverse events in a severe-AUD population, which is itself informative. The finding most worth carrying into any recovery-framed product is that the therapy arm carried the largest effect, not the drug alone. The broader argument about drug-assisted interruption of addictive patterns belongs to psychedelic_assisted_addiction_interruption.)

8. In cocaine use disorder, a single infusion alongside behavioural therapy increased abstinence — from one trial of fifty-five people (Emerging). Fifty-five cocaine-dependent adults were randomised to a single forty-minute intravenous infusion of ketamine (0.5 mg/kg) or midazolam during a five-day inpatient stay, alongside a five-week course of mindfulness-based relapse prevention. Primary outcomes were end-of-study abstinence and time to relapse. Forty-eight point two percent of the ketamine group maintained abstinence over the last two weeks of the trial, versus 10.7 percent on midazolam. Single site, fifty-five participants, one infusion, no replication identified. (Dakwar E, et al. "A Single Ketamine Infusion Combined With Mindfulness-Based Behavioral Modification to Treat Cocaine Dependence." Am J Psychiatry 2019;176(11):923-930. Emerging. A companion single-infusion trial in alcohol use disorder with motivational enhancement therapy exists from the same group and we did not read it, so no numbers from it appear here. No claim about ketamine and smoking cessation appears in this entry: we located no source for one. Cui bono note: forty-eight percent versus eleven percent is the single most quotable number in the ketamine-for-addiction literature, and it comes from one fifty-five-person trial at one site with one infusion. It is repeated constantly by clinics without the sample. Report the number with the sample attached in the same sentence or do not report it.)

9. Bipolar depression and obsessive-compulsive disorder are real signals in fifteen-person crossover trials, and nothing more (Experimental). In bipolar depression, a double-blind randomised crossover placebo-controlled study in fifteen subjects with bipolar I or II depression maintained on therapeutic lithium or valproate gave a single intravenous infusion (0.5 mg/kg) or placebo on two test days two weeks apart; depressive symptoms and suicidal ideation improved significantly within forty minutes, with seventy-nine percent responding to ketamine and zero percent to placebo at some point during the trial. It was framed as a replication of an earlier crossover trial from the same group. In OCD, a randomised double-blind placebo-controlled crossover trial in fifteen drug-free adults with near-constant obsessions gave two forty-minute infusions (saline and ketamine 0.5 mg/kg) at least a week apart; one week after infusion, four of eight ketamine patients met treatment-response criteria (at least thirty-five percent Y-BOCS reduction) versus zero of seven on placebo. (Bipolar: "Replication of ketamine's antidepressant efficacy in bipolar depression: a randomized controlled add-on trial." Biol Psychiatry 2012, PMID 22297150. OCD: Rodriguez CI, et al. "Randomized Controlled Crossover Trial of Ketamine in Obsessive-Compulsive Disorder: Proof-of-Concept." Neuropsychopharmacology 2013. Experimental — fifteen participants each; neither supports a clinical recommendation. Cui bono note: investigator-initiated academic work, not industry-driven — but fifteen-person crossover trials generate the "ketamine treats X" headlines that clinics convert into service lines. In a crossover design with a strongly psychoactive drug, blinding is close to impossible, which inflates apparent effects. This entry lists no further indications; anything beyond these is thinner still.)

The aftercare and integration evidence, which is where this entry's name comes from and where the evidence runs out.

10. The supervised-monitoring requirement is real, specific, and regulator-mandated (Foundational as a regulatory fact). Because of the risks of sedation and dissociation, the approved esketamine product's risk-management programme requires that it be administered under the direct observation of a healthcare provider in a certified healthcare setting, that patients be monitored for at least two hours at each treatment session followed by an assessment of clinical stability before leaving, and that certified settings and pharmacies do not dispense it directly to patients for home use. Two hours is a floor, not a fixed release time. Separately, the label states the product may impair attention, judgment, thinking, reaction speed and motor skills, and instructs patients not to drive or operate machinery until the next day following a restful sleep. (SPRAVATO (esketamine) nasal spray, CIII — FDA prescribing information and REMS, NDA 211243. Flag: the programme has been modified over time, particularly its patient-registry component; the in-clinic, no-take-home and at-least-two-hour requirements are stable across revisions, but any statement about registry enrolment should be checked against the current label. Cui bono note: this is a regulator requiring a manufacturer to restrict its own product's convenience, which is the funding shape least likely to be inflated. Note the contrast that follows from it: everything else in the aftercare protocol below has nothing like this behind it.)

11. That paid integration sessions are necessary is a commercial position, not a finding (not tiered as evidence). No trial identified in the construction of this entry isolates structured integration coaching as a component and demonstrates that it adds benefit over doing the reflective work yourself, with an existing therapist, or with the treating clinician. The nearest relevant evidence points to structured psychological work mattering — in the alcohol trial the arm combining ketamine with manualised relapse-prevention therapy produced the largest single cell effect — but that therapy was a research protocol delivered inside a randomised trial, the abstract reports no formally significant drug-by-therapy interaction, and the whole study was ninety-six people. That supports taking structured psychological support seriously. It does not establish that a purchased integration package is required, and the claim that self-directed integration is inadequate is, as far as this entry can determine, asserted overwhelmingly by the sector that sells the alternative. (No primary source supports the necessity claim; it is stated here as an absence, which is what it is. Cui bono note: integration coaching is an unregulated service category with no licensure requirement, priced per session, sold to people in acute distress immediately after a dissociative experience — a population with unusually low capacity to evaluate the offer. The opposite error also exists and is worth naming: dismissing psychological support entirely, when the one recovery trial in this entry found the therapy arm carrying more of the effect than the drug arm.)

12. The "neuroplasticity window" is clinical convention built on rodent timelines, and the one human test of the mechanism contradicted it (Emerging for the mechanism, not tiered as a clinical parameter). In rats, ketamine rapidly activated the mTOR pathway, increasing synaptic signalling proteins, and blockade of mTOR signalling completely blocked ketamine-induced synaptogenesis and the behavioural response in rodent despair models — the finding that underwrites the whole plasticity-window story. Even within rodent work the result is route-dependent: rapamycin infused directly into medial prefrontal cortex blocks the effect, whereas systemic single-dose pretreatment has been reported not to. And in humans it went the other way. A randomised controlled trial in twenty depressed patients found that rapamycin did not block ketamine's antidepressant effect — it prolonged it: two weeks after administration, response was forty-one percent and remission twenty-nine percent with rapamycin plus ketamine, versus thirteen percent (p equals 0.04) and seven percent (p equals 0.003) with placebo plus ketamine. No human trial has randomised the timing or content of post-dose activity and measured the outcome. (Li N, Lee B, Liu RJ, et al. "mTOR-dependent synapse formation underlies the rapid antidepressant effects of NMDA antagonists." Science 2010;329(5994):959-964, PMID 20724638 — rats. Abdallah CG, Averill LA, Gueorguieva R, et al. "Modulation of the antidepressant effects of ketamine by the mTORC1 inhibitor rapamycin." Neuropsychopharmacology 2020;45(6):990-997, PMID 32092760 — humans, n=20. Emerging. Academic and publicly funded on both sides; no manufacturer interest driving the mechanism claim. Cui bono note: flagging this evidence as "rodent" is necessary and not sufficient. An entry that presents mTOR-dependent synaptogenesis as the established mechanism, caveated only as animal work, still implies a translation the human data actively contradict. The window is a useful reason to clear your calendar. It is not a measured biological parameter, and it should not be the basis of a priced service.)

The risk evidence.

13. In heavy chronic recreational use, ketamine causes a severe, characteristic uropathy, and the mechanism is direct urothelial toxicity from luminal exposure rather than a systemic effect (Strong Evidence). A systematic review and meta-analysis of forty-five articles covering 4,921 patients found urinary frequency in 77.1 percent, urgency in 69.9 percent, suprapubic pain in 60.4 percent, and hydronephrosis — upper urinary tract involvement — in 30.2 percent. Mean functional bladder capacity was 95.23 millilitres and mean maximum flow rate 8.69 millilitres per second, both markedly impaired. Cystoscopy and biopsy showed urothelial denudation, inflammatory changes and inflammatory cell infiltration. The founding description of the entity was nine daily ketamine users presenting with severe dysuria, frequency, urgency and gross haematuria, in whom CT showed small bladder capacity with marked wall thickening and perivesical inflammation and all nine had severe ulcerative cystitis at cystoscopy. Damage is driven by direct bladder exposure rather than circulatory effects, proceeding through mitochondrial stress, oxidative injury, urothelial barrier disruption, inflammation and progressive fibrosis. Severe end-stage management escalates through oral medication, intravesical therapy, hydrodistension, augmentation cystoplasty and cystectomy. (Chan EOT, Chan VWS, Tang TST, et al. "Systematic review and meta-analysis of ketamine-associated uropathy." Hong Kong Med J 2022;28(6):466-474, PMID 36464318. Shahani R, Streutker C, Dickson B, Stewart RJ. "Ketamine-associated ulcerative cystitis: a new clinical entity." Urology 2007;69(5):810-812, PMID 17482909 — nine cases, daily users, no prevalence information and nothing about therapeutic dosing. Zhou J, Scott C, Miab ZR, Lehmann C. "Current approaches for the treatment of ketamine-induced cystitis." Neurourol Urodyn 2023;42(4):680-689 — a narrative review of management, supporting statements about how the condition is treated and not about incidence, therapeutic risk or causation. Mechanism corroborated by Andrade C, J Clin Psychiatry 2025;86(4):25f16083. Strong Evidence for the recreational picture. A separately circulating pair of ranges — 44 to 77 percent lower-tract and 8 to 30 percent upper-tract, attributed to a thirty-seven-study meta-analysis — is deliberately not welded to the forty-five-study review above; it is a different source we did not trace. Cui bono note: academic urology groups working in the population where the harm was first characterised, with no commercial sponsor apparent.)

14. At therapeutic doses the picture is different, and the independent review concludes the opposite of what the popular claim asserts (Moderate). A systematic review of twenty-seven studies of ketamine for psychiatric disorders — fifteen randomised trials, seven retrospective cohorts, five open-label trials, mostly depressive disorders — found urological symptoms reported in zero to 24.5 percent of patients receiving treatment, that symptoms tended to be mild or moderate in severity, that rates in randomised trials were generally at or below control conditions, and that continuous outcome measures did not show significant changes from baseline to follow-up. The review's stated conclusion is that, based on the limited data available, ketamine treatment does not appear to be associated with elevated risk of urological symptoms, and that further long-term studies are required. The exposure contrast is the reason: recreational uropathy cases involved doses exceeding a gram a day for a median of about thirty-five months, against therapeutic protocols of intravenous racemic 0.5 mg/kg two to three times weekly, intranasal esketamine twenty-eight to a hundred milligrams twice weekly, or oral racemic. A review notes cumulative therapeutic doses are roughly ninety to two hundred times lower than in heavy recreational use. The approved product's own label states no cases of esketamine-related interstitial cystitis were observed in any of its studies, which included treatment for up to a year. (Kerr-Gaffney J, Tröger A, Caulfield A, Ritter P, Rucker J, Young AH. "Urological symptoms following ketamine treatment for psychiatric disorders: A systematic review." J Psychopharmacol 2025;39(10):1103-1113, PMID 40583492. Supported by Andrade C, J Clin Psychiatry 2025;86(4):25f16083; Leo RJ, et al., Prim Care Companion CNS Disord 2026;28(2):25br04121; SPRAVATO prescribing information, Warnings and Precautions. Moderate. Correction of record: the 24.5 percent figure is the upper bound of a range whose lower bound is zero, drawn from a single arm of a single trial. It is not a pooled estimate, it is not a central tendency, and this review must not be cited as evidence that therapeutic-dose cystitis risk is elevated — it concludes the reverse. Cui bono note: independent academic review, King's College London and Dresden, with no manufacturer sponsorship identified, which makes its no-elevated-risk conclusion harder to dismiss rather than easier. One honest flag for a reader: the senior authors work in a group with substantial psychopharmacology and psychedelic-trial relationships, so this is not a disinterested corner of the field — though the review's wording is conservative and unusually candid about its own limits.)

15. The therapeutic reassurance is real and thin, and the review says so about itself (Moderate — read this immediately after the claim above). The same review states six limitations that matter more than its headline. Median duration of follow-up for urological symptoms was four weeks, against a recreational injury that takes a median of about thirty-five months to appear. Only fifteen percent of studies were rated low risk of bias. Eleven studies specified no systematic assessment of urological symptoms at all, reporting only spontaneously arising adverse-event data. Only two studies conducted laboratory urinary analyses, so subclinical urothelial damage would be invisible. Only five assessed treatment beyond six months, giving insufficient evidence to conclude whether long-term ketamine treatment is associated with serious urological symptoms. And some studies enrolled only prior responders, excluding people who had already dropped out over side effects. The clinical risk-reduction advice that accompanies this literature — hydration on treatment days, frequent voiding, fortnightly or monthly urine screening for protein, cells and blood, and routine symptom enquiry during maintenance — is itself the tell: clinicians would not be screening urine if the question were closed. (Kerr-Gaffney J, et al., J Psychopharmacol 2025;39(10):1103-1113, limitations section. Monitoring recommendations: Andrade C, J Clin Psychiatry 2025;86(4):25f16083. Moderate. Verification scope: the review's abstract was verified directly in this pass and carries the 0–24.5 percent range, the fifteen-percent low-risk-of-bias figure, the null continuous-measure finding and the no-elevated-risk conclusion. The four-week median follow-up, the eleven-studies-without-systematic-assessment count and the only-two-with-laboratory-urinalysis count sit in the paper's full-text limitations section, which was not re-read in this pass. They are retained because they cut against this entry's own reassurance — the conservative direction — but they are flagged rather than presented as re-verified. Cui bono note: this is the counterweight that clinic marketing omits when it quotes the headline conclusion, and it is also the counterweight that prohibition-era framing omits when it treats the headline conclusion as manufactured.)

16. Uropathy substantially improves on cessation, but not universally and not always completely (Moderate). Approximately seventy-five percent of recreational users showed substantial symptom reduction within three months of stopping, with the remainder including cases requiring surgery. The pooled review concludes flatly that abstinence is important for all patients with ketamine-associated uropathy — improvement required abstinence regardless of which treatment was used. Clinical descriptions note symptomatic improvement during abstinence but incomplete recovery in some, symptoms that fluctuate with use, and rapid worsening on relapse. Write it as: most people improve markedly within about three months of stopping, and a minority are left with lasting damage. Do not write "reversible on cessation" unqualified; that is the version that gets a reader to discount the risk. (Andrade C, J Clin Psychiatry 2025;86(4):25f16083; Chan EOT, et al., Hong Kong Med J 2022;28(6):466-474; Leo RJ, et al., Prim Care Companion CNS Disord 2026;28(2):25br04121. Moderate.)

17. Ketamine cholangiopathy is a real described entity, confined to chronic heavy use (Moderate). Presentation is abdominal pain in seventy-one percent of reported cases, with jaundice, elevated liver enzymes and cholestasis; imaging shows fusiform dilation of the common bile duct and irregular narrowing of intrahepatic ducts with strictures, radiologically resembling secondary sclerosing cholangitis. A cross-sectional survey of 297 chronic users found liver injury in 9.8 percent. Median duration of use in reported cases was about twenty-four months, and affected patients are typically young, mean age around twenty-five to twenty-six, majority male. Roughly eighty-eight percent improved after stopping while about twelve percent had persistent cholangiopathy, with biochemical recovery documented after around sixty days of abstinence in some cases, and degree of bile duct dilatation correlating with duration of use. (Waheed HW, Ashraf MR, Sajjad T, et al. "Hepatobiliary Complications Associated With Ketamine Use: Clinical Insights and Future Directions." Cureus 2025;17(5):e84974. Moderate, with a source-vehicle flag: Cureus is a low-barrier peer-reviewed journal, and the specific figures here — the 9.8 percent, the 88/12 split — should be traced to the underlying cross-sectional study and case-report synthesis before being leaned on hard. Every source checked describes this in recreational or chronic-abuse contexts, and no hepatobiliary injury has been reported specifically in therapeutic psychiatric dosing. Do not imply a clinical-dosing counterpart exists.)

18. Abuse liability, tolerance and physical dependence are real and labelled — and the reassuring trial data are narrower than they are usually reported (Strong Evidence for the liability; Emerging for the "no illicit transition" claim). The approved product's prescribing information is explicit: it is a Schedule III controlled substance and may be subject to abuse and diversion; individuals with a history of drug abuse or dependence are at greater risk and require careful consideration before treatment; physical dependence has been reported with prolonged use of ketamine; and tolerance has been reported with prolonged use of ketamine. The boxed warning names abuse and misuse, and the product is available only through a restricted programme. Against that, the trial literature is genuinely reassuring but narrower than the popular version. A systematic review of sixteen studies covering 2,174 depression patients found that four patients exhibited clear signs of tolerance to the antidepressant effect or dependence on the drug, while the majority did not — a low number, but not none, and tolerance and dependence are not the same category as transition to illicit use. A 2026 review states that existing substance-use-disorder trials have not reported illicit transition, diversion, or ketamine use disorder when delivered under rigorous clinical protocols — a statement about SUD trials specifically, not about the whole clinical literature. One dedicated prospective measurement followed twenty-three patients through eight intranasal esketamine sessions using a liking-and-craving questionnaire and found no cravings in 78.3 percent initially, 86.96 percent reporting no desire to use more than prescribed, and no increase across the course; its authors flag the small sample, the unvalidated instrument, and the exclusion of people with active substance use disorders. (SPRAVATO prescribing information, Boxed Warning and Warnings and Precautions. Ingrosso G, Cleare AJ, Juruena MF. "Is there a risk of addiction to ketamine during the treatment of depression? A systematic review." J Psychopharmacol 2025;39(1):49-65, PMID 39688236. "Treating addiction with an addictive drug: the ketamine paradox revisited." Front Psychiatry 2026, doi:10.3389/fpsyt.2026.1866092. Gutierrez G, Vazquez G, Ravindran N, et al. Ther Adv Drug Saf 2025;16:20420986251347360. What this entry may say: in supervised clinical trials, transition to illicit ketamine use has not been reported, dose escalation reflects clinical titration rather than drug-seeking, and craving measures are mostly neutral — but these trials largely excluded people with active addiction and followed patients only briefly, so this is reassuring rather than conclusive. What it may not say: that there are no cases full stop, or that clinical use carries no dependence risk. Cui bono note: much of the reassurance-oriented abuse-liability work is produced by groups with clinical or commercial stakes in ketamine services; the systematic review above is academic. A prescribing label, by contrast, is written by the sponsor under regulatory review — which for risk statements cuts in the trustworthy direction, since a sponsor has no incentive to invent warnings about its own product.)

19. Higher benzodiazepine doses may blunt ketamine's antidepressant effect — a real signal, from small retrospective data, with inconsistent timing across studies (Emerging). A retrospective analysis of forty-seven inpatients with treatment-resistant depression, pooled from two randomised trials and all given a single 0.54 mg/kg intravenous infusion, identified a cutoff of more than eight milligrams diazepam-equivalent daily dose separating responders from non-responders (sensitivity eighty percent, specificity eighty-five percent, p less than 0.001). Response — at least a fifty percent MADRS reduction within one week — was eight percent in higher-dose benzodiazepine users versus thirty-five percent in lower-dose users. The impairment appeared at days three and seven (p equals 0.04 and 0.02) but not at twenty-four hours; Hedges' g was 0.67 at day three and 0.78 at day seven; adjusted odds ratio 9.7 (95% CI 1.2 to 69.7, p equals 0.03). The proposed mechanism is that benzodiazepines increase inhibition of pyramidal neurons and thereby counteract the disinhibition ketamine produces by acting on GABAergic interneurons. The authors' own limitations: small sample, retrospective, no blood benzodiazepine levels, baseline anxiety differences between groups, post-hoc. (Andrashko V, Novak T, Brunovsky M, Klirova M, Sos P, Horacek J. "The Antidepressant Effect of Ketamine Is Dampened by Concomitant Benzodiazepine Medication." Front Psychiatry 2020;11:844, PMID 33005153. Emerging. Two things this entry does not assert: a separate secondary analysis of an intravenous ketamine study reportedly found the attenuation at day one but not day three — the opposite time pattern — and esketamine data reportedly show no meaningful blunting at twenty-four hours; we did not retrieve either paper and will not cite a time course we have not read. Academic, Czech National Institute of Mental Health; no commercial interest in a finding that reduces ketamine's apparent efficacy. This belongs in a conversation with the prescribing clinician and must never prompt anyone to stop or reduce a benzodiazepine on their own — abrupt benzodiazepine reduction carries seizure risk, and that harm is far better established than this interaction.)

The regulatory and market picture.

20. Regulators have twice warned specifically about compounded and at-home ketamine (regulatory fact, not tiered). On 16 February 2022 the FDA alerted health care professionals about potential risks associated with compounded ketamine nasal spray. On 10 October 2023 it warned patients and health care providers about potential risks associated with compounded ketamine, covering compounded products including oral formulations for the treatment of psychiatric disorders. Its substance: compounded ketamine places patients at risk for serious adverse events, misuse and abuse, especially where there is home use; the agency has received adverse event reports from patients taking compounded ketamine outside health care settings; at-home administration is risky specifically because no provider is on site to monitor for serious outcomes from sedation and dissociation; safe and effective dosing of ketamine for any psychiatric disorder has not been established, because ketamine is not approved for these indications; and the agency stated it is not aware of evidence suggesting compounded ketamine is safer, more effective or faster-acting than approved medications. The alert explicitly names telemedicine platforms and compounders as the supply route. (FDA compounding alerts, 16 February 2022 and 10 October 2023. Verification caveat, stated openly: fda.gov returned errors to our retrieval tooling on every attempt, so these titles, dates and wording are confirmed via a regulatory-law analysis linking the primary URLs plus two independent search-index summaries in agreement. A human should open the two FDA pages and check the wording before this entry is published externally. On market size: we could not find a primary source with hard figures for the growth of at-home, telehealth or compounded oral ketamine — the best available academic source states that the 2020 pandemic accelerated adoption of at-home models and triggered federal scrutiny, but provides no prevalence numbers. No figure for clinic count, market size or patient count appears in this entry, because none could be traced. The FDA alert is itself the evidence that the sector grew enough to warrant one.)

21. Scheduling across three jurisdictions, including the injection provision that is usually reported as a myth and is not (regulatory fact, not tiered). In the United States, ketamine is a Schedule III controlled substance under the federal Controlled Substances Act, scheduled in 1999. In the United Kingdom, it is Class B under the Misuse of Drugs Act 1971, reclassified upward from Class C with effect from 10 June 2014. A ketamine preparation designed for administration by injection is treated as Class A — this is a genuine statutory rule, at Schedule 2 Part I paragraph 6 of the Act, which captures any preparation designed for injection that includes a substance specified in the relevant Part II paragraphs, and it is a general provision applying to all such substances rather than a ketamine-specific rule. In Australia, ketamine is Schedule 8, a Controlled Drug, requiring state or territory authority to prescribe. Separately, on 8 January 2025 the UK Home Office asked its statutory advisory council whether ketamine should be moved to Class A, citing an estimated 299,000 people aged sixteen to fifty-nine reporting last-year use in the year ending March 2023, the largest number on record; on 28 January 2026 the council advised that ketamine should remain Class B, declining the upgrade, on the grounds that acute toxicity and death rates remain consistent with Class B, that evidence reclassification reduces use or harm is limited, that many acute harms are significantly influenced by concurrent use of other drugs, that reclassification could increase stigma and discourage help-seeking, and that higher penalties could disproportionately affect young and vulnerable people. It recommended a public-health and whole-systems approach instead. That recommendation was not unanimous within the council, and this entry states that rather than borrowing more authority from the advice than it carries. (US: 21 CFR 1308.13. UK: Misuse of Drugs Act 1971, Schedule 2 Part II para 1(a), as inserted by SI 2014/1106 in force 10 June 2014; injection provision at Schedule 2 Part I para 6. Australia: Poisons Standard, Schedule 8; TGA final decision 2023. ACMD, "Ketamine: an updated review of use and harms," 28 January 2026. Primary law and statutory advice, no funding interest. Note for tone: the council's own reasoning is anti-moralising, and this entry reflects that rather than editorialising on top of it.)

22. Whether the branded spray and the generic molecule are equivalent is genuinely unsettled, and is being tested by an independently funded trial that reports in 2030 (not tiered — no results exist). A registered comparative-effectiveness trial, sponsored by Yale University and funded by the Patient-Centered Outcomes Research Institute through a reported twelve and a half million dollar grant, is randomising four hundred participants one-to-one to intravenous racemic ketamine or intranasal esketamine, eight treatments over four weeks, in collaboration with Emory University, the University of Michigan and three community sites. It began recruiting on 27 January 2025 and its estimated primary and study completion date is 31 December 2030. Separately, the published head-to-head meta-analytic comparison does not find the two equivalent, reporting greater response and remission for racemic ketamine than for esketamine — and that comparison is itself contested: it carries a published erratum and drew a formal Letter to the Editor with an author response, which is exactly the treatment this entry gives to every other favourable finding and must not be dropped merely because this one points away from the branded product. (ClinicalTrials.gov NCT06713616, protocol published as the EQUIVALENCE trial, PMID 41780747. Bahji A, Vazquez GH, Zarate CA Jr. "Comparative efficacy of racemic ketamine and esketamine for depression: A systematic review and meta-analysis." J Affect Disord 2021;278:542-555 — racemic response RR 3.01 and remission RR 3.70 versus esketamine 1.38 and 1.47, with lower dropout for racemic; see also the published erratum, PMID 33229028, and the Letter to the Editor and author response, PMID 33957360 and PMID 36574849. The trial is recruiting with no results; it may be cited as evidence that the question is genuinely open and being tested, never for any result, and the entry must not imply near-term resolution — completion is roughly five years out. Cui bono note: the trial is the strongest funding shape in the entire entry. An independently funded comparative-effectiveness test of an approved branded product against its cheaper generic-molecule competitor, with no sponsor incentive in either direction, is the study design that is hardest to buy and rarest to see. A widely circulated claim that real-world response rates for both esketamine and compounded racemic ketamine fall in a similar sixty to seventy-five percent range is deliberately absent from this entry: no primary source carries that range, peer-reviewed pooled real-world figures run materially lower at roughly fifty to fifty-five percent response, and the range traces to clinic and telehealth marketing content.)

Mechanism

This section owns enough mechanism to make the use-case map, the durability problem, the dissociation caution and the bladder risk intelligible — not the full pharmacology.

The antidepressant chain, and where it stops being human. Ketamine is a non-competitive NMDA receptor antagonist. The mainstream account runs: blockade preferentially affects NMDA receptors on GABAergic inhibitory interneurons, disinhibiting pyramidal neurons and producing a transient surge of glutamate; that surge activates AMPA receptors; AMPA activation drives release of brain-derived neurotrophic factor, which acts on the TrkB receptor; TrkB signalling activates the mTOR pathway; and mTOR activation produces new synaptic proteins and new synaptic connections in prefrontal cortex. The evidence for the last link is rodent: in rats, ketamine rapidly activated mTOR and increased synaptic signalling proteins, and blocking mTOR completely blocked both the synaptogenesis and the antidepressant-like behaviour in despair models. That is where honest reading has to slow down, for three reasons. Antidepressant-like behaviour in a rodent despair model is not depression. Even within rodent work the blockade result depends on how the blocker is delivered. And when the mechanism was finally tested in people, giving depressed patients an mTOR inhibitor did not abolish ketamine's effect — it made the effect last longer. A mechanism whose one human test produced the opposite of the predicted result is a mechanism you can describe and cannot lean on. In particular, it cannot be converted into a prescription for what you should do in the hours after a dose.

Why the effect wanes, on the account's own terms. If the antidepressant action is a burst of synaptic remodelling triggered by a transient pharmacological event, then a decaying effect is exactly what the model predicts: the trigger is over in hours, and whatever it built is subject to the same ordinary attrition as any other synaptic change. Nothing in the mechanism suggests a single exposure installs a permanent state. This is the mechanistic reason the treatment is structurally a maintenance treatment, and it is why the interesting question is not "how do I extend the window" but "what durable changes can be made while the window is open" — a question the mechanism does not answer and no human trial has tested.

Dissociation is not a side effect that happens to accompany the mechanism; it is close to the mechanism. The same NMDA blockade that produces the glutamate surge produces the altered sense of self, body and reality that patients describe, and the regulator treats sedation and dissociation as the two core monitored risks — the reason for at least two hours of observation. This matters practically in one specific direction: a person with a history of depersonalisation or derealisation is being offered a drug whose therapeutic action is pharmacologically adjacent to the experience they are trying to get away from. That is a distinct conversation with a clinician, not a general caution. See depersonalisation_derealisation_management.

The acute cardiovascular effect has a known shape. Blood pressure rises at all recommended doses, peaks approximately forty minutes after administration, and lasts approximately four hours. That specific time course is why the monitoring window is at least two hours rather than fifteen minutes, and it is why the absolute contraindications concern vessels that a transient pressure surge could rupture.

The bladder mechanism is local, not systemic, and that is the whole reason therapeutic and recreational exposure differ so much. Ketamine uropathy is driven by direct exposure of the bladder lining to ketamine and its metabolites in urine, not by a circulating effect — proceeding through mitochondrial stress, oxidative injury, urothelial barrier disruption, inflammation, and progressive fibrosis with loss of bladder capacity, and in severe cases upper tract involvement. Because the injury depends on cumulative luminal exposure, a use pattern of grams per day for years and a use pattern of a supervised dose once or twice a week are not the same exposure in any meaningful sense — a review puts the cumulative difference at roughly ninety to two hundred fold. That is the mechanistic reason the therapeutic literature does not reproduce the recreational picture, and it is also the reason hydration and frequent voiding are the risk-reduction advice clinicians give: both reduce the concentration and dwell time of what the bladder lining is exposed to. Note that this is a mechanistically sensible practice, not a tested one.

Risks And Contraindications

• BLADDER AND URINARY RED FLAGS — these warrant prompt medical assessment, not self-monitoring, and not a decision about your own dosing. Needing to urinate much more often. Sudden urgency. Waking repeatedly at night to urinate. Pain or burning on urination. Pain above the pubic bone. Blood in the urine, even if it comes and goes. The intermittency is the trap — blood that appears once and then stops is easy to dismiss, and the clinical descriptions specifically note symptoms that fluctuate with use and worsen rapidly on resumption. These are the recognised early features of ketamine uropathy, and the prevalence ordering in the heavy-use literature matches: frequency, then urgency, then suprapubic pain. Because this is a controlled substance, the action here routes to a clinician; it is explicitly not a threshold to titrate your own use against.
• HEPATOBILIARY RED FLAGS, in the same category. Persistent abdominal pain — the presenting feature in seventy-one percent of reported cholangiopathy cases — jaundice, or abnormal liver blood tests. Reported in chronic heavy use, not in therapeutic psychiatric dosing, and a reason to be seen rather than to wait.
• Absolute contraindications, from the approved product's label — meaning it must not be given at all. Aneurysmal vascular disease, including thoracic and abdominal aorta, intracranial and peripheral arterial vessels. Arteriovenous malformation. History of intracerebral haemorrhage. Hypersensitivity to esketamine, ketamine or excipients. The logic is direct: blood pressure rises at all recommended doses, peaking around forty minutes and lasting around four hours, and a transient pressure surge in a weakened or malformed vessel risks rupture or bleed. Scope flag: these are the esketamine label's contraindications. Racemic ketamine used off-label in psychiatry has no equivalent psychiatric label; clinics generally apply the same screening by analogy, which is sensible practice rather than a single official standard covering both.
• Careful individual assessment, not an absolute bar: other cardiovascular and cerebrovascular conditions, including uncontrolled hypertension. The label's wording is that patients with other cardiovascular and cerebrovascular conditions should be carefully assessed before prescribing.
• Psychosis-spectrum history is a screening item requiring an explicit benefit-risk judgement — and it is commonly overstated as a flat contraindication. The label states that given the potential to induce dissociative effects, patients with psychosis should be carefully assessed and treatment initiated only if the benefit outweighs the risk. It is in Warnings and Precautions, not in Contraindications. Many clinical protocols and trial exclusion criteria do exclude active or historical psychotic disorders outright, which is a defensible practice statement and should be attributed as practice rather than as label.
• A history of depersonalisation or derealisation is a distinct and specific caution. Dissociation is not an incidental adverse effect here — it is close to the therapeutic mechanism and it is one of the two risks the mandatory monitoring period exists for. Being offered a drug whose action is pharmacologically adjacent to the state you are trying to escape is a real conversation to have before the first dose. depersonalisation_derealisation_management owns the management side.
• Acute effects to expect and be monitored for. Blood pressure increases at all recommended doses, peaking around forty minutes and lasting around four hours. Sedation — in clinical trials, forty-eight to sixty-one percent of treated patients developed sedation, and 0.3 to 0.4 percent experienced loss of consciousness. Dissociation. Nausea. Impaired attention, judgment, thinking, reaction speed and motor skills, with the instruction not to drive or operate machinery until the next day following a restful sleep — which also makes falls a real consideration in the hours afterwards. The boxed warning names respiratory depression as a monitored risk, which is why the observation period is what it is.
• Additive central nervous system depression with alcohol, benzodiazepines and opioids — compounded sedation and compounded suppression of breathing. This is class-level pharmacodynamics rather than a ketamine-specific discovery, and it is consistent with the boxed warning's naming of sedation and respiratory depression. Stated at mechanism level, because we did not read the label's drug-interactions section directly. Real-world corroboration from the UK statutory review: many acute harms experienced by ketamine users are likely to be significantly influenced by using other drugs at the same time.
• Abuse liability, tolerance and physical dependence are real and are on the label. It is a controlled substance and may be subject to abuse and diversion; individuals with a history of drug abuse or dependence are at greater risk; physical dependence and tolerance have both been reported with prolonged use. Do not let a clinical framing imply the molecule loses its liability when a clinician hands it to you. The supervised-trial data are reassuring — very few dependence signals across more than two thousand depression patients, and no reported transition to illicit use in supervised substance-use trials — but those trials largely excluded people with active addiction and followed people only briefly.
• The benzodiazepine interaction, handled safely. There is a plausible mechanism and a real signal from small retrospective data that higher benzodiazepine doses are associated with a weaker or shorter-lived antidepressant response to ketamine, with inconsistent timing across studies. This sits alongside, and does not replace, the additive-sedation problem above — the two are separate concerns and both are real. Raise it with the prescribing clinician. Do not stop or reduce a benzodiazepine on your own under any circumstances — abrupt reduction carries seizure risk, and that harm is far better established than this interaction.
• At-home and compounded ketamine sit outside the safety architecture everything above assumes. The regulator's own stated reason is that no health care provider is on site to monitor for the serious outcomes of sedation and dissociation, that safe and effective dosing for any psychiatric disorder has not been established, and that it has received adverse event reports from patients taking compounded ketamine outside health care settings. That is not a moral position about the sector; it is what the monitoring requirement is for.

Controversy

Nature: this entry carries two live controversies with the same underlying shape — a treatment whose necessary components have never been isolated, surrounded by parties who profit from asserting that their component is the necessary one. The sharper of the two is pharmaceutical: whether the patented single-enantiomer nasal spray is meaningfully better than the cheap generic racemic molecule it was derived from, or whether the approval architecture reflects what could be patented rather than what works best. The second is the service layer: whether professional integration is a necessary part of treatment or a product sold into a moment of maximal suggestibility. Error is possible at both poles of both, and the same reader can be harmed by overclaim (paying indefinitely for a subscription framed as a cure, or for a coaching package framed as a requirement) and by dismissal (writing off the one treatment that works for depression that has resisted everything else, or writing off psychological support because some of it is sold badly).

Position A — "The approved nasal spray is the properly evidenced product; everything else is unregulated improvisation."
• Best evidence: esketamine is the only ketamine product approved for treatment-resistant depression anywhere in the mainstream regulatory system; it has a randomised active-controlled short-term trial with a statistically significant primary result, a randomised-withdrawal maintenance trial supporting relapse prevention with substantial hazard ratios in both strata, a dedicated trial programme in acutely suicidal patients that met its depression endpoint, and a formal risk-management architecture requiring supervised administration, at least two hours of monitoring, and no home dispensing. Its label states no cases of product-related interstitial cystitis were observed in studies including treatment for up to a year. Compounded racemic ketamine has none of that, and the regulator has twice warned about it by name.
• Where it overreaches: the positive short-term trial's effect was a four-point difference on a depression scale, sponsored by the manufacturer — and it was the only one of the three short-term phase 3 trials to separate from control on its primary endpoint. A new antidepressant normally needs two positive adequately powered short-term trials; this approval leaned instead on the randomised-withdrawal trial, a design that is structurally flattering to any drug tested in it, and the approval was argued about at the advisory committee for exactly that reason. The suicidality trials did not meet their suicidality endpoint, and the label says so. And the published head-to-head meta-analytic comparison of the two agents does not favour the branded one — it favours racemic ketamine. The claim that the approved product is the better medicine rather than the patentable one has not been demonstrated.

Position B — "Esketamine is a patent play; the generic molecule is as good or better and vastly cheaper."
• Best evidence: racemic ketamine is generic, cheap, decades old, on the WHO essential medicines list, and unpatentable, which is precisely why no manufacturer ran the registration programme for it in depression. The one published meta-analytic head-to-head reports greater response and remission for racemic ketamine than for esketamine, with lower dropout. And the field's own institutions evidently regard the question as open: an independent public research funder is spending a reported twelve and a half million dollars on a four-hundred-participant randomised comparison of exactly these two agents, which would be a strange thing to fund if the branded product were established as superior.
• Where it overreaches: the meta-analytic comparison is indirect and heterogeneous, not a head-to-head randomised trial — and it carries a published erratum and drew a formal Letter to the Editor challenging it, which is the whole reason the comparison is now being run properly. "Racemic is as good" does not license "so buy it from a telehealth company and take it at home" — the trial evidence for racemic ketamine is supervised intravenous administration in clinical settings, and none of it underwrites unsupervised oral dosing, which is what the generic-is-fine argument is most often used to sell. And the comparative trial does not report until December 2030, so anyone claiming the question is resolved today — in either direction — is ahead of the evidence by about five years.

The adjacent controversy — is paid integration necessary? The claim that self-directed integration is inadequate, and that professionally facilitated integration is what converts a drug experience into a durable change, is asserted with great confidence and, as far as this entry can determine, almost entirely by people who sell that service. It has never been isolated in a trial. The one adjacent result cuts a subtler way than either camp would like: in the alcohol trial, the cell combining ketamine with structured psychological therapy produced the largest effect, which is real support for structured psychological work — and the therapy in question was a manualised relapse-prevention protocol delivered by a research team, not a coaching package, and the trial reports no formally significant interaction between the drug and the therapy. The dismissive position ("integration is just journaling with a price tag") is also wrong, because it slides from "unproven as a paid product" to "psychological support does not matter," which the same trial contradicts.

The funding and bias dimension — cui bono, both ways. Toward overclaim, four incentives converge and all four are visible in this entry's source list. Ketamine clinics and telehealth prescribers have a direct interest in borrowing esketamine's approval to legitimise off-label racemic and oral products, in converting proof-of-concept trials into service lines, in reframing waning effect as "maintenance" rather than as a limitation, and in borrowing the safety record of supervised anaesthetic dosing to reassure patients about unsupervised dosing at home. The integration-coaching sector sells an unregulated service, per session, to people in acute distress in the hours after a dissociative experience. The manufacturer funded the registration trials, its press framing of the suicidality programme reads to a non-specialist as a claim the regulator's own label refuses, and the registration programme's two failed short-term trials are almost never mentioned in the same breath as the one that worked. And the most quotable figures in the field — a forty-eight percent versus eleven percent abstinence contrast from fifty-five people, a sixty to seventy-five percent real-world response range with no primary source at all — circulate stripped of their samples or with no source to strip. Toward dismissal, prohibition-era framing treats a WHO essential medicine primarily as a drug of abuse, waves away primary endpoints that were genuinely met, and treats any repeat exposure as inevitably harmful when supervised maintenance has a real relapse-prevention evidence base — and the UK's own statutory advisory council, reviewing the harms in 2026, declined to reclassify upward partly because stigma discourages help-seeking. The cleanest reading of the whole field: the best-evidenced use of ketamine is surgical anaesthesia in resource-limited settings, which has no commercial interest behind it at all, and the least-evidenced claims are the ones with the most money attached.

Realised Position: Ketamine is a real drug with real effects and one genuinely settled use that has nothing to do with mental health. For depression that has resisted adequate treatment, the acute effect is real and modest, it is worth taking seriously, and it should be entered with eyes open about two things. The first is that "FDA-approved" is thinner here than it sounds — one of three short-term trials met its primary endpoint, and the approval leaned on a withdrawal design. The second is the thing that actually shapes the experience: it wears off, and every protocol that works is a repeat-exposure protocol, which means the question "what happens in a year" is a cost question, an exposure question and a safety question simultaneously, and it deserves an answer before the first dose rather than after the twelfth. Do not accept the suicidality claim; the label refuses it. Do not accept the cure framing; the pharmacology refuses it. On branded versus generic, hold the question open — it is genuinely unsettled, it is being tested properly and independently for the first time, and the answer arrives in 2030. On the bladder, hold both halves: the catastrophic picture belongs to heavy chronic recreational use, the therapeutic literature does not reproduce it and there is reasonable evidence of absence rather than mere absence of evidence, and nobody has watched maintenance patients long enough with good enough monitoring to declare years of repeat dosing safe — so know the urinary red flags and act on them. On integration, our position is the plainest one in this entry: reflecting carefully on what happened, writing it down, and doing that work with someone who knows you is worth doing, and there is no evidence you have to buy it. If someone tells you that self-directed integration wastes the treatment, that sentence has never been tested and the person saying it usually sells the alternative. And for recovery specifically — the use case that matters most to the people most likely to read this — the honest headline from the best trial available is not "ketamine treats addiction." It is that ninety-six people, three infusions and six months later, the arm that combined the drug with structured psychological therapy did best, one of the two primary outcomes was null, and the authors called it a proof of concept. That is worth pursuing with a clinician. It is not worth reorganising your recovery around.

Cross-Pillar Connections

Ketamine touches mood, addiction recovery, pain, dissociation and autonomic state, so its connections run wide — and most of them are handoffs rather than overlaps.
• Mental (psychedelic_assisted_addiction_interruption): the hard boundary for the recovery material. This entry owns the ketamine-specific trials — the alcohol trial's abstinence result and null relapse co-primary, the cocaine trial's sample-attached contrast — and defers the entire broader argument about whether drug-occasioned experiences can interrupt addictive patterns. Do not re-argue that case here; route it there.
• Mental (psilocybin_assisted_therapy_clinical): the adjacent trial architecture, and the useful contrast. The classic psychedelic literature has wrestled longer and more publicly with blinding, expectancy and the drug-versus-therapy component problem — the same problem that makes the integration question unanswerable here. Read the two together to see why "the drug plus the therapy worked" is such a hard sentence to decompose.
• Mental (depersonalisation_derealisation_management): the specific caution this entry flags and does not manage. Dissociation is close to ketamine's therapeutic mechanism, not an incidental effect, which makes a history of depersonalisation or derealisation a distinct pre-treatment conversation. Everything about recognising, managing and recovering from those states belongs there.
• Mental (anhedonia_reward_recovery): what the acute effect is often described as lifting, and what has to be rebuilt if it is going to last past the dose. The durability problem in this entry is precisely the argument for that entry mattering.
• Mental (depression_lifestyle_interventions): the durable levers. Ketamine's effect wanes; the ordinary interventions do not need a maintenance infusion. This entry owns the drug and defers the rest of the treatment picture entirely.
• Mental (chronic_stress_management) and Mental (hpa_axis_dysregulation): the background physiology of the states ketamine is being used against, and the reason a fast pharmacological change sits on top of a slower system rather than replacing it.
• Mental (autonomic_nervous_system_balance) and Mental (vagal_tone_practices): the post-dose settling practices. Blood pressure rises peak around forty minutes and last around four hours, and the hours afterwards are for down-regulation rather than activation — but the practices themselves belong to those entries and are not reinvented here.
• Cross-pillar (fibromyalgia_lifestyle_and_management): a direct routing consequence of the pain guidelines. Fibromyalgia is explicitly in the weak-or-no-evidence list for ketamine infusions, so if that is the condition, the conservative management entry is the better use of time and money.
• Cross-pillar (low_back_pain_evidence_and_management): the same routing, for the same reason — low back pain is also in the weak-or-no-evidence list, and the guideline is the reason to send the reader there rather than toward an infusion clinic.

What would change our mind

Industry bias note

Sources (34)

• WHO Model List of Essential Medicines, 23rd list, 2023 (WHO/MHP/HPS/EML/2023.02) — ketamine, injection, under general anaesthetics; listed as an intravenous anaesthetic since 1985. (No commercial interest; generic and cheap. Uses described: induction and maintenance of anaesthesia, procedural sedation, analgesia, with particular importance in resource-limited settings where piped oxygen, ventilators and reliable electricity are unavailable. Correction of record: an earlier draft dated the listing to 1984 in four places. No source could be found for 1984; the available sources give 1985. Corrected throughout — do not reinstate 1984 in a future reconciliation pass. The relative-haemodynamic-stability and preserved-airway-reflex properties are stated generically in this entry — no specific textbook or guideline citation was verified for them. WHO Expert Committee on Drug Dependence scheduling-review outcomes are referenced in secondary sources but the primary document was not reached and is not cited.)↗
• Popova V, Daly EJ, Trivedi M, et al. "Efficacy and Safety of Flexibly Dosed Esketamine Nasal Spray Combined With a Newly Initiated Oral Antidepressant in Treatment-Resistant Depression: A Randomized Double-Blind Active-Controlled Study" (TRANSFORM-2). Am J Psychiatry 2019;176(6):428-438. doi:10.1176/appi.ajp.2019.19020172.↗ pubmed.ncbi.nlm.nih.gov/31109201↗/" target="_blank" rel="noopener">PMID 31109201↗. (Industry-funded: Janssen Research and Development, the manufacturer. A registration trial supporting its own product's approval.) — 39 outpatient referral centres; 435 screened, 227 randomised, 197 completed the 28-day double-blind phase; primary outcome MADRS change to day 28; difference of least-squares means −4.0, SE 1.69, 95% CI −7.31 to −0.64; discontinuation for adverse events 7% versus 0.9%. Context that must travel with it: this is the only one of three short-term phase 3 trials in the registration programme that separated from control on its primary endpoint.
• Fedgchin M, et al. TRANSFORM-1 — the fixed-dose short-term companion trial, pubmed.ncbi.nlm.nih.gov/31290965↗/" target="_blank" rel="noopener">PMID 31290965↗ — did not reach statistical significance on its primary endpoint; the third short-term trial, TRANSFORM-3, likewise did not. (Verification scope: the one-of-three record and the approval's reliance on the randomised-withdrawal trial were confirmed from the FDA Psychopharmacologic Drugs Advisory Committee briefing record and from independent published commentary. The TRANSFORM-1 and TRANSFORM-3 primary papers were not read in this pass, so no figures from either appear anywhere in this entry — only the fact that neither met its primary endpoint. A human should confirm journal and pagination for TRANSFORM-1 and TRANSFORM-3 before external publication.)
• SPRAVATO (esketamine) nasal spray, CIII — FDA prescribing information and SPRAVATO REMS. NDA 211243. (Regulatory document; manufacturer-authored under FDA review. Authoritative for what the regulator requires be said; for risk statements this cuts in the trustworthy direction, since a sponsor has no incentive to invent warnings.) — indication for treatment-resistant depression in adults, as monotherapy or with an oral antidepressant (original approval 5 March 2019; monotherapy added by supplemental approval 21 January 2025); boxed warning naming sedation, dissociation, respiratory depression, abuse and misuse, and suicidal thoughts and behaviours; at least two hours of post-dose monitoring, administration under direct observation in a certified setting, no dispensing for home use; no driving or hazardous activity until the next day following restful sleep; blood pressure increases at all recommended doses, peaking approximately 40 minutes after administration and lasting approximately 4 hours; sedation in 48–61% of treated patients and loss of consciousness in 0.3–0.4%; contraindications limited to aneurysmal vascular disease, arteriovenous malformation, history of intracerebral haemorrhage and hypersensitivity; psychosis handled in Warnings and Precautions as a careful-assessment item, not as a contraindication; Limitations of Use stating that effectiveness in preventing suicide or in reducing suicidal ideation or behaviour has not been demonstrated; statement that no cases of esketamine-related interstitial cystitis were observed in studies including treatment for up to a year; explicit statement that the product is not approved as an anaesthetic. Flag: the programme has been modified since 2019, particularly its registry component; cite a specific label revision, and re-check anything about registry enrolment against the current label. The label's drug-interactions section was not read directly; the additive-CNS-depression statement in this entry is stated at mechanism level rather than quoted.↗
• Daly EJ, et al. "Efficacy of Esketamine Nasal Spray Plus Oral Antidepressant Treatment for Relapse Prevention in Patients With Treatment-Resistant Depression: A Randomized Clinical Trial" (SUSTAIN-1). JAMA Psychiatry 2019;76(9):893-903. pubmed.ncbi.nlm.nih.gov/31166571↗/" target="_blank" rel="noopener">PMID 31166571↗; NCT02493868. (Janssen-sponsored.) — phase 3 randomised-withdrawal design; 297 patients randomised in the maintenance phase after 16 weeks of esketamine treatment — 176 in stable remission, 121 in stable response without remission — each stratum randomised 1:1 to continue esketamine plus oral antidepressant or switch the spray to placebo while continuing the oral antidepressant; stable remitters relapsed 26.7% versus 45.3%, hazard ratio 0.49 (95% CI 0.29–0.84); stable responders relapsed 25.8% versus 57.6%, hazard ratio 0.30 (95% CI 0.16–0.55). These figures were previously withheld in this entry as unread and have now been verified against the primary record. This is also the trial the original approval leaned on most heavily, which is a fact about the approval and not a criticism of the trial.
• Ionescu DF, et al. "Esketamine Nasal Spray for Rapid Reduction of Depressive Symptoms in Patients With Major Depressive Disorder Who Have Active Suicide Ideation With Intent: Results of a Phase 3, Double-Blind, Randomized Study" (ASPIRE II). Int J Neuropsychopharmacol 2021;24(1):22-31. pubmed.ncbi.nlm.nih.gov/32861217↗/" target="_blank" rel="noopener">PMID 32861217↗. (Janssen-sponsored, with press framing that reads to a non-specialist as a suicidality claim the label refuses.) — 230 randomised, 227 dosed and analysed; esketamine 84 mg or placebo twice weekly for 4 weeks on top of comprehensive standard of care (hospitalisation for at least five days plus newly initiated or optimised oral antidepressants); primary endpoint met — MADRS at 24 hours, LSM difference −3.9 (SE 1.39), 95% CI −6.60 to −1.11, two-sided p=0.006 (esketamine −15.7 ±11.56 versus placebo −12.4 ±10.43); key secondary endpoint not met — both arms improved rapidly on the revised Clinical Global Impression of Severity of Suicidality and the between-group difference was not statistically significant, with the authors themselves pointing at psychiatric hospitalisation as a likely reason both arms moved. ASPIRE I is named but no numeric effect is published from it: only a pooled/press summary was read.
• Cohen SP, Bhatia A, Buvanendran A, et al. "Consensus Guidelines on the Use of Intravenous Ketamine Infusions for Chronic Pain Management" — American Society of Regional Anesthesia and Pain Medicine, American Academy of Pain Medicine, and American Society of Anesthesiologists committees on pain medicine and on standards and practice parameters. Reg Anesth Pain Med 2018;43(5):521-546, published 1 July 2018. (Multi-society consensus, notably unflattering to the treatment; about as clean a source as this field offers.) — moderate evidence for CRPS (22 mg/h over 4 days, or 0.35 mg/kg/h over 4 hours daily for 10 days; benefit up to 12 weeks); weak for spinal cord injury pain, short-term only; weak or no evidence for mixed neuropathic pain, phantom limb pain, postherpetic neuralgia, fibromyalgia, cancer pain, ischaemic pain, migraine and low back pain; excluding CRPS, no evidence for intermediate- or long-term pain improvement. Verification limit: dated 2018 with a stated review point of 2023; we did not verify whether a superseding version exists.↗
• Feder A, et al. "A Randomized Controlled Trial of Repeated Ketamine Administration for Chronic Posttraumatic Stress Disorder." Am J Psychiatry 2021;178(2):193-202. pubmed.ncbi.nlm.nih.gov/33397139↗/" target="_blank" rel="noopener">PMID 33397139↗; doi:10.1176/appi.ajp.2020.20050596.↗ (Single academic site, n=30. The midazolam comparator partly blinds for the psychoactive experience, which is a genuine methodological strength most psychedelic-adjacent trials lack.) — six infusions of ketamine 0.5 mg/kg or midazolam 0.045 mg/kg over two consecutive weeks; CAPS-5 total 11.88 points lower (SE 3.96), d=1.13 (95% CI 0.36–1.91); response 67% versus 20%; no serious adverse events; median time to loss of response among responders 27.5 days. Replication status not verified — a later multi-site trial exists in the literature and was not read; this entry asserts neither replication nor failure to replicate.
• Grabski M, et al. "Adjunctive Ketamine With Relapse Prevention-Based Psychological Therapy in the Treatment of Alcohol Use Disorder" (KARE). Am J Psychiatry 2022;179(2):152-162. pubmed.ncbi.nlm.nih.gov/35012326↗/" target="_blank" rel="noopener">PMID 35012326↗; doi:10.1176/appi.ajp.2021.21030277.↗ (The recovery-relevant trial; must be reported with both hands.) — 96 patients with severe alcohol use disorder (35 women; mean age 44.07), 2×2 randomisation (three weekly ketamine infusions 0.8 mg/kg IV over 40 minutes or saline; psychological therapy or alcohol education), 6-month follow-up; percentage days abstinent significantly greater with ketamine, mean difference 10.1% (95% CI 1.1–19.0); largest cell effect ketamine plus therapy versus saline plus education 15.9% (95% CI 3.8–28.1); no significant difference in relapse rate — the second primary outcome was null; no formally significant drug-by-therapy interaction reported in the abstract; wide confidence intervals, authors describe it as proof of concept; well tolerated with no serious adverse events attributed to study drug.
• Dakwar E, et al. "A Single Ketamine Infusion Combined With Mindfulness-Based Behavioral Modification to Treat Cocaine Dependence: A Randomized Clinical Trial." Am J Psychiatry 2019;176(11):923-930. doi:10.1176/appi.ajp.2019.18101123.↗ — 55 cocaine-dependent adults, single 40-minute infusion of ketamine 0.5 mg/kg or midazolam during a 5-day inpatient stay alongside 5 weeks of mindfulness-based relapse prevention; 48.2% versus 10.7% maintained abstinence over the final two weeks; single site, single infusion, no replication identified. (A companion single-infusion alcohol trial from the same group, Am J Psychiatry 2020, was not read and no figures from it appear.)
• "Replication of ketamine's antidepressant efficacy in bipolar depression: a randomized controlled add-on trial." Biol Psychiatry 2012. pubmed.ncbi.nlm.nih.gov/22297150↗/" target="_blank" rel="noopener">PMID 22297150↗. — 15 subjects with bipolar I or II depression on therapeutic lithium or valproate; double-blind randomised crossover, single IV infusion 0.5 mg/kg versus placebo on two test days two weeks apart; improvement within 40 minutes; 79% responded to ketamine, 0% to placebo at some point during the trial.
• Rodriguez CI, et al. "Randomized Controlled Crossover Trial of Ketamine in Obsessive-Compulsive Disorder: Proof-of-Concept." Neuropsychopharmacology 2013. — 15 drug-free adults with near-constant obsessions; two 40-minute infusions (saline and ketamine 0.5 mg/kg) at least a week apart; one week post-infusion, 4 of 8 ketamine patients met response criteria (≥35% Y-BOCS reduction) versus 0 of 7 on placebo.↗
• Li N, Lee B, Liu RJ, Banasr M, Dwyer JM, Iwata M, Li XY, Aghajanian G, Duman RS. "mTOR-dependent synapse formation underlies the rapid antidepressant effects of NMDA antagonists." Science 2010;329(5994):959-964. pubmed.ncbi.nlm.nih.gov/20724638↗/" target="_blank" rel="noopener">PMID 20724638↗. [SPRAGUE-DAWLEY RATS] (Academic/NIH-supported; no manufacturer interest in the mechanism claim.) — ketamine rapidly activated mTOR, increasing synaptic signalling proteins; blockade of mTOR signalling completely blocked ketamine induction of synaptogenesis and behavioural responses in rodent despair models. The endpoint is antidepressant-like behaviour in a rodent model, not depression, and the blockade result is route-dependent even within rodent work.
• Abdallah CG, Averill LA, Gueorguieva R, et al. "Modulation of the antidepressant effects of ketamine by the mTORC1 inhibitor rapamycin." Neuropsychopharmacology 2020;45(6):990-997. pubmed.ncbi.nlm.nih.gov/32092760↗/" target="_blank" rel="noopener">PMID 32092760↗. [HUMAN, n=20] — rapamycin did not block ketamine's antidepressant effect; it prolonged it: at two weeks, response 41% and remission 29% with rapamycin plus ketamine versus 13% (p=0.04) and 7% (p=0.003) with placebo plus ketamine. This is the human result that must accompany any statement of the mTOR mechanism; flagging the animal work as "rodent" is necessary and not sufficient.
• Chan EOT, Chan VWS, Tang TST, Cheung V, Wong MCS, Yee CH, Ng CF, Teoh JYC. "Systematic review and meta-analysis of ketamine-associated uropathy." Hong Kong Med J 2022;28(6):466-474. doi:10.12809/hkmj209194.↗ pubmed.ncbi.nlm.nih.gov/36464318↗/" target="_blank" rel="noopener">PMID 36464318↗. (Academic urology, Chinese University of Hong Kong / Prince of Wales Hospital, working in the population where the harm was first characterised; funding statement not read.) — 45 articles, 4,921 patients; urinary frequency 77.1%, urgency 69.9%, suprapubic pain 60.4%, hydronephrosis 30.2%; mean functional bladder capacity 95.23 mL, mean maximum flow rate 8.69 mL/s; urothelial denudation and inflammatory infiltration on biopsy; treatment escalating through oral medication, intravesical therapy, hydrodistension and bladder reconstruction; abstinence necessary for improvement regardless of treatment. A separately circulating "37 studies, 44–77% lower-tract and 8–30% upper-tract" set of ranges is NOT welded to this review — it is a different, untraced source and is not used.
• Shahani R, Streutker C, Dickson B, Stewart RJ. "Ketamine-associated ulcerative cystitis: a new clinical entity." Urology 2007;69(5):810-812. doi:10.1016/j.urology.2007.01.038.↗ pubmed.ncbi.nlm.nih.gov/17482909↗/" target="_blank" rel="noopener">PMID 17482909↗. (Academic clinical case series, Department of Surgery, St. Michael's Hospital and University of Toronto; no commercial interest.) — nine cases, daily ketamine users, severe dysuria, frequency, urgency and gross haematuria; investigation by urine culture, microscopy and cytology plus CT, cystoscopy and bladder biopsy; CT showing small bladder capacity, marked wall thickening and perivesical inflammation; severe ulcerative cystitis at cystoscopy. A nine-patient uncontrolled series in daily users: it establishes that the entity is real and carries no prevalence information and nothing about therapeutic dosing. It must not be allowed to do prevalence work.
• Zhou J, Scott C, Miab ZR, Lehmann C. "Current approaches for the treatment of ketamine-induced cystitis." Neurourol Urodyn 2023;42(4):680-689. doi:10.1002/nau.25148.↗ (Academic review, Dalhousie University group; no manufacturer interest.) — treatment approaches from conservative and pharmacological management through hydrodistension, augmentation cystoplasty and cystectomy. A narrative review of treatment options, most underlying evidence uncontrolled: it supports statements about how the condition is managed and how severe end-stage management gets, and supports nothing about incidence, therapeutic risk or causation.
• Kerr-Gaffney J, Tröger A, Caulfield A, Ritter P, Rucker J, Young AH. "Urological symptoms following ketamine treatment for psychiatric disorders: A systematic review." J Psychopharmacol 2025;39(10):1103-1113. doi:10.1177/02698811251350267.↗ pubmed.ncbi.nlm.nih.gov/40583492↗/" target="_blank" rel="noopener">PMID 40583492↗; PMCID PMC12572349. (Academic — King's College London and Dresden; no manufacturer sponsorship identified, which makes its no-elevated-risk conclusion harder to dismiss. Honest flag: the senior authors work in a group with substantial psychopharmacology and psychedelic-trial relationships, so this is not a disinterested corner of the field, though the review's wording is conservative and unusually candid about its own limits.) — 27 studies (15 RCTs, 7 retrospective cohorts, 5 open-label), mostly depressive disorders; urological symptoms reported in 0%–24.5% of patients, tending to be mild or moderate, generally at or below control rates in RCTs; continuous outcome measures showed no significant change from baseline; conclusion: "ketamine treatment does not appear to be associated with elevated risk of urological symptoms; however, further long-term studies are required." Limitations in the authors' own terms: median follow-up for urological symptoms 4 weeks; only 15% of studies at low risk of bias; 11 studies with no systematic urological assessment; only two with laboratory urinalysis; only five assessing treatment beyond six months; some enrolling only prior responders. Verification scope: the abstract was verified directly and carries the range, the 15% low-risk-of-bias figure, the null continuous-measure finding and the conclusion. The four-week median, the eleven-studies count and the two-urinalysis count come from the full-text limitations section, which was not re-read in this pass; they are retained because they cut against this entry's own reassurance, and flagged rather than presented as re-verified. Correction of record: 24.5% is the upper bound of a 0%–24.5% range from a single arm of a single trial, not a pooled estimate. An earlier draft of this entry presented it as "roughly a quarter" and cited this review as evidence of elevated therapeutic-dose risk — the exact reverse of its conclusion. That claim is cut and the correction is stated in the entry body so it cannot be silently reintroduced.
• Andrade C. "Ketamine-Associated Uropathy During Therapeutic and Nontherapeutic Use: Prevalence, Clinical Features, Mechanisms, and Strategies for Risk Reduction." J Clin Psychiatry 2025;86(4):25f16083. doi:10.4088/JCP.25f16083.↗ — mechanism as direct bladder exposure rather than a circulatory effect, through mitochondrial stress, oxidative injury, urothelial barrier disruption, inflammation and progressive fibrosis; approximately 75% of recreational users showing substantial symptom reduction within three months of cessation, with the remainder including cases requiring surgery; risk-reduction advice of hydration on treatment days, frequent voiding, fortnightly or monthly urine screening for protein, cells and blood, and routine symptom enquiry during maintenance; independent read that the 2025 systematic review presented no convincing evidence of uropathy arising in therapeutic contexts.
• Leo RJ, Blaes SL, Jaffri MA, et al. "Recreational ketamine-induced uropathy: recognition and management in primary care and psychiatric practice." Prim Care Companion CNS Disord 2026;28(2):25br04121. doi:10.4088/PCC.25br04121.↗ — classic lower urinary tract symptoms named as frequency, urgency, nocturia and intermittent haematuria, with dysuria and suprapubic pain as additional warning signs; symptoms fluctuating with use and worsening rapidly on relapse; incomplete recovery in some; progressive bladder wall thickening; discontinuation described as the only consistently effective strategy for preventing irreversible renal injury; therapeutic ketamine under medical supervision described as carrying minimal risk with only one confirmed published case of cystitis from therapeutic use; escalation case material including a patient reaching 3 g/day through tolerance and another taking 2–6 g daily over three years.
• Waheed HW, Ashraf MR, Sajjad T, et al. "Hepatobiliary Complications Associated With Ketamine Use: Clinical Insights and Future Directions." Cureus 2025;17(5):e84974. doi:10.7759/cureus.84974.↗ (Source-vehicle flag: Cureus is a low-barrier peer-reviewed journal. The specific figures — the 9.8%, the 88%/12% split — should be traced to the underlying Hong Kong cross-sectional study and the case-report synthesis before being leaned on hard.) — abdominal pain in 71% of reported cases, with jaundice, elevated enzymes and cholestasis; fusiform common bile duct dilation and irregular intrahepatic narrowing resembling secondary sclerosing cholangitis; liver injury in 9.8% of a 297-person chronic-user cross-sectional survey; median use around 24 months, mean age around 25–26, majority male; roughly 88% improving after cessation with about 12% persistent; biochemical recovery after around 60 days of abstinence in some cases. Described exclusively in recreational or chronic-abuse contexts; no hepatobiliary injury reported in therapeutic psychiatric dosing.
• Ingrosso G, Cleare AJ, Juruena MF. "Is there a risk of addiction to ketamine during the treatment of depression? A systematic review of available literature." J Psychopharmacol 2025;39(1):49-65. doi:10.1177/02698811241303597.↗ pubmed.ncbi.nlm.nih.gov/39688236↗/" target="_blank" rel="noopener">PMID 39688236↗. (Academic, King's College London.) — 16 studies, 2,174 patients; four patients exhibited clear signs of tolerance to the antidepressant effect or dependence on the drug, the majority did not; conclusion emphasising the relative safety of ketamine treatment for adults with depression alongside the importance of medically supervised administration, vigilant monitoring and judicious dosing.
• "Treating addiction with an addictive drug: the ketamine paradox revisited." Front Psychiatry 2026. doi:10.3389/fpsyt.2026.1866092.↗ — conclusion that existing substance-use-disorder trials have not reported illicit transition, diversion or ketamine use disorder when delivered under rigorous clinical protocols. This is a statement about SUD trials specifically, not about the whole clinical trial literature, and this entry does not broaden it.
• Gutierrez G, Vazquez G, Ravindran N, et al. "Abuse liability for esketamine in a cohort of patients undergoing an acute treatment course to manage treatment-resistant depression." Ther Adv Drug Saf 2025;16:20420986251347360. doi:10.1177/20420986251347360.↗ (Secondary analysis of a multicentre observational study of a marketed product — check its sponsor before quoting its percentages.) — 23 patients across eight intranasal esketamine sessions using a liking-and-craving questionnaire; no cravings in 78.3% initially, 86.96% reporting no desire to use more than prescribed, no increase across the course; authors flag the small sample, the unvalidated instrument, and the exclusion of people with active substance use disorders.
• Andrashko V, Novak T, Brunovsky M, Klirova M, Sos P, Horacek J. "The Antidepressant Effect of Ketamine Is Dampened by Concomitant Benzodiazepine Medication." Front Psychiatry 2020;11:844. doi:10.3389/fpsyt.2020.00844.↗ pubmed.ncbi.nlm.nih.gov/33005153↗/" target="_blank" rel="noopener">PMID 33005153↗. (Academic, Czech National Institute of Mental Health; no commercial interest in a finding that reduces ketamine's apparent efficacy.) — retrospective analysis of 47 inpatients with treatment-resistant depression pooled from two randomised trials, single 0.54 mg/kg IV infusion; ROC-identified cutoff of >8 mg diazepam-equivalent daily dose (sensitivity 80%, specificity 85%, p<0.001); response 8% in higher-dose versus 35% in lower-dose benzodiazepine users; impairment at days 3 and 7 (p=0.04, p=0.02) but not at 24 hours; Hedges' g 0.67 and 0.78; adjusted OR 9.7 (95% CI 1.2–69.7, p=0.03); authors' limitations of small sample, retrospective design, no blood levels, baseline anxiety differences and post-hoc analysis. Two related findings are named as unread and NOT cited: a reported secondary analysis of an intravenous ketamine study finding the attenuation at day 1 but not day 3 — the opposite time pattern — and a reported esketamine null at 24 hours.
• Bahji A, Vazquez GH, Zarate CA Jr. "Comparative efficacy of racemic ketamine and esketamine for depression: A systematic review and meta-analysis." J Affect Disord 2021;278:542-555. — racemic ketamine showing greater response (RR 3.01) and remission (RR 3.70) than esketamine (RR 1.38 and 1.47), and lower dropout (RR 0.76 versus 1.37). Indirect comparison, not a head-to-head randomised trial, which is precisely why the head-to-head is now being run. It also carries a published erratum (pubmed.ncbi.nlm.nih.gov/33229028↗/" target="_blank" rel="noopener">PMID 33229028↗) and drew a formal Letter to the Editor with an author response (pubmed.ncbi.nlm.nih.gov/33957360↗/" target="_blank" rel="noopener">PMID 33957360↗, pubmed.ncbi.nlm.nih.gov/36574849↗/" target="_blank" rel="noopener">PMID 36574849↗) — this entry names that contestation rather than leaning on the paper one-sidedly because its conclusion points away from the branded product.
• ClinicalTrials.gov NCT06713616. "Comparative Effectiveness of Racemic Ketamine Versus Esketamine (Spravato®) for Depression." Lead sponsor Yale University; collaborator and funder the Patient-Centered Outcomes Research Institute (reported $12.6 million grant); collaborating with Emory University, the University of Michigan and three community sites; protocol published as the EQUIVALENCE trial, pubmed.ncbi.nlm.nih.gov/41780747↗/" target="_blank" rel="noopener">PMID 41780747↗. (The strongest funding shape in this entry: an independently funded comparative-effectiveness test of an approved branded product against its cheaper generic-molecule competitor, with no sponsor incentive in either direction.) — RECRUITING, no results. Estimated enrolment 400; randomised 1:1 to intravenous racemic ketamine (infused over 40 minutes, maximum 60 mg daily dose) or intranasal esketamine (56–84 mg per label), 8 treatments over 4 weeks; start 27 January 2025; estimated primary and study completion 31 December 2030. Citable only as evidence that the comparison is genuinely unsettled and being tested, never for any result, and never in a way that implies near-term resolution.
• FDA. "FDA Alerts Health Care Professionals About Potential Risks Associated with Compounded Ketamine Nasal Spray," 16 February 2022. FDA. "FDA Warns Patients and Health Care Providers About Potential Risks Associated with Compounded Ketamine," 10 October 2023. (Regulatory; no commercial funding. ⚠ VERIFICATION CAVEAT: fda.gov returned errors to our retrieval tooling on every attempt. Titles, dates and quoted wording are confirmed via a regulatory-law analysis linking the primary FDA URLs plus two independent search-index summaries in agreement. A human should open both FDA pages and check the wording before this entry is published externally.) — the 2023 alert covering compounded products including oral formulations for psychiatric disorders: risk of serious adverse events, misuse and abuse especially with home use; adverse event reports from patients taking compounded ketamine outside health care settings; at-home administration risky because no provider is on site to monitor for sedation and dissociation; safe and effective dosing of ketamine for any psychiatric disorder not established; agency not aware of evidence that compounded ketamine is safer, more effective or faster-acting than approved medications; telemedicine platforms and compounders named as the supply route.↗
• Scheduling instruments. US: Controlled Substances Act, 21 CFR 1308.13 — Schedule III, scheduled 1999. UK: Misuse of Drugs Act 1971, Schedule 2 Part II para 1(a) — Class B, as inserted by The Misuse of Drugs Act 1971 (Ketamine etc.) (Amendment) Order 2014, SI 2014/1106, in force 10 June 2014; injection provision at Schedule 2 Part I para 6 — any preparation designed for administration by injection which includes a substance specified in the relevant Part II paragraphs is treated as Class A. This is a genuine statutory rule and a general provision applying to all such substances, not a ketamine-specific one. Australia: Poisons Standard (SUSMP), Schedule 8 Controlled Drug; TGA, "Notice of final decisions not to amend the current Poisons Standard in relation to ketamine, esketamine and methysergide," 2023. (Primary law; no funding interest.)↗
• Advisory Council on the Misuse of Drugs. "Ketamine: an updated review of use and harms," published 28 January 2026 (gov.uk); UK Home Office, "Home Office requests review of ketamine classification," 8 January 2025 (gov.uk). (UK statutory advisory body; no commercial funding. The council's reasoning is itself anti-moralising and this entry reflects it rather than editorialising on top.) — Home Office request citing an estimated 299,000 people aged 16–59 reporting last-year use in the year ending March 2023, the largest on record; council advised ketamine remain Class B, declining the Class A upgrade, on the grounds that acute toxicity and death rates remain consistent with Class B, that evidence reclassification reduces use or harm is limited, that many acute harms are significantly influenced by concurrent use of other drugs, that reclassification could increase stigma and discourage help-seeking, and that higher penalties could disproportionately affect young and vulnerable people; recommending a public-health and whole-systems approach and greater support for police and healthcare professionals instead. The recommendation was not unanimous within the council, and this entry states that rather than quoting the advice as a settled institutional verdict.↗
• Aslam AM, Shinozuka K, Muir O, Tabaac BJ. Am J Ther 2025;32(3):e242-e246. doi:10.1097/MJT.0000000000001951.↗ (Cited for context only, and specifically for what it does NOT contain.) — states that the onset of the COVID-19 pandemic in 2020 accelerated adoption of at-home ketamine treatment models, triggering DEA and FDA scrutiny; provides no figures on the prevalence of at-home, telehealth or compounded oral ketamine and no data on clinic growth, and its own base is two survey studies. No market-size, clinic-count or patient-count figure appears anywhere in this entry, because none could be traced to a primary source.
• *Claims cut for citation failure and named so they are not reintroduced:** (1) that a 2025 systematic review found roughly a quarter of ketamine-treated psychiatric patients developed lower urinary tract symptoms — REFUTED; the review reports a 0%–24.5% range and concludes no elevated risk. (2) That real-world response rates for both esketamine and compounded racemic ketamine run 60–75% — UNVERIFIABLE; no primary source, peer-reviewed pooled figures materially lower, range traces to clinic and telehealth marketing. (3) Racemic infusion effect sizes of approximately 1.46 at 24 hours and 0.68 at one week — UNVERIFIABLE; not traceable to a named primary meta-analysis. (4) Any claim about ketamine and smoking cessation — no source located. (5) "One to two weeks back to baseline" as a durability figure — traceable only to secondary summaries; the shape is retained, the number is not.↗
• *Corrections applied in the August 2026 adversarial pass, named so they are not reversed: (a) the WHO essential-medicines listing date moved from 1984 to 1985 in all four places it appeared. (b) The esketamine registration record was completed: only one of the three short-term phase 3 trials met its primary endpoint, and the approval leaned on the randomised-withdrawal trial — previously absent, now carried in the evidence claim, the bad-advice list, the controversy, the tier justification and the bias analysis. (c) The withdrawal trial's relapse percentages and hazard ratios, previously withheld as unread, were verified and added. (d) ASPIRE II, the chronic-pain guidelines and the PTSD trial were given complete author and pagination citations, all figures re-verified unchanged. (e) The meta-analysis favouring racemic ketamine is now reported with its erratum and its published challenge. (f) The UK advisory council's recommendation is now reported as not unanimous**.↗
• Funding notation: the entry's firmest claims run in the opposite direction from the money. Ketamine's essential-medicine status comes from an intergovernmental body listing a cheap generic nobody markets. The finding that therapeutic dosing does not appear to elevate urological risk comes from an independent academic review with no manufacturer sponsorship. The chronic-pain grading that is most damaging to infusion clinics comes from a three-society consensus. The comparison that will finally settle branded versus generic is publicly funded. The claims this entry had to cut all traced to manufacturer press framing, clinic marketing, or nowhere at all.*↗