Ketamine for the treatment of major depressive disorder and bipolar depression: A review of the literature

The investigators performed a literature search of EBSCO, PsycINFO and MEDLINE through December 2015 using the keywords "ketamine" and "depression" with no year limits. Inclusion criteria required studies to be published in English in peer-reviewed journals, to be randomised controlled trials of ketamine, to enrol subjects diagnosed with a major depressive episode according to DSM-III, DSM-IV or DSM-5 criteria, and to administer ketamine at either 0.5 mg/kg IV or 50 mg intranasally. Case reports and case series were excluded. Seven trials met the inclusion criteria: five in MDD and two in bipolar depression. Study selection was performed by two authors (S.G., K.K.) with final selection carried out by S.G. and verified by T.M. The review therefore concentrates on randomised designs, including crossover trials and parallel-group trials with either inert placebo or an active control (midazolam) and captures key outcome measures used across the trials, chiefly changes in clinician-rated depression scales (Hamilton Depression Rating Scale, Montgomery–Åsberg Depression Rating Scale (MADRS)). Where reported, the review notes trial-level design features such as infusion parameters (0.5 mg/kg over 40 minutes for IV trials), timing of rating assessments, and whether concomitant psychotropic medications were allowed.

Across the seven included randomised trials, ketamine produced rapid antidepressant effects in treatment-resistant unipolar and bipolar depression, though sample sizes were generally small and durability of effect was limited. Ketamine versus inert placebo (crossover trials): An 18-subject randomised, double-blind crossover study in treatment-resistant unipolar depression administered 0.5 mg/kg IV over 40 minutes and assessed Hamilton Depression Rating Scale scores up to 7 days. Using a linear mixed model on the 14 subjects who completed both phases, the trial observed significant drug, time and drug-by-time effects (drug F = 58.24, P < .001), with superiority of ketamine emerging by 110 minutes and persisting through day 7. Reported adverse events more frequent with ketamine included perceptual disturbances, confusion, transient blood pressure increases and dizziness; most resolved within about 80 minutes. A single-centre crossover trial in 18 inpatients with bipolar I or II depression found a similar rapid antidepressant effect; this study allowed maintenance lithium or valproate in some participants and achieved near-target power for day 1 outcomes. A replication attempt that enrolled fewer participants than planned reported a significant drug-by-time interaction in intent-to-treat analyses, but extracted text for that trial is truncated and does not fully report all outcome details. Intranasal ketamine: An intranasal trial randomised 18 subjects in a crossover design to 50 mg ketamine versus saline using a mucosal atomisation device. MADRS scores at 24 hours were significantly improved after ketamine compared with placebo (t = 4.39; P < .001), with a mean difference of 7.6 points (95% CI 3.9–11.3). Repeated-measures analyses showed greater improvement with ketamine over the 7-day follow-up (F = 28.10; P < .001). Adverse effects such as feeling strange, memory complaints and fatigue were transient and resolved within about 4 hours. The extracted text notes that some intranasal trial participants were taking concurrent antidepressant or psychotropic medications, which may complicate attribution of effects. Ketamine versus active control (midazolam): A two-centre, double-blind trial by Murrough and colleagues randomised 73 subjects (48 to ketamine 0.5 mg/kg IV, 25 to midazolam 0.045 mg/kg IV; both infused over 40 minutes). The primary outcome was MADRS score at 24 hours. At 24 hours the ketamine group had a mean MADRS of 14.77 (95% CI 11.3–17.80) versus 22.72 (95% CI 18.85–26.59) for midazolam, yielding a significant difference (P < .001). Adjusted analyses also favoured ketamine (95% CI for between-group difference 3.20–12.7). Effects were most pronounced early but began to wane by day 3 and were no longer significantly different by day 7. Adverse events more common with ketamine included dizziness, blurred vision, nausea, poor coordination and restlessness; 17% of ketamine recipients reported significant transient dissociative symptoms that resolved within a few hours. Cognitive effects: A related Murrough trial assessing neurocognitive performance randomised 62 participants to ketamine or control; the extracted text does not clearly report the ketamine dose in this fragment. Analyses showed significant improvements with ketamine versus midazolam in processing speed (F = 6.58; P = .013), verbal learning (F = 6.80; P = .012) and visual learning (F = 6.48; P = .014), with no changes in working memory or reasoning. Baseline slower processing speed was associated with later antidepressant response. The study did not clearly report responder counts in the extracted text and apparently did not obtain MADRS scores after infusion, according to the authors' account of limitations. Across trials, the rapid onset of antidepressant effect (within hours, often peaking by 24 hours) and short-lived duration (typically days to about one week after a single infusion) were consistent findings. Adverse events were generally transient but included dissociative and perceptual effects as well as haemodynamic changes. The extracted text does not provide pooled effect sizes beyond the single-study numerical results described above.

Grady and colleagues conclude that ketamine shows promise for rapid symptom reduction in treatment-resistant unipolar and bipolar depression, but emphasise important caveats. The discussion reiterates that a single subanaesthetic infusion of ketamine (0.5 mg/kg over 40 minutes) can produce antidepressant effects within 2–4 hours, often maximal at 24 hours and commonly persisting up to 7 days after a single dose. The authors cite a meta-analysis by Romeo and colleagues (as reported in the extracted text) that found statistically significant antidepressant effects versus placebo from day 1 through day 7 in treatment-resistant depression; subgroup analyses reported sustained efficacy through day 7 for unipolar depression but only through day 4 for bipolar depression. Safety and implementation risks are highlighted: ketamine carries its own adverse-effect profile (psychotic-like and dissociative symptoms, transient blood pressure and heart-rate increases, blurred vision, drowsiness) and has abuse potential. The authors also express concern about unregulated clinical use, including direct-to-consumer access via compounding pharmacies and intranasal or sublingual formulations delivered outside structured clinical settings. Methodological limitations of the evidence base are noted, including small sample sizes, limited generalisability beyond treatment-resistant inpatients, potential unblinding due to perceptual effects, heterogeneity in allowance of concomitant psychotropic medications, and short follow-up durations. For future research, the authors recommend larger, multisite, more generalisable trials, investigation of repeated or multiple-infusion regimens, studies including patients with suicidality, trials testing ketamine as an adjunct to ongoing antidepressants or mood stabilisers (reflecting usual clinical practice), and research to define optimal administration frequency to maximise benefit while managing risks. The authors present ketamine as a potentially useful option for patients who have exhausted other pharmacotherapeutic avenues, but they stress the need for more robust evidence to guide routine clinical use.