Ketamine for treatment of mood disorders and suicidality: A narrative review of recent progress

Abstract

BACKGROUND:

Mood disorders are a leading cause of morbidity. Many patients experience treatment-resistant depression (TRD), and suicide rates are rising. Faster-acting and more effective antidepressant medications are needed. Four decades of research has transformed the use of ketamine from an anesthetic to an outpatient treatment for major depressive disorder (MDD). Ketamine is a N-methyl-d-aspartate (NMDA) receptor antagonist and has been shown to rapidly improve mood symptoms and suicidal ideation by targeting the glutamate system directly.

METHODS:

We used the PubMed database to identify relevant articles published until September 1, 2020. We focused on meta-analyses, randomized controlled trials, and original observational studies. We included relevant studies for depression, MDD, TRD, bipolar disorder, anxiety, posttraumatic stress disorder (PTSD), suicide, ketamine, and esketamine.

RESULTS:

Both racemic ketamine and esketamine have been shown to rapidly treat depression and suicidality. There is evidence that ketamine can be helpful for anxiety and PTSD; however, more research is needed. Intranasal esketamine has been FDA approved to treat depression.

CONCLUSIONS:

This narrative review describes the evolution of ketamine to treat mood disorders and suicidality. We provide the evidence supporting recent developments using esketamine as well as unresolved issues in the field, such as dosing and safety.

INTRODUCTION

Major depressive disorder (MDD) is a leading contributor to the global burden of disease with the rate of suicide rapidly increasing.^1,2 Approximately 60% of patients do not respond adequately to antidepressant medications, and up to 30% of patients fail ≥2 antidepressant trials, a condition termed treatment-resistant depression (TRD).³ For more than 60 years, medications approved for the treatment of depression targeted the monoamine system. While demonstrating efficacy in clinical trials, the practical utility of monoamine antidepressants does not meet patient needs in alleviating their depressive symptoms. A combination of adverse effects, limited efficacy, and delayed onset of action have led to the search for more effective and faster-acting antidepressant approaches. Additionally, finding effective antidepressant treatment for bipolar depression that has little risk of precipitating affective switch to mania is a well-known challenge.⁴ The discovery that ketamine exerts robust and rapid antidepressant effects resulted in a paradigm shift for treating patients with MDD. The data for use in bipolar depression is also growing. Through research examining ketamine over the past 20 years, investigators now have a better understanding of the role of the glutamatergic system in the pathophysiology of mood disorders. This narrative review explores pivotal clinical studies and meta-analyses for efficacy of ketamine in the treatment of mood disorders and suicidal ideation, and discusses recent developments and unresolved issues in the field.

Ketamine for depressive disorders

Ketamine, a derivative of the psychoactive compound phencyclidine (PCP) and an N-methyl-d-aspartate (NMDA)-receptor antagonist, has been used as an anesthetic for more than 50 years.^4,5 Over the past 20 years it has been thoroughly investigated in preclinical and clinical studies for the rapid treatment of depression.⁶ The initial clinical study by Berman et al⁷ (2000) provided proof-of-concept data that the glutamatergic system could be clinically targeted with racemic ketamine—henceforth referred to as “ketamine”—to treat depressive symptoms. In that study, participants with depression displayed significant improvement in their depressive symptoms within 72 hours following IV ketamine administration compared with placebo infusion. This small study was replicated in a larger sample in 2006 by Zarate et al.⁸ This was a landmark trial that ushered in an era of ketamine research for psychiatric indications. Zarate et al⁸ investigated the efficacy of IV ketamine in participants with TRD (defined in this study as a failure of ≥2 antidepressant agents). Participants received a single 40-minute infusion of ketamine with a randomized crossover design to also receive a saline infusion control. A very rapid and robust antidepressant effect of IV ketamine was observed with a similar time course to the initial study by Berman et al⁷ (ie, >50% of patients responded to a single ketamine infusion at 24 hours).

In light of concerns regarding the integrity of trial blinding because ketamine has obvious adverse effects such as dissociation, nausea, and sedation, in 2013 Murrough et al⁹ conducted a 2-site, double-blind, collaborative study evaluating ketamine’s rapid antidepressant response compared with midazolam, an active control medication. Participants with TRD experiencing a major depressive episode were randomly assigned to receive a single 40-minute IV infusion of ketamine or midazolam. Participants with TRD, who in this study averaged 5 antidepressant failures, had very similar response rates to previous ketamine, both at 24 and 72 hours. Approximately 67% of participants with TRD responded to ketamine, while the midazolam group had a 28% response rate at 24 hours (>2-fold response of ketamine vs active control midazolam). However, by 7 days, the antidepressant effects of ketamine were no longer statistically significant compared with midazolam. A meta-analysis showed racemic ketamine to be superior to other NMDA antagonists.¹⁰

In 2015, Newport et al¹¹ conducted a meta-analysis of studies of participants with TRD who underwent a single infusion of ketamine. Seven trials encompassing 147 participants treated with ketamine produced a rapid and transient antidepressant effect with an odds ratio approaching 10 for treatment response and 14.47 for remission of symptoms at 24 hours. Since that report, other meta-analyses have been conducted. One such analysis included >200 participants receiving a single ketamine infusion.¹² This study also indicated that a single infusion of ketamine was effective at 24 hours and that those effects dissipated by 7 days. Most of these studies used an IV dose of 0.5 mg/kg infused over approximately 40 minutes. This dose was based on historical studies derived mainly from psychotomimetic models of schizophrenia. This prompted the National Institute of Mental Health (NIMH) to sponsor a 6-site, single infusion, dose evaluation study known as the NIMH RAPID Trial.¹³ Participants with TRD were randomized to 1 of several ketamine doses, including a very low dose of 0.1 mg/kg, 0.2 mg/kg, the standard 0.5 mg/kg, and a high dose, 1 mg/kg, using midazolam as the control. At the primary endpoint, the 0.5 mg/kg and 1 mg/kg doses were both effective in reducing depressive symptoms compared with midazolam. The intermediate dose of 0.2 mg/kg did not separate from midazolam, and surprisingly 0.1 mg/kg also produced an antidepressant response, making the findings somewhat difficult to interpret. Of note, this was one of the first studies that had a negative result of any kind with ketamine in the treatment of depressive disorders. As suspected, the higher doses had higher levels of dissociation and 0.1 mg/kg had minimal levels of dissociation.¹³ Although this was the largest randomized control trial (RCT) of ketamine published to date (N = 99), each dose arm included an average of approximately 20 participants, with some demographic differences across the patient groups.

Although the response rate to ketamine in TRD is high and substantiated in the research community, not all patients respond, and there has been increasing interest in whether dissociation, route of administration, or biological factors such as brain-derived neurotrophic factor (BDNF) correlate with treatment response. Zarate et al⁸ had shown a correlation between the degree of dissociation and short-term response to ketamine, although the data on this is mixed.¹⁴

In addition to dosing and dissociation, the route of administration was a question for the efficacy of ketamine. A dose-finding study in Australian patients with late-life TRD found that subcutaneous injection of ketamine was safe and efficacious, and that higher doses had better overall response rates.¹⁵ That study also found that doses ≥0.2 mg/kg were significantly more effective than midazolam, and repeated treatments within 1 week resulted in a higher likelihood of remission and a longer time to relapse. An Asian study by Su et al¹⁶ showed that 2 doses of ketamine exhibited a dose-related response relationship to IV ketamine in the treatment of depressive symptoms.¹⁶ This study was of particular interest because most of the participants had the lower activity BDNF genotypes (Val/Met variants). Given that a dose-response relationship to ketamine existed, with no clear evidence of reduced efficacy in this population, it brought into question the impact of preclinical findings that BDNF was necessary for an antidepressant response. Zarate et al⁸ also showed that the use of a single dose, 0.5 mg/kg, in addition to lithium or valproic acid was effective in reducing anhedonia in patients with bipolar depression, paving the way for expanding indications beyond unipolar depression.¹⁷ Some preclinical work with lithium also suggested that it may enhance the efficacy of ketamine¹⁸; however, a clinical study by Costi et al¹⁹ failed to show sustained improvement with lithium after remission with ketamine treatment.

Interestingly, although anxious and nonanxious patients with bipolar depression had significant antidepressant responses to ketamine, the group of patients who were anxious and depressed had less of a response compared with the group of patients who were not anxious. Given that anxiety has been shown to be a predictor of poor treatment response to other treatments in bipolar depression,²⁰ this finding was not surprising and suggests a need for further investigations into novel treatment approaches of predominantly anxious bipolar depression.²¹

Most of the ketamine studies to date have used a single dose in research settings, although off-label use in the community has begun and many centers are advertising multi-dose protocols. Popular protocols are twice-weekly treatments for 2 to 4 weeks or 3 times weekly for 3 weeks, stemming from early studies such as Murrough et al.²² This was an open-label trial with 24 participants with TRD who received 6 infusions over 2 weeks, a treatment schedule analogous to electroconvulsive therapy (ECT) in the United States.²² They observed that most patients responded to ketamine, and the individuals who initially responded well to ketamine generally maintained their treatment response for 18 days until relapse. Since then, there have been several multi-infusion trials, including an open-label trial at the Mayo Clinic²³ and a notable study using intranasal (IN) ketamine in Australia,²⁴ among others with varying inclusions of mood and anxiety disorder populations.^25,26 Community-based clinicians have now begun to administer ketamine in many forms with a widely accepted regimen of 2 to 3 times per week.^27,28

An open-label study examining the effect of repeated ketamine infusions for the treatment of patients with TRD and comorbid posttraumatic stress disorder (PTSD) in a sample of veterans had a remission rate of 80% for PTSD, and a response rate of 93.3% for TRD. PTSD responders had a median time to relapse of 41 days. MDD responders had a median time to relapse of 20 days. These findings stimulated the exploration of ketamine use for nondepressive disorders by other groups.^29–32 Interestingly, although repeated ketamine infusions were associated with transient increases in dissociative symptoms, no participant reported worsening of PTSD symptoms over the study duration. Additionally, a recent study from the Minneapolis VA documented that a progressive accrual of ketamine response occurred with each infusion over time, although the veteran population tends to have a different pattern of response to psychopharmacology treatments.³³ This study evaluated the efficacy of increased doses of IV ketamine per week in sustaining antidepressant effects in patients with TRD. In a multicenter, double-blind study, adults (age 18 to 64 years) with TRD were randomized to receive IV ketamine (0.5 mg/kg) or IV saline, administered over 40 minutes, 2 or 3 times weekly for up to 4 weeks. This was the first controlled study looking at repeated doses of ketamine versus repeated doses of a saline control. It found that response rates were 69% in the 2-times-per-week infusion schedule and 54% in the 3-times-per-week dosing schedule, suggesting that 2 times per week may be sufficient for >50% response, which would likely be cost-effective, and 3 times per week could be used for nonresponders.³⁴ Dissociative symptoms occurred transiently and seemed to attenuate with repeated dosing, which is intriguing and also helped to address whether there is a need for dissociation for treatment response.

Importantly, a recent study by Phillips et al³⁵ provided evidence that maintenance ketamine can create lasting antidepressant effects. They performed a double-blind crossover study and found that participants with TRD who responded to IV ketamine 3 times per week for 6 doses and were then given a weekly dose for 4 additional weeks had ongoing efficacy. It is notable that approximately one-third of the participants were nonresponders in the first and second phases, which may illustrate the severity of the TRD in this study.

Recently, McIntyre et al³⁶ conducted a large retrospective analysis of patients with MDD and bipolar depression who received IV ketamine at their Canadian Rapid Treatment Center of Excellence and found significant improvement in mood, suicidality, and overall psychological functioning with 4 ketamine infusions.³⁶ Interestingly, they also found that patients who were depressed with anxious features, irritability, and mixed states also improved from baseline.³⁷

Ketamine for suicidality

Suicide is an increasing public health crisis. Wilkinson et al³⁸ conducted a systematic review and individual participant data meta-analysis examining the effects of a single dose of ketamine on suicidal ideation (SI) in studies of participants with TRD, although 1 included study was for the treatment of PTSD. Overall, the effect of ketamine was positive in reducing SI compared with saline or midazolam, suggesting a rapid anti-suicidal effect lasting up to 7 days.

Grunebaum et al³⁹ examined a group of participants with MDD and clinically significant SI as indicated by scores on the Scale for Suicidal Ideation (SSI). They found that there was a significant decrease in SI 24 hours after ketamine infusion.³⁹ Participants were then given traditional antidepressant treatments and followed for 6 weeks. As seen previously, if a participant responded initially to the ketamine treatment, they would remain without SI for the study duration. Another RCT examined self-report data from patients endorsing SI within a cohort of participants with varied primary diagnoses receiving ketamine or midazolam.⁴⁰ Scores on the Beck Scale for Suicidal Ideation were not different between treatment groups at 24 hours; however, a significant reduction was seen at 48 hours in patients treated with ketamine compared with the midazolam control treatment group (P = .047). Also, the Montgomery-Åsberg Depression Rating Scale Suicidality Item (MADRS-SI) score was not significantly different in the ketamine group compared with the midazolam treatment group at 24 hours (P = .05), and the ketamine treatment effect was not significant at the end of the 7-day assessment. Another study in patients with bipolar depression indicated a nonsignificant difference of approximately 6 points on the SSI after ketamine infusion as compared with midazolam.⁴¹

Intranasal esketamine for depression and suicidality

To date, most clinical studies of ketamine have been conducted using IV administration of the racemic compound. Janssen Pharmaceuticals, Inc., developed an intranasal formulation of the pure S-isomer of ketamine, referred to as esketamine, which received US FDA approval for TRD in March 2019 and for patients at imminent risk for suicide and endorsing acute SI and/or self-injurious behaviors in August 2020. This medication can only be administered in certified health care settings via a Risk Evaluation and Mitigation Strategy (REMS) program. The Phase II trial for the TRD indication in patients with TRD experiencing a major depressive episode was recently published.⁴² It was a dose-response study that investigated 3 fixed doses of intranasal esketamine (28 mg, 56 mg, and 84 mg) and placebo. The highest esketamine dose, 84 mg, appeared clinically equivalent to a ketamine dose of 0.3 mg/kg, likely because of a combination of metabolism and potency.⁴³ Similar to some of the racemic studies with IV ketamine, there was a dose-response relationship with esketamine. Although there have been fewer studies of esketamine, the clinical trials are an order of magnitude larger than that of the ketamine trials. Meta-analyses of ketamine studies generate similar amounts of participants to esketamine trials and can provide a means of comparison.^35,41,42,44

Although no direct comparison clinical trials have been conducted, results generally support the notion that esketamine and racemic ketamine (S and R-ketamine) have similar efficacy, based on the higher affinity of esketamine for the NMDA receptor. Esketamine is approximately 3 times more potent than arketamine (R-ketamine) for NMDA receptors.⁵ For example, 0.20 mg/kg and 0.40 mg/kg esketamine correspond to a racemic ketamine dose equivalent of approximately 0.31 mg/kg and 0.62 mg/kg, respectively.⁵ The explanation for a <2:1 dose difference is likely due to the observation that metabolism of esketamine tends to be higher in the absence of arketamine.

Canuso et al⁴⁵ (Janssen Pharmaceuticals, Inc.) published a report in the American Journal of Psychiatry examining patients in the emergency room setting admitted for inpatient psychiatry treatment with imminent risk of SIB. This study found that 4 hours after receiving esketamine, 21% of patients had a resolution of SI compared with 10% on placebo.⁴⁵ One day later, approximately 40% of patients had resolution of SI compared with 6% on placebo.⁴⁵ The patients were subsequently followed as outpatients and given 2 additional intranasal esketamine applications per week for 4 weeks, although there was no statistical separation occurring between esketamine and controls at Day 25.⁴⁵ This was the first proof-of-concept study that indicated esketamine as an approach that could mitigate SI for high-risk patients. In another study, ketamine also rapidly reduced SI (within 1 day) on both the clinician-administered and self-report outcome measures.³⁸

Popova et al⁴⁶ recently published results from the SUSTAIN-1 trial. In this Phase III study, they investigated relapse rates among responders to esketamine. Participants who had responded acutely to esketamine received either esketamine or placebo nasal spray maintenance. Researchers observed over twice the rate of relapse in participants who received the placebo control compared to esketamine. Recently, Papakostas et al⁴⁷ conducted a meta-analysis of double-blind RCTs to determine the effectiveness of esketamine augmentation for MDD. After their analysis of data from 774 patients, they concluded that adjunctive esketamine for patients with MDD who are either treatment-resistant or acutely suicidal is an effective treatment strategy.

Safety considerations

Ketamine use has a known risk of dissociation and other psychological adverse effects, although they tend to be transient and resolve by 90 minutes post-infusion. There are also known concerns regarding hemodynamic responses (blood pressure values >180/100 mmHg, or heart rate >110 bpm in 30% of patients with MDD), and respiratory suppression. “Ketamine bladder” has been described in people who have abused ketamine and has been observed extensively in eastern Asia, where ketamine abuse is relatively prevalent.^48–51 Current concerns of ketamine bladder are hemorrhagic cystitis and painful urination. Also, nausea is a fairly frequent occurrence and has been observed in 30% to 40% of patients; therefore, some clinicians have decided to pretreat patients with ondansetron.⁵²

Rodrigues et al⁵³ recently reaffirmed much of these findings, demonstrating tolerability with <5% of participants withdrawing from treatment due to adverse events. They found that treatment-emergent hypertension is a frequent effect of ketamine infusions (approximately 44%) with blood pressure increases ≥165/100 mmHg, and they needed to intervene pharmacologically in 12% of these patients.⁵³ In their study, the most frequently reported adverse events included drowsiness (56.4%), dizziness (45.2%), dissociation (35.6%), and nausea (13.3%)—consistent with previous lines of investigation.⁵³ They also noted that dissociation severity was reduced and plateaued over the course of treatment compared to the first infusion.⁵³

Lastly, antidepressant treatment has a risk of affective switch to mania and therefore care must be taken to monitor this rare and serious complication.^54,55 There have been published case reports that implicate ketamine in treatment-emergent manic events when used to treat pain as well as when used as an antidepressant.^56,57 Although a handful of controlled studies in bipolar depression have not shown affective switch to mania, those studies generally allow for the concurrent use of mood-stabilizing medications, therefore leaving room for additional study during phase IV post-marketing surveillance.

Recommendations for treatment

A survey of outpatient ketamine clinics indicated that the most common off-label indications for ketamine use were MDD (72%), bipolar disorder (15%), and PTSD (6%).²⁸ Despite the relatively small sample sizes of individual studies, lack of long-term efficacy data, and limited data on safety from individual studies, meta-analyses have provided sufficient evidence and excitement for increased use of ketamine as an off-label treatment for TRD and other psychiatric disorders.

There is little guidance for off-label ketamine use in the treatment of psychiatric illness. The treatment data are mainly derived from clinical trials using structured research settings and limited dosing in research participants. In 2017, an American Psychiatric Association (APA) Task Force released a consensus statement and recommendations for clinical practice. The consensus was that there is not enough information and data to issue formal treatment guidelines and that there should be a list of best practices for patient selection. The APA recommended avoiding ketamine use for patients who are actively psychotic or abusing substances, although there are some intriguing reports of ketamine efficacy in psychotic depression.^58,59 Appropriate informed consent is also important given the adverse effects, likely need for repeated treatments, and alternative validated treatments. Prior to receiving off-label ketamine treatments, patients should have tried or had a discussion of current standard FDAapproved antidepressants, as well as ECT, transcranial magnetic stimulation, lithium, and second-generation antipsychotics to augment antidepressant response, especially for individuals with TRD.

Although there have been many advancements in the science of ketamine and its clinical applications, many practical questions remain. The APA Task Force consensus statement discussed who are “the right practitioners” to offer ketamine treatment to patients with TRD.⁶⁰ Some ketamine clinics are managed by nurse anesthetists or emergency room physicians who have limited psychiatric training. Considering access to care, should ketamine be administered solely by psychiatrists? The Task Force deferred taking a stance on this issue, other than saying that the clinician should be trained in Advanced Cardiovascular Life Support and be capable of handling medical complications. What is the best treatment delivery modality? The data are still emerging; IV, subcutaneous, intranasal, and oral each have their benefits and drawbacks. We will need to determine the dose, route, and rate of administration that optimizes positive outcomes while minimizing adverse events. Another consideration is patient monitoring after treatment. Most clinicians advise patients not to drive home afterwards, though this may decrease access to care. These are important considerations because protocols and approaches in the ketamine field are changing, often creating more questions than answers. The APA will need to issue updated consensus statements that will articulate recent clinical trial data, risks, monitoring, and the informed consent process.

Future directions and challenges

The impact of ketamine use on cognition is an area of recent interest. Long-term cognitive impairment has been reported with high and frequent doses of ketamine (>4 times per week), typically seen when abused.⁶¹ Frequent use has been associated with impairments in working memory, episodic memory, and aspects of executive function as well as reduced psychological well-being.⁶¹ Murrough et al⁶² reported that adverse effects on processing speed were observed after a single ketamine infusion. In contrast, a study that thoroughly examined the cognitive effects of repeated low-dose ketamine found no significant concerns within 4 weeks.³³ Long-term studies examining cognitive outcomes are ongoing. There are valid historical concerns regarding the abuse potential of ketamine,⁶³ and many investigators and clinicians are aware.⁶⁰ There are ongoing studies evaluating the abuse potential of ketamine when administered at low dosages and in structured clinical settings. That said, if frequent self-administered intranasal esketamine becomes a reality, then certainly further study and monitoring for abuse and cognitive adverse effects are needed. While ketamine may be beneficial to some patients with mood disorders and SI, it is important to consider the limitations of the available data and the potential risks associated when considering ketamine as a widespread treatment option.⁶⁴

The TABLE^{7–9,13,15,16,19,22–26,39–42,46,64–68} summarizes pivotal trials for ketamine and esketamine use in mood disorders and suicidality.

TABLE.

Summary of pivotal trials for ketamine and esketamine use in mood disorders and suicidality

Study	Study design	Dose	Control	Criteria	Route	Participants	Duration	Mono vs adjunctive	Outcome
Berman et al7 (2000)	RCT	0.5 mg/kg over 40 minutes	Saline	MDD (TRD not specified)	IV	7	Single infusion	Adjunctive	Significant decrease in HAM-D scores 72 hours later
Zarate et al8 (2006)	RCT, crossover	0.5 mg/kg over 40 minutes	Saline	TRD; failed ≥2 pharmacotherapy trials	IV	18	Single infusion; 1 week with each condition	Monotherapy after wash-out	Decrease in HAM-D scores 24 hours later
Murrough et al9 (2013)	RCT 2 sites	0.5 mg/kg over 40 minutes	Midazolam	TRD; failed ≥3 pharmacotherapy trials	IV	73	Single infusion	Adjunctive	Decrease in MADRS scores 24 and 72 hours later
Murrough et al22 (2013)	Open-label	0.5 mg/kg over 40 minutes	None	TRD; failed ≥2 pharmacotherapy trials	IV	24	6 infusions over 12 days	Monotherapy after wash-out	70.8% (17/24) response at 2 hours to 13 days; 75% relapse risk by 83 days
Ghasemi et al26 (2014)	RCT	0.5 mg/kg over 45 minutes 3x, every 48 hours	ECT	TRD; failed ≥2 pharmacotherapy trials	IV	18	3 infusions or ECT treatments each 48 hours apart	Adjunctive	Greater decrease in BDI and HAM-D scores compared with ECT 24 hours after treatment, 72 hours, and 1 week after last treatment
Diamond et al65 (2014)	Open-label	0.5 mg/kg over 40 minutes; 3 times/3 weeks or 6 times/3 weeks	ECT	Unipolar and BD TRD; failed ≥2 pharmacotherapy trials	IV	28	3 weekly infusions or 2 times/week for 3 weeks	Adjunctive	29% reached response criteria at day 21 (4 to 7 days after last treatment); only 2/22 had memory impairment
Lapidus et al66 (2014)	RCT, crossover	50 mg	Saline	MDD	IN	18	Single treatment	Adjunctive	44% responded at 24 hours
Murrough et al40 (2015)	RCT	0.5 mg/kg over 40 minutes	Midazolam	SI	IV	24	Single infusion	Adjunctive	Decrease on MADRS-SI at 24 hours BSI effect emerges at 48 hours
Vande Voort et al23 (2016)	Open-label, continuation study	0.5 mg/kg over 100 minutes; up to 6 infusions 3 times weekly then weekly for 4 weeks	None	TRD with SI	IV	12	2 weeks acute phase, 4 weeks continuation phase	Adjunctive	7/12 responded, 5/12 remitted in acute phase; 4/5 lost remission post continuation; 1 completed suicide later
Singh et al67 (2016)	RCT	Esketamine 0.2 or 0.4 mg/kg	Saline	TRD; failed ≥2 pharmacotherapy trials	IV	30	2 infusions	Adjunctive	Significant reduction in MADRAS with both doses compared to placebo at 24 hours
George et al15 (2017)	RCT; Australia	0.1, 0.2, 0.3, 0.4, and 0.5 mg/kg 1 week apart	Midazolam	Late-life TRD MDD or BD	SQ	15	Single injection	Adjunctive	7/14 remitted; ≥0.2 mg/kg ketamine better than midazolam
Su et al16 (2017)	RCT; Taiwan	0.2 or 0.5 mg/kg over 40 minutes	Saline	TRD; failed ≥2 pharmacotherapy trials	IV	71	Single infusion	Adjunctive	Dose-related HAM-D improvement with ketamine
Grunebaum et al41 (2017)	RCT	0.5 mg/kg over 40 minutes	Midazolam	SI in BD	IV	16	Single infusion; weekly follow-up for 6 weeks	Adjunctive	Significant reduction at weeks 1 and 5 compared with baseline
Gálvez et al24 (2018)	RCT	100 mg	Midazolam	TRD	IN	10 (5 completed)	8 treatments over 4 weeks: 3/week for 2 weeks then weekly for 2 weeks	Adjunctive	Safety and tolerability not achieved in 50% of patients
Albott et al25 (2018)	Open label	0.5 mg/kg over 40 minutes; 6/12 days	None	Comorbid PTSD/TRD	IV	15	6 infusions over 12 days	Adjunctive	80% PTSD remission for a median of 41 days, 93% TRD remission for a median of 20 days
Grunebaum et al39 (2018)	RCT	0.5 mg/kg over 40 minutes	Midazolam	SI in MDD	IV	80	Single infusion	Adjunctive	Greater SI decrease in ketamine group
Daly et al42 (2018)	RCT	Esketamine 28, 56, or 84 mg	Placebo	TRD; failed ≥2 pharmacotherapy trials	IN	67	Variable administration for 8 weeks	Adjunctive	Dose-related decrease in MADRS
Costi et al19 (2019)	RCT	0.5 mg/kg	Ketamine + lithium vs placebo	TRD; failed ≥2 pharmacotherapy trials	IV	34	Multiple infusions over 11 days	Adjunctive	No difference between lithium and placebo at 28 days
Popova et al46 (2019)	RCT, 39 outpatient sites	Esketamine 56 or 84 mg twice weekly	Placebo	TRD; failed ≥2 pharmacotherapy trials	IN	197	28 days	Adjunctive	MADRS with esketamine + antidepressant was significantly greater than with antidepressant + placebo at Day 28
Fedgchin et al68 (2019)	RCT	Esketamine 56 or 84 mg twice weekly for 4 weeks	Placebo	TRD; failed ≥2 pharmacotherapy trials	IN	346	28 days	Adjunctive	Esketamine (56 and 84 mg) + antidepressant each significantly better than placebo + antidepressant
Daly et al64 (2019)	RCT	56 or 84 mg every 1 to 2 weeks	Placebo (denatonium benzoate)	MDD after esketamine response	IN	455 initially, 297 maintenance	16 weeks	Adjunctive	Esketamine + antidepressant was superior to placebo + antidepressant at 76 weeks
Fava et al13 (2020)	RCT 6 sites	0.1, 0.2, 0.5, 1 mg/kg over 40 minutes	Midazolam	TRD; failed ≥2 pharmacotherapy trials	IV	99	Single infusion	Monotherapy following wash-out	0.5 mg/kg and 1 mg/kg effective

BD: bipolar disorder; BSI: Beck Scale for Suicidal Ideation; ECT: electroconvulsive therapy; HAM-D: Hamilton Depression Rating Scale; IN: intranasal; IV: intravenous; MADRS: Montgomery-Åsberg Depression Rating Scale; MADRS-SI: Montgomery-Åsberg Depression Rating Scale–Suicidality Item; MDD: major depressive disorder; PTSD: posttraumatic stress disorder; RCT: randomized control trial; SI: suicidal ideation; SQ: subcutaneous; TRD: treatment-resistant depression.

CONCLUSIONS

While IV ketamine and intranasal esketamine are effective for patients with TRD and/or SI, there are many unanswered questions. We remain uncertain about whether there are special populations of patients that would particularly benefit from ketamine treatment or should be excluded. We do not know the predictors of biologic and clinical response. For example, recent studies have shown that individuals with a higher body-mass index had a higher initial response to IV ketamine.⁶⁹ Additionally, somewhat counterintuitively, individuals with a higher childhood trauma burden also displayed a better response.⁷⁰ Findings such as these need broader validation, yet aid in the phenotyping of patients to help determine who may have better outcomes from treatment with ketamine and esketamine and triage earlier in their treatment course—perhaps saving years of morbidity.

We need to strike a balance between safety, efficacy, and the potential for abuse. We do not know if there are augmentation strategies that can prolong the effect of ketamine, though recent investigation has demonstrated that esketamine augmentation improves outcomes in patients receiving oral antidepressant medications.^48,67 Many of these concerns can be best addressed using realworld patients and by having data captured in registry formats. Ruppert McShane, MD, FRCPsych, from the United Kingdom and Colleen Loo, MBBS, FRANZCP, MD, from Australia have been developing national registries. There is a Canadian Rapid Treatment Center of Excellence led by Roger McIntyre, MD, FRCPC. Additionally, the Centers of Psychiatric Excellence (COPE) is a coalition of 6 outpatient psychiatric clinics in the United States that have also developed a database registry. It is important to underscore a cautious approach to interpreting clinical data and be aware of any foreseeable potentially dangerous adverse effects while we explore the use of rapid-acting antidepressants. Ultimately, the goal is to safely and appropriately disseminate novel treatments such as rapid-acting antidepressants to reduce the burden of depression and suicide worldwide.