Side-effects associated with ketamine use in depression: a systematic review

Major depressive disorder is highly prevalent and existing monoaminergic antidepressants are effective in only about 50% of patients, with a protracted onset of action. In this context, ketamine has emerged as a promising rapidly acting antidepressant, but uncertainty remains about its safety profile. Short and colleagues note that most clinical trials to date have been small, short-term, frequently assessed a single dose only, and were not designed to detect rare, cumulative, or long-term harms. Evidence from other populations exposed to ketamine (for example, people with chronic pain or recreational users) has linked repeated use to urinary, hepatic, cognitive, and dependence-related harms, raising concern about the generalisability of short-term depression trial safety data. This paper therefore set out to systematically aggregate and evaluate reporting of side-effects associated with ketamine when used to treat depression. The primary objective was to identify and categorise adverse effects reported after single (acute) and repeated (cumulative and long-term) dosing in studies of adults with unipolar or bipolar depression, and to assess the extent and quality of side-effect surveillance in the published literature.

The review followed PRISMA guidelines. Eligible studies had to report original findings in adult humans with a validated diagnosis of unipolar or bipolar depression, include one or more doses of ketamine by any route, and report changes in depression as a primary outcome. Exclusions were animal studies, non-original or mechanistic reports, and studies in paediatric or adolescent populations. Two authors (JF and WS) conducted a two-step literature search (title/abstract screening followed by full-text review); disagreements were resolved by consensus and consultation with a senior author. The investigators intentionally included a broad range of study designs—including randomised trials, open-label studies, case series, and letters reporting case material—to increase the chance of detecting rare or delayed adverse effects. Data extracted from each article included study design and sample characteristics, ketamine administration details (route, dose, and number of doses), pre-treatment health screening, pre-existing morbidity, concomitant medications, and timing and method of side-effect assessment. Each article was quality-appraised with an appropriate checklist (appendix). For analysis, adverse effects were categorised into psychiatric, psychotomimetic/dissociative, cardiovascular, neurological (including cognitive), and other systems. The authors applied the Cochrane Adverse Effects Methods Group framework to examine selective outcome reporting, withdrawal/dropout, and the presence of control groups, and distinguished active surveillance (structured enquiry) from passive monitoring (spontaneous reports). Timeframes were defined as acute (immediately after a single dose), cumulative (after repeated doses), and long-term (at least 2 weeks after the last dose). A meta-analysis had been planned a priori, but after data extraction the investigators judged quantitative pooling inappropriate because studies were clinically heterogeneous (different routes, doses, comparators), used inconsistent reporting formats (often qualitative), and had variable risk of bias. Consequently, the review presents a qualitative synthesis of side-effect reporting across the included studies.

The review included 60 studies comprising 899 patients who received at least one dose of ketamine. Most study reports did not include a placebo or other control arm. Assessment of side-effects was concentrated on the acute period: 55 (92%) studies assessed acute side-effects, 24 (40%) assessed cumulative effects after repeated dosing, and 12 (20%) reported any long-term follow-up. Across studies that included a comparator, adverse events in all categories were reported more frequently in patients who received ketamine than in controls. Psychiatric side-effects were described in 23 (38%) studies and psychotomimetic or dissociative effects in 43 (72%) studies; some of these used structured scales. Timepoints for measurement varied between studies, but reported changes were generally transient. The most common acute psychiatric reactions were anxiety, agitation or irritability, euphoria or mood elevation, delusions or unusual thoughts, panic, and apathy. Withdrawal from trials for psychiatric reasons most commonly reflected worsening mood (12 participants), followed by anxiety (6 participants) and suicidal ideation (5 participants); less frequent causes included panic attacks and irritability. One isolated suicide attempt was reported in a single study. Dissociation was the predominant psychotomimetic phenomenon, followed by perceptual disturbances, derealisation, hallucinations, and depersonalisation. No long-term psychotomimetic effects were reported, but most assessments for these phenomena were limited to the short term (usually within 4 hours post-dose). Dissociation led to withdrawal in two participants. Intravenous administration was associated with a higher frequency of psychotomimetic or dissociative reports (72% of intravenous studies) compared with non-intravenous routes (36%). Cardiovascular changes were reported in 23 (38%) studies; the most common findings were increased blood pressure and heart rate. Other cardiovascular events included palpitations or arrhythmia, chest symptoms, orthostatic dizziness, and, less commonly, decreased blood pressure or heart rate. Five participants were withdrawn because of cardiovascular side-effects. Most cardiovascular effects occurred during or immediately after intravenous dosing and generally resolved within 90 minutes. Neurological side-effects most often consisted of headache and dizziness; less common reports included sedation, faintness, poor coordination, and tremor. Cognitive complaints—poor memory, poor concentration, confusion—were reported but typically only over the short term, and prospective cognitive assessment was lacking in most studies. Other adverse effects were described in 32 (53%) studies and encompassed gastrointestinal, ocular, respiratory, and urological symptoms; blurred vision and nausea were the most frequently reported. Urinary adverse effects were evaluated in only five studies. Liver function tests were specifically reported in seven studies, with two noting abnormalities after ketamine dosing. Only two studies explicitly enquired about dependence or abuse potential. Regarding assessment methods, structured side-effect scales (for example, SAFTEE, PRISE) were used in 15 (25%) studies. Most studies relied on passive monitoring and ad hoc reporting rather than active, systematic enquiry. When systematic reporting was present, it occurred in only a minority of studies: psychiatric or psychotomimetic effects were systematically reported in 25 (42%) studies, cardiovascular effects in 19 (32%), neurological or cognitive effects in 14 (23%), and other side-effects in 14 (23%). Randomised controlled trials tended to report similar rates of acute side-effects as other designs, but rarely assessed or reported longer-term risks such as cognition, urinary tract effects, or dependence.

Short and colleagues conclude that acute side-effects after ketamine treatment for depression are common but usually transient, and that the existing literature provides inadequate active surveillance of adverse effects. The main findings highlighted were: frequent acute reactions (psychiatric, psychotomimetic, cardiovascular, neurological, and other), predominant use of passive monitoring in studies, higher frequency of reported side-effects with intravenous administration, and insufficient data to draw conclusions about cumulative or long-term harms because most trials assessed only single doses with short follow-up. The authors place these results in the context of reports from other populations exposed to ketamine. Repeated or frequent ketamine use in recreational users and some clinical groups has been linked with ulcerative cystitis and other urological toxicity, liver injury, cognitive deficits (short-term and long-term memory impairments), and dependence-related behaviours. The review summarises case reports and series describing bladder dysfunction, renal complications in severe cases, and hepatotoxicity after prolonged or repeated infusions. Experimental human studies have demonstrated dose-dependent acute impairment of short-term memory after intravenous ketamine, whereas findings from clinical treatment populations are mixed but limited by short follow-up durations. Safety concerns in patients with medical comorbidity are also emphasised. Given common acute blood-pressure elevations after ketamine, the authors note that WHO contraindications for ketamine (moderate-to-severe hypertension, congestive cardiac failure, prior stroke) are pertinent and that clinicians should exercise caution, particularly with repeated dosing and rapid delivery routes. Dependency risk is discussed: preclinical and clinical evidence indicates rapid development of tolerance (tachyphylaxis) and in recreational settings craving and compulsive use are reported, although fewer than 15 well-documented human dependence cases have been described in the past two decades. Key limitations acknowledged by the investigators include heterogeneity in study designs and reporting, inconsistent and often qualitative side-effect descriptions, variable timepoints, and insufficient numerical detail to permit meta-analysis or robust incidence estimates. Because most included studies used passive or ad hoc reporting, the true frequency and longer-term trajectory of many adverse effects remain uncertain. The authors recommend that future ketamine trials for depression adopt active surveillance and standardised, structured scales to assess side-effects across acute, between-dose, and long-term periods, and implement pre-treatment screening for comorbid conditions and concomitant medications. They note ongoing work to develop a Ketamine Side Effect Tool and a Ketamine Safety Screening Tool, and highlight outstanding questions about differences between racemic ketamine and single isomers or metabolites with respect to adverse-effect profiles.

Ketamine has pharmacological properties that make it a potentially useful treatment for refractory depression, but the evidence base for its safety is incomplete. Acute side-effects after single-dose use are common yet generally transient. By contrast, higher doses and repeated administration have been associated—outside depression trials—with potentially serious and sometimes persistent toxicities, including urological and hepatic injury, cognitive impairment, and craving or dependence. These longer-term risks have not been adequately assessed in studies of ketamine for depression, most of which evaluated single-session treatments with only short-term follow-up. The safety of long-term, repeated ketamine dosing in depression therefore remains uncertain. The authors recommend further large-scale clinical trials that include multiple doses, long-term follow-up, careful monitoring, and systematic reporting of all potential side-effects before ketamine is adopted widely as a clinical antidepressant.