Ketamine as the prototype glutamatergic antidepressant: pharmacodynamic actions, and a systematic review and meta-analysis of efficacy

Major depressive disorder (MDD) and bipolar affective disorder (BPAD) are common, often persistent illnesses for which current monoaminergic antidepressants frequently show limited efficacy, particularly in treatment-resistant depression (TRD). Interest has grown in non-monoaminergic agents, and ketamine — a dissociative anaesthetic and NMDA (N-methyl-D-aspartate) glutamate receptor antagonist — has attracted attention because it appears to produce unusually rapid antidepressant and antisuicidal effects. Understanding ketamine's actions might therefore provide new biological insight into depression and point toward a new class of glutamatergic antidepressants. Caddy and colleagues set out to review mechanistic data on how ketamine might exert antidepressant effects and to systematically collate and meta-analyse clinical trial evidence for its efficacy up to January 2013. The review aimed to summarise pharmacokinetic and pharmacodynamic actions, evaluate clinical trial results (including effects on mood and suicidal ideation), and assess efforts to prolong benefit through augmentation or combination with electroconvulsive therapy (ECT) or surgery anaesthesia.

Caddy and colleagues performed a systematic literature search up to January 2013 across electronic databases (PSYCINFO among them) and by manual reference searching. Search terms included a broad set of hallucinogen- and depression-related keywords (for example, "ketamine", "lsd", "psilocybin", combined with "depression"/"antidepressant"/"bipolar"). The team sought randomized and non-randomized trials of hallucinogens for depression but prioritised ketamine-related studies for detailed synthesis. Pre-specified inclusion criteria required adult participants (≥18 years) with MDD or BPAD diagnosed by structured interview (DSM or ICD). Studies had to assess mood symptoms and/or suicidality using validated rating scales; investigations of addiction or recreational use were excluded. Only journal articles were eligible and case reports with fewer than three participants were excluded. No specific comparator was required for inclusion. Data extraction captured participant characteristics (age, gender, illness duration), diagnostic and trial inclusion/exclusion criteria, details of interventions (route, dose, schedule), response/remission definitions, outcome measures (for example MADRS, HDRS), and follow-up. To aid interpretation the authors grouped studies into three thematic categories: ketamine only, ketamine plus a second drug, and ketamine in combination with ECT or surgery. For quantitative synthesis, Hamilton Depression Rating Scale (HDRS) data from five controlled studies were entered into two cross-study meta-analyses using OpenMeta[Analyst]. The authors assessed differences between ketamine and placebo at baseline, 60–80 minutes post-infusion, and 210–230 minutes post-infusion. Because means and standard deviations were not consistently reported, values were in some cases read from published graphs. The extracted text does not report a formal risk-of-bias assessment approach or detailed search dates beyond the stated 2013 cut-off.

Study identification and overall sample. The extracted text reports that 11,138 records were identified by the search and that, after selection, 26 studies were ultimately reported as fulfilling inclusion criteria with a total of 629 participants. Elsewhere the paper summaries refer to 22 RCTs and non-RCTs totalling 629 participants; the extraction is inconsistent on the exact study count but consistently reports a pooled participant total of 629. Ketamine-only studies. Eleven open-label (no control) studies administering ketamine to all participants were described, enrolling 206 participants with major depressive episodes. Most studies used intravenous ketamine at 0.5 mg/kg (one study used a 0.2 mg/kg bolus). The Montgomery–Åsberg Depression Rating Scale (MADRS) was the common primary outcome, with response defined as ≥50% score reduction and remission typically defined as MADRS <10. Open-label single-dose trials reported rapid and statistically significant mood improvement within hours (for example significant MADRS changes at ~230 minutes post-infusion in several studies). Three open-label multiple-infusion series reported higher sustained response rates through the dosing period: one study delivering six infusions over 12 days reported up to 100% response by study end and 88.9% remission, while another thrice-weekly 12-day protocol reported a 70.8% response rate among completers. Relapse after repeated dosing occurred on average around 19 days (SD 13) following final infusion in one series; in another study median time to relapse was 18 days (IQR 11–27). Effects on suicidal ideation were reported in several open-label and small controlled trials. Some studies found rapid reductions in suicidal thinking that in some cases occurred even in participants who did not meet antidepressant response criteria. Adverse events were generally transient: common effects included short-lived dissociative/psychotomimetic symptoms (measured with scales such as the BPRS or CADSS), transient hypertension and tachycardia, with no long-term adverse-event follow-up reported in the trials. Controlled studies and meta-analysis. Five studies compared IV ketamine (0.5 mg/kg) with saline placebo. As in the open-label work, antidepressant effects were rapid but typically short-lived. For example, one trial reported 79% remission at 24 hours with 35% maintaining remission at one week. Caddy and colleagues meta-analysed HDRS differences between ketamine and placebo: at 60–80 minutes the mean difference was −3.406 (p < 0.001; 95% CI −6.303 to −0.509), and at 210–230 minutes the mean difference was −5.371 (p < 0.001; 95% CI −6.574 to −4.168). Heterogeneity was high at 60–80 minutes (I2 = 79%); at 210–230 minutes the model also showed statistically significant benefit despite some baseline disparities between groups in one study (reported heterogeneity p = 0.531 in the extracted text). The authors note that blinding was likely compromised in many trials because of ketamine's distinctive acute effects. Ketamine plus a second drug. Three studies tested whether anticonvulsant or neuroprotective agents (lamotrigine or riluzole) given alongside ketamine could prolong benefit. Ibrahim et al. randomised 42 participants to riluzole versus placebo following an initial ketamine infusion and found no between-group difference in time to relapse; mean time to relapse across groups was 13.2 days. A Duncan substudy also reported no significant riluzole effect when started 4–6 hours after ketamine. Mathew et al. used a two-stage design (pre-treatment with lamotrigine versus placebo before ketamine, then randomized riluzole versus placebo in responders) and found lamotrigine did not limit side effects or augment efficacy, and riluzole did not prolong time to relapse (mean time to relapse 24.4 days riluzole vs 22.0 days placebo; log-rank p = 0.68). Ketamine in ECT or surgical anaesthesia. Several studies explored ketamine as an adjunct to or anaesthetic for ECT, and one trial examined ketamine as part of general anaesthesia for orthopaedic surgery in depressed patients. Results were mixed but suggested short-lived additional benefit. A randomized trial by Loo et al. (n = 51) found a small early clinical advantage for subanaesthetic ketamine augmentation of ECT in the first week that disappeared by treatment end; Abdallah et al. terminated a trial early for lack of between-group differences. Studies using ketamine as the anaesthetic agent for ECT reported earlier improvement in depressive scores compared with other anaesthetics up to several sessions, but not sustained across longer ECT courses. In a single-surgery cohort, depressed patients receiving ketamine as part of their anaesthetic (combined with propofol and fentanyl) had statistically significant improvement in HDRS scores one day after surgery and reduced postoperative pain compared with those who did not receive ketamine. Overall synthesis. Across the reviewed literature ketamine produced rapid antidepressant and antisuicidal effects in most reported studies. Several RCTs recorded very high early response rates (one cluster of RCTs reported 71–79% response at 24 hours). However, benefit was frequently transient, with relapse typically occurring within days to a few weeks. Methodological limitations were emphasised: small sample sizes, few fully double-blind randomized placebo-controlled trials (six studies reported as adopting that gold standard), and concerns about compromised blinding due to acute drug effects.

Caddy and colleagues interpret the assembled evidence as consistently showing that ketamine can induce rapid and clinically meaningful reductions in depressive symptoms and suicidal ideation, including in treatment-resistant samples. They highlight mechanistic lines of evidence that may underlie these effects: NMDA antagonism on GABA interneurons leading to increased glutamate release and relative AMPA receptor activation; downstream changes in intracellular signalling and expression of proteins implicated in synaptic plasticity (notably mTOR and BDNF); modulation of circadian gene expression; interactions with σ1 receptors; and potential links to adipokine signalling. Animal and cellular studies linking ketamine to rapid synaptogenesis and to BDNF/mTOR signalling are presented as biologically plausible mediators of the clinical effects. At the same time the authors emphasise important limitations and uncertainties. The antidepressant effect of ketamine is often transient, typically lasting days to weeks, and attempts to prolong benefit pharmacologically (for example with riluzole or lamotrigine) have been unsuccessful in the reviewed trials. Practical barriers to clinical implementation include the requirement for IV administration with anaesthetic support in most studies, the potential for dissociative/psychotomimetic effects that compromise blinding and raise safety concerns, and questions about abuse potential if easier formulations (for example oral) were used. They also note that the trials generally excluded people with psychosis or current substance dependence, limiting generalisability, and that long-term adverse effects were not followed up in the reviewed studies; separate MRI data from chronic illicit use suggest possible cortical atrophy, but this was not established in therapeutic settings. The authors position ketamine as a valuable prototype that has advanced understanding of glutamatergic mechanisms in depression and stimulated development of more selective NMDA receptor subunit antagonists (for example NR2B antagonists) that might capture therapeutic benefits without ketamine's liabilities. They conclude that larger, longer-term randomized controlled trials are required to clarify optimal dosing, durability of benefit, safety, and the clinical role ketamine (or next-generation glutamatergic agents) might play in treating depressive disorders. The extracted text emphasises that practical, ethical and longer-term efficacy questions remain unresolved and that ketamine's principal future role may be as a prototype for novel glutamatergic antidepressants rather than as a routine treatment in standard psychiatric practice.