Ketamine administration in depressive disorders: a systematic review and meta-analysis

Major depressive disorder (MDD) and bipolar disorders are major contributors to global disability and carry large social costs. Despite available treatments, a substantial proportion of patients do not respond to conventional medication, motivating investigation of novel therapies. Ketamine, an anaesthetic with N-methyl-D-aspartate receptor (NMDAR) antagonist properties, has been reported to have rapid antidepressant effects in clinical studies, but results across trials have been inconsistent and clinical questions remain regarding indications, dose, duration of effect, impact on suicidal ideation, and safety. Fond and colleagues therefore undertook a systematic review and meta-analysis of randomised controlled trials to assess whether ketamine administration reduces depressive symptomatology. The primary focus was on efficacy in MDD, bipolar depression, treatment-resistant depression (non-ECT trials), and when ketamine is used as the anaesthetic agent during electroconvulsive therapy (ECT). Secondary aims included characterising duration of effect, effects on suicidal ideation, possible dose–response relationships, and safety/tolerability.

The investigators performed a PRISMA-based systematic search across multiple databases (PubMed/MEDLINE from 1966 to September 2013, Embase from 1980, PsycINFO from 1806, BIOSIS from 1926, Science Direct from 2006, and ProQuest Dissertations) using terms for ketamine, depression-related disorders and controlled clinical trials. There were no language or date restrictions. Eligible studies were randomised controlled trials comparing ketamine administration (single or repeated doses, alone or with other anaesthetic agents) against control conditions in participants diagnosed with a depressive disorder, with outcomes assessed on validated depression rating scales. Trials without a comparator or those where a standardised mean difference (SMD) could not be calculated were excluded. Study selection and data extraction were conducted by multiple reviewers with disagreements resolved by consensus. Extracted items included authorship, publication year, design, sample size, diagnostic composition (MDD versus bipolar), rating scales used (HDRS or MADRS), whether patients were drug-free at baseline, co-administration with ECT or other anaesthetics, and ketamine dose. Methodological quality was assessed using Cochrane risk-of-bias markers (sequence generation, allocation concealment, blinding, incomplete outcome data) and trials were further classified by a four-tiered evidence scheme (class I–IV). For quantitative synthesis the researchers calculated standardised mean differences (SMDs) and 95% confidence intervals based on between-group pre–post change scores, and they pooled estimates using fixed- and random-effects models. Heterogeneity was quantified with I2, with I2>50% considered large. Publication-bias assessments included funnel plots, Rosenthal’s fail-safe N and Egger’s regression intercept. Prespecified subgroup and sensitivity analyses examined effects by diagnosis (unipolar vs bipolar), resistance status, drug-free versus medicated status, co-administration with other anaesthetics, and placebo-controlled versus non-placebo-controlled designs. Analyses used Comprehensive Meta-Analysis software (version 2.0). In total, 13 RCTs met inclusion criteria: nine non-ECT trials (including some crossover designs) and four ECT trials.

Thirteen randomised trials were included: nine non-ECT studies (conducted 2000–2013) and four ECT studies (2012–2013). Across non-ECT trials the pooled sample comprised 192 patients with MDD and 34 with bipolar depression; some crossover studies contributed within-subject data. Ketamine hydrochloride was used in all trials (no S-ketamine in the included studies) at doses of 0.4, 0.5, 0.8 or 1 mg/kg, with 0.5 mg/kg the most common regimen. Depression severity was assessed with HDRS in five trials and MADRS in four trials. Several studies enrolled patients who were drug-free for at least 2 weeks, while others enrolled medicated patients and some specifically included pharmacologically treatment-resistant cases. Primary efficacy in non-ECT trials: pooled analysis showed a large and statistically significant reduction in depressive symptoms for ketamine versus control (SMD = -0.99; 95% CI -1.23 to -0.75; p<0.001), with low heterogeneity (I2 = 2.9%). Sensitivity analysis excluding two studies at high risk of bias yielded a similar result (SMD = -1.10; 95% CI -1.39 to -0.81; p<0.01). Tests for small-study/publication bias showed a reasonably symmetrical funnel plot, Egger’s intercept p = 0.76, and a Rosenthal fail-safe N of 138. Subgroup findings: ketamine was effective in MDD (127 cases versus 112 controls; SMD = -0.91; 95% CI -1.19 to -0.64; p<0.01; I2 = 4.4%) and in bipolar depression (27 cases versus 27 controls; SMD = -1.34; 95% CI -1.94 to -0.75; p<0.01; I2 = 24.1%), though bipolar data derived from a small number of participants and two studies from the same centre. Ketamine showed benefit in both drug-free studies (SMD = -1.15; 95% CI -1.52 to -0.79; I2 = 0.0%) and studies where participants remained on antidepressant or mood-stabilising treatment (SMD = -0.87; 95% CI -1.19 to -0.55; I2 = 27.1%). Trials that enrolled pharmacological non-responders also demonstrated efficacy. ECT trials: four studies totalling 118 patients (58 cases, 60 controls) were pooled. Ketamine used in ECT anaesthesia (various regimens including combinations with thiopental or propofol) was associated with a modest but statistically significant improvement in depression scores compared with other anaesthetics (SMD = -0.56; 95% CI -1.10 to -0.02; p = 0.04), but heterogeneity was moderate to high (I2 = 52.4%) and study designs and timing of outcome assessment varied. Time course and suicidal ideation: in non-ECT trials symptom reduction was rapid, with differences detectable as early as 40 minutes post-infusion and effects persisting in reported studies for about 2–3 days; only two trials provided such duration data. Suicidal ideation was evaluated in five RCTs, typically via the suicide item on depression scales. Three trials reported significant reductions in suicidal thoughts after ketamine versus control; one trial reported no effect on suicidal ideation despite improvement in other depressive symptoms. One study reported reduced suicidal ideation from 40 minutes up to day 10. The authors note that assessments were heterogeneous and were not adjusted for overall depression change. Safety and adverse events: common acute side-effects included transient dissociation, dry mouth, dizziness, blurred vision, headache, nausea, tachycardia and blood-pressure elevation. More pronounced cardiovascular events (tachycardia or hypertension requiring treatment) were reported with higher doses (0.8–1 mg/kg), and some ECT studies reported severe hypertension with 0.8 mg/kg. Eight of 47 ketamine recipients (17%) in one comparison experienced notable dissociative symptoms that resolved within 2 hours. Two infusions were halted for haemodynamic changes in one trial (one hypertensive episode, one transient hypotension with bradycardia). Manic or hypomanic switches were uncommon but were observed in isolated cases in bipolar samples. Overall tolerability was described as acceptable within the trial settings, but the authors caution about cardiovascular risks.

Fond and colleagues interpret their meta-analysis as supporting ketamine’s rapid antidepressant efficacy across a range of depressive disorders. They highlight consistently large pooled effect sizes in non-ECT randomised trials and a smaller but statistically significant benefit when ketamine is used for ECT anaesthesia, while noting heterogeneity among the ECT studies. The investigators consider the evidence positive for MDD and suggestive for bipolar depression, although the bipolar dataset was small and derived mainly from two low-risk-of-bias studies from a single centre. The authors emphasise the rapid onset of effect—observable within 40 minutes—and short-term duration of benefit reported (about 2–3 days in the limited trials that measured this), underscoring the need for research into maintenance strategies such as repeated dosing. They describe preliminary and promising findings for reduction in suicidal ideation, but caution that suicidal outcomes were measured heterogeneously (usually via single suicide-items on depression scales) and were not adjusted for overall changes in depressive symptoms, so a specific anti-suicidal effect remains unproven. Key limitations acknowledged by the study team include the small number of RCTs (13) and the limited diversity of ketamine doses studied (0.4–1 mg/kg), which precluded dose–response analysis. The participant populations were generally middle-aged with longstanding mood disorders; most trials excluded people with substance dependence and those with psychotic histories, limiting generalisability. The bipolar evidence base was small and the ECT literature was heterogeneous in anaesthetic regimens and timing of outcome assessment. Safety concerns were highlighted, particularly transient dissociative effects and the potential for clinically significant cardiovascular events at higher doses; the authors therefore advise caution when considering ketamine for patients with cardiovascular disease. Finally, they note uncertainty about whether an anaesthesiologist is required for ketamine administration in psychiatric settings. The investigators recommend further randomised trials focusing on bipolar depression, systematic study of repeated or maintenance dosing schedules to assess longer-term efficacy and safety, exploration of alternative administration routes (for example intramuscular), and more rigorous assessment of suicidal outcomes and cardiovascular risk management.

The meta-analysis concluded that ketamine administration produces significant short-term reductions in depressive symptoms in both non-ECT and ECT settings, though ECT-study designs were heterogeneous. Promising preliminary evidence was observed for reductions in suicidal ideation, but it remains unclear whether this effect is independent of overall antidepressant response. The authors advise cautious use of ketamine in patients with cardiovascular disease and state that middle- and long-term efficacy and safety remain unknown, warranting additional trials in bipolar depression, ECT settings, and studies of repeated/maintenance dosing.